Economics and Business Review, vol. 1(15), no. 4, 2015 by EBReview

Economics and Business Review

ISSN 2392-1641

Volume 1 1 (15) 2 2015 Volume (15)Number Number 4 2015 CONTENTS ARTICLES A turnpike theorem for non-stationary Gale economy with limit technology. A particular case Emil Panek Product market cooperation under efficient bargaining with different disagreement points: a result Domenico Buccella Banks, non-bank companies and stock exchange: do we know the relationship? Binam Ghimire, Rishi Gautam, Dipesh Karki, Satish Sharma Measuring the usefulness of information publication time to proxy for returns Itai Blitzer Business tendency survey data. Where do the respondents’ opinions come from? Sławomir Kalinowski, Małgorzata Kokocińska Does outward FDI by Polish multinationals support existing theory? Findings from a quantitative study Marian Gorynia, Jan Nowak, Piotr Trąpczyński, Radosław Wolniak The complex relationship between intrinsic and extrinsic rewards Orni Gov Improvement of the communication between teachers and students in the coaching programme and in a process of action research Michal Lory BOOK REVIEWS Barney G. Glaser, Choosing Classic Grounded Theory: a Grounded Theory Reader of Expert Advice, CA: Sociology Press, Mill Valley 2014 (Gary Evans)

Poznań University of Economics Press

Editorial Board Ryszard Barczyk Witold Jurek Cezary Kochalski Tadeusz Kowalski (Editor-in-Chief) Henryk Mruk Ida Musiałkowska Jerzy Schroeder Jacek Wallusch Maciej Żukowski International Editorial Advisory Board Udo Broll – School of International Studies (ZIS), Technische Universität, Dresden Wojciech Florkowski – University of Georgia, Griffin Binam Ghimire – Northumbria University, Newcastle upon Tyne Christopher J. Green – Loughborough University John Hogan – Georgia State University, Atlanta Bruce E. Kaufman – Georgia State University, Atlanta Steve Letza – Corporate Governance Business School Bournemouth University Victor Murinde – University of Birmingham Hugh Scullion – National University of Ireland, Galway Yochanan Shachmurove – The City College, City University of New York Richard Sweeney – The McDonough School of Business, Georgetown University, Washington D.C. Thomas Taylor – School of Business and Accountancy, Wake Forest University, Winston-Salem Clas Wihlborg – Argyros School of Business and Economics, Chapman University, Orange Jan Winiecki – University of Information Technology and Management in Rzeszów Habte G. Woldu – School of Management, The University of Texas at Dallas Thematic Editors Economics: Ryszard Barczyk, Tadeusz Kowalski, Ida Musiałkowska, Jacek Wallusch, Maciej Żukowski • Econometrics: Witold Jurek, Jacek Wallusch • Finance: Witold Jurek, Cezary Kochalski • Management and Marketing: Henryk Mruk, Cezary Kochalski, Ida Musiałkowska, Jerzy Schroeder • Statistics: Elżbieta Gołata, Krzysztof Szwarc Language Editor: Owen Easteal • IT Editor: Piotr Stolarski

ISSN 2392-1641

POZNAŃ UNIVERSITY OF ECONOMICS PRESS ul. Powstańców Wielkopolskich 16, 61-895 Poznań, Poland phone +48 61 854 31 54, +48 61 854 31 55, fax +48 61 854 31 59 www.wydawnictwo-ue.pl, e-mail: wydawnictwo@ue.poznan.pl postal address: al. Niepodległości 10, 61-875 Poznań, Poland Printed and bound in Poland by: Poznań University of Economics Print Shop Circulation: 300 copies

Volume 1 (15) Number 4 2015 CONTENTS ARTICLES A turnpike theorem for non-stationary Gale economy with limit technology. A particular case Emil Panek.......................................................................................................................................... 3 Product market cooperation under efficient bargaining with different disagreement points: a result Domenico Buccella............................................................................................................................. 14 Banks, non-bank companies and stock exchange: do we know the relationship? Binam Ghimire, Rishi Gautam, Dipesh Karki, Satish Sharma...................................................... 25 Measuring the usefulness of information publication time to proxy for returns Itai Blitzer........................................................................................................................................... 46 Business tendency survey data. Where do the respondents’ opinions come from? Sławomir Kalinowski, Małgorzata Kokocińska............................................................................... 69 Does outward FDI by Polish multinationals support existing theory? Findings from a quantitative study Marian Gorynia, Jan Nowak, Piotr Trąpczyński, Radosław Wolniak......................................... 84 The complex relationship between intrinsic and extrinsic rewards Orni Gov......................................................................................................................................... 102 Improvement of the communication between teachers and students in the coaching programme and in a process of action research Michal Lory.................................................................................................................................... 126 BOOK REVIEWS Barney G. Glaser, Choosing Classic Grounded Theory: a Grounded Theory Reader of Expert Advice, CA: Sociology Press, Mill Valley 2014 (Gary Evans)...................................................... 154

Economics and Business Review, Vol. 1 (15), No. 4, 2015: 3–13 DOI: 10.18559/ebr.2015.4.1

A turnpike theorem for a non-stationary Gale economy with limit technology. A particular case1 Emil Panek2

Abstract : This paper is a continuation of the work by Panek [2013a], [2013b] and Panek and Runka [2011], with direct reference to [2013b]. The difference between this paper and paper [2013b] reveals when applying a different optimality criterion with the help of which growth processes are assessed. The optimality criterion in [2013b] is the value of production in the last period of a fixed planning horizon T = {0, 1, …, t1} measured in von Neumann equilibrium prices. In this paper we take the total value of output (also measured in von Neumann prices) produced in the whole horizon T as a growth criterion. The aim of the paper is to prove that a change of growth criterion does not deprive the economy of its turnpike properties. It was pointed out that a Gale economy with a changeable technology convergent to the limit technology also preserves similar properties if we accept the social utility function (with standard properties) as a growth criterion. The paper consists of three sections. In Section 1 we present a non-stationary Galetype model with a limit technology. Section 2 contains proof of the so-called ‘weak’ turnpike theorem, whilst in Section 3 we prove a ‘strong’ version of the turnpike theorem for the economy under investigation. The paper closes with final comments. Keywords : non-stationary Gale economy, limit technology, von Neumann equilibrium, turnpike. JEL codes : C10.

1. A model of non-stationary Gale economy with limit technology The model that we shall consider has been presented in detail in Panek [2013b]. Here we give a brief description of the facts that are necessary for understanding of what follows. 1

Article received 10 December 2014, accepted 15 September 2015. Poznań University of Economics, Department of Mathematical Economics, al. Nie podległości 10, 61-875 Poznań, Poland, emil.panek@ue.poznan.pl. 2

Economics and Business Review, Vol. 1(15), No. 4, 2015

The economy we consider works in (discrete) time periods t = 0, 1, 2, … producing and/or utilizing n goods. By x(t) = (x1(t), …, xn(t))  0 we denote an input vector in period t, by y(t) = (y1(t), …, yn(t))  0 we denote an output vector in period t. A pair (x(t), y(t)) is said to describe a (technologically) feasible production process in period t. The set of all feasible production processes is denoted by Z(t) and it is called a production space in period t. Production spaces are defined for all integers t ≥ 0 and are assumed to meet the following conditions, see Panek [2003; Ch. 5], Panek [2013b; Sect. 1]: (G1) ∀(x, y) ∈ Z(t) ∀ λ ≥ 0 (λ(x, y) ∈ Z(t)). (G2) ∀(xi, yi), i = 1, 2 ((x1 + x2, y1 + y2) ∈ Z(t)). (G3) ∀(x, y) ∈ Z(t) (x = 0 ⇒ y = 0). (G4) ∀(x, y) ∈ Z(t) (x'  x ⇒ (x', y) ∈ Z(t)). (G5) ∀(x, y) ∈ Z(t) (0  y'  y ⇒ (x, y') ∈ Z(t)). (G6) Z(t) is a closed subset of R2n. (G7) Z(t) ⊆ Z(t + 1) ⊆ Z, where Z is the smallest closed subset of R2n containing all production spaces Z(t) such that the conditions (G1)–(G6) are satisfied for Z (with Z in place of Z(t)). Production spaces satisfying the conditions (G1)–(G6) are called Gale’s production spaces. The inclusion (x, y) ∈ Z(t) (equivalently (x(t), y(t) ∈ Z(t))) means that it is possible to produce output y from input x in period t. The set Z ⊂ R+2n mentioned in condition (G7) is called the limit technology. An economy whose production spaces Z, Z(t), t = 0, 1, 2, …, fulfill the conditions (G1)–(G7) is called a non-stationary Gale economy with limit technology.  Theorem 1. If production spaces meet conditions (G1)–(G7), then

(

)

(x , y ) ∈ Z ⇔ ∃( x(t ), y(t ) ) ∈ Z (t ), t = 0, 1, … lim ( x (t ), y (t ) ) = (x , y ) . t

Proof. See Panek [2013b; Theorem 1].  Let us fix a production process (x, y) ≠ 0. Then, according to (G3), x ≠ 0. The number α(x, y) = max{α|αx  y} is called the technological efficiency rate of the feasible production process (x, y). The function α is continuous and positively homogenous of degree 0 in Z\{0} (and in Z(t)\{0}), see Panek [2003; Ch. 5, Theorem 5.2]. The number α M , t = max α(x , y ) ( x , y ) ∈ Z (t ) (1) (x,y) ≠ 0

E. Panek, A turnpike theorem for a non-stationary Gale economy with limit technology

is called an optimal technological efficiency rate in a non-stationary Gale economy in period t, and the number α M = max α(x , y ) (2) (x, y )∈ Z (x,y) ≠ 0

is called an optimal technological efficiency rate in a non-stationary Gale economy with limit technology.  Theorem 2. Under assumptions (G1)–(G7):    α( x, y) = αM , t  , (i) ∀t ∃( x (t ), y (t ) ) ∈ Z (t ) α ( x (t ), y (t ) ) = ( x max , y ) ∈ Z (t )   (x,y) ≠ 0     ∃( x , y ) ∈ Z  α ( x , y ) = max α(x , y ) = α M  , (x, y )∈ Z   (x,y) ≠ 0  

(

)

(2i) ∀t αM , t ≤ αM , t +1 ≤ αM . Proof. See Panek [2013b; Theorem 2].  A pair of vectors ( x (t ), y (t ) ) satisfying the claim of Theorem 2 is said to be an optimal production process in a non-stationary Gale economy in period t. Similarly a pair ( x , y ) is said to be a limit optimal production process in a nonstationary Gale economy. In addition we assume:

(

)

(G8) ∃( x , y ) ∈ Z α ( x , y ) = α M ∧ y > 0 .

An economy satisfying property (G8) is called limit-regular. If a Gale economy meets (G8), then ∃ ( x , y ) ∈ Z ( α M x = y > 0 ). (3)

Whenever we speak of a limit-regular Gale economy we mean an economy for which condition (3) is met. Let us observe that if ( x , y ) ∈ Z is a limit optimal production process satisfying the condition, then3 α −1 y x y = M−1 = = s > 0. (4) x y αM y 3

Here and on: a = ∑ ai. i

Economics and Business Review, Vol. 1(15), No. 4, 2015

The vector s characterizes output and input structures in a limit optimal production process ( x , y ). The half-line

N = { λs | λ > 0} is called a turnpike (or von Neumann’s ray) for a non-stationary Gale economy with limit technology. Let p = (p1, …, pn) ≥ 0 denote a vector of goods’ prices and let us fix a production process (x, y) ≠ 0. The number4 β(x , y , p) =

p, y p, x

(if defined) is called economic efficiency rate of the process (x, y) at prices p.  Theorem 3. If a Gale economy is limit-regular, then ∃ p ≥ 0 ∀ (x , y ) ∈ Z ( p , y − α M p , x ≤ 0 ) (5)

and

β ( x , y , p ) = max β ( x , y , p ) = α ( x , y , p ) = α M. (x, y )∈ Z (x,y) ≠ 0

Proof. See Panek [2003; Ch. 5, Theorems 5.3, 5.4].  A process ( x , y ) ∈ Z and prices p are said to characterize a non-stationary Gale economy with limit technology in a von Neumann equilibrium. In such an equilibrium the economic efficiency rate and the greatest possible technological efficiency rate are equal. The prices are called von Neumann equilibrium prices (in a non-stationary Gale economy with limit technology). The next condition, (G9) ∀(x , y ) ∈ Z \ {0} ( x ∉ N ⇒ β ( x , y , p ) < α M ) ensures that the production turnpike is unique – that is presented in the following theorem.  Theorem 4. In a Gale economy satisfying conditions (G1)–(G9) there a unique production turnpike exists. Proof. The existence of a turnpike N is ensured by conditions (G1)–(G8). Let us suppose that N' is a turnpike different from N. Then N ∩ N' ≠ ∅ (since if N ∩ N' ≠ ∅, then N = N'). By the definition of the turnpike N:

  x ∃ ( x , y ) ∈ Z  α M x = y > 0 ∧ = s .   x   4

a = ∑ ai bi. 〈a, b〉 denotes the scalar product of vectors a, b ∈ R2n: 〈a, b〉 i

E. Panek, A turnpike theorem for a non-stationary Gale economy with limit technology

Similarly for the turnpike N':

 x' ∃ ( x' , y' ) ∈ Z  αM x' = y' > 0 ∧ = s' ≠  x' 

 s ,  

p , y' = αM p, x' .

(*)

By (G9) we get β ( x' , y' , p ) =

(

p , y' p , x'

) <α

i.e. p , y' < α M p , x' , which contradicts with (*). The contradiction terminates the proof.  A next consequence of condition (G9) (and the other conditions) is presented in the theorem below (in literature it is known as Radner’s Lemma).  Theorem 5. If conditions (G1)–(G9) are satisfied, then

 x  p, y ∀ε > 0 ∃δε > 0 ∀( x , y ) ∈ Z  − s ≥ ε ⇒ β ( x, y, p ) = ≤ α M − δε . (6)  x  p, x   Proof. See Radner [1961]; Takayama [1985; Ch. 7]; Panek [2003; Ch. 5, Lemma 5.2].  Theorem 5 states that the economic efficiency rate and the technological efficiency rate are equal on the turnpike only. Off the turnpike, the economic efficiency rate is less than its maximal value, which applies only at the turnpike.

2. Dynamics. “Weak” turnpike effect in a non-stationary Gale economy under a special growth condition Let us fix any time period t1 < +∞. The set T = {0, 1, …, t1} is called a horizon (of the functioning) of the economy. The period t1 is also the length of the economy’s horizon T. Let (x(t), y(t)) ∈ Z(t), t ∈ T. In a Gale economy the next period’s inputs originate from the directly preceding period:5 x(t + 1)  y(t), t = 0, 1, …, t1 – 1, 5

In this sense the economy is closed, see Gale [1956].

Economics and Business Review, Vol. 1(15), No. 4, 2015

which due to (G5) leads to:

(y(t), y(t + 1)) ∈ Z(t), t = 0, 1, …, t1 – 1.

(7)

Let us fix an initial output vector y0: y(0) = y0 ≥ 0.

(8)

A sequence of output vectors { y(t )}t1= 0 satisfying conditions (7)–(8) is said to be a (y0, t1) – feasible growth process in a non-stationary Gale economy. It is clear that if conditions (G1)–(G6) hold, then ∀y0 ≥ 0 ∀t1 < +∞ exist there (y0, t1) – feasible growth processes. In the work of Panek [2013b] we were interested in solving the following dynamic programming problem – the maximization of value of the last period’s t1 output (at von Neumann prices): t

max p , y(t1 ) subject to: (7)–(8). In the current paper we will deal with the properties of growth processes solving the problem of maximization of the value of output (again at von Neumann prices) throughout the whole horizon T:6 t1

max ∑ p , y(t )

(9)

subject to: (7)–(8).

(10)

t =0

(

)

The problem has a solution which we denote by { y *(t )}t1= 0 and call y 0 , t1 , p – the optimal growth process in a non-stationary Gale economy. The last assumption essential for the proof of turnpike properties of optimal growth processes in the light of the criterion (9) states that there a feasible growth process exists starting from the initial state (8) and reaching the turnpike in a finite time:    t (G10) There exists ( y 0 , t ) – feasible growth process { y (t )}t = 0, where t < t1, such that t

  α ( y (t ), y (t + 1) ) = αM.7 (11) 6 A similar problem in a non-stationary von Neumann economy was formulated and solved in Panek [2013a]. 7 The condition is not needed if the initial output y0 in (8) is positive.

E. Panek, A turnpike theorem for a non-stationary Gale economy with limit technology

 Theorem 6 (A „weak” turnpike theorem). If in a non-stationary Gale economy with limit technology satisfying conditions (G1)–(G10) the optimal turnpike growth rate αM meets the inequality αM > 1, then for any ε > 0 there exists a natural number kε, such that the numt ber of periods in which an ( y 0 , t1 , p) – optimal growth process { y *(t )}t1= 0 satisfies the inequality

y *(t ) − s ≥ ε (12) y *(t ) is not greater than kε. The number kε is independent of the horizon T length. t Proof. Let us be given a ( y 0 , t1 , p) – optimal growth process { y *(t )}t1= 0 which is a solution of the problem (9)–(10). Then, according to (5), (7), (G7):

p , y * (t + 1) ≤ α M p , y * (t ) , t = 0, 1, …, t1 – 1.

(13)

Suppose that for an ε > 0 in periods τ1 < τ2 < … < τk (0 ≤ τi < t1, i = 1, …, k) condition (12) is satisfied. Then due to Theorem 5 (condition (6)):

p , y *( t + 1) ≤ ( αM − δε ) p, y *(t) , t = τ1, …, τk.

(14)

From (13) and (14) we obtain an upper bound for the criterion (9) (value of production throughout the horizon T): t1

∑

t =0

k t1 − k  p , y * (t ) ≤ p , y 0  ∑ αtM + αtM1 −k ∑ (α M − δε )τ. τ =1  t = 0 

(15)

  t 0   (t )}t = 0 t < t y ( y , t ) By (G10) there is a period and a – feasible growth process { 1      such that α ( y (t ), y (t + 1) ) = αM. From the fact that ( y (t ), y (t + 1) ) ∈ Z (t + 1), and due to (11) and (G5)



( y (t ), α

  y (t ) ) ∈ Z (t + 1),

  which by Theorem 4 entails that y (t ) ∈ N , and thus y (t ) = σ s for some σ > 0. By t the condition (G7) we get a (y0, t1) – feasible growth process { y(t )}t1= 0:

 t = 0, 1, …, t ,  y (t ), y(t ) =  t −t  σ sαM , t = t + 1, …, t1 ,

Economics and Business Review, Vol. 1(15), No. 4, 2015

and we can bound the growth criterion (9) below by: t1

∑ t =0

p , y *(t ) ≥ C1 + σ p, s

∑α

 t = t +1

 t −t M

(16)

 t

where C1 = p , y + ∑ p, y (t ) > 0. The constant C1 is independent of t1. From 0

t =1

(15) and (16) we obtain the inequality: t1

∑α

σ p, s

 t = t +1

 t −t M

k  t1 −k  p , y 0  ∑ αtM + αtM1 −k ∑ (α M − δ ε ) τ  , τ =1  t = 1 

hence t1 −k

∑α t =1

t M

+ αtM1 −k ∑ (αM − δ ε )τ > τ =1

σ p, s p, y 0

∑α

 t = t +1

 t −t M

> 0,

and finally 

(α − δ )k − 1 αtM1 −t σ p , s αt1 −k . α M M + αtM1 −k (α M − δ ε ) M ε > αM − 1 α M − δε − 1 αM − 1 p, y 0

Taking a sufficiently small δε > 0 for which it holds αM – δε – 1 > 0, after simple manipulations we arrive at the following condition:

  σ p, s , (17) 0 ∀ t1 ≥ k  A(t1 , k ) + B(t1 , k ) > C2 = > 0   p , y   where

α (αt1 −k − 1) A(t1 , k ) = M t − tM+ 1 , (αM1 − 1)

αtM1 −k (αM − δε )(αM − 1) (αM − δε )k − 1  B(t1 , k ) = . (αtM1 −t − 1)(α M − δε − 1)

Due to the facts, A(t1, k) → 0 as k → +∞ (t1 ≥ k) and B(t1, k) → 0 as k → +∞ (t1 ≥ k), we conclude that the number k in (17) is bounded. To be more specific a natural number kε exists (depending on ε but not on the horizon T = {0, 1, …, t1}

E. Panek, A turnpike theorem for a non-stationary Gale economy with limit technology

length) such that the number of periods k, for which the inequality (12) is satisfied is not greater than kε. 

3. “Very strong” turnpike effect A simple consequence of the „weak” turnpike theorem in a non-stationary Gale economy with limit technology under the growth criterion (9) is Theorem 7.  Theorem 7 (A „very strong” turnpike theorem). If conditions (G1)–(G10) are satisfied and a y 0 , t1 , p – optimal growth t process { y *(t )}t1= 0 reaches the turnpike N in a period t < t1, i.e.

(

)

  α ( y *(t ), y *(t + 1) ) = α M, (18)  then ∀ t ∈{t + 1, …, t1 − 1} ( y *(t ) ∈ N ).   Proof.8 By (18), y *(t ) ∈ N , i.e. y *(t ) = σs for some σ > 0. The optimal p rocess t { y *(t )}t1= 0 meets the condition (13), hence  t

p , y *(t ) = ∑ p , y *(t ) +

∑

t =0

 t

∑

 t = t +1

p , y *(t ) ≤

≤ ∑ p , y *(t ) + σ p , s

t =0

∑α

 t =t +1

 t −t M

. (19)

 Assume that y*(τ) ∉ N for some period τ ∈{t + 1, …, t1 − 1}, then, according to Theorem 5 (condition (6)), there is δε > 0 such that

p , y *(τ + 1) ≤ (α M − δ ε ) p , y *(τ ) , which together with (19) leads to the inequality: t1

∑

t =0

 t

   t1 p , y *(t ) ≤ ∑ p , y *(t ) + σ p, s  ∑ αtM−t − δε αMτ− t  .  t = t + 1  t =0    t

On the other hand, a process { y (t )}t = 0 of form:

 y *(t ), y (t ) =  t − t σsα M , 8

 t = 0, 1, …, t ,  t = t + 1, …, t1 ,

This proof mimicks the proof of Theorem 2 in Panek and Runka [2011].

(20)

Economics and Business Review, Vol. 1(15), No. 4, 2015

is (y0, t1) – feasible, so t1

∑

t =0

 t1

p , y *(t ) = ∑ p , y *(t ) + t =0 t1

∑

 t = t +1  t

p , y *(t ) ≥

≥ ∑ p , y (t ) = ∑ p , y *(t ) + σ p , s

t =0

∑α

 t = t +1

 t −t M

. (21)

From (20) and (21) we obtain, t1

∑α

 t = t +1

 t −t M



− δε αMτ− t ≥

 t− t M

∑α

 t = t +1

from which it follows that δε ≤ 0. The contradictions terminate the proof. 

Final comments The results contained in Theorems 6 and similar ones contained in earlier papers, e.g. in Panek and Runka [2011], can be easily generalized. They are valid also in the case where the growth criterion (9) is replaced with one of the following: t1

max ∑ u ( y(t ) ),

(I)

t =0

u: R+n → R+1 is an increasing continuous homogenous of degree 1 of the utility function, positive on the turnpike such that for a positive number a it holds:, ∀y  0 ( u( y ) ≤ a p , y ),

or max U ( y(0), y(1), …, y (t1 ) ),

(II)

U the composition of an increasing continuous function g: R+1 → R+1 and the t1

linear function p , ∑ y(t ) : t =0

t1  U ( y(0), y(1), …, y (t1 ) ) = g  p , ∑ y (t )  t =0 

 .  

E. Panek, A turnpike theorem for a non-stationary Gale economy with limit technology

In the literature devoted to research on the properties of optimal growth processes in stationary versions of a Gale economy (under constant technology) a ‘strong’ version of turnpike property (theorems) is also mentioned, which is a kind of link between the weak and the very strong turnpike theorems. If a strong turnpike property result holds in the model we presented needs to be further investigated.

References Gale, D., 1956, The Closed Linear Model of Production, in: Kuhn, H.W., Tucker, A.W. (eds.), Linear Inequalities and Related Systems, Princeton University Press, Princeton: 285–303. Panek, E., 2003, Ekonomia matematyczna, Wydawnictwo Akademii Ekonomicznej w Poznaniu, Poznań. Panek, E., 2013a, Niestacjonarny model von Neumanna z graniczną technologią, Studia Oeconomica Posnaniensia, vol. 1, no. 1 (250): 49–68. Panek, E., 2013b, „Słaby” i „bardzo silny” efekt magistrali w niestacjonarnej gospodarce Gale’a z graniczną technologią, Przegląd Statystyczny, nr 60 (3): 291–303. Panek, E., Runka, H.J., 2011, Efekt magistrali w gospodarce Gale’a. Wersja szczególna, w: Panek, E. (red.), Matematyka i informatyka na usługach ekonomii. Modelowanie zjawisk gospodarczych, Wydawnictwo Uniwersytetu Ekonomicznego w Poznaniu, Poznań: 220–231. Radner, R., 1961, Path of Economic Growth that Are Optimal with Regard to Final States: A Turnpike Theorem, Rev. Econ. Studies, XXVIII, no. 2: 98–104. Takayama, A., 1985, Mathematical Economics, Cambridge University Press, Cambridge.

Economics and Business Review, Vol. 1 (15), No. 4, 2015: 14–24 DOI: 10.18559/ebr.2015.4.2

Product market cooperation under efficient bargaining with different disagreement points: a result1 Domenico Buccella2

Abstract : This paper analyzes the effects of product market cooperation in a duopoly with homogeneous goods. Labor unions and firms are locked in bilateral monopoly relations and bargaining within the industry takes place under the efficient bargaining (EB) model. The paper discusses the role of different firms’ disagreement payoff on bargaining outcomes and their effect on social welfare components. It shows that the disagreement point of firms can play a crucial role in solving potential conflicts of interest between unions and firms concerning bargaining issues. Keywords : efficient bargaining, cooperation, unionized oligopoly, social welfare. JEL codes : D43, J51, L13.

Introduction The interactions between product and labor markets are at the core of the functioning of advanced economies. Recently the impact of product market cooperation on negotiations between firms and unionized labor has attracted the attention of economists and antitrust authorities because of the consequences for social welfare. The literature has developed a large number of studies related to the impact on welfare of labor union activities. The research considered both the conduct of negotiations with firm-level unions [e.g. Horn and Wolinsky 1988; Symeonidis 2008, 2010] and industry-wide unions [e.g. Mukherjee 2010]. These works primarily focus on the right-to-manage (RTM) model [e.g. Nickell and Andrews 1983; Dobson 1994, 1997; Naylor 2002; López and Naylor 2004]: unions and firms first negotiate the wages; then, once the wages are fixed, the firms retain the right to choose employment levels. In comparison, the analysis carried out on the efficient bargaining (EB) model, where unions and firms negotiate simultaneously over wages and employment [e.g. McDonald and Solow 1

Article received 08 November 2014, accepted 15 September 2015. Kozminski University, Department of Economics, Jagiellońska 57/59, 03-301 Warszawa, Poland, buccella@kozminski.edu.pl. 2

D. Buccella, Product market cooperation under efficient bargaining

1981; Oswald 1985; Espinosa and Rhee 1989; Bughin 1995, 1996] is scant. Kraft [2006] points out that several empirical studies have shown that negotiations conducted under the EB model have been practiced [MaCurdy and Pencavel 1986; Bughin 1993; more recently, Dobbaleare and Mairesse 2011]. The aim of this paper is to shed light on the impact of the EB agenda on product and labor market interactions. Notable exceptions are represented by Symeonidis [2010] and Buccella [2014]. Symeonidis [2010] briefly discusses the case of firm-level negotiations with cross-ownership in an industry where the EB model is the reference bargaining framework. The author obtains the standard results that an increase in the degree of market cooperation, by restricting industry output, decreases the overall social welfare. The firms’ disagreement payoff is represented by a fraction of the monopoly output. Using a conjectural variation model Buccella [2014] analyzes the impact of market competition intensity on profits, union rent, consumer surplus and social welfare in a duopoly with differentiated goods where an industry-wide union conducts negotiations with the firms simultaneously, although separately, according to the EB model. The author shows that the consumer surplus definitely declines when the market becomes less competitive. On the other hand the industry-wide union and the duopolists mutually benefit from the restriction of market competition: the industry profits and the union utility increase. With regard to welfare the impact of the market competition on its overall level depends on the degree of product differentiation: with perfect substitute goods, Cournot competition maximizes social welfare; for more differentiated goods, intermediate levels of market competition maximizes welfare; for virtually independent goods, a less competitive market unambiguously decreases the social welfare. The paper further extends Symeonidis’ [2010] analysis taking into consideration the situation where, in the case of breakdown of negotiations in one bargaining unit, the other firm can operate at the anticipated duopoly equilibrium level of output [e.g. Horn and Wolinsky 1988]. The paper demonstrates that, under the specific assumptions of the linear demand function, constant returns to scale technology and risk-neutral unions, the firms’ disagreement payoff is not crucial in determining the relationship between product market cooperation, consumer surplus and social welfare; however, it has a key role in shaping the potential conflict of interest between firms and unions during the bargaining process. In fact whilst increasing product market cooperation decreases wages when the firms’ disagreement payoff during negotiations is based on monopoly output, the opposite holds true in the case of the anticipated duopoly equilibrium output as a disagreement point. As a consequence both industry profits and union rents increase to the detriment of consumer surplus. This result mirrors Buccella [2014], therefore suggesting that, in certain cases, the EB framework may solve the conflict of interest between firms and unions, regardless of the form of product market cooperation.

Economics and Business Review, Vol. 1(15), No. 4, 2015

The remainder of the paper is organized as follows. Section 1 describes the model and presents the results. The next section closes with plausible directions for further research on the subject.

1. The model and the results Consider an industry with two firms, denoted 1 and 2, competing in homogeneous goods à la Cournot. Labor, l, is the only factor of production with a constant returns to scale technology. For simplicity, it is assumed that each worker produces one unit of the goods, l = q: hence, output and employment levels are equal. In the economy there is a continuum of identical consumers who have preferences over goods Q and y, characterized by a separable utility function V(Q; y). The representative consumer maximizes V(Q; y) = U(Q) + y, linear in the numeraire goods y, with respect to quantities subject to the budget constraint pQ + y = M, with Q non-negative and M the exogenous consumer’s income. The utility function U(Q) is assumed continuously differentiable and satisfies the desirable standard properties of the consumer theory [see e.g. Singh and Vives 1984]. Given the quasi-linearity of the function V(Q; y), there are no income effects on the duopolistic sector. Therefore for an adequately high level of income, the optimization problem of the representative consumer is to choose Q to maximise U(Q) – pQ + M. The utility maximization problem leads to the ∂U (inverse) demand function p = = p(Q).3 ∂Q Following the usual specification of Dixit [1979] and Singh and Vives [1984], to obtain explicit demand functions the representative consumer is assumed to have a quadratic utility function over the homogeneous goods produced by the two firms and a linear function of a numeraire good, y. Thus the preferences of Q2 the representative consumer over Q are U (Q) = aQ − . From this quadratic 2 utility function, the maximization problem of the representative consumer subject to the budget constraint pQ + y = M leads to the linear (inverse) market 3

The paper analyzes only the standard case when homogenous goods are normal. Thus the paper abstracts from cases where the homogenous goods might fall into the categories of Giffen and Veblen goods. The rationales for this choice are as follows. Giffen goods (where the income effect dominates the substitution effect) are relatively rare and the only empirical evidence of the existence of such goods can be found in Jensen and Miller [2008]. Anecdotal evidence indicates that Veblen goods (consumers are willing to pay a higher price for functionally equivalent goods) might be empirically considerable in markets for luxury goods [Bagwell and Bernheim 1996]. Empirical studies on Veblen effects are, e.g. Clark and Oswald [1996], Bowles and Park [2005] and Carlsson, Johansson-Stenman and Martinsson [2007].

D. Buccella, Product market cooperation under efficient bargaining

  demand curve p = a – Q, where p is the price and Q = ∑ qi =  ∑ li  , i = 1, 2 i  i  total output. Both firms are unionized and the bargaining structure is firm-specific. In other words the firms and the unions are cast into bilateral monopoly relations. The EB model characterizes negotiations. Thus unions and management at each bargaining unit simultaneously negotiate wages and employment. The firm i’s profits can be written as

Πi = πi + λπ j = ( p − wi )qi + λ( p − w j )q j, i, j = 1, 2; i ≠ j. (1) As per Symeonidis [2008, 2010] the parameter λ ∈ [0, 1) represents the degree of firms’ cross-ownership or, alternatively, the degree of product market cooperation [Mukherjee 2010]. The union utility takes the following form

Ωi = (wi − w0 )qi, i = 1, 2,

(2)

that is each union is risk-neutral. A different interpretation of (2) is that unions assign an equivalent weight to wage and employment in their preferences (neutrally oriented unions). The positive utility derives from the fact that negotiated wages wi lie above the reservation wages w0, or what workers receive if unemployed. In the present model, and without a loss of generality, w0 equals zero. Under the above outlined assumptions, it can be derived that the expression of the consumer surplus is [Mukherjee 2010] CS =

Q2 . (3) 2

The social welfare of the economy is given by the sum of the industry profits, unions’ utility and consumer surplus, that is,

 Q SW = ∑ Πi + ∑ Ωi + CS = Q  a − , i = 1, 2. 2 

(4)

Under the EB framework, the parties set wi and qi to maximize the following generalized Nash Product

max Gi (wi , qi ) = (Ωi )α (πi + λπ j − λπ *j )1−α, (5) where α ∈ (0, 1) is the relative bargaining power, assumed equal across units, and λπj* is the firm i’s disagreement point in the case of negotiations’ failure. The

Economics and Business Review, Vol. 1(15), No. 4, 2015

union i’s disagreement point equals zero. As known [Horn and Wolinsky 1988], the disagreement payoff of the bargaining parties may have different specifications. This paper investigates the two most common assumptions considered in literature. First, the analysis focuses on the situation where, in the case of disagreement, firm j can produce the monopoly output. In this case, the firm * − w *jM )q *jM , where q*jM and w*jM are i’s disagreement payoff is λπ *jM = λ (a − q jM the equilibrium monopoly output and wage, respectively. Second, the analysis investigates the situation where, in the case of disagreement, firm j produces at the anticipated duopoly equilibrium level of output. In this case, firm i’s disagreement payoff is λπ *jD = λ (a − q *jD − qiD * − w *jD )q *jD , where q*jD and q*iD are the firms j and i’s equilibrium quantities, and w*jD is wage.

1.1. Partial monopoly output as the firms’ disagreement point: a discussion After the derivation of first-order conditions for maximization of (5), it can a be derived that the firm i’s equilibrium output is q*i = , and the equilib( λ + 3) αa rium wage w*i = , i = 1, 2 (see Appendix). As Symeonidis [2010] dis( λ + 3) cusses, when bargaining is over wages (input price) and employment (quanti∂w * ∂Q * < 0, i < 0 ∀α ∈ (0, 1), and ties), straightforward calculations show that ∂λ ∂λ ∂wi* > 0 ∀λ ∈[0, 1). The first is the standard result: an increase in the degree of ∂α product market cooperation decreases total output in the industry. Given (3) the qualitative effects on consumer’s surplus of an increase in λ is identical. The second finding establishes that the equilibrium wage decreases as λ increases. For given wages an increase in the degree of cooperation between firms causes a fall in total industry output. On the one hand, this increases firms’ profits and oligopoly rents: unions may demand higher wages, on the other, the fall in production levels due to coordination leads to a lower demand for labor at each firm. This poses a downward pressure on wages. However the latter effect dominates the former and the final result is a wage reduction. As expected the equilibrium wage is increasing in the union’s relative bargaining strength. Reinserting the equilibrium wages and output into the relevant equations, the values for the industry profits, total unions’ rents, consumer surplus and social welfare are obtained Π = Π1 + Π2 =

2a 2 (1 − α + λ)(1 + λ) 2a 2 α , Ω Ω Ω , = + = 1 2 (3 + λ)2 (3 + λ)2 2a 2 2a 2 (2 + λ) CS = SW , . = (3 + λ)2 (3 + λ)2

D. Buccella, Product market cooperation under efficient bargaining

∂Π ∂Π ∂Ω ∂Ω ∂CS ∂SW Analytical inspection shows that > 0, < 0, < 0, > 0, < 0 and <0 ∂λ ∂α ∂λ ∂α ∂λ ∂λ ∂CS ∂SW < 0 in the relevant parameters’ space. The above results can be summa, < 0 and ∂λ ∂λ rized in the following proposition. Proposition 1. In an efficient bargaining framework if the firms’ disagreement point is a fraction of the monopoly output, a higher degree of product market cooperation: 1) decreases industry output and wages; and 2) increases industry profits at the expenses of unions’ rents and consumer surplus, decreasing overall social welfare.

1.2. Partial anticipated duopoly equilibrium output as the firms’ disagreement point Let us now consider the case of partial anticipated duopoly equilibrium output as the firm i’s disagreement point. From the first-order conditions for a maximization of (5), it is obtained that the equilibrium output is q*i = , ( λ + 3) i = 1, 2, identical to the previous case. Consequently total industry output equals ∂Q * 2a αa(1 + λ) < 0. However the equilibrium wages are wi* = Q* = , with , ∂λ ( λ + 3) ( λ + 3) ∂w * i = 1, 2 (see Appendix). The differentiation shows that i > 0 ∀α ∈ (0, 1), as ∂α ∂wi* expected; however, > 0 ∀λ ∈[0, 1): the equilibrium wage increases as the ∂λ degree of product market cooperation increases. For given wages an increase in the degree of cooperation between firms triggers a fall in total output. As before there are two effects on negotiated wages. On the one hand, firms’ cooperation increases profits and oligopoly rents: unions, therefore, may claim higher wages, on the other the reduction in production levels decreases labor demand. Nonetheless if negotiations break down when the disagreement point is the partial anticipated duopoly output, the industry output will be higher than in the case of partial monopoly output as the disagreement point. As a consequence unions ask for higher wages. This result complements Symeonidis [2010] who shows that in the presence of bilateral monopoly relations, if the firms’ disagreement point is the partial monopoly output, under the EB framework the impact of higher product market cooperation (or cross-ownership) on negotiated wages in equilibrium is negative. Substituting the equilibrium output and the equilibrium wage into the relevant equations, the following values for the industry profits and total union rents are obtained

20 Π = Π1 + Π2 =

Economics and Business Review, Vol. 1(15), No. 4, 2015

2a 2 (1 − α )(1 + λ )2 2a 2α (1 + λ ) , Ω Ω Ω , = + = 1 2 (3 + λ)2 (3 + λ)2 2a 2 2a 2 (2 + λ) CS = SW , . = (3 + λ)2 (3 + λ)2

An analytical inspection reveals that, in the relevant parameters’ space, the fol∂Π ∂Π ∂Ω ∂Ω ∂CS ∂SW > 0, < 0, > 0, > 0, < 0 and <0 lowing comparative statics hold: ∂ λ ∂ α ∂ λ ∂ α ∂ λ ∂ λ ∂CS ∂SW < 0. It is extremely significant to remark that industry profits as well as , < 0 and ∂λ ∂λ union utility are now increasing in λ: firms and unions find mutual benefit in increasing the degree of product market cooperation. That is both bargaining parties raise their relative gains to the detriment of consumers. Thus the technical conditions of the firms are relevant in determining the bargaining outcomes. In fact, if the firms are quantity constrained and cannot produce more than QC ∈ [λπ*M, λπ*D], negotiations may end with a situation of either conflict or common interest between labor and management. These results can be summarized in the following proposition. Proposition 2. In an efficient bargaining framework, if each firm’s disagreement point is a fraction of the anticipated duopoly output in equilibrium, a higher degree of the product market cooperation: 1) decreases industry output; 2) increases wages; and 3) increases industry profits and union rents at the expenses of consumer’s surplus, decreasing overall social welfare.

Conclusions In an industry characterized by the EB model with firms and unions locked into bilateral monopoly relations this paper has investigated the impact of different degrees of product market cooperation on profits, union rents, consumer surplus and social welfare and analyzes the role of different firms’ disagreement payoffs. The analysis has shown that the firms’ disagreement point plays a key role in solving potential conflicts of interest between unions and firms in respect of bargaining issues. In fact the noteworthy finding is that, when the disagreement payoff is based on the partial anticipated duopoly equilibrium output, firms and unions mutually gain from product market cooperation because both industry profits and unions’ utility increase. Nevertheless the robustness of these findings needs to be checked for different conjectures concerning the strategic behaviour of firms in the product market. A more competitive product market environment (for instance, price competition à la Bertrand with differentiated products) may reverse many of the results obtained here. This issue is left for future research.

D. Buccella, Product market cooperation under efficient bargaining

Appendix When the disagreement payoff of the firms is based on the partial anticipated monopoly output, the maximization problem in (5) is

max Gi (wi , qi ) = (wi qi )α × * − w jM * )q jM *  × (a − qi − q j − wi )qi + λ( a − qi − q j − w j )q j − λ(a − q jM

1− α

i, j = 1, 2 i ≠ j. (A.1)

The first-order conditions from the maximization of (A.1) are

∂Gi αGi (1 − α)Gi qi + = − =0 ∂wi wi (a − qi − q j − wi )qi + λ(a − qi − q j − w j )q j − λ(a − q jM  w q * * * − ) jM jM   ∂Gi αGi (1 − α)Gi qi = − = 0, ∂wi wi (a − qi − q j − wi )qi + λ(a − qi − q j − w j )q j − λ(a − q jM  w q * * * − ) jM jM   (1 − α)Gi a − 2qi − wi − (1 + λ)q j  ∂Gi αGi = + =0 ∂qi qi (a − qi − q j − wi )qi + λ( a − qi − q j − w j )q j − λ( a − q jM  * * * ) − w q jM jM   (1 − α)Gi a − 2qi − wi − (1 + λ)q j  ∂Gi αGi = + = 0, ∂qi qi (a − qi − q j − wi )qi + λ( a − qi − q j − w j )q j − λ( a − q jM  * * * − ) w q jM jM  

which, solved for wi, lead to wi = α  a − qi − (1 + λ)q j  (rent-sharing curve),

wi = a − (2 − α)qi − (1 + λ)qj (contract curve) i, j = 1, 2 i ≠ j.

(A.2) (A.3)

Equating (A.2) and (A.3), firm i’s production as a function of the rival firm’s output is qi =

a − (1 + λ)q j 2

, i, j = 1, 2 i ≠ j. (A.4)

Solving the system of equations in (A.4), the equilibrium output and employa ment at each firm is q*i = , i = 1, 2, and the further substitution in (A.2) ( λ + 3) αa leads to the equilibrium wage w*i = as reported in the main text. ( λ + 3)

Economics and Business Review, Vol. 1(15), No. 4, 2015

On the other hand when the disagreement payoff of the firms is based on the partial anticipated duopoly output, the maximization problem in (5) becomes

max Gi (wi , qi ) = (wi qi )α 1−α

(a − qi − q j − wi )qi + λ(a − qi − q j − w j )q j − λ(a − q *jD − qiD * − w *jD )q *jD  ,  (A.5) i, j = 1, 2 i ≠ j, where q*jD and q*iD are the firms j and i’s duopoly equilibrium quantities, and w*jD is the equilibrium wage. The first-order conditions from the maximization of (A.5) are

∂Gi αGi = + ∂wi wi –

(1 − α)Gi qi

(a − qi − q j − wi )qi + λ(a − qi − q j − w j )q j − λ(a − q*jD − qiD* − w jD * )q *jD  

= 0,

∂Gi αGi = + qi ∂qi +

(1 − α)Gi a − 2qi − wi − (1 + λ)q j  (a − qi − q j − wi )qi + λ(a − qi − q j − w j )q j − λ(a − q jD * − qiD* − w jD * )q jD *  

= 0.

Considering that at equilibrium (a − qi − q j − w j )q j = (a − q *jD − qiD * − w jD * )q *jD, those conditions reduce to α ∂Gi α(wi qi ) (a − qi − q j − wi )qi  = ∂wi wi

1−α

−

α ∂Gi α(wi qi ) (a − qi − q j − wi )qi  = ∂qi qi

(1 − α)qi (wi qi )α (a − qi − q j − wi )qi  (a − qi − q j − wi )qi   

1−α

= 0,

1−α

(1 − α)(wi qi )α (a − qi − q j − wi )qi 

1−α

a − 2qi − wi − (1 + λ)q j    = 0. (a − qi − q j − wi )qi   

D. Buccella, Product market cooperation under efficient bargaining

Solving the above system of equations for wi, it is obtained wi = α(a – qi – qj) (rent sharing curve) i, j = 1, 2 i ≠ j, (A.6) wi = a − (2 − α)qi − 1 + λ(1 − α)q j (contract curve) i, j = 1, 2 i ≠ j. (A.7) Equating (A.6) and (A.7), the resulting firm’s i reaction function to the rival firm’s production is as in (A.4). Therefore the equilibrium output and employa ment at each firm is q*i = , i = 1, 2; however, the further substitution of ( λ + 3) αa(1 + λ) the output expression in (A.6) leads to the equilibrium wage wi* = ( λ + 3) reported in the main text.

References Bagwell, L.S., Bernheim, B.D., 1996, Veblen Effects in a Theory of Conspicuous Consumption, American Economic Review, vol. 86, no. 3: 349–373. Bowles, S., Yongjin, P., 2005, Emulation, Inequality, and Work Hours: Was Thorsten Veblen Right?, Economic Journal, vol.115, no. 507: F397–F412. Bughin, J., 1993, Union-Firm Efficient Bargaining and a Test of Oligopolistic Conduct, Review of Economics and Statistics, vol. 75: 563–567. Bughin, J., 1995, Bargaining over Employment as a Firm Strategic Choice, Review of Industrial Organization, vol. 10: 723–735. Bughin, J., 1996, Trade Unions and Firms’ Product Market Power, The Journal of Industrial Economics, vol. 44, no. 3, 289–307. Buccella, D., 2014, Product Market Competition with Differentiated Goods and Social Welfare in the Presence of an Industry-wide Union, Portuguese Economic Journal, vol. 13, no. 2: 131–140. Carlsson, F., Johansson-Stenman, O., Martinsson, P., 2007, Do You Enjoy Having More than Others? Survey Evidence of Positional Goods, Economica, vol. 74: 586–598. Clark, A.E., Oswald, A.J., 1996, Satisfaction and Comparison Income, Journal of Public Economics, vol. 61, no. 3: 359–381. Dixit, A., 1979, A Model of Duopoly Suggesting a Theory of Entry Barriers, The Bell Journal of Economics, vol. 10: 20–32. Dobbalaere, S., Mairesse, J., 2011, Panel Data Estimates of the Production Function and Product and Labor Market Imperfection, Journal of Applied Econometrics, vol. 28, no. 1: 1–46. Dobson, P.W., 1994, Multifirm Unions and the Incentive to Adopt Pattern Bargaining in Oligopoly, European Economic Review, vol. 38: 87–100. Dobson, P.W., 1997, Union–Firm Interaction and the Right to Manage, Bulletin of Economic Research, vol. 49, no. 3: 213–229. Espinosa, M., Rhee, S., 1989, Efficient Wage Bargaining as a Repeated Game, Quarterly Journal of Economics, vol. 104: 565–588.

Economics and Business Review, Vol. 1(15), No. 4, 2015

Horn, H., Wolinsky, A., 1988, Bilateral Monopolies and Incentives for Merger, RAND Journal of Economics, vol. 19, no. 3: 408–419. Jensen, R.T., Miller, N.H., 2008, Giffen Behavior and Subsistence Consumption, American Economic Review, vol. 98, no. 4: 1553–77. Kraft, K., 2006, Wage Versus Efficient Bargaining in Oligopoly, Managerial and Decision Economics, vol. 27: 595–604. López, M.C., Naylor, R.A., 2004, The Cournot–Bertrand Profit Differential: A Reversal Result in a Differentiated Duopoly with Wage Bargaining, European Economic Review, vol. 48: 681–696. MaCurdy T.E., Pencavel, J.H., 1986, Testing between Alternative Models of Wage and Employment Determination in Unionized Markets, Journal of Political Economy, vol. 94: S3–S39. McDonald, I.M., Solow, R., 1981, Wage Bargaining and Employment, American Economic Review, vol. 71, no. 5: 896–908. Mukherjee, A., 2010, Product Market Cooperation, Profits and Welfare in the Presence of Labor Union, Journal of Industry, Competition and Trade, vol. 10: 151–160. Naylor, R.A., 2002, Industry Profits and Competition under Bilateral Oligopoly, Economic Letters, vol. 77: 169–175. Nickell, S., Andrews, M., 1983, Unions, Real Wages and Employment in Britain 1951–79, Oxford Economic Papers, vol. 35: 183–206. Oswald, A.J., 1985, The Economic Theory of Trade Unions: An Introductory Survey, The Scandinavian Journal of Economics, vol. 87, no. 2: 160–193. Singh, N., Vives, X., 1984, Price and Quantity Competition in a Differentiated Duopoly, RAND Journal of Economics, vol. 15: 546–554. Symeonidis, G., 2008, Downstream Competition, Bargaining, and Welfare, Journal of Economics & Management Strategy, vol. 17, no. 1: 247–270. Symeonidis, G., 2010, Downstream Merger and Welfare in a Bilateral Oligopoly, International Journal of Industrial Organization, vol. 28, no. 3: 230–243.

Economics and Business Review, Vol. 1 (15), No. 4, 2015: 25â&#x20AC;&#x201C;45 DOI: 10.18559/ebr.2015.4.3

Banks, non-bank companies and stock exchange: do we know the relationship1? Binam Ghimire,2 Rishi Gautam,3 Dipesh Karki,4 SatishÂ Sharma4

Abstract : This paper investigates the role played by the banking sector in founding, sustaining and developing stock exchange markets. The paper has constructed data on market capitalisation separately for banks and non-bank companies. We apply cointegration techniques developed by Engle and Granger [1987] and Johansen [1988] and we bootstrap the variables to examine the nature of relationship between the banks and stock markets. We found that banks have played an important role in the development of stock exchanges. Further, the empirical analysis made amongst ten developed and developing exchanges suggests aÂ listing of non-bank companies important for development of stock markets. These findings have also been verified by analysing the data of an exchange not included in the test. Keywords : financial development, banks, stock exchanges, cointegration, bootstrap. JEL codes : C22, C15, G10, G21.

Introduction This paper provides evidence that in many countries the financial institutions, mainly the banks and stock markets, are closely related and banks play an important role in the establishment, management and daily trading of stock markets. The motivation for this paper is the gap in the literature regarding the nature of the relationship between the two financial institutions. Many empirical papers have provided evidence that both banks and stock markets are important for economic growth [Chakraborty and Ray 2006; Deidda and Fattouh 2008; 1

Article received 27 March 2015, accepted 15 September 2015. We thank Professor Jackie Harvey of Newcastle Business School for all her valuable advice and guidance on this paper. 2 Newcastle Business School, Northumbria University, Newcastle upon Tyne, NE1 8ST, U.K.; corresponding author: binam.ghimire@northumbria.ac.uk. 3 Shaker Dev Campus, Tribhuvan University, Kathmandu, Nepal. 4 Newcastle Business School, Northumbria University, Newcastle upon Tyne, U.K.

Economics and Business Review, Vol. 1(15), No. 4, 2015

Levine 2002; Levine and Zervos 1998]. Studies have also shown the complementary role between banks and stock markets [Demirgüç-Kunt and Levine 1996; Garcia and Liu 1999; Li 2007] whilst Deidda and Fattouh [2008] provide evidence of a diminishing impact of banks upon economic growth as stock markets develop. The literature as such is not clear and not explicit enough on the relationship between the two financial institutes. Moreover economic growth literature also includes variables capturing the effect of both banks and stock exchanges as independent variables. Given the substitute and complementary nature of the relationship mentioned in literature, keeping them as independent variables can be empirically questionable which further enhances the scope of this paper. The paper is organised as follows. Section two reviews the literature on banks and markets. The new variables and data constructed for this paper will be discussed in section three. Section four describes the methodology used, discusses the results and also includes an analysis of a further case study: Nepal Stock Exchange. Finally, section five outlines the main conclusions.

1. Literature review Atje and Jovanovic [1993] carried out the first cross-country growth analysis involving banks and stock markets and found a positive influence of stock markets but a negative one for banks. The debate on the comparative advantages of bank based and market based economies became more intense in the subsequent literature. Some of them included a comparison of financial systems in different countries, particularly developed countries, whilst others investigated the nature of the relationship between the banks and markets. Allen [1993] recommends bank-based systems for traditional industries where there is consensus about policies, and market-based systems for dynamic industries where wide agreement is lacking. Allen and Gale [1995] while expanding over Allen [1993] provide a quantitative measure of the share of banks and markets for Germany and United States. For example, they show that the ownership of publicly listed banks during the period 1990–1991 was 8.9% in Germany compared to only 0.3% in the USA [Allen and Gale 1995: 188, Table 3]. The authors continue to explore the bank and market based economies and in Allen and Gale [1995] they show the theoretical model in which they find that a bank based system may perform better than one which is market based. The authors suggest that the German financial system, with its reliance on financial intermediary market, can minimize the risk (using the reserve held by the bank) better than the US financial system that relies more on financial markets. Levine [1997] compares the close bond between banks and industrialists in bank-based economies such as Germany and Japan, and the greater liquidity and risk sharing opportunities in market-based countries such as the United

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange

Kingdom and the United States. The author finds the bank-based financial structure of Japan superior to that of the United States but raises concern over the available quantitative measure that differentiates an economy into bank-based and market-based. The author doubts whether Japan is a bank-based economy since it has one of the best developed stock markets in the world. The author therefore suggests the need for further research with new quantitative measures of financial structure and functioning of financial system [p. 719]. Allen and Gale [2000] provided a more comprehensive explanation of five developed economies (France, Germany and Japan as bank-based and United Kingdom and United States as market-based) and their effect in resource allocation and economic development. The authors find both banks and markets important for a good financial system. Nevertheless they stress the need for more research in the area to understand the advantages and disadvantages of the different types of financial institutions [Allen and Gale 2001]. Some empirical works have suggested a complementary role played by banks and markets. Boyd and Smith [1996] suggest that stock markets and banks may act as complements rather than as substitute sources of capital. Similar to Boyd and Smith [1996], Demirgüç-Kunt and Levine [1996] find that across countries the level of stock market development is positively correlated with development of financial intermediaries. Demirgüç-Kunt and Levine [1996] use data on 44 developed and emerging markets from 1986 to 1993 and find that large stock markets are more liquid, less volatile and more internationally integrated than smaller markets. The authors find developed markets having developed intermediaries. Thus they conclude that stock markets and financial intermediaries complement each other and therefore they grow together as they develop. Boot and Thakor [1997] explain the interaction between banks and markets. They make models of a financial system based on three types of informational asymmetries. The first one is about imperfect knowledge on the quality of investment projects available to borrowers. This is better handled by financial markets as markets are better at pricing the value of the firms. The second is the post-lending moral hazard and the third is uncertainty that a borrower would be prone to moral hazard. The second and third informational asymmetries are better handled by banks as banks continue to retain information about the borrowers. The authors therefore present an optimal combination of banks and markets as a better financial system. The authors also find that when the borrowers gain reputation (at the expense of the bank), the capital market expands. Garcia and Liu [1999] use seven countries in Latin America, six countries in East Asia, and two developed industrial countries (United States and Japan) in their empirical analysis regarding the macroeconomic determinants of stock market surge in the period 1980–1995. They argue that the more developed banking sector in East Asian economies led to growth in the size of the market in the region. The authors use stock market capitalization as a measure of

Economics and Business Review, Vol. 1(15), No. 4, 2015

stock market development. They find the stock market as a complement rather than substitute for the banking sector. Similarly, Li [2007] finds development of financial intermediaries have a positive association with the size of equity markets. The author uses 33 developed and developing countries. The author finds the stock market of less developed countries growing much faster in size than the developed countries in the sample, whereas more developed countries enjoyed faster growth in trading activity than the developing countries. Levine [2002] could not find support for either a bank-based or a marketbased financial system, instead favoured for overall financial development importantly influenced by legal system. Chakraborty and Rays’ [2006] findings are similar to Levine [2002] as they were also unable to find one type of system superior to another although they suggested bank-based system are more beneficial to industrial countries. Deidda and Fattouh [2008] find both banks and stock markets important for growth. However, in their study, the growth impact of bank development (measured by domestic credit to private sector to GDP) is lower when the level of stock market development (measured by turnover ratio) is higher. Minier [2009] finds that opening a stock exchange is important for growth. The author finds that countries experience higher economic growth during the first 5 years of the existence of a stock exchange. However, the longer-term results, according to Minier, [2009] are more ambiguous. Lee [2012] looked at the relative merits of bank-based and market-based financial systems. The author used a subsample analysis in which it is found that for all six countries in the sample, which includes both bank and market based economies, the banking sector played an important role in the early years of economic growth. A further analysis by the author shows that the banking sector and the stock market were complementary to each other in each country in the process of economic growth except for the United States, where the two sectors were mildly substitutable. Ghimire and Giorgioni [2013] apply the World Bank Enterprise survey data in their study of finance and growth relationship. They identify internal finance as an important variable which is omitted in the finance and growth literature. Similar to some existing literature [Beck and Levine 2004; Loayza and Rancière 2006; Saci, Giorgioni, and Holden 2009], the authors provide evidence of a negative impact of banks’ development on economic growth in the shortterm. However, they could not establish whether stock markets contribute to growth in a significant manner when the effect of internal finance was included. Vithessonthi [2014] examines how bank risk varies with changes in financial market development in a broad data set of 52 publicly listed commercial banks in five South East Asian countries over a 23-year period between 1990 and 2012. The author finds that higher degrees of financial market development are associated with weaker bank capital positions and are positively related to higher degrees of bank revenue diversification.

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange

2. New variables and data In the literature on finance and growth variables approximating the role of banks and stock markets are included in the models to be estimated and are considered as independent variables. In this paper we examine whether this assumption is correct. The argument is that banks may play an important role in the founding, managing and development of stock exchanges in certain circumstances. In every country of the world, banks (mostly commercial) have been established much earlier than stock markets. Therefore the possibility that banks can play an important role in the actual setting up and running of a stock exchange should not be discounted. For example banks can own brokerage houses, shares of banks themselves are floated on stock markets (as parts of initial placement offers IPOs, further issuance or privatisation programmes) and banks may act as brokers or financial investment advisors. In addition banks might be instrumental in setting up stock markets in a number of other ways (providing initial capital, management and operators). Banks could establish and/or own stock markets. As shown in Table1, which is based on the elaboration of the authors from data provided by Minier [2009], this is frequently the case. Table 1. Role of banks in stock markets Country

Date of Establishment

Founder

Croatia

1991

25 banks and insurance companies

Estonia

1996

Founded by a collection of commercial banks, brokerage firms and state actors

Fiji

1994

Stock exchange established as a wholly owned subsidiary of the Fiji Development Bank

Iceland

1996

Joint venture of several banks and brokerage firms, at the initiative of the Central Bank

Jamaica

1968

Privately founded by four founding members in 1968 with involvement of the Bank of Jamaica

Korea

1956

Trading began in March 1956 with 12 issues; established jointly by banks, insurance and securities companies under government sponsorship

Nicaragua

1994

Government, banks and private companies met in 1990 to discuss forming a stock market

Taiwan

1961

Established late 1961 with capital from government operated and privately owned banks and enterprises

To move one step further and provide some empirical evidence of this possible relationship, we collect, separately for banks and for companies other than

Economics and Business Review, Vol. 1(15), No. 4, 2015

banks, data on one of the variables most commonly utilised to approximate the development of stock exchanges: market capitalisation (usually as a proportion of GDP in the literature). Data on market capitalisation of banks (henceforth “BANKCAP”) and market capitalisation of all other companies except banks (henceforth “NONBANKCAP”) have been collected for the exchanges in countries listed in Table 2. Table 2 also includes the Age of the Exchange and the average dollar capitalisation 18 year period for all 10 sample countries. Table 2. Average annual market capitalization for the period 1991–2008 and age of the exchange between 1991–2008 for sample countries Country

Market Capitalisation (million US Dollars)

Age of Exchange

Hong Kong

505,366

117

Korea

332,734

Malaysia

171,536

Singapore

170,331

Thailand

90,488

Indonesia

61,690

Pakistan

18,614

Kenya

3,784

Sri Lanka

2,990

Bangladesh

2,221

The year of establishment is obtained from the websites of stock exchanges of the countries. Details available in Appendix Table A.1. The age of exchange is calculated by subtracting from 2008. Market capitalisation above is from Standard and Poor’s, Emerging Stock Markets Factbook and supplemental S and P data downloaded from the UK Data Archive.

Given the complexity of collecting data for market capitalisation for banks and non-bank companies which is very time consuming, we restricted our sample to 10 countries. The motivation for selecting the countries listed in Table 2 is to include exchanges at different levels of development in the sample which is discussed below: 1. Exchanges in Hong Kong and Singapore are some of the most developed in the world. Additionally, they are very old and therefore represent well established stock exchanges. 2. Exchanges in countries such as Korea, Malaysia and Thailand are not very old but are regarded as systematically more developed exchanges than many (Australia, Canada and many in Europe) in the world [Demirgüç-Kunt and Levine 1996].

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange

3. Exchanges in Bangladesh, Pakistan, Sri Lanka are not very new (for example Dhaka Stock Exchange was established two years before the Korea Exchange and the stock exchange in Pakistan was established before the Korea Exchange) but they are not as developed as some others in the sample. 4. Finally, the exchange in Kenya is relatively new and has fewer listed companies and aÂ low market capitalisation. It is important to emphasise that the variables BANKCAP and NONBANKCAP are new in the literature. As such we construct the annual time series of the data for the two variables over the period 1991 to 2008. We start from 1991 as data before this period are not available for many countries in the sample. To compute the two variables, market capitalisation data for the composite (or all indexes) to obtain maximum possible stocks have been downloaded from the main exchange of the country using Datastream and then manually separated into BANKCAP and NONBANKCAP. Table 3 shows the number of banks and non-bank companies listed in the exchanges of the countries (June 2008 is randomly selected as the time period to collect data related to numbers of banks and non-bank companies). A scrutiny of the number of banks listed in the exchange (Table 3) reveals that in less developed exchanges, the proportion of banks is higher. The order of the countries is based on the percentage of banks in the exchange in ascending order. Table 3. Number of listed companies and the % of banks in total Country

Number of banks listed on the exchange

Number of nonÂ â&#x20AC;&#x2018;bank cos. listed on the exchange

% of banks

Malaysia

992

0.6

Hong Kong

1021

1.27

Thailand

464

1.94

Korea

745

2.01

Singapore

215

2.79

Pakistan

234

7.69

Sri Lanka

233

7.73

Indonesia

346

8.38

Bangladesh

235

17.87

Kenya

35.71

The highest percentage of banks in the total number of listed companies is from Kenya followed by Bangladesh. The exchanges in these countries are rela-

Economics and Business Review, Vol. 1(15), No. 4, 2015

tively newer and less developed (according to total market capitalisation) than other exchanges in the sample. In countries with stock exchanges classified as the most developed and fastest growing such as Malaysia, Singapore, Hong Kong and Korea [Demirgüç-Kunt and Levine 1996], banks play a small role (less than 3%). In Sri Lanka, Pakistan and Indonesia the percentage of market capitalisation explained by banks ranges above 7 percent. We first calculated the correlation coefficient between average capitalisation and age for the countries in Table 2 and found it to be 0.64, which is high i.e. older stock exchanges tend to be more developed. Therefore we identified and constructed a new variable “AGEXCHANGE” as a proxy representing the development of the exchange. This variable is admittedly less sophisticated than a variable capturing market capitalisation or a variable capturing the number of firms quoted on a market, although AGEXCHANGE is not immune from criticism. The correlations between the variables BANKCAP and NONBANKCAP for the exchanges in the sample countries are presented in Table 4. Table 4. Correlation coefficient between the variables BANKCAP and NONBANKCAP (annual data 1991–2008 17 observations) Country

Correlation

Bangladesh

0.94

Hong Kong

0.84

Indonesia

0.95

Kenya

0.97

Korea

0.97

Malaysia

0.81

Pakistan

0.89

Singapore

0.91

Sri Lanka

0.96

Thailand

0.80

As noted in Table 4 the correlations are positive and very high for each country in the sample. However in the case of established exchanges BANKCAP represents a very small portion of the total market capitalisation. For effective analysis we now present the line chart for both the variables. Figure 1 presents the line chart for the variables BANKCAP and NONBANKCAP for the markets of Hong Kong, Korea, Malaysia, and Singapore. Banks do not seem to have a prominent role as one would expect

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange Hong Kong 12000

Korea 900000 800000

10000

700000

8000

600000 500000

6000

400000

4000

300000 200000

2000

100000

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Singapore 900

800

700

600

500

400

300

200

100

0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

900

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Malaysia

Figure 1. Line chart for the variables BANKCAP (dashed line) and NONBANKCAP (solid line) for Hong Kong, Korea, Malaysia and Singapore (values in trillion local currency)

in large and diversified exchanges, although BANKCAP is more important in Hong Kong. In Figure 2 the line charts for Indonesia, Pakistan, Sri Lanka and Thailand are presented. Banks appear to be playing a larger role, although in both Thailand and Sri Lanka the gap between BANKCAP and NONBANKCAP widens after around 2001. Finally, the line graphs of Bangladesh and Kenya in Figure 3 show that the two variables have moved quite close. In the case of low developed or new exchanges such as Kenya we believe that banks play a very prominent role in shaping the size of the exchange as banks seem to represent a significant portion of total market capitalisation. In fact, in Bangladesh, during the period 2001–2007, BANKCAP has been higher than NONBANKCAP, whilst in Kenya, BANKCAP has almost been equal to NONBANKCAP throughout the entire period 1991–2008. It may be noted that compared to other countries in the sample, these countries have less developed exchanges.

Economics and Business Review, Vol. 1(15), No. 4, 2015 Indonesia

Pakistan 3000

1600000

2500

1400000 1200000

2000

1000000 800000

1500

600000

1000

400000

500 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

200000

Sri Lanka

Thailand

800

6000

700

5000

600

4000

500 400

3000

300

2000

200

0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

1000

100

Figure 2. Line chart for the variable BANKCAP (dashed line) and NONBANKCAP (solid line) for Indonesia, Pakistan, Sri Lanka and Thailand (values in trillion local currency)

By looking both at the number of banks against the number of non-bank companies (Table 3) and at the market capitalisation of banks against the market capitalisation of other companies it is reasonable to suggest aÂ testable hypothBangladesh

Kenya

900

350

800

300

700

250

600

200

500 400

150

300

100

200

100 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Figure 3. Line chart for the variable BANKCAP (dashed line) and NONBANKCAP (solid line) for Bangladesh and Kenya (values in trillion local currency)

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange

esis that in less developed and newer (i.e. more recently established) markets the stakes of banks are relatively higher than in more developed and longer established markets. This is investigated in the next section.

3. Methodology and results This paper applies the Johansen [1988] and the Engle-Granger [1987] two step methods of cointegration. Johansen [1988] is applied as the main test for the variable BANKCAP and NONBANKCAP. As an additional test, Engle-Granger [1987] is carried out where the variable AGEEXCHANGE is used. Two variables will be cointegrated to test if they have a long term, or equilibrium, relationship between them. So the test will help to establish if there is cointegration between banks and exchanges, confirming evidence of any longrun relationship. The basic objective of the cointegration is to test the null hypothesis that ϕ =1 in yt = ϕyt – 1 + ut

(1)

against the one-sided alternative ϕ <1. So we have, H0: series contains a unit root vs. H1: series is stationary. The regression equation is: Δyt = ψyt – 1 + ut,

(2)

so that a test of ϕ = 1 is equivalent to a test of ψ = 0 (since ϕ – 1 = ψ). We first apply Johansen [1988] cointegration test which produces two statistics: maximal eigenvalue of the stochastic matrix and the test based on the trace of the stochastic matrix. This test is based around an examination of the π matrix, where π can be interpreted as a long-run coefficient matrix. The test for co-integration between the variables is calculated by looking at the rank of the π matrix through its eigenvalues. π can be defined as the product of two matrices: π = αβ'. (3) The matrix β gives the cointegrating vectors, whilst a gives the amount of each cointegrating vector entering each equation of the vector error correction model. Here the trace test is a joint test where the null hypothesis is that the number of cointegrating vectors is less than or equal to r, against a general alternative that there are more than r. Whereas the maximum Eigenvalue test conducts separate tests on the individual eigenvalues and the null hypothesis

Economics and Business Review, Vol. 1(15), No. 4, 2015

is that the number of cointegrating vectors is r, against an alternative of (r + 1). The two statistics are: λTrace (r ) = −T

∑ in(1 − λˆ ), (4)

i = r +1

λMax (r , r + 1) = −T ln(1 − λˆr +1 ), (5)

where λˆi is the estimated value for the ith ordered eigenvalue. Trace and Maximum Eigenvalue statistics are calculated for all countries in the sample. In order to further support the Johansen test, the two step cointegration, using Engle and Granger’s [1987], procedure will also be performed. To achieve this, the residuals from the regression equation are calculated. On the residuals, unit root tests (ADF5) are applied to find the t statistics. It is expected that if the t-statistics, are higher this will indicate a closer association between the variables and vice versa. If the hypothesis is true then t-statistics for less developed stock exchanges will be higher. On the other hand, a developed stock exchange should have a lower t-statistics value i.e. associated to a lesser extent. The correlation coefficient is calculated between t-statistics and variables representing stock exchange development (AGEXCHANGE). If the correlation is negative then it would imply that in countries with highly developed stock exchanges, the banks and stock market will have less cointegration as compared to countries with less developed exchanges. As a further robustness check, bootstrapping6 of the variables (AGEX CCHANGE and ADF t-statistics) can be done to find the level of confidence interval between the two7. Finally, if the difference is stationary (from the cointegration test), this should imply that banks are dominant for the stock market (and the country does not have a well developed exchange). Cointegration does not seek the causality. However in the exchange banks are only one out of many industries in a country. So it should not be the only element responsible for the growth of a stock market. In other words, if the difference between total market capitalization and bank stock capitalization remains stationary, this practically means that bank stocks are the main element in the stock market contributing to its growth. The Johansen cointegration results are reported in Appendix Table A2. The test shows that there is cointegration amongst BANKCAP and NONBANKCAP for countries namely Bangladesh, and Kenya (Trace test at 8%). The cointegra 5

The unit root test is conducted for both intercept and trend. ADF is the augmented DickeyFuller test. 6 Bootstrapping with replacement. 7 Bootstrapping is done in MATLAB. MATLAB is a numerical computing environment maintained by the MathWorks (http://www.mathworks.com/).

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange

tion for Thailand can be established at 9% from the Trace test and 8% from the Maximum Eigenvalue test. The countries that have no cointegration are Hong Kong, Indonesia, Korea, Malaysia, Pakistan, Singapore, and Sri Lanka. Both the Trace and Maximum Eigenvalue tests support the results. The results are similar to those predicted through the line graphs. This means stock exchanges that have a larger share of banks are cointegrated. The paper also runs the Engle and Granger [1987] two step method of calculating cointegration. The OLS equation is run and the series for residual is derived in which the ADF test is performed. It is found that that the variables are not cointegrated for many countries. However it is well established that the Engle and Granger [1987] method can be unreliable in the case of a small sample. Hence, the t-statistics computed from ADF tests of the residuals are taken. A higher t-statistics will mean the variables are more correlated and vice versa. The ADF t-statistics values of the unit root test on the residuals using Engle and Granger [1987] are made available in Table 5. Table 5. ADF t-statistics Country

ADF t-statistics

Bangladesh

–1.403

Hong Kong

–1.491

Indonesia

–1.962

Kenya

–2.574

Korea

–2.719

Malaysia

–1.740

Pakistan

–3.085

Singapore

–1.279

Sri Lanka

–2.659

Thailand

–3.554

In addition, the re-sampling of the AGEXCHANGE and ADF t-statistics vectors a 1000 times is done to consider the variation in the resulting correlation coefficients. The correlation coefficient is computed on each sample and obtained the histogram that is shown in Figure 4 is obtained. The histogram shows that nearly all the estimates lie on the interval [–1 to –0.2]. Next bootstrapping is done for the pairs consisting of t-stats of ADF test and GEXCHANGE (to construct a confidence interval). After bootstrapping the correlation coefficient 5,000 times (this also helps in minimising error bias in small time series data) at 95% confidence interval, lower/upper limit of –0.2103 and –0.8577 respectively are obtained.

Economics and Business Review, Vol. 1(15), No. 4, 2015 350 300 250 200 150 100 50 0 –1

–0.8

–0.6

–0.4

–0.2

0.2

0.4

0.6

0.8

Figure 4. Histogram of correlation between the variables

The above analysis implies an evidence for an inverted relation between the t-stats of the ADF test and AGEEXCHANGE. In other words when the age of the exchange is high the t-statistics are low and therefore there is no cointegration and vice versa. This evidence of a negative correlation between t-statistics and AGEEXCHANGE implies that in countries with highly developed stock exchanges, the banks and stock market will have less co-integration as compared to countries with less efficient exchanges. The implication is that in less developed stock exchanges the variables BANKCAP and NONBANKCAP are more cointegrated suggesting stationarity of the relationship. In practical terms this means that bank stocks are the main element in the stock market contributing to its growth. In the exchange banks are therefore the dominant players and the exchange may be less developed.

4. Confirming the result The variables used are the first of its kind in literature and the methods applied are not without limitations. In order to gather further support for the findings (i.e. least developed exchanges have banks as dominant players in the market) the paper has collected the market capitalisation data of Nepal Stock Exchange (NEPSE8). Nepal is not the sample country of this empirical investigation. Hence the testing of the results obtained should be unbiased when tested for NEPSE. 8

NEPSE is the only stock exchange of Nepal. It was established in 1983.

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange

The numbers of listed companies in the exchange (as of July 2009) after separation into financial and non-financial sectors are presented below. The number of listed financial and non-financial companies in the exchange (as of July 2009) is presented below in Table 6. Table 6. Number of listed companies in NEPSE Types

Financial intermediaries

Non financial

Total number of listed companies

133

Category

Number listed

Commercial banks

Finance companies

Development banks

Insurance companies

Hotels

Manufacturing & processing

Others

HydroPower

Tradings

Source: NEPSE.

The line graph of the data for BANKCAP and NONBANKCAP is provided in Figure 5 next. Throughout the six years the contribution of banks in total market capitalisation has remained stable at 82% on average. The paper has also collected data on the average annual Value traded and Number of shares traded for one year on aÂ monthly basis for all companies listed in the exchange. It is found that on average financial institutions represented 85.84% of the total value traded. Similarly, almost 91% of the total numbers of shares were traded on account of financial institutions. Finally, the Johansen co-integration test is carried out for NEPSE. The results show cointegration between the market capitalisation of financial and non-financial stocks (Appendix Table A3).

Economics and Business Review, Vol. 1(15), No. 4, 2015 450000 400000 350000 300000 250000 200000 150000 100000 50000 Apr-09

Dec-08

Aug-08

Apr-08

Dec-07

Apr-07

Aug-07

Dec-06

Apr-06

Aug-06

Dec-05

Aug-05

Apr-05

Dec-04

Aug-04

Apr-04

Dec-03

Aug-03

Month and year

Figure 5. BANKCAP (Dashed line) and NONBANKCAP (Solid line) – NEPSE. Amount in million local currency

The sudden increase in the capitalization of non-bank companies in August 2008 is due to the listing of the telecommunication company (Nepal Doorsanchar Company Ltd.) into NEPSE. Nepal Doorsanchar Company Ltd. was listed in the exchange on 25th August 2008 (The Himalayan Times, 26 August 2008, http://www.thehimalayantimes.com/)

Conclusions To summarise, it is found that more developed exchanges have poor cointegration with banks’ development. The hypothesis that under-developed exchanges will have a higher level of cointegration has been confirmed by the high 95% confidence interval of correlation coefficient. This implies that the less developed exchanges are relying mainly upon banks and hence do not have a developed stock exchange. Hence, our paper finds it important to list non-bank companies for the development of exchanges. It is however important to note that our results are the only tentative explanation of the relationship between the two institutes. The methods applied have limitations and given our limited database a future research with extended time series will be beneficial to further confirm our results.

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange

Appendix Table A1. Name of the exchange, establishment date, and Datastream code used to download the data Bangladesh

Hong Kong

Bangladesh has two stock exchanges namely the Dhaka Stock Exchange (DSE) and the Chattagong Stock Exchange (CSE). The former was established in 1954 as “East Pakistan Stock Exchange Ltd”. The name was changed to Dhaka Stock Exchange in 1964. (http://www.dsebd.org/ilf.php). CSE was established in 1995 and has relatively fewer numbers of companies as compared to DSE

Hong Kong is the most investor friendly place in the world

In order to download the data on capitalisation, Datastream provided market capitalisation for “all quoted shares” and has been used to download the data. The mnemonic in Datastream is FBANG Date of establishment of exchange: 1954

As the first exchange, the Association of Stockbrokers in Hong Kong was established in 1891. At present the stock exchange is known as Hong Kong Exchanges and Clearing Limited (HKEx). It is a merger of The Stock Exchange of Hong Kong Limited (SEHK), Hong Kong Futures Exchange Limited (HKFE) and Hong Kong Securities Clearing Company Limited (HKSCC) (http://www.hkex.com.hk/eng/exchange/ corpinfo/history/history.htm) In order to download the data on capitalisation, Datastream provided market capitalisation for “all domestic and foreign shares” and has been used to download the data. The mnemonic in Datastream is FHKQ Date of establishment of the exchange: 1891

Indonesia

Kenya

The first Stock Exchange in Indonesia was built in Batavia (currently known as Jakarta) by the Dutch East Indies Company in 1912

In Kenya until 1963 the trading of shares was limited to European communities. In 1988 the first privatisation through Nairobi Stock Exchange took place when it sold the 20% share of the Kenya Commercial bank. So we take 1988 as the date of establishment of NSE (http://www.nse.co.ke/newsite/inner. asp?cat=ahistory)

Later new stock exchanges were established in Semarang and Surabaya. Surabaya Stock Exchange was merged into the Jakarta Stock Exchange (JSX). As a result, JSX changed its name into the Indonesia Stock Exchange (http://www.idx.co.id/) The data for “Jakarta Composite Index” is downloaded using Datastream where the mnemonic is LJAKCOMP

The stock market capitalisation for “Nairobi Stock Exchange Index” is obtained using Datastream where the mnemonic is LNSEINDX

Date of establishment of the exchange: 1912

Date of establishment of the exchange: 1988

Korea

Malaysia

The Daehan Stock Exchange, the predecessor of the Korea Stock Exchange (KSE), was established in 1956. In 1962 the KSE reorganized into a joint stock corporation. The

The first formal securities’ business organisation in Malaysia was the Singapore Stockbrokers’ Association that was established in 1930

Economics and Business Review, Vol. 1(15), No. 4, 2015

Korea Exchange was established in 2005 as a merger of the Korea Stock Exchange, the KOSDAQ and the Korea Futures Exchange. The Korea Exchange is one of Asia’s largest exchanges with around 1,800 listed companies (http://eng.krx.co.kr/m9/m9_1/ m9_1_3/UHPENG09001_03.html) The data for “KOSPI Composite Index constituents” is obtained using Datastream where the mnemonic is LKORCOMP

The Malayan Stock Exchange was established in 1960 and the public trading of shares commenced Currency interchangeability between Malaysia and Singapore ceased in 1973 and the Stock Exchange of Malaysia became Kuala Lumpur Stock Exchange Berhad. On April 14, 2004, the name was changed to Bursa Malaysia Berhad (http://www.klse. com.my/website/bm/about_us/the_organisation/history.html) The data for “Malaysia all quoted securities” is obtained using Datastream where the mnemonic is FMALQ

Pakistan

Singapore

Karachi Stock Exchange is the premier stock exchange of the country. It was established in 1947 with 5 listed companies. The total number of companies listed is 651 as of March 2010. KSE has now 4 indices namely KSE 100, KSE 30, KSE All Share Index and KMI 30 (http://www.kse.com.pk/)

Singapore Stockbrokers’ Association was established in 1930 (http://www.klse.com. my/website/bm/about_us/the_organisation/ history.html)

Interchange of currency between Malaysia and Singapore ceased in 1973 and the exchange became the Stock Exchange of “All stock Pakistan stocks” data is obtained Singapore. The Singapore International from Datastream. The mnemonic is PAKALL Monetary Exchange was a futures’ exchange that was established in 1984. The Singapore Stock Exchange established on 1st December 1999 resulted from the merger of the two financial institutions - the Stock Exchange of Singapore and the Singapore International Monetary Exchange (http://www.sgx.com/ wps/portal/corporate/cp-en/about_sgx) In the Datastream “Singapore All Quoted Securities” is taken to download the data. The mnemonic is FSINQ Sri Lanka

Thailand

The exchange for a specific purpose (when British Planters needed funds to set up Tea Plantations in Sri Lanka in the 19th century) was established a long time ago; however the formal exchange market, Colombo Stock Exchange, was established only in 1985 (http://www.cse.lk/welcome.htm)

The first stock exchange was established in 1962 privately. A more formal exchange was established in 1975 and the name given was The Securities exchange of Thailand. On January 1, 1991 its name was formally changed to “The Stock Exchange of Thailand”, SET (http://www.set.or.th/en/ about/overview/history_p1.html)

In order to download the data on capitalization, Datastream provided “Research Stocks” In the Datastream, “Stock Exchange of is used where the mnemonic is FSRILA Thailand (S.E.T.)” is taken to download the data. The mnemonic is LBNGKSET

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange

Table A2. Results of Johansen [1988] Cointegration test Country Bangladesh

Kenya

Hongkong

Indonesia

Korea

Malaysia

Pakistan

Singapore

SriLanka

Test type and detail Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None * At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None * At most 1 Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None At most 1 No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None * At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None * At most 1 Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None * At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None * At most 1

P-values

0.00 0.34

0.08 0.77

0.04 0.77

0.35 0.65

0.31 0.65 0.17 0.56

0.15 0.56

0.16 0.69

0.11 0.69

0.11 0.98

0.03 0.98

0.28 0.54

0.82 0.98

0.61 0.98

0.42 0.33

0.64 0.33

Singapore

44 SriLanka

Thailand

Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) Economics and Business Review, None At most 1 Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None * At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None * At most 1 Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None * At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None * At most 1

0.82 0.98

Vol. 1(15), No. 4, 2015 0.61 0.98

0.42 0.33

0.64 0.33

0.09 0.49

0.08 0.49

Table A3. Results of Johansen [1988] Cointegration Test: Nepal Country Nepal

Test type and detail Unrestricted Cointegration Rank Test (Trace) No. of CE(s) None At most 1 Unrestricted Cointegration Rank Test (Maximum Eigenvalue) No. of CE(s) None* At most 1

P-values

0.0671 0.6196

0.0409 0.6196

References Allen, F., 1993, Stock Markets and Resource Allocation, in: Mayer, C., Vives, X. (eds.) Capital Markets and Financial Intermediation, Cambridge University Press, Cambridge. Allen, F., Gale, D., 1995, A Welfare Comparison of Intermediaries and Financial Markets in Germany and the US, European Economic Review, no. 39: 179–209. Allen, F., Gale, D., 2000, Comparing Financial Systems, MIT Press, Cambridge, MA. Allen, F., Gale, D., 2001, Comparative Financial Systems: A Survey [online], Working Papers, no. 15, Wharton School Center for Financial Institutions, http://fic.wharton.upenn.edu/fic/papers/01/0115.pdf [access: 9.05.2014]. Atje, R., Jovanovic, B., 1993, Stock Markets and Development, European Economic Review, no. 37: 632–640. Beck, T., Levine, R., 2004, Stock Markets, Banks, and Growth: Panel Evidence, Journal of Banking & Finance, no. 28: 423–442. Boot, A., Thakor, A., 1997, Financial System Architecture, Review of Financial Studies, no. 10: 693–733. Boyd, J., Smith, B., 1996, The Coevolution of the Real and Financial Sectors in the Growth Process, The World Bank Economic Review, no. 10: 371–396. Chakraborty, S., Ray, T., 2006, Bank-based versus Market-based Financial Systems: A Growth-theoretic Analysis, Journal of Monetary Economics, no. 53: 329–350.

B. Ghimire, R. Gautam, D. Karki, S. Sharma, Banks, non-bank companies and stock exchange

Deidda, L., Fattouh, B., 2008, Banks, Financial Markets and Growth, Journal of Financial Intermediation, no. 17: 6–36. Demirgüç-Kunt, A., Levine, R., 1996, Stock Market Development and Financial Intermediaries: Stylized Facts, World Bank Economic Review, no. 10: 291–321. Engle, R., Granger, C., 1987, Co-Integration and Error Correction: Representation, Estimation, and Testing, Econometrica, no. 55: 251–276. Garcia, V., Liu, L., 1999, Macroeconomic Determinants of Stock Market Development, Journal of Applied Economics, no. 2: 29–59. Ghimire, B., Giorgioni, G., 2013, Finance and Growth: An Investigation into the Role of Internal, Bank and Equity Finances, Poznań University of Economics Review, no. 14: 31–46. Johansen, S., 1988, Statistical Analysis of Cointegration Vectors, Journal of Economic Dynamics and Control, no. 12: 231–254. Lee, B., 2012, Bank Based and Market Based Financial Systems: Time-series Evidence, Pacific-Basin Financial Journal, no. 20: 179–197. Levine, R., 1997, Financial Development and Economic Growth: Views and Agenda, Journal of Economic Literature, no. 35: 688–726. Levine, R., 2002, Bank-based or Market-based Financial Systems: Which Is Better?, Journal of Financial Intermediation, no. 11: 398–428. Levine, R., Zervos, S., 1998, Stock Markets, Banks, and Economic Growth, American Economic Review, no. 88: 537–558. Li, K., 2007, The Growth in Equity Market Size and Trading Activity: An International Study, Journal of Empirical Finance, no. 14: 59–90. Loayza, N., Rancière, R., 2006, Financial Development, Financial Fragility, and Growth, Journal of Money, Credit & Banking, no. 38: 1051–1076. Minier, J., 2009, Opening a Stock Exchange, Journal of Development Economics, no. 90: 135–43. Saci, K., Giorgioni, G., Holden, K., 2009, Does Financial Development Affect Growth?, Applied Economics, no. 41: 1701–1707. Vithessonthi, C., 2014, The Effect of Financial Market Development on Bank Risk: Evidence from Southeast Asian Countries, International Review of Financial Analysis, no. 35: 249–260.

Economics and Business Review, Vol. 1 (15), No. 4, 2015: 46–68 DOI: 10.18559/ebr.2015.4.4

Measuring the usefulness of information publication time to proxy for returns1 Itai Blitzer2

Abstract : This paper deals with investors’ reaction to financial reports submitted by firms to the stock exchange, and specifically measures the influence of publication timing on investors: by using the proximity of the publication date to the regulated publication deadline as an independent variable the study examines whether deadline proximity causes a change in investors’ reaction (as reflected in share returns). Understanding the connection between the publication date and investors’ reaction contributes to the general understanding of financial reports and to the understanding of investors as recipients of those reports. Methodology: quantitative analysis is used based on empirical data collected at the Tel-Aviv Stock Exchange from financial reports published over the fiscal years 2009– 2013. The data include quarterly and yearly financial reports and share performance for the corresponding periods. By bundling report publications made within a specific proximity to the deadline and comparing them with investors’ reaction, non-parametric tests reveal a statistically significant correlation between the publication date deadline proximity to the share performance (returns). The conclusion of this research is that the time of financial report publication has an influence on investors’ reaction. This suggests that investors react to financial reports not just based on their intrinsic information content but also in respect of their publication time. Keywords : stock performance, predictive analytics, financial report publications. JEL codes : G14, L25, M41.

Introduction The aim of this paper is to study the nature of relationship between the following two variables: –– Deadline proximity: measuring a time-span between a financial report publication date and the deadline for publishing (which is regulated by the Israel Securities Authority). 1 2

Article received 05 February 2015, accepted 15 September 2015. TradeSoft Ltd., 9 Bavli St., Tel Aviv, Israel, itai.blitzer@gmail.com.

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

–– Investors’ reaction to publications: measuring the aggregate response of investors to publications as reflected by share performance movements (returns). In addition to establishing the said relationship between the two variables, the article also examines the reasons for this and concludes with the outcomes deriving therefrom. The conclusion presented by this paper is that the date in which financial reports are published (in relation to the regulated deadline) has an impact on investors, and in turn influences share prices. Results demonstrate the existence of a trend, statistically significant, in which investors’ reaction is related to the deadline proximity (days left to the financial report presentation deadline date at the date on which the report is filed) of the publication. This conclusion suggests that investors’ reaction to publications made by firms is not limited to the intrinsic content of information contained in the publications themselves but is rather more influenced by external effects such as the deadline proximity of the published report. As will be elaborated in the conclusion section, the conclusions contribute to the general understanding of financial reports and investors’ reaction by benchmarking TASE as a stock exchange of smaller proportions and with a less regulated environment than US based stock exchanges upon which most of the relevant research has been conducted. The Israel Securities Authority regulations for financial report publication state that reports are to be sent by firms to the stock exchange after their period of reference (quarter/year) has passed, similarly to the mandatory 10-Q and 10-K filings which are regulated by the Securities and Exchange Commission at the US based stock exchanges. Given 2–3 months to produce the reports, public firms traded at TASE are required to publish the reports within two months following the quarter’s end in the case of quarterly reports, or within three months following the quarter’s end in the case of annual reports (Tabel 1). Table 1. Publication regulated deadline dates per quarter Quarter

Quarter end date

Report delivery deadline

31 March

31 May

30 June

31 August

31 September

30 November

31 December

31 March

Over the five year period examined for this research (2009 to 2013), nearly 16% of the firms have filed their reports to the Tel Aviv Stock Exchange on the last possible day for publication, 50% of firms do so in the last five days before the publication deadline date. As opposed to the latter 50% which are con-

Economics and Business Review, Vol. 1(15), No. 4, 2015

Percentage of documents delivered

% 14 12 10 8 6 4 2 0

–30

–28 –26 –24 –22 –20 –18 –16 –14 –12 –10

–8

–6

–4

–2

Days left for deadline

Figure 1. Distribution of reports filed for publication at TASE during last month before the regulated deadline

sidered „late filers”, more than 20% of firms prefer to present the report early, meaning more than two weeks before the publication deadline date (Figure 1). Quarterly financial reports can be viewed as a channel of information between firms and investors. It is quite obvious that the exact same item of news, when read in a certain channel and not in another, might produce a differing extent of reactions. The volatility of reactions to news stories has been demonstrated in previous research [Das and Chen 2007; Antweiler and Frank 2006]. News and information always come within a context. Whilst both channel and information might be identical, the different context in which information arrives may cause a great deal of difference in aggregate investors’ reaction. Controlling for channels, previous research [Baumeister et al. 2001] noted that negative oriented information has a bigger impact and is more thoroughly processed and absorbed than positive oriented information across a wide range of contexts. The strong form of the efficient market theory [Fama 1970] states that all future information is already reflected in a stock’s price (including inside information). An outcome for this form of efficiency is that one can never “beat the market”, since the latter already incorporates everything expected to happen within the stock price. If so, we should question whether the market reacts to the data in the financial report or to the difference between that data and the prior assumptions held by the reader before reading. Previous research [Morris et al. 2005] demonstrated that whilst holding information about stock price trends constant, expectations of investors toward trend continuation are influenced by the language used to describe the trend. If the efficiency itself is subject to the environment should we not question the outcomes deriving from fundamental analysis showing a larger differentiation in environments with different (higher) information dissemination? Put

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

simply, the ability to extract data quickly and efficiently from the report is as important as the actual data contained in it. Therefore we should treat the speed of information dissemination as an influencer on performance. This paper illustrates such a case to demonstrate this concept which can be implemented in many relations and forms. Another possible interpretation can be found within the relationship to investors (rather than managers who control publications): as stated by [BędowskaSójka 2014], “At a time of high volatility informed traders are willing to place large orders because high volatility provides a sufficient camouflage of their information”. The last days before the deadline date can be treated as a highly volatile time span (as it encompass filings for the majority of firms). This can lead in turn to the encouragement of stealth investors to choose these days, thus changing the composition of investors between the filing groups (early/late). Tel-Aviv Stock Exchange (TASE): Being the only public market in Israel for trading securities, it lists almost all of the Israeli market’s public firms. Although an impressive 60% of which are also traded in other stock exchanges around the world, TASE holds a very important role in Israel’s economy. As of January 2015 TASE consists of over 567 public firms, over a thousand mutual funds, 180 exchange traded funds and also corporate and government bonds. TASE is a privately held firm, its 26 owners are banks and large investment companies which are the only ones allowed to coordinate the exchange and ask for operation fees for their services. Of these 26 members, there are three foreign banks (Barclays, Citibank, HSBC), and four foreign investment firms (Merrill Lynch, UBS, Citi Group and Deutsche Securities). All other members are Israeli. The aggregated market value of all TASE market firms is estimated, as of January 2015, at 810 Billion NIS, which were equivalent to about 207 Billion US dollars at the time of estimation. TASE offers its users a computerized system called TACT (similar to the US system named “EDGAR”). It is a fully automated trading system which allows collaboration and seamless integration of real time information. All products, including shares, bonds, treasury bills and derivatives, are traded using this system. The positive impact of adopting this system on trade composition has been demonstrated in several research papers [Shapiro 1986; Amihud, Hauser, and Kirsh 2003; Kalay, Wei, and Wohl 2002]. TASE allows dual listing, in many cases the New York Stock Exchange or the NASDAQ are the preferred choice by firms. Foreign holdings of international investors in TASE tend to be fixed at around 11% during the last decade. A note on location and market volume: questioning whether the Tel Aviv stock exchange qualifies as a representative stock exchange for performance prediction can be answered through similarity with research done at the major stock exchanges, measured by trade volume and market cap (NYSE, NASDAQ, JPX, EURONEXT, LSE). In order to qualify research conducted at the Tel Aviv Stock Exchange this paper deals with the same metrics, measurements and time

Economics and Business Review, Vol. 1(15), No. 4, 2015

frames as previous research focusing on stock exchanges in developed markets [Iqbal and Farooqi 2011; Li 2006; Hirschey, Palmrose, and Scholz 2003]. The paper is divided into four sections. The first section includes a literature review of both parent and immediate disciplines. The second section deals with methodology, with special focus placed on data sampling, manipulation and preliminary steps taken in processing the data. The third section demonstrates and describes the findings. The last section summarizes the conclusions, which support the correlation between stock performance and report filing timing and include a discussion of the possible reasons as to why this correlation exists, and suggestions for future research.

1. Literature review Numerous articles have been written over the effects which financial reports have on company figures. These effects were researched by looking into two main verticals: a quantitative vertical mostly dealing with timing to performance ratios and a qualitative vertical which deals with text analysis. From the quantitative point of view researchers who dealt with the concept of timing [Mackinlay 1997] did it in an indirect fashion, through examining presentation timing of restatements (presenting additional data over the same time period in which existing data already exists), whilst other researchers [Hirschey, Palmrose, and Scholz 2003; Anderson and Yohn 2002] differ by focusing on the efficiency of dissimilation of new data for a new time period. At the qualitative vertical a great deal of research has already been conducted to establish a relationship between the quality of financial reports and their impact on investment decisions, thus placing emphasis on the narrative composition of the reports presented. These research papers [Biddle and Hilary 2006; Abarbanell 1991] are usually single-dimensioned from the timing point of view, (meaning they only deal with presenting new data), and focus on qualitative measures such as text and fundamentally related attributes. This research can be categorized amongst other research done in the quantitative vertical. However it differs to a rather large extent from the timing based research group, since the data used are actually based on metadata and are not firsthand stock exchange data. What this means is that comparison data which was used for correlation testing was actually calculated (days from deadline) and is not considered a part of, or is in any of the financial data. Specific research over late filed reports (after the deadline) observed a worse stock performance [Alford, Jones, and Zmijewski 1994; Bagnoli, Kross, and Watts 2002], as expected in regard to the correlation discovered in this paper. Such research was also conducted in relation to Form 10-Ks filed after SEC deadline, revealing a worse financial performance of a firm compared with previous year performance and expected performance [Li and Ramesh 2009].

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

It can be argued that the very existence of quarterly and early reports illustrates a somewhat ambiguous side of Fama’s “Efficient market hypotheses” [Fama 1970]. The efficient market hypothesis in its strongest form assumes that all investors are exposed to the same data. Even when accepting this hypothesis it is also clear that no two investors really value a stock in the same way – the exact same information given to two different investors can be treated in a completely opposite manner. This is of importance when making a distinction between institutional investors and non-institutional investors as indicated in previous research [Lakonishok et al. 1991; Musto 1999], which related increased price reactions around calendar quarter-ends with the incentives of institutional investors to window dress (to improve the appearance of the portfolio performance before presenting it, the manager will sell stocks with large losses and purchase high flying stocks near the end of the quarter). Such actions are also documented in post-EDGAR research [Carhart et al. 2002; Morey and O’Neal 2006]. When trying to measure the impact which early delivery of reports has on stock performance we should question whether the variance composition of different investors from various aspects (institutional/non institutional, bearish/ bullish, fast/slow moving) is distributed evenly (homogenously) throughout the range. This is because early presentation may possess an indirect influence over the stock performance not due to its intrinsic properties, but rather through the different types of investors reacting to its data, meaning that the prices are affected by a mediator variable (which is the reaction of different investors to the delivery of reports). Previous research [Dontoh and Ronen 1993], which examined abnormal trading volume surrounding the filing of periodic financial reports had also showed lack of homogeneity of investor type. On viewing the bigger picture, the very essence of a financial report is not just about the data included, but is also about the aggregate reaction of all investors to it. Creating overreaction or under reaction is not “efficiency” in terms of data dissimilation but rather an after-effect. Stating that “all investors are exposed to the same data” as the efficient market hypotheses argues [Fama 1970] may become an unfulfilled prerequisite, due to the reason that the very reaction of aggregate investors is also considered data, that should also be “exposed to all investors” in the same way. Previous research [Piotroski 2000] has added another important observation to the discussion about fundamental analysis showing that the effectiveness of a report filing should also be measured with regard to the ability of investors to successfully process its information quickly: “More importantly, the effectiveness of the fundamental analysis strategy to differentiate value firms is greatest in slow information-dissemination environments.” [Piotroski 2000]. More recent research [Ball and Shivakumar 2008; Li and Ramesh 2009] also questions the value of information conveyed by the SEC form 10-K, suggesting a possible connection between earlier filings and information asymmetry.

Economics and Business Review, Vol. 1(15), No. 4, 2015

Referring specifically to the Tel Aviv Stock Exchange, it’s relatively small size and composition may act as an environmental attribute allowing analysts and investors to better process the early delivered reports (in relation to other global stock exchanges), meaning that one can relate to some extent to TASE as a fast information-dissemination environment in contrast to other (bigger and global) stock exchanges. Although research [Leuz and Wysocki 2008] has indicated that an increase in the quantity of public information can help reduce uncertainty about firm value, this should be viewed in consideration of the dissemination attributes of the information examined. It should be noted that in an average quarter over 200 reports are sent to the Tel Aviv Stock Exchange within a time span of two days (“late delivered” group). This might present a problem to the non-institutional investors (due to lack of resources), creating the exact situation referred to in Piotroski’s research – quick processing as an attribute of effectiveness [Piotroski 2000]. Another notable dimension is risk – non-institutional investors may expect analysts or economic journalists to extract the highlights for them, but nevertheless this poses an increasing risk for investors (as opposed to reports sent early, long before the deadline), so the volatility increases and relative stock performance moves downwards. Metadata can help us understand the existing data in its proper context and enhance analytical performance: the simplest example is a report delivered after the deadline. Whether it is for conscious reasons (such as keeping the stock price high as long as possible before breaking problematic news to the public), or just failing to meet the delivery deadline due to administrative reasons, the very reason a report was handed in late can direct an investor into a different type of reading, resulting in a shift in aggregate reaction from liberal to conservative (and eventually, bullish to bearish). Previous research [Alford, Jones, and Zmijewski 1994] found that late filers typically face significant economic events which account for the delayed filing of reports. Previous research [Kahneman, Knetsch, and Thaler 2008] referred to the objectiveness of an investor when buying rather than selling and relates it to the endowment effect. This effect was originally meant to describe a phenomenon in which people give higher financial value to products they actually own than to another product (similar or identical) that they do not own. His research found that the willingness to accept is always higher than the willingness to pay. The question is, does having a financial report qualify as ownership of a certain product? Does the very reason of a having (meaning getting a report before the deadline) a report, even without reading it, lower investor risk and volatility factors, simply due to the fact he is no longer “waiting for news” regarding an investment, but already has all the data the firm supplies? Research by [Brown and Warner 1980] studied the impact of the annual earnings announcement over share prices. They concluded that even when a report has been proven to contain useful and valuable information it was not

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

integrated immediately it was disclosed. This conclusion is aligned with this paper’s conclusion that as the approach of non-homogeneous report presentation (meaning delivery is made on different days for different firms) affects the market, and most of all affects the investor’s ability to make a proper analysis in an orderly fashion, rather than taking an educated guess when facing hundreds of reports in a very short time frame. This result also concurs with previous research [Meier and Schaumburg 2006] indicating that institutional investors may possess incentives to trade on dates near the quarter-end for various reasons, so there is an end of quarter bias in favour of institutional investors.

2. Methodology This study uses quantitative analysis of empirical data collected at the Tel-Aviv Stock Exchange from financial reports published over the fiscal years 2009– 2013. The data include quarterly and yearly financial reports and share performance for the corresponding periods. In order to understand the nature of the connection between the publication and the investors’ reaction to the publication, a process of bundling report publications (made within a specific proximity to the deadline) and comparing them with investors’ reaction is conducted. After the bundling process a battery of statistical tests are run against the data to check for the existence of a statistically significant correlation between the publication date deadline proximity to share performance (returns). Steps taken in order to sample and prepare the data for non-parametric tests: –– Sampling: collecting data for financial report publications and daily stock quotes. See elaboration in Section 2.1. –– Classification: splitting the data by the type of report, create a distinction between the following types: native report, original report, fix to report, unified report. See elaboration in Section 2.2. –– Standardizing: creating a relative scale for measurement of investors’ reaction as reflected by the stock returns. A standardized share performance measure creates an estimation of how well a share performed with relation to the entire market performance on a specific day. See elaboration in Section 2.3. –– Bundling: publications made by public firms were bundled by time-span proximity to the publication deadline. See elaboration in Section 2.4. –– Filtering: Removing extreme entries from the corpus and dataset of inactive firms and deadline proximity outliers. See elaboration in Section 2.5. After conducting the preliminary steps, a total of 10,632 financial reports were included amongst all the timing (proximity) groups. Two measures were used in this process. The first was a figure of non-standardized (see Section 2.3) share price changes during the selected period, which reflect investor response

Economics and Business Review, Vol. 1(15), No. 4, 2015

to the publication on the day following the publication. The second was a figure of a standardized (see Section 2.3) value of stock price changes, which allowed controlling for market-wise stock movements (influences on all traded stocks), which are not an outcome of a specific publication. After obtaining both standardized and unstandardized figures a comparison between the groups was conducted. The results examine whether there are statistically significant changes amongst the groups in relation to the time left for the report delivery deadline (i.e. deadline proximity).

2.1. Sampling This research uses data gathered from the following two sources: –– Quarterly and yearly reports: financial reports which were delivered by public firms to the Tel-Aviv Stock Exchange during the years 2009–2013. –– Daily return values (share performance quotes) for each firm whose shares were offered for trade at the Tel-Aviv Stock Exchange during the years 2009–2013. Sampling financial report publications: Downloading the corpus of financial reports was done using the TASE filing access framework. A total of 9,687 individual reports were downloaded for the years 2009–2013 (averaging at 1,937 reports for each year of the selected period). Tagging: Every report was given a unique report identifier, as well as the corresponding firm identification within TASE, and the corresponding share identification. Dating: Each of the reports examined was attributed a date indication, referring to the date it was sent to the stock exchange (thus released to the general public) and the corresponding quarter and year for which the content is relevant. Sampling daily return values: Downloading the daily return values was done using the TASE data access framework. Approximately 400 thousand share performance figures were sampled, representing all individual TASE quotes for the five year period mentioned (in a daily resolution), for each firm whose shares were offered for trade during any part of the selected period (2009–2013). Tagging: Every performance entity consisted of a share identification, a relevant trade date and a number representing the change from the previous day in percent.

2.2. Classification As part of the working process at TASE, public firms can file several types of reports. As customary in respect of restatements, fixing reports can be issued by firms to relate to the same period of time to which earlier publications already refer. In order to control situations where two reports refer to the same period of time, a classification process was conducted for each report. This

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

process allowed the creation of a distinction between the following four types of financial reports: native report, original report, fix to report, unified report. The simplest type of report, which is called a „native” report, is a report introduced by a firm for publication on TASE. Some reports are issued with fixes on a later date. In this situation, the original file status is changed from „native” to „original”, and the file containing the fix is given the status „fix”. Another document is created „on the fly”, titled „unified”, which contains both the text of the „original” document and of the „fix” document (Figure 2). Native

Fix

A single document, refering to a specific quarter

Original

A document refering to the same quarter as the native document

Unified

Representing the first native document which was allocated to this quarter

Combined text of all documents refering to the same quarter

Firgure 2. Types of documents in the corpus of financial report publications

The dataset has a grand total of 9,687 files, of which 5,679 are native, 1,512 fix, and 1,248 of both original and unified. The average of reports per firm in these 20 quarters is 27.7 (meaning fix publications are relatively common). Native and original files are distributed evenly between quarters, varying between 24.87% and 25.13% per quarter (Table 2). Table 2. Inventory of financial report publications per period Row labels

Fix

Native

Original

Unified

Grand total

2009

298

1,135

237

1,907

296

462

296

438

305

432

137

238

100

575

2010

376

1,085

298

2,058

295

460

299

448

307

424

221

185

160

726

2011

376

1,072

319

2,086

263

492

302

446

264

530

Economics and Business Review, Vol. 1(15), No. 4, 2015

Tab. 2 continued 4

149

223

123

618

2012

292

1,144

247

1,930

272

512

809

423

311

422

127

525

573

2013

170

1,242

147

1,706

318

410

317

415

310

430

4 Grand total

297

415

1,512

5,679

1,248

9,687

2.3. Standardizing Creating a standardized scale for relative share performance: trade data are available to the general public in any given time through share quotes. The most common performance indicators are the opening and closing figures, followed by the low and high figures and the volume of trade. These figures are standalone figures, measuring the absolute figures of a specific share. A standardized share performance measure provides an estimation of how well a share performed in relation to the entire market performance on a specific day. This measure is especially relevant with regard to the deadline proximity method conducted in this paper, since the goal is not comparing a single share through different times, but rather to compare groups of shares (bundled by deadline proximity). Several stock exchanges have an index indicating the overall performance, for example, the NASDAQ Composite Index. In order to avoid the inherent bias in which share prices are skewed on a certain day which shows overall good (or bad) results for the entire stock population (“green days” or “red days”), returns were standardized in relation to the entire stock population performance. This allows control of the share performance changes which are not single share related, but a cross-share trend (which affects several shares). The standardization was conducted on a daily basis (single day resolution), using the following method: –– Using the unstandardized performance figure (acronym: UPF) as a base value (calculated as the percentage difference between the opening and closing price). –– For each day an average and a standard deviation were computed for every daily UPF.

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

–– A standardized figure for each quote was calculated by extracting the average of UPF from the selected UPF and dividing the outcome by the standard deviation of the UPF population for the selected day. The standardized data was computed in accordance with all stock changes for every day, meaning the average and standard deviation were taken by using not just stocks of firms which issued a report, but all the active stocks available for trade.

2.4. Bundling In order to establish a connection between the time of filing and the investors’ reaction to the filing, we need to test if the performance of stocks is different when controlling for the proximity of the actual publication date with the regulated publication deadline. In order to allow this comparison publications made by public firms were bundled by time-span proximity to the publication deadline. For an entire month prior to the deadline, 31 separate groups were created each containing data for a specific proximity (single day to 31 days). The bundling process includes the following steps: –– Creating 31 different publication groups for each time-span daily group of the month prior to the regulated deadline (filtering returns which do not fall within the 31 groups will be done at the next stage, see Item 2.5). –– Classify each publication to a certain group based on the deadline proximity of the report. –– Calculate returns for each group (average and median for both standardized and unstandardized groups).

2.5. Filtering The downloaded corpus of financial reports and the respective share returns for each firm were filtered as follows: –– Filtering of inactive firms. –– Filtering by deadline proximity. After filtering both inactive firms and time related outliers the data showed the following characteristics: on average, 25% of the firms had sent their reports to TASE on the last two days possible, whilst 50% of firms did so in the last 5 days. On the other hand, 20% of firms presented the report more than two weeks before the deadline and 1% of reports were delivered late, filed on the following day after the deadline. Filtering of inactive firms Population for filtering: –– Firms which have actively traded stocks at TASE in less than 19 out of the 20 quarters of the period examined (2009–2013) were excluded from the dataset.

Economics and Business Review, Vol. 1(15), No. 4, 2015

–– Firms which ceased trading for any reason during the five year period tested for this research were excluded from the dataset. Exclusion of reports was made for all four types of reports and their respective stock performance values. Filtering by deadline proximity Population for filtering: –– Publications which were made earlier than 31 days before the regulated publication deadline were excluded from the corpus and were considered as deadline proximity outliers. –– Publications which were made later by more than one day after the regulated publication deadline were excluded from the corpus and were considered as deadline proximity outliers.

2.6. Justification for methodology Bundling by days (deadline proximity): similar methodology was adopted in research by [Hirshleifer, Lim, and Teoh 2009] which used time based clustering to find heightened market movements over particular times in which information related to a large number of firms. In addition, research [Alford, Jones, and Zmijewski 1994] conducted on US based stock exchanges also used day based bundling by defining the equivalent SEC form 10-Ks as “early” in cases where it was filed at least five calendar days prior to the SEC deadline and bundling filings by categories (early, late and on-time). Using a standardized scale: selecting the autonomous stock performance and the standardized (share wise) value was made in previous research [Kloptchenko et al. 2004; Tetlock, Saar-tsechansky, and Macskassy 2007] in the same manner, thus measuring both an independent stock performance ratio and a unified figure matching all the actual stock price movements on a specific day. A standardized trend model for each firm’s earnings was also adopted in previous research [Bernard and Thomas 1989] in timing related correlation discovery, using a standardized trend model (for each firm’s earnings). A similar standardization of stock prices was taken by previous research [Tetlock, Saar-tsechansky, and Macskassy 2007; Kraft, Vashishtha, and Venkatachalam 2014], which showed success in capturing movements which were relative to other active stock movements during a selected period (a single day in this case). Justification for classification by filing types: this method had been used by [Hranaiova and Byers 2007] whilst including re-statements to research market response reactions to financial restatements. Creating unified documents was implemented to prevent treating the fix documents in the same way as the original ones which may cause unnecessary bias, especially due to the fact that data for the same quarter was presented previously so the original timing would not be overwritten. Applying different predictive force based on the timing of

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

notification was also adopted in relation to restatement effects on investor reliance on earnings [Anderson and Yohn 2002]. Testing a five year period has also been chosen in previous research conducted at the Tel Aviv Stock Exchange [Amihud, Hauser, and Kirsh 2003]. As in previous research [Hirschey et al. 2005], data does not include any cases of overlapping events or performance observation.

3. Findings 3.1. Descriptive statistics The following tests were all conducted using the IBM SPSS statistics’ framework: –– Shapiro-Wilk test of Normality, –– Kolmogorov-Smirnov test of normality, –– Levene’s test for homogeneity of variance, –– Kendall-Tau Correlation, –– Jonckheere’s trend test. Case processing summary: a total of 10,632 cases were processed for standardized and unstandardized stock performance groups combined, for the period of 31 proximity groups (as elaborated in the methodology analysis sub chapter) (Table 3). Table 3. Case processing summary Valid

Missing

Total

percent

Standardized returns

5,316

100

5,316

100

Unstandardized returns

5,316

100

5,316

100

The overall mean of standardized share returns for the selected period was 0.878, with a standard deviation of 1.087, whilst the mean of unstandardized share returns was 0.002 (as expected for a long, five year period), with a standard deviation of 0.0339 (Table 4). The Shapiro-Wilk test of Normality showed that for every year tested (2009–2013), the distribution was not normal with p < 0.05. The same p < 0.05 was shown for the Kolmogorov-Smirnov test of normality (Table 5). Therefore, due to lack of normality in the data, non-parametric tests were used in order to examine correlation and trend. In order to use Kendall-Tau and

Economics and Business Review, Vol. 1(15), No. 4, 2015

Table 4. Mean and standard deviation for both standardized and unstandardized returns

Valid

Days before deadline

Standardized returns

Unstandardized returns

5,316

–

–7.660

0.878

0.002

6.718

1.088

0.034

45.132

1.183

0.001

Missing Mean Std. deviation Variance

Table 5. Normality test: Shapiro-Wilk and Kolomogorov-Smirnov Kolmogorov-Smirnov

Shapiro-Wilk

Year

Statistic

Signifi cance

Statistic

Signifi cance

2009

0.190

302

0.0001

0.007

302

0.0001

2010

0.137

1,320

0.0001

0.778

1,320

0.0001

2011

0.148

1,329

0.0001

0.799

1,329

0.0001

2012

0.188

1,346

0.0001

0.743

1,346

0.0001

2013

0.166

1,019

0.0001

0.843

1,019

0.0001

2009

0.200

302

0.0001

0.601

302

0.0001

2010 Unstandardized 2011 returns 2012

0.167

1,320

0.0001

0.735

1,320

0.0001

0.175

1,329

0.0001

0.740

1,329

0.0001

0.202

1,346

0.0001

0.730

1,346

0.0001

2013

0.189

1,019

0.0001

0.812

1,019

0.0001

Standardized returns

Table 6. ANOVA test between based on absolute differences from the median, between standardized and unstandardized returns

Median difference, standardized

Median difference, unstandardized

Sum of Squares

Mean Square

Sig.

between groups

1.174

1.439

0.055

within groups

4,311

5,284.000

0.816

total

4,347.188

5,315.000

between groups

0.03

0.001

1.116

0.301

within groups

1,523.000

5284

0.001

total

4,552.000

5,315.000

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

Jonckheere’s trend tests, homogeneity of variance is a prerequisite. Levene’s test for homogeneity of variance was conducted in a robust way based on medians instead of means. Results were F(1.439), P = 0.055 for the standardized group and F(1.116), P = 0.301 for the unstandardized group (Table 6). The homogeneity test did not show any significance (p > 0.05) for both standardized and unstandardized stock performance, so the null hypothesis was retained, which means variance is evenly distributed between groups. As there is no normality in the distribution of stock performance (both standardized and unstandardized) Levene’s test however showed an equal distribution of variance between the groups (including both standardized and unstandardized). The prerequisites are met for Kendall-Tau and Jonckheere’s trend tests. These tests are non-parametric and are used for non-normal distributions (as Shapiro-Wilk results shows), but require equal variance (as Levene’s test shows). The final step was to obtain results from the Kendall-Tau correlation. The value of r tau was –0.047 with p < 0.05 for standardized performance and value of r tau = –0.02 with p < 0.05 for unstandardized performance (see elaboration on the conclusions section). Both figures from Kendall Tau were statistically significant at p < 0.05 (Table 7). Table 7. Kendall Tau correlation coefficient and significance Kendall’s tau

Standardized returns

Unstandardized returns

Deadline proximity

Correlation coefficient

Standardized returns

Unstandardized returns

Deadline proximity

0.745

–0.047

0.000

Sig (2-tailed) N

5,316.000

Correlation coefficient

0.745

–0.020

Sig (2-tailed)

0.000

5,316.000

Correlation coefficient

–0.047

–0.020

1.000

Sig (2-tailed)

0.000

0.042

5,316.000

0.042

5,316.000

A crosscheck with Jonckheere’s trend test was conducted to make sure the p-values of the Kendall-Tau were computed correctly. As seen in Table 7 in the appendix, the p-values indeed showed the same p-values as the Kendall Tau correlation, so the significance of the relation between parameters (days to deadline and stock performance) was again proven to exist for both standardized and unstandardized sets (Table 8).

Economics and Business Review, Vol. 1(15), No. 4, 2015

Table 8. Trend test results showing same p-value as the Kendall Tau for crosscheck purposes Number of levels in proximity days

Standardized returns

Unstandardized returns

5,316

Observed J-T Statistic

6,247,522.500

Mean J-T Statistic

6,565,615.00

Std. Deviation of J-T Statistic

64,367.317

63065.497

Std. J-T Statistic

–4.942

–2.037

Asymp. Sig. (2-tailed)

0.000

0.042

Note on nonparametric tests: Since the normality test for distribution has indicated that data was not distributed normally (not following the normal/Gaussian distribution), non-parametric tests had to be made. Non-parametric rank tests have been reported in previous research [Corrado 1989; Campbell and Wasley 1996] to be more powerful than the parametric t-tests for detecting abnormal daily returns and trading volume. Since the particular non-parametric tests needed have required homogeneity of variance to be present at the source data, the dataset had to be manipulated in order to allow the Levene’s test to run against medians. In order to run Levene’s test for homogeneity of variance based on median values instead of means the following process took place: first, computing and aggregating a median value for each of the groups, done twice (for both standardized and unstandardized figures). Then the difference between the actual reading (daily stock performance) and the median was computed and an absolute value from the latter was also obtained. Levene’s test was performed on the absolute figures and succeeded in retaining the null hypothesis which stated that the variances are equal throughout the groups. This step served as the prerequisite for statistical tests that require such homogeneity of variance which were conducted later – Kendall Tau and Jonckheere’s trend test.

3.2. Inferential statistics The results of the non-parametric tests using the data collected from the Tel Aviv Stock Exchange for the years 2009–2013 show the following: –– Delivery timing of financial reports (early/late publication) has an impact on investors’ reaction (as reflected in share returns). –– There is a correlation between the publication date deadline proximity and the share performance. –– The correlation found is statistically significant, showing a negative relation between deadline proximity and returns. Publications with larger proximity

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

Standardized performance

are likely to gain a larger reaction amongst investors rather than publications with smaller proximity to the publication’s regulated deadline. These results align with previous research conducted on the US market: significant market reactions to SEC form 10-K early filings were indicated in several studies [Qi, Woody, and Haw 2000; Asthana and Balsam 2001; Griffin 2003; Asthana, Balsam, and Sankaraguruswamy 2004]. Based on the results, in cases where a report was delivered early to the stock exchange, the relative share performance of the firm will be more likely to have a performance that ranked higher in relation to other stocks. The effect is a relatively linear trend, with a negative correlation between share performance and publication deadline proximity. This negative correlation can be seen through the statistical measures (Kendall-Tau test and Jonckheere’s trend test) and is also visible visually (Figure 3 and 4).

Days to due date

Standardized performance

Figure 3. Standardized values in relation to days left before the deadline, on the X axes, from left to right are the days left for presentation

Days to due date

Figure 4. Unstandardized values in relation to days left before the deadline, on the X axes, from left to right are the days left for presentation

Economics and Business Review, Vol. 1(15), No. 4, 2015

Conclusions This paper deals with investors’ reaction to the publication of financial reports made by firms to the stock exchange. Specifically this study measures the influence of publication timing on investors: by using the proximity of the publication date to the regulated publication deadline as an independent variable, this study examines whether deadline proximity causes a change in investors’ reaction (as reflected in share returns). The conclusion of this paper is that the date on which financial reports are published (in relation to the regulated deadline) has an impact on investors, and in turn, influences share prices. Results demonstrate the existence of a trend, statistically significant, in which investors’ reaction is related to the deadline proximity (days left to the deadline for financial report presentation and the date on which the report is filed) of the publication. Additional conclusions deriving from the test results: Investors’ reaction to publications issued by firms is not limited to the intrinsic content of information contained in the publications themselves, but rather can be influenced by external attributes such as the deadline proximity of the published report. The earlier a report is presented (i.e. far from the deadline date), the more chance of the respective share to show a better performance, compared to the performance it would have shown when presentation would have been late rather than early. These results align with previous research [Choudhary, Markley, and Schlotzer 2009] conducted on the US market, showing that earlier SEC form 10-K’s timing are associated with a decrease in measures of information asymmetry which is observed around the publication date and may lead to a decrease in returns. Combining the two conclusions suggests an interesting connection between information asymmetry and deadline proximity. When referring to share performance in relation to the other share returns made the same day (standardized measure) the correlation between deadline proximity and returns has twice the magnitude, as illustrated in Table 7. This means that shares where their corresponding reports were filed early (late) proved more likely to outperform (under-perform) the average of the entire stocks on the market within the day of publication. The magnitude of this effect drops as the time of publication moves towards the deadline date. The heightened magnitude measured over the standardized values (compared with the unstandardized values) suggests that measuring the deadline proximity effect should check for additional variables influencing the entire population of public firms in order to filter their effects. These conclusions contribute to the general understanding of financial reports and investors’ reaction, by benchmarking TASE as a stock exchange of smaller proportions and a less regulated environment than the US based stock exchanges upon which most of the relevant research has been conducted.

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

Observing the effect of deadline proximity over stock reaction on the TASE reinforces the similar results on US based exchanges and provides evidence as to the global nature of the phenomenon described.

Recommendations for future research Several researchers have presented a different approach towards the topic of this paper Menike and Man [1995] has researched whether the reports’ reflection on stock performance is industry related and he performed an industry based segmentation for measuring the report publication impact over performance whilst checking a specific industry. This might be an interesting direction to follow, thus taking both deadline proximity and industry into consideration. An industry based approach can be also supported by recent research [Dzikowska and Jankowska 2012], which shows particular industries to be prone to greater influence in times of crisis. A heterogeneous division of influence (and risk) between public firms may also lead to the strategic management of publications, and in turn prove useful as a proxy for return changes. Research conducted at the Karachi Stock Exchange by [Iqbal and Farooqi 2011] found no abnormal return in the period after the time of publication (such as earnings’ announcements or the filing of financial reports). The research did not make a distinction between the deadline proximity of different publications, but rather treated those as a fixed variable. Reviewing the Karachi Stock Exchange data in the light of the deadline proximity method may shift their conclusion of no filing to stock correlation and supply additional weight to the method presented here. Previous research [Hirshleifer, Lim, and Teoh 2009] conducted on the US market indicated that investors react less to earnings’ announcements when they face a large number of competing announcements on that day. Although earnings’ announcements are not bound by regulatory deadlines, studying the relationship between the timing of the announcements and investors’ reaction in conjunction with proximity to other fixed dates (quarter end, fiscal years) may produce interesting observations.

References Abarbanell, J.S., 1991, Do Analysts’ Earnings Forecasts Incorporate Information in Prior Stock Price Changes?, Journal of Accounting and Economics, no. 14(2): 147–165. Alford, A.W., Jones, J.J., Zmijewski, M.E., 1994, Extensions and Violations of the Statutory SEC Form 10-K Filing Requirements, Journal of Accounting and Economics, no. 17: 229–254. Amihud, Y., Hauser, S., Kirsh, A., 2003, Allocations, Adverse Selection, and Cascades in IPOs: Evidence from the Tel Aviv Stock Exchange, Journal of Financial Economics, no. 68(1): 137–158.

Economics and Business Review, Vol. 1(15), No. 4, 2015

Anderson, K.L., Yohn, T.L., 2002, The Effect of 10K Restatements on Firm Value, Information Asymmetries, and Investors Reliance on Earnings, Information Asymmetries, and investors Reliance on Earnings, September. Antweiler, W., Frank, M.Z., 2006, Do US Stock Markets Typically Overreact to Corporate News Stories?, available at SSRN 878091. Asthana, S., Balsam, S., 2001, The Effect of EDGAR on the Market Reaction to 10K Filings, Journal of Accounting and Public Policy, no. 20: 349–372. Asthana, S., Balsam, S., Sankaraguruswamy, S., 2004, Differential Response to Small versus Large Investors to 10-K Filings on EDGAR, The Accounting Review, no. 79: 571–589. Bagnoli, M., Kross, W., Watts, S.G., 2002, The Information in Management’s Expected Earnings Report Date: A Day Late, Penny Short, Journal of Accounting Research, no. 40: 1275–96. Ball, R., Shivakumar, L., 2008, How Much New Information Is There in Earnings?, Journal of Accounting Research, no. 46(5): 975–1016. Baumeister, F., Bratslavsky, E., Finkenauer, C., Vohs, K.D., 2001, Bad Is Stronger than Good, Review of General Psychology, no. 5, 323–370. Będowska-Sójka, B., 2014, Intraday Stealth Trading. Evidence from the Warsaw Stock Exchange, Poznań University of Economics Review, vol. 14, no. 1: 5–19. Bernard, V.L., Thomas, J.K., 1989, Post-earnings-announcement Drift: Delayed Price Response or Risk Premium, Journal of Accounting Research, no. 27: 1–36. Biddle, G.C., Hilary, G., 2006, Accounting Quality and Firm-level Capital Investment, The Accounting Review, no. 81(5): 963–982. Brown, S.J., Warner, J.B., 1980, Measuring Security Price Performance, Journal of Financial Economics, no. 8(3): 205–258. Campbell, C.J., Wasley, C.E., 1996, Measuring Abnormal Daily Trading Volume for Samples of NYSE/ASE and NASDAQ Securities Using Parametric and Nonparametric Test Statistics, Review of Quantitative Finance and Accounting, no. 6 (3): 309–326. Carhart, M.M., Kaniel, R., Musto, D.K., Reed, A.V., 2002, Leaning for the Tape: Evidence of Gaming Behavior in Equity Mutual Funds, Journal of Finance, no. 57 (2): 661–693. Carter, M.E., Soo, B.S., 1999, The Relevance of Form 8-K Reports, Journal of Accounting Research, no. 37 (1): 119–132. Choudhary, P., Merkley, K., Schloetzer, J., 2009, Timeliness of Form 10-K, Firm Financial Performance and Information Uncertainty, Working Paper. Corrado, C.J., 1989, A Nonparametric Test for Abnormal Security-price Performance in Event Studies, Journal of Financial Economics, no. 23 (2): 385–395. Das, S.R., Chen, M.Y., 2007, Yahoo! for Amazon: Sentiment Extraction from Small Talk on the Web, Management Science, no. 53(9): 1375–1388. Dontoh, A., Ronen, J., 1993, Information Content of Accounting Announcements, The Accounting Review 68 (4): 857–869. Dzikowska, M., Jankowska, B., 2012, The Global Financial Crisis of 2008–2009 and the Fortune Global 500 Corporations. Looking for Losers among the Biggest-exploratory Study, Poznań University of Economics Review, vol. 12, no. 3: 99–124. Fama, E.F., 1970, Efficient Capital Markets: A Review of Theory and Empirical Work, The Journal of Finance, no. 25.2: 383–417. Griffin, P., 2003, Got Information? Investor Response to Form 10-K and Form 10-Q EDGAR Filings, Review of Accounting Studies, no. 8: 433–460.

I. Blitzer, Measuring the usefulness of information publication time to proxy for returns

Hirschey, M., Palmrose, Z.V., Scholz, S., 2003, Long-term Market Underreaction to Accounting Restatements, University of Kansas, Lawrence, KS. Hirshleifer, D., Lim, S.S., Teoh, S.H., 2009, Driven to Distraction: Extraneous Events and Underreaction to Earnings News, The Journal of Finance, no. 64(5): 2289–2325. Hranaiova, J., Byers, S.L., 2007, Changes in Market Responses to Financial Statement Restatement Announcements in the Sarbanes-Oxley Era, available at SSRN 1319354. Iqbal, J., Farooqi, F.A., 2011, Stock Price Reaction to Earnings Announcement: The Case of an Emerging Market, available at MPRA: 30865. Kahneman, D., Knetsch, J.L., Thaler, R.H., 2008, The Endowment Effect: Evidence of Losses Valued More than Gains, Handbook of Experimental Economics Results, no. 1, 939–948. Kalay, A., Wei, L., Wohl, A., 2002, Continuous Trading or Call Auctions: Revealed Preferences of Investors at the Tel Aviv Stock Exchange, The Journal of Finance, no. 57(1): 523–542. Katz, S.B., 2001, Language and Persuasion in Biotechnology Communication with the Public: How not to Say What You’re not Going to Say and not Say It, AgBioForum, no. 4 (2): 93–97. Kloptchenko, A., Eklund, T., Karlsson, J., Back, B., Vanharanta, H., Visa, A., 2004, Combining Data and Text Mining Techniques for Analysing Financial Reports, Intelligent Systems in Accounting, Finance and Management, no. 12(1): 29–41. Kraft, A.G., Vashishtha, R., Venkatachalam, M., 2014, Real Effects of Frequent Financial Reporting, available at SSRN 2456765. Lakonishok, J., Shleifer, A., Thaler, R., Vishny, R., 1991, Window Dressing by Pension Fund Managers, American Economic Review, no. 81 (2): 227–231. Leuz, C., Wysocki, P., 2008, Economic Consequences of Financial Reporting and Disclosure Regulation: A Review and Suggestions for Future Research, Working Paper, University of Chicago and Massachusetts Institute of Technology. Li, E., Ramesh, K., 2009, Market Reaction Surrounding the Filing of Periodic SEC Reports, The Accounting Review, no. 84 (4): 1171–1208. Li, F., 2006, Do Stock Market Investors Understand the Risk Sentiment of Corporate Annual Reports?, SSRN Electronic Journal, 898181. MacKinlay, A.C., 1997, Event Studies in Economics and Finance, Journal of Economic Literature: 13–39. Meier, I., Schaumburg, E., 2006, Do Funds Window Dress? Evidence for U.S. Domestic Equity Mutual Funds, Working Paper, University of Montreal and Northwestern University. Menike, M.G. P.D., Man, W., 1995, Stock Market Reactions to the Release of Annual Financial Statements Case of the Banking Industry in Sri Lanka, European Journal of Business and Management, no. 5(31): 75–86. Morey, M.R., O’Neal, E., 2006, Window Dressing in Bond Mutual Funds, Journal of Financial Research, no. 29 (3): 325–347. Morris, M.W., Sheldon, O.J., Ames, D.R., Young, M.J., 2005, Metaphor in Stock Market Commentary: Consequences and Preconditions of Agentic Descriptions of Price trends, Working Paper, Columbia University, Cornell University, and University of California, Los Angeles.

Economics and Business Review, Vol. 1(15), No. 4, 2015

Musto, D.K., 1999, Investment Decisions Depend on Portfolio Disclosures, Journal of Finance, no. 54 (3): 935–952. Piotroski, J.D., 2000, Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers, Journal of Accounting Research: 1–41. Qi, D., Woody, W., Haw, I., 2000, The Incremental Information Content of SEC 10-K Reports Filed under the EDGAR System, Journal of Accounting Auditing and Finance, no. 15: 25–46. Tetlock, P.C., Saar-tsechansky, M., Macskassy, S., 2007, More than Words?: Quantifying Language to Measure Firms’ Fundamentals, The Journal of Finance, no. 63(3): 1437–1467.

Economics and Business Review, Vol. 1 (15), No. 4, 2015: 69–83 DOI: 10.18559/ebr.2015.4.5

Business tendency survey data. Where do the respondents’ opinions come from1? Sławomir Kalinowski,2 Małgorzata Kokocińska2

Abstract : The article deals with the topic of business tendency survey data. The aim of the article was to research whether the respondents’ opinions reflect the variability of factors influencing the economic situation of enterprises. For the research the business tendency surveys of Central Statistical Office of Poland in 2000–2015 for manufacturing industry have been used. Based on the outcome of earlier research a general hypothesis has been assumed which states that the opinions of enterprises’3 containing qualitative variables are substantially influenced by quantitative variables which are based on the economic situation of businesses. The research has been done using the methods of time series analysis (Census II and Hodrick-Prescott filter) and the interdependence analysis method (Pearson correlation coefficient with time lags and Granger causality test). Keywords : business fluctuations, business tendency survey, qualitative data, business performance measures. JEL codes : E320.

Introduction A business tendency survey is considered to be a tool to examine the situation of the economy on the basis of the rational expectations concept [Muth 1961]. According to this there is a relationship between the variability of economic reality with the opinions and approaches of business entities being the most important subjects of this reality. Hence the interest in business tendency surveys and their relationship with the variability of the economic dynamics is increasing. In Great Britain the first surveys were conducted amongst the companies of the Confederation of British Industry. They started in 1958 [Klein and 1

Article received 29 December 2014, accepted 15 September 2015. Poznań University of Economics, Department of Microeconomics, al. Niepodległości 10, 61-875 Poznań, Poland; corresponding author, slawomir.kalinowski@ue.poznan.pl. 3 Those answering questions in the Central Statistical Office’s business tendency surveys are managers or entrepreneurs. In the article they are defined as respondents. 2

Economics and Business Review, Vol. 1(15), No. 4, 2015

Moore 1981a]. The first business tendency surveys in Germany were carried out in 1965 [Striegel 1965]. They combined the opinions of enterprises from the manufacturing sector and consumers. In the following years the research included other sectors of the economy. Similar research started in Poland in the Institute of Economic Development of the Warsaw School of Economics in 1993. Basing their results on two independent attempts were made simultaneously to build a synthetic business tendency indicator [Stanek 1993; Matkowski 1993] according to the methodology used in the EU. The other institutions in Poland which conducted or still carry out business tendency surveys are the Central Statistical Office, IpsosDemoskop and The Institute of Research on Market Economy [Matkowski 2002]. It is worth mentioning that the Poznan University of Economics was also an important research centre as far as the application of qualitative variables as elements of the leading indicators forecasting economic growth [Rekowski 2003]. The objects of lasting interest were the relationships between the results of the surveys and the variability of quantitative data describing the situation of the economy. If the results of the surveys are intended to forecast the economic situation we have to be sure that the opinions of managers and consumers reflect actual economic processes. The paper is put together as follows. Section 1 reviews the literature on the informational capacity of business tendency surveys’ qualitative data. Section 2 presents the methodology and the data used. Empirical results are presented and discussed in Section 3. Chapter 4 was devoted to the regression analysis verifying the variables selection. Conclusions are drawn in the next Section.

1. Results of previous research The research into the influence of quantitative variables on the qualitative variables from business tendency surveys has been done ever since the latter have been used to forecast the fluctuations of economic dynamics. One of the first was the study by Jochems and De Wit [1959]. Its results confirmed the usefulness of qualitative data to forecast economic dynamics. In another early research Theil used the regression approach in order to compare the qualitative data from the „Munich Business Test” with quantitative data [1952]. The outcome was also positive for the prognostic capacity of qualitative variables. An important thorough study comparing qualitative with quantitative data was the research by Klein and Moore published in two articles[1981a, 1981b]. They tested the outcome of business tendency surveys carried out amongst the companies incorporated in the Confederation of British Industry. In the first article they showed the research results into the coordination of business ten-

S. Kalinowski, M. Kokocińska, Business tendency survey data

dency cycles in the UK economy and the results of the surveys concerning the volume of new orders. Using the method of turning points they carried out research which enabled them to conclude from the evidence presented that surveys of respondents’ views on new orders can assist greatly in assessing current economic developments. All of the companies that they examined conformed closely to the growth cycles in the UK economy and to quantitative data on the volume of new orders. In the second article Klein and Moore examined the qualitative data on inventories, profits and business confidence expressed in the answers to the question: „are you more, or less, optimistic than you were four months ago about the general business situation in your industry”?4 The evidence in the second study demonstrated that in every case the net balances from the survey, whether in original or cumulated form, conform closely to growth cycles in the UK. Another important research carried out at the time was a study by Carlson and Parkin. It created one of the standards of the method of research into the interrelations of qualitative and quantitative data [1975]. It is based on the probabilistic approach. It was used by Müller whilst examining the information content of qualitative survey data [2009]. This study was based on data analysis from particular companies. The author, contrary to other approaches, did not use the data aggregated for the whole sector. He compared qualitative and quantitative data for each company separately. The main research comprised almost 13 thousand monthly observations in the period of 04.2005–11.2006. „The analysis of firm-level data on quantitative and qualitative questions presented in this paper allows more direct tests of the rationality assumption. It has shown that survey responses are very reliable. This outcome can be taken as strong support for the basic assumption of quantification methods applied to qualitative survey data” [Müller 2009, p. 9]. An example of research showing the predictive ability of qualitative data for the Polish manufacturing sector are the studies carried out by Research Institute for Economic Development [Adamowicz, Dudek, and Walczyk, 2002; Adamowicz, Klimkowska, and Walczyk 2011]. The outcome of the studies was the acceptance of the usefulness of quantitative variables for the short term forecast of the production sold by industry. In particular the RIED business indicator (general index elaborated by the Institute) and „expected production” have revealed their prediction power. All the research presented has indicated the substantial prediction power of qualitative data from business performance surveys. This study concentrates on how real processes in companies and their vicinity are reflected in the respondents’ replies. The other distinctive feature of this research the consideration of a much wider spectrum of quantitative variables. 4

These issues correspond with qualitative variables I, OM, GES i FES from the study presented here.

Economics and Business Review, Vol. 1(15), No. 4, 2015

2. Qualitative and quantitative variables. Research hypotheses The period of the analysis is 2000–2015. All the variables used in the research are based on quarterly data. Such a uniform time period has been used because of the lack of accessible monthly quantitative data. The results of business tendency surveys were calculated as arithmetic means from the three monthly data respectively. Similarly monthly data on the employment in manufacturing industry were adjusted. Quarterly indices of prices of sold production were calculated as a geometric mean from the respective monthly data. All the variables expressed in monetary units were changed into a series of fixed prices from the first quarter of 2000. All the time series of qualitative and quantitative variables were subjected to seasonal adjustment with the help of Census II with correction due to unusual events. The series of variables under the influence of long term trends were checked using the Hodrick-Prescotta filter [Hodrick and Prescott 1997]. It referred to GDP, the sold production of industry and manufacturing industry employment. In the last phase the indices of variables were calculated dividing the value from a given quarter by the value from the analogical quarter of the previous year. Six qualitative variables were examined. They come from business tendency surveys conducted by Polish Central Statistical Office: –– GES – current, general economic situation of a company, –– FES – forecast economic situation of a company, –– I – current level of inventories of a company, –– R – level of receivables of a company, –– P – current ability of a company to pay its liabilities, –– FP – future ability of a company to pay its liabilities. The respondents can choose from one of three replies to answer the question about the current situation of a company (GES): a) good, b) satisfactory, c) bad. In the case of the question about the predicted situation of a company (FES) the respondents should consider a three-month-perspective. They can choose from one of the three answers: a) it will improve, b) it will be the same, c) it will worsen. The values of these qualitative variables are calculated as the difference between a relative number of a. answers and the relative number of c. answers. The number of b. answers is omitted [Walkowska 2011]. The original time series of the variables GES and FES, in order to eliminate the calculation problems with indices of negative values, have been linearly transformed by adding 100%. The basis for the variable I value was the question from the business tendency sur-

S. Kalinowski, M. Kokocińska, Business tendency survey data

vey about inventories. It says: “what is the current level of available stocks in your company?”. The respondents can choose from one of the following replies: a) too high, b) appropriate to demand, c) too low. The values of the qualitative variable I are calculated based on the indices equal to the difference between the relative number of the c. replies and the relative number of the a. replies. In conformity with the intention of the survey the a. answer is supposed to indicate the deterioration of the economic situation resulting in the lower demand for the products of a company. If we adopt such an interpretation we have to exclude the possibility of choosing the first answer in a situation of stock mismanagement, irrespective of the business tendency. In other words we reject instances in which the stocks are too high because the management of companies have overestimated the growth in demand for their products. The question from the survey concerning the variable R reads: „how is the level of your general receivables in your company changing”? Respondents can choose from one of the following answers: a) it is increasing, b) it remains constant, c) it is decreasing. The way in which the questions and the replies are formulated does not make it possible to reflect the influence of economic fluctuations on the level of a company’s receivables. The fact that the term “general receivables” does not exist in the terminology of financial reports can be ignored. Unfortunately we cannot accept the mistake in the question about the level of receivables nor about the promptness of their collection. In general the level of receivables increases if revenue increases and it decreases if the sales are falling. The improvement in the business tendency brings about the increase of sales revenues resulting in the increase of receivables. The index “the level of receivables of a company” (dependent upon the R variable) published in the business tendency survey results, is counted as a difference between the percentage share of the respondents who chose the c. answer and those who chose the a. answer. So its positive value means that those whose receivables decreased outnumber those with a rising level. The question should deal with the promptness or waiting time for the incoming payment. Then no problems of interpretation will arise. The issue of liabilities on deliveries is dealt with in two questions. One is about the current situation, the other about the future. The opinion on the current situation is expressed by the answers to the following question: “how is your company’s ability to pay financial liabilities on an ongoing basis changing?”: –– it is improving, –– it is the same, –– it is getting worse.

Economics and Business Review, Vol. 1(15), No. 4, 2015

The index „the ability to pay financial liabilities on an ongoing basis” used as the basis to calculate the P variable, is calculated as the difference between the percentage share of respondents choosing a. answer and the respondents choosing c. answer. This reflects correctly the way in which economic fluctuations can influence a company’s ability to meet its liabilities. The positive value indicates economic acceleration and the negative means a slowdown. The other factor from the indices examining the opinions of respondents on the payment of liabilities, deals with forecast. It is the basis used to build the FP variable. Here the respondents answer the question: “how will your company’s ability to meet financial liabilities change in the following three months”? They can choose from the same set of answers as in the previous question. Also, as far as this element of the survey is concerned, the accuracy of the formulation and the ease of interpretation of the can be considered. The set of quantitative variables will contain the macroeconomic data of the Polish economy and the financial parameters of companies of the Polish manufacturing industry:5 –– GDP – business cycle of gross domestic product, –– OM – operating margin, –– IPS – business cycle of manufacturing industry production sold, –– IE – business cycle of manufacturing industry employment, –– IPI – manufacturing industry price index, –– IT – inventory turnover in days (average for manufacturing industry), –– RT – receivables collection period in days (average for manufacturing industry), –– PT – payables payment period in days (average for manufacturing industry). The procedure of using quantitative variables has already been presented. We have only to indicate that a substantial positive influence is expected of the quantitative variables GDP, IPS, IE and IPI on the qualitative variables GES and FES. The issues connected with the definition of the applied financial indicators should be explained in more detail. The first is the operating margin (OM). In order to avoid the possibility of calculating the changes from negative values, the values of the indicator were increased by one:

OM =

S EBIT = 1+ , OC OC

where: OM – operating margin, 5 The data source was two publications by the Central Statistical Office: Biuletyn Statystyczny [Statistical Bulletin] and Wyniki finansowe podmiotów gospodarczych [Financial Performance of Enterprises].

S. Kalinowski, M. Kokocińska, Business tendency survey data

S – sales revenues, OC – total operating costs, EBIT – earnings before interest and taxes. The bigger the value of the variable OM the better the economic situation of the company. This variable should positively influence the qualitative variable GES. The second is the inventory turnover in days. It has been defined as: IT =

I ⋅ 90, OC

where: IT – inventory turnover in days, I – average level of inventory (the arithmetic mean from the beginning and end of a quarter), OC – quarterly total operating costs. The increasing value of the IT indicator means rising problems with sales. Its increase can mean economic slowdown. One can expect a negative correlation between the value of IT and the values of qualitative variables GES and I. The third financial indicator whose volatility should influence the opinion on the economic situation of the company is the receivables collection period in days: RT =

R ⋅ 90, S

where: RT – receivables collection period in days, R – average level of receivables (arithmetic mean from the beginning and end of the quarter), S – quarterly revenues from sales. The higher the value of the indicator, the longer the time between the moment of sales and the time of collection of the receivables due for this sale. Normally the extension of this period means problems with cash availability in the time of economic downturn. It is possible to formulate a hypothesis about a negative relationship between the values of the RT variable and the value of the qualitative variable GES. Due to the incorrect formulation of the question about receivables it is difficult to formulate a hypothesis about the factual relationship between the qualitative variable R and the quantitative variable RT. The prevailing fall in the level of receivables in companies cannot be linked with a shortening of the period of collection of receivables. If sales revenues fall faster than receivables the time of the collection is longer. Such a situation often occurs in a situation of the economic slowdown.

Economics and Business Review, Vol. 1(15), No. 4, 2015

According to the general rule used by the authors of the research, the rise of the qualitative indicator means an improvement in the business tendency. It should then expect a negative relationship between the variable RT and the values of the qualitative variables GES and R. A quantitative variable dealing with the ability to meet liabilities is the payables payment period in days PT =

P ⋅ 90, OC

where: PT – payables payment period in days, P – average level of short-term liabilities on deliveries and services (arithmetic mean from the beginning and end of a quarter). In periods of economic acceleration one should expect a decrease in the value of this indicator, in times of economic slowdown, the payment period of liabilities will be longer. The deciding factor is the change in cash accessibility which strongly depends on economic dynamics. The research hypothesis in the area of the reflection of the actual situation in respondents’ replies about the payables payment ability, shows the negative correlation between the quantitative variable PT and the values of the qualitative variables GES, P and FP. Table 1. Expected signs of Pearson’s correlation coefficients Variables

GDP

IPS

IPI

GES

–

FES

–

The summary of the considerations about the relationships between quantitative variables (columns) and qualitative (lines) is included in Table 1. It is necessary to emphasise that the signs in line R were selected based on the general rules for the formulation of survey questions and the calculation of qualitative indices, consisting of a positive relationship between economic dynamics and the survey results. It is worth mentioning that the expected signs of correlation inside the set of qualitative variables are positive because of the uniformity of the relationships with quantitative variables.

S. Kalinowski, M. Kokocińska, Business tendency survey data

3. Empirical research results The research was carried out using by two methods. Firstly, the cross correlation coefficients for qualitative and quantitative variables were calculated. These values were considered statistically significant, in which case the probability of accepting the zero hypothesis about the lack of correlation is smaller than the significance level α = 0.05. Such instances are marked in red in the tables. In the second stage of the research the Granger causality test [Granger, 1969] was used to find the reasons for the different answers to a survey question. Here the level of significance equal α = 0.05 was also taken. Table 2 shows the correlation coefficients of the variable GES and the quantitative variables applied with no time shifts and with 1 to 4 quarterly time lags. Time lags mean that quantitative precede qualitative data. For example a two quarter time lag means the relationship between OM from the first quarter and the GES from the third quarter of the same year. The values r from the first line indicate the accordance with the assumed hypotheses. In the case of variables GDP, OM and IPS the correlation is strong. For the variables IE, and RT it is moderate and weak in the rest of the cases.6 However it is worth underlining that for the lags in 1–4 quarters the correlation of the variables GES and IT rises to moderate. In almost each instance the correlation is statistically significant. Table 2. Correlation coefficients r between qualitative variable GES and quantitative variables with quarterly lags Lags

GDP

IPS

0.6808

0.6503

0.6605

0.5295

0.3528 –0.3341 –0.5190 –0.2595

0.5567

0.7404

0.6459

0.3290

0.2823 –0.5014 –0.5620 –0.2125

0.3263

0.7259

0.4737

0.0858

0.1639 –0.5745 –0.4462 –0.1513

0.0316

0.6221

0.2027 –0.1435

0.0216 –0.5561 –0.2490 –0.0868

–0.2757

rwas ranking

IPI

0.4595 –0.0794 –0.3184 –0.1316 –0.4861 –0.0653 –0.0382 1

The lag of the series of quantitative variables by one quarter has strengthened the power of the correlation relationship in case of the variables OM, IT and RT. Unfortunately a weaker relationship has moved the variable IE to the class of weak correlations. Almost all the correlations of the variable GES with 6

The classification of the correlation intensity was taken after: [Green, Salkind, and Akey 2000].

Economics and Business Review, Vol. 1(15), No. 4, 2015

the quantitative variables lagged by one quarter were statistically significant, excluding PT. The analysis of the further lags up to four quarters has demonstrated a strong or moderate correlation of the opinion on “the current general economic situation of a company” (GES) with the operating margin (OM) and inventory turnover (IT). Moreover the variable GES is strongly correlated with the changes in production sold of manufacturing industry (IPS) from the previous quarter. Significant moderate correlations for the lags by two quarters occur also in the case of the variables RT, IPS and IT. In the last case the statistical significance was maintained for the lag by three quarters. For the general evaluation of the potential influence of quantitative variables on respondents’ opinions about the current situation of their companies a mean weighted coefficient of statistically significant correlations was built: rwas = 0.4rt −1 + 0.3rt −2 + 0.2rt −3 + 0.1rt −4 ,

where: rwas – mean weighted coefficient of statistically significant correlations, rt–i – s tatistically significant correlation coefficient with a quantitative variable lagged by i quarters (with a sign compatible with the hypothesis formulated previously). Thus the calculated weighted means enabled the creation of a ranking of the significance of quantitative variables for the respondents’ opinions shown in the last line of Table 2. The variables OM, IT, RT and IPS have potentially the most important significance for the respondents’ opinion about their current financial situation. Variables IPI, IE and PT are least reflected in these opinions. The differences in the significance of variables connected with current asset management meant the need to compare the quantitative variables IT, RT and PT with their corresponding qualitative variables I, R, P and FP. Inventory turnover was very significant in the opinion about the current economic situation of companies (GES). The collection period of receivables (RT) was slightly less significant. In case of the time of payables payment (PT) the relationship was marginal. Based on these observations and the general rule of question formulation in survey research, it is possible to form a hypothesis about a negative correlation between the pairs of variables IT and T, RT and R, PT and P also PT and FP. The strongest relationship can be expected in case of inventories, whereas it is moderate in case of receivables and the weakest in case of delivery liabilities. The hypothesis about the lack of negative correlation between the variables I and IT can be rejected for all the quarterly lags. Respondents take into account the changes in inventory turnover in the last four quarters whilst answering the question about inventory. At the highest level it refers to the quantitative variable lagged by one and two quarters.

S. Kalinowski, M. Kokocińska, Business tendency survey data

Table 3. Correlation coefficients r between qualitative variables I, R, P, FP and corresponding quantitative variables with quarterly lags Lags

I and IT

R and RT

P and PT

FP and PT

–0.4772

0.4767

–0.0323

–0.0973

–0.6454

0.4116

–0.0323

–0.1067

–0.6527

0.2392

–0.0004

–0.0808

–0.5605

0.0409

0.0517

–0.0127

–0.4378

–0.1210

0.0940

0.0622

Unfortunately the authors’ reservations concerning the formulation of the questions about receivables were fulfilled in the contrary to the expected correlation sign of the variables R and RT. Although the variable RT lagged by one quarter was moderately correlated with the variable R in a statistically significant way, the relationship, contrary to the intention of the authors of the survey, was positive. The negative correlation in the relationship of the quantitative variable RT with the variables describing the current economic situation of companies (GES) led us to expect a negative correlation of the variables R and RT, similarly in the case of the qualitative and quantitative variables describing the inventories. The correlation between the quantitative variable PT and the qualitative variables P and FP to the lags of two quarters, shows compatibility with the hypothesis formulated earlier. Unfortunately the strength of the relationship is small or demonstrating the lack of correlation. We can make no statement about the statistical significance. One can suppose that the opinions concerning the ability to meet liabilities is not associated with its promptness. In other words managers and entrepreneurs do not associate the actual lengthening of the liabilities payment time with a worsening of the potential ability in this area. The relationships between the quantitative and qualitative variables have been also examined by the Granger causality test. A one-sided zero hypothesis has been analysed, which shows that the chosen quantitative variables are not the cause of quantitative variables in the Granger sense. I In this research the level of significance was assumed to be α = 0.05. The analysis of causality in the Granger sense confirmed the statistically significant influence of the variables OM, IT, IPS, RT and IE on the respondents’ opinions about the general current economic situation of their companies (GES). Except for the receivables turnover and industry production sold, the causality of the listed quantitative variables showed one quarterly lag. A new phenomenon was a statistically significant causation of cyclic changes in gross domestic product (GDP) for the lags of 2, 3 and 4 quarters. This factor in the Granger causality test turned out to have the biggest influence on the respond-

Economics and Business Review, Vol. 1(15), No. 4, 2015

Table 4. Following variables do not Granger cause GES Lags

GDP

IPS

F-statistics

p-value

F-statistics

p-value

F-statistics

p-value

F-statistics

p-value

2.4882

0.1207

14.2434

0.0004

0.4704

0.4958

16.7966

0.0001

14.7178

0.0000

1.3532

0.2677

5.7385

0.0057

0.4260

0.6554

13.3584

0.0000

1.8468

0.1516

2.2225

0.0979

1.2589

0.2993

9.1954

0.0000

0.9721

0.4324

1.9754

0.1149

1.1476

0.3469

Lags

IPI

F-statistics

p-value

F-statistics

p-value

F-statistics

p-value

F-statistics

p-value

18.9200

0.0001

2.6112

0.1120

0.4834

0.4899

1.2962

0.2600

0.3183

0.7288

4.2300

0.0201

0.6273

0.5382

0.5251

0.5947

3.9210

0.0141

1.9485

0.1347

0.8188

0.4900

1.5002

0.2267

1.5458

0.2057

1.1798

0.3328

0.3178

0.8645

1.6528

0.1781

Lags

FES

F-statistics

p-value

F-statistics

p-value

F-statistics

p-value

F-statistics

p-value

1.9694

0.1663

4.9061

0.0311

28.1153

0.0000

33.9372

0.0000

4.5114

0.0158

3.2565

0.0468

3.9927

0.0246

2.2871

0.1121

1.7600

0.1677

2.1770

0.1032

4.0481

0.0122

6.0859

0.0014

1.4072

0.2473

1.6736

0.1732

1.9049

0.1265

2.8869

0.0331

ents’ evaluations, who appeared to be strongly under the influence of announcements about the macroeconomic situation. The Granger causality test did not allow a rejection of the hypothesis about the lack of influence of PT and IPI. The time of liabilities payment and the index of prices of sold production were characterized by weak results as well as in the correlation analysis. To check the coherence of the qualitative variables the causality of qualitative variables I, R, P and FES for the variable GES were examined. It became apparent that the opinion about the level of inventory from two quarters ago, the opinion about the ability to meet liabilities from a quarter or two ago was the cause, in the Granger sense, of the changeability of opinions about the current economic situation of companies. One cannot say the same about the opinions on the level of receivables. Again this is due to the incorrect formulation of the survey question. Between the variable GES and the variables I and P a statistically significant positive correlation was noted ( rI = 0.6917 and rP = 0.7981 respectively). The cohesion of the answers to the survey questions is very high. At the same time the same relationship for the variable R, although statistically significant, has a negative sign and only a moderate power (rR = –0.5612). All

S. Kalinowski, M. Kokocińska, Business tendency survey data

the other qualitative variables are characterized by a statistically significant positive correlation. The Granger causality test revealed the coherence of the answers of the respondents to the questions about the economic situation of companies. It turned out that the respondents’ opinions about the future economic situation of companies in the coming three months (FES) cause the changing of opinions about the actual situation in the next quarter (GES). The hypothesis about the lack of a correlation between the underlying quantitative variable describing the level of companies’ economic activity (IPS) and the quantitative variables linked by net working capital (IT, RT and PT) can be rejected. In each case the correlation was statistically significant and had a negative sign, compatible with the theoretical expectation (rI = –0.4320, rR = –0.6591 and rP = –0.2991 respectively). The lack of a relationship between the quantitative variables could not be the reason for the lack of importance of the variable PT on the opinions about the current and future economic situation of enterprises.

4. Regression analysis In order to check the explanatory power of the quantitative variables indicated as far as the qualitative dependent variable GES is concerned two regression functions were estimated. First with the explanatory variables chosen using the correlation method and the second using the outcomes of the Granger causality tests. The explanatory variables that were correlated or appeared in the regression equation with the opposite sign from the theoretical point of view were excluded. The correlation analysis indicated all quantitative variables with the exception of PT. For almost every variable the highest correlation coefficient was for one quarter time lag. Only for inventory turnover during two quarters was the time shift taken into account. The result was: GESt = –7.9880 + 2.7857GDPt–1 + 6.2180OMt–1 + ε. The determination coefficient R2 = 0.7349. The value of the measure is satisfying. All the explanatory variables’ parameters are statistically important at the significance level α = 0.001. The Granger causality test excluded the following time series from the set of possible explanatory variables: PT and IPI. He variables OM, IE, IT were taken with a one quarter time lag and the variables GDP, IPS and RT with two quarters. The model is: GESt = –7.6182 + 1.2298IEt–1 + 7.3999OMt–1 + ε.

Economics and Business Review, Vol. 1(15), No. 4, 2015

The determination coefficient is also satisfying (R2 = 0.6559) All the explanatory variables’ parameters are statistically important at the significance level α = 0.001. In both models the operating margin is the main explanatory variable. It is coherent with the correlation ranking outcomes. The Granger test revealed the importance of industry employment, which was confirmed by the presence of the variable in the second model.

Conclusions The study conducted has shown that managers and entrepreneurs generally reflect the changeability of the quantitative data describing the financial situation of their companies in their opinions. The Granger causality test has revealed an exceptionally strong influence of the information about the changes in GDP on their opinions concerning their company performance. The role of the variable is also underlined by the first regression model. A strong influence of the exogenous quantitative variable which is forecast on the basis of qualitative variables raises a question if we deal with a depending loop. The volatility of GDP strongly influences opinions about the economic situation of companies, which in turn is used to predict economic dynamics. The research identified endogenous quantitative variables which, in a statistically significant way, influence respondents’ opinions about the current and future economic situation of their companies. The analysis of correlations with time lags has revealed the influence of operational profitability, inventory turnover, the receivables collection period and manufacturing industry production sold. Industry employment, the price index of sold production and turnover period of payables were characterised by a weaker relation. The Granger causality test has reduced to five the number of endogenous quantitative variables having influence on the evaluation of companies’ current economic situation. These variables were: operational profitability, industry production sold, manufacturing industry employment, inventory and receivables turnover. The regression analysis revealed the exceptional explanatory power of the operating margin. This quantitative variable appeared in both models as the factor with the highest influence. The respondents, when asked about the general economic situation of their companies, mostly have in mind operational profitability.

S. Kalinowski, M. Kokocińska, Business tendency survey data

References Adamowicz, E., Dudek, S., Walczyk, K., 2002, Wykorzystanie metody testu koniunktury do prognoz krótkoterminowych, in: Diagnozy i prognozy stanu gospodarki w świetle wyników badań koniunktury, Prace i Materiały Instytutu Rozwoju Gospodarczego, nr 73, SGH, Warszawa: 6–28. Adamowicz, E., Klimkowska, J., Walczyk, K., 2011, Wahania koniunkturalne w Polsce, in: Zmiany aktywności gospodarczej w świetle wyników badań koniunktury, Prace i Materiały Instytutu Rozwoju Gospodarczego, nr 87, Warszawa: 9–32. Carlson, J., Parkin, M., 1975, Inflation Expectations, Economica, no. 42:166: 123–38. Granger, C.W.J., 1969, Investigating Causal Relations by Econometric Models and Crossspectral Methods, Econometrica, no. 37 (3): 424–438. Green, S., Salkind, N., Akey, T., 2000, Using SPSS for Windows: Analyzing and Understanding Data, Prentice Hall, New Jersey. Hodrick, R.E., Prescott, C., 1997, Postwar U.S. Business Cycles: an Empirical Investigation, Journal of Money, Credit, and Banking, no. 29 (1): 1–16. Jochems, D.R., De Wit, G.M., 1959, The Macro-Economic Relationship between Business Tendency Surveys and Numerical Data, Bericht des Okonometrischen Instituts der Nederlandsche Economische Hoogeschool, no. 5904. Klein, P.A., Moore, H.G., 1981a, Industrial Surveys in the UK: Part I New Order, Applied Economics, no. 13: 167–179. Klein, P.A., Moore, H.G., 1981b, Industrial Surveys in the UK: Part II Stocks Profits and Business Confidence Over Business Cycle, Applied Economics, no. 13: 465–480. Matkowski, Z., 1993, In Search for an Economic Sentiment Indicator for Poland: the RIED Approach and Preliminary Results, Workshop on Business Surveys in Transition Countries, Poznań, reprinted in Prace i materiały IRG SGH, vol. 36, Warszawa. Matkowski, Z., 2002, Wskaźniki klimatu gospodarczego jako narzędzie oceny stanu gos podarki, Ekonomista, nr 1, Warszawa: 9–40. Müller, Ch., 2009, The Information Content of Qualitative Survey Data, OECD Journal: Journal of Business Cycle Measurement and Analysis, vol. 1: 1–12. Muth, J., 1961, Rational Expectations and the Theory of Price Movements, Econometrica, no. 29: 315–335. Rekowski, M. (ed.), 2003, Wskaźniki wyprzedzające jako metoda prognozowania ko niunktury w Polsce, Wydawnictwo Akademii Ekonomicznej, Poznań. Stanek, K., 1993, Koncepcja barometru gospodarczego koniunktury, w: Badania koniun ktury gospodarki Polski, IRG SGH, Warszawa. Striegel, W.H., 1965, Die Konjunkturumfragen des IFO-Instituts für Wirtschaftsforschung, Allgemeines Statistisches Archiv, Nr. 2. Theil, H., 1952, On the Shape of Microvariables and the Munich Business Test, Revue de l’Institut International de Statistique, nr 20: 105–120. Walkowska, K. (ed.), 2011, Badanie koniunktury gospodarczej. Zeszyt metodologiczny zaopiniowany przez Komisję Metodologiczną GUS, Główny Urząd Statystyczny, Warszawa.

Economics and Business Review, Vol. 1 (15), No. 4, 2015: 84–101 DOI: 10.18559/ebr.2015.4.6

Does outward FDI by Polish multinationals support existing theory? Findings from a quantitative study1 Marian Gorynia,2 Jan Nowak,3 Piotr Trąpczyński,2 Radosław Wolniak4

Abstract : This study identifies and investigates the relationships between FDI motives and establishment modes in the context of outward foreign direct investment (OFDI) undertaken by companies from Poland. It is based on survey data collected in 2013 from Polish companies engaged in FDI. A number of hypotheses, derived from pertinent outward FDI literature, are tested using quantitative analysis tools. Its main findings show that almost half of the surveyed firms preceded their first FDI project with exports, but at the same time a similar proportion of firms had no host-country experience prior to their initial FDI project. The study indicates the leading role of market-seeking motives in Polish OFDI followed by efficiency-seeking and strategic assetseeking motives, with only a marginal role being left to resource-seeking. Keywords : outward foreign direct investment, FDI motives, FDI establishment modes, emerging multinationals, Central and Eastern Europe, Poland. JEL codes : F21, F23, P20.

Introduction Outward foreign direct investment (OFDI) is a phenomenon with important economic consequences. The investing firms can develop economies of scale and scope which lead to their increased profitability and market power. Secondly, investing firms can enhance their capabilities through asset acquisition abroad and the synergistic combination of the acquired assets with their own capabili 1

Article received 02 January 2015, accepted 15 September 2015. This study has been financed with the funds of the National Science Centre, granted with the decision no. DEC-2012/07/N/ HS4/00283. Piotr Trąpczyński is supported by the Foundation for Polish Science (FNP). 2 Poznań University of Economics, Faculty of International Business and Economics, Department of International Competitiveness, al. Niepodległości 10, 61-875 Poznań, Poland; corresponding author: piotr.trapczynski@ue.poznan.pl. 3 IBD Business School, Panieńska 9, 03-704 Warsaw, Poland. 4 University of Warsaw, Faculty of Economic Sciences, Department of Development Economics, Długa 44/50, 00-241 Warsaw, Poland.

M. Gorynia, J. Nowak, P. Trąpczyński, R. Wolniak, Does outward FDI…

ties, thus also possibly contributing to an upgrade of domestic firms [Buckley et al. 2010]. Moreover Visser [2006] argues that OFDI normally leads to a shift from lower to higher-skilled jobs at home. However the empirical evidence regarding the impact of OFDI on domestic investment and employment remains mixed [Gorynia et al. 2014]. Hence this topic is subject to debate by scholars and policy-makers alike. The volume of OFDI by firms originating from Central and Eastern European (CEE) economies has raised the question about number of issues related to the specific character of this category of firm, including their competitiveness [Jaworek, Szałucka, and Szóstek 2009], foreign expansion motives [Rosati and Wiliński 2003] or entry modes [Gorynia et al. 2013a]. Amongst CEE-related studies a macroeconomic perspective has prevailed [Antalóczy and Éltető 2003; Bohata and Zeplinerova 2003; Varblane, Reiljan, and Roolaht 2003; Svetličič and Jaklič 2003; Gorynia, Nowak, and Wolniak 2008, 2010; Kalotay 2004; Radlo and Sass 2012], although studies combining macro-level with firm-level analysis can also be found [Svetličič, Rojec, and Trtnik 2000; Rugraff 2010; Zemplinerová 2012; Gorynia et al. 2013a, b]. All of them point to the geographic concentration of CEE firms’ outward FDI in neighbouring European countries. Prominent amongst them are the reports on emerging multinationals from Hungary, Poland, Russia and Slovenia that have been published in recent years within the Emerging Market Global Players (EMGP) Project [Kaliszuk and Wancio 2013; Kuznetsov 2013; Jaklič and Svetličič 2009]. They recognize that the vast majority of these firms have become regional, rather than global, players and many of them remain state-controlled. They focus on Europe and Central Asia as their FDI destinations. Their transnationality index (TNI) is not high: it ranges on average from 30% for firms from Poland to 53% for firms from Hungary. Although these companies’ FDI motives and modes were not extensively covered by those reports certain patterns can nevertheless be discerned. The predominant FDI motive is market-seeking (even among Russian multinationals), followed by resource- and efficiency-seeking. Strategic-asset seeking is quite important only for Russian multinational enterprises (MNEs). The most common FDI mode is acquisition although there seems to be a growing importance of green-field investment. Zemplinerová [2012] concludes that Czech companies invest in other countries not only to exploit their firm-specific advantages but also to access new markets. In general, although investing companies may pursue multiple motives, market- and efficiency-seeking predominate. Svetličič, Rojec and Trtnik’s [2000] study reveals a predominance of market-seeking motives of Slovene OFDI, however their study also points to the emergence of OFDI in Slovenia as a result of lack of ownership advantages of local firms and their desire to improve competitiveness through FDI. Despite the existence of several studies devoted to FDI motives of emerging multinationals from the CEE region none of them systematically examines how these motives affect strategic choices. Whilst studies in the Asian con-

Economics and Business Review, Vol. 1(15), No. 4, 2015

text devoted to the relationship between FDI motives and the specific modes of entry have indicated that the choice of acquisitions is frequently related to the strategic asset-seeking motive [Cui and Jiang 2010; Yamakawa, Peng, and Deeds 2008] This relationship has not been examined across a broader set of different FDI motives. The present paper addresses this research gap by attempting to examine the relationship between different FDI motives and FDI establishment modes (understood as the dichotomous choice between acquisitions and greenfield investments). By explicitly accounting for the underlying motives of FDI projects and attempting to link them with the selected establishment modes this study contributes to the ongoing debate in international business research on the specific character of emerging multinational motivation for FDI. It also addresses a fundamental issue of the internationalization process of emerging multinationals by considering FDI projects in the context of the earlier foreign operations of the sample firms. The paper starts with a review of key literature on OFDI with focus on its motives and entry modes as well as the character of the internationalization trajectories of emerging multinationals. The ensuing sections introduce research hypotheses and present the results of their verification by using quantitative analysis. The final sections offer discussion about the relevance and implications of the findings, and finally, a summary of final conclusions and suggestions for future studies.

1. Firm internationalization, FDI motives and establishment modes: the theoretical background Any study of FDI motives and FDI modes should first be placed in a broader context of firm internationalization, addressing the question whether firms precede their direct investment abroad with non-equity entry modes, such as exporting, licensing, franchising or contract manufacturing. According to the Uppsala Model developed by Johanson and Wiedersheim-Paul [1975] and Johanson and Vahlne [1977, 1990], firm internationalization is a sequential and gradual process of increasing resource commitments in foreign markets. These authors postulated that internationalizing firms will first select foreign countries with market conditions and cultures similar to those of their home country and introduced the concept of “psychic distance” between home and host countries. Thus, according to the Uppsala Model, FDI modes are expected to be used by firms that have already internationalized through non-equity entry modes, notably by exporting [Fonfara 2011; Trąpczyński and Wrona 2013]. However critics of the Uppsala and similar models point to their weaknesses and limited explanatory power [e.g. Turnbull 1997] and call for the

M. Gorynia, J. Nowak, P. Trąpczyński, R. Wolniak, Does outward FDI…

development of a theory with better predictive ability and more in line with recent changes in the international business environment [Vissak, Ibeh, and Paliwoda 2007]. In the context of the present study it is therefore germane to also investigate how the FDI establishment modes of the companies studied relate to their non-equity entry modes. Although there are many specific motives for undertaking FDI, Dunning [1998] conveniently classified them, and the respective types of MNE activity, into the following four groups: resource seeking; market seeking; efficiency seeking; and strategic-asset seeking. Dunning argues that resource and market seeking motives typically characterize initial FDI, while those of efficiency and strategic asset-seeking characterize sequential FDI. He also argues that “[…] as strategic asset-acquiring investment has become more important, the locational needs of corporations have shifted from those to do with access to markets, or to natural resources, to those to do with access to knowledge-intensive assets and learning experiences which augment their existing O specific advantages”5 [Dunning 1998: 54]. It should be pointed out in this context that the link between FDI motives and FDI establishment modes has not been an intensively researched issue in IB studies as compared to other determinants of choice between green-fields, acquisitions or joint ventures [see e.g. Barkema and Vermeluen 1998; Larimo 2003; Hennart and Reddy 1997; Chen and Hennart 2002; Buckley and Casson 1998; Gorg 2000; Kogut and Singh 1988; Padmanabhan and Cho 1995; Slangen and Hennart 2008]. More importantly the said relationship does not have a long and intensive research tradition in the context of developing country MNEs as it does for those in developed country. Multinational enterprises from emerging economies may lack ownership advantages to be exploited and sustained abroad and thus may seek strategic assets abroad in order to close this gap and enhance their international competitiveness [see e.g. Cui and Jiang 2010; Yamakawa, Peng, and Deeds 2008; Makino, Lau, and Yeh 2002]. Mathews [2006] proposed in his „LLL” (linkage, leverage, learning) concept that emerging multinationals skip conventional stages of foreign expansion in order to catch up with international rivals in the sphere of technology. Likewise Luo and Tung [2007] proposed in their „springboard” approach that the latecomer disadvantage of firms from emerging markets can be overcome by acquisitions of intangible assets from developed country MNEs. According to the imbalance theory of Moon and Roehl [2001] FDI is undertaken to increase productivity of existing assets or to acquire complementary assets in order to balance out the asset portfolio. Imbalance stands for a situation whereby the latecomer does have certain firm-specific assets for which the current marginal value is below the market rate. However whilst there have been several studies in the Asian context highlighting the relationship between the intention of acquiring strategic assets and the preference of emerging mul 5

Ownership advantages.

Economics and Business Review, Vol. 1(15), No. 4, 2015

tinationals for selecting acquisitions in more developed countries as an establishment mode [see e.g. Deng 2007; Luo and Tung 2007] this relationship has not been investigated in a systematic manner for a broader range of different motives. In their qualitative study based on a sample of foreign investors in Poland, Gorynia et al. [2007, 2012] shed some light on this relationship and postulated further research using larger samples and quantitative methods. FDI motives (resource seeking, market seeking, efficiency seeking, strategic asset seeking)

Basic FDI establishment modes (acquisitions and green-fields)

Non-equity internationalization modes (exporting, licensing, franchising, contract manufacturing)

Figure 1. Analytical framework for studying FDI motives and modes

To summarize, the authors’ analytical framework, presented in Figure 1, is derived from conventional IB literature in order to provide a basis for analyzing the context of firms from an advanced emerging market – Poland. Thus whilst the strategic decisions in the internationalization process discussed above are not a novel, nor context-specific phenomenon in themselves, it is their application to the Polish context on which the present study will focus. The analytical framework shows the variables investigated in this paper and their basic relationships which will be formalized in the form of hypotheses in Section 3.

2. Research hypotheses Based on the literature review presented in Sections 1 and 2, as well as the findings from a qualitative study carried out by the authors [Gorynia et al. 2013a, b], several hypotheses were formulated. They were grouped into three subject areas: FDI in the internationalization process; FDI motives and FDI motives vs. FDI establishment modes. The first area deals with the place of FDI in the firm’s internationalization process. Consistent with the stages of the internationalization models reviewed in the previous section and the results of the previous qualitative study Polish companies tend to expand – with several exceptions – gradually, by preceding equity entry modes with exports, also in line with previous research in the CEE context [see e.g. Antalóczy and Éltető 2003; Śliwiński 2012]. This phenomenon has been increasingly visible since the opening of former centrally planned economies in the CEE region at the beginning of the 1990s. Whilst initially Slovenia displayed the highest dynamics of the emerging outgoing

M. Gorynia, J. Nowak, P. Trąpczyński, R. Wolniak, Does outward FDI…

investment Poland already appeared as the new dominant source of OFDI in the early 2000s. This gradual expansion could be attributed to several factors including inter alia Poland›s large domestic market size, allowing local firms to benefit from economies of scale. Moreover whilst Poland gradually liberalized its foreign trade and investment policy the introduction of support measures for the internationalization of domestic firms remained limited [Gorynia et al. 2013b]. This evolutionary behaviour can also be interpreted by the exploitation of previous business connections, frequently established before the transition process in the region [Bartosik-Purgat and Schroeder 2014]. In fact the studies of Rosati and Wilinski [2003] and Gorynia, Nowak, and Wolniak [2011] reveal a geographic concentration of Polish OFDI in neighbouring European countries. The experience of doing business in the CEE region, but also in more developed EU economies, is often perceived by managers as a key advantage as compared to foreign competitors.6 Hence, the following hypothesis is formulated: H1: FDI in a given host country is preceded by non-equity entry modes. Strong evidence was found in both the authors’ previous study [Gorynia et al. 2013a, b] and other empirical research [see e.g. Kaliszuk, Błaszczuk-Zawiła, and Wancio 2012; Svetličič, Rojec, and Trtnik 2000; Zemplinerová 2012] that the most important FDI motive for investors from CEE countries was the market seeking one, whilst efficiency-seeking and strategic asset-seeking motives were less prevalent. Jaworek, Szałucka, and Szóstek [2009] found that for Polish outward investors the search for new markets was the dominant motive whilst other motives varied with the host country’s level of development. Likewise Radło [2012] found a predominance of market-seeking motives for the largest Polish companies investing abroad and confirmed the geographic concentration of their investment in Europe. This is in line with the notion of Dunning and Lundan [2008] that such a motive usually appears in strategies of firms embarking on a move into foreign markets in the early, extensive stages of the internationalization process. Accordingly, the second hypothesis is as follows: H2: The main motives for undertaking FDI fall into the market-seeking category, followed by efficiency-seeking and strategic asset-seeking motives. Although empirical literature on the relationship between FDI motives and FDI establishment modes remains scarce the argument appearing in research on emerging-market MNEs is that these enterprises seek strategic assets abroad through acquisitions to overcome their lack of firm-specific advantages [see e.g. Peng 2012; Cui and Jiang 2010; Svetličič, Rojec, and Trtnik 2000]. The authors’ previous study also indicated that there might be an interdependency between the two variables, which is especially visible for Polish investments in targets located in more advanced economies [Gorynia et al. 2013a, b]. On the 6

For a review of experience effects on internationalization success see: [Trąpczyński 2013].

Economics and Business Review, Vol. 1(15), No. 4, 2015

other hand a significant portion of investments made by CEE firms, including those from Poland, has geographically focused on neighbouring country markets, which are either similar or less developed in economic and institutional terms. The authors also found cross-case patterns between efficiencyseeking motives and FDI establishment modes. In FDI projects driven by these motives the firms studied preferred green-field investments as establishment modes allowing them to exploit firm-specific advantages and to better adjust the scale of operations relative to the home country and, if applicable, to other host country operations. The above arguments lead to the formulation of the following two hypotheses: H3a: The dominance of efficiency-seeking motives increases the propensity to choose green-field investment as the FDI establishment mode. H3b: The dominance of strategic asset-seeking motives increases the propensity to choose acquisition as the FDI establishment mode.

3. Research methods 3.1. Data collection Data were obtained from a sample of firms investing abroad and registered in Poland. In order to maximize the available pool of Polish foreign investors several data sources were combined such as Amadeus, Kompass Poland, BPR Benchmark Poland and Deal Watch, as well as press articles and company reports. The triangulation of sources led to the creation of a proprietary database of 910 firms. Between May and June 2013 an invitation to participate in the study with a link to a web-based survey was sent to top managers directly responsible for foreign operations or other managers with a request to forward it to responsible colleagues. The survey, originally written in Polish, was designed for a larger project on Polish OFDI and thus contained a number of aspects not investigated in this study, including location choice criteria or foreign affiliate performance. Due to frequent concerns about technical reliability, response rates or security of electronic surveys [Kim and Gray 2008], an IT services agency was entrusted with the preparation of the survey, its execution and repeated reminders. An automated survey management system was supported by a substantial number of personal contacts with the sample firms in order to identify and persuade appropriate respondents to take part in the survey. Moreover additional interviews and secondary sources, including annual reports, were used to complete missing survey data. Thus a total sample of 60 complete surveys was obtained, which corresponds to a usable response rate of 6.6%.

M. Gorynia, J. Nowak, P. Trąpczyński, R. Wolniak, Does outward FDI…

3.2. Sample structure Whilst the sample size can by no means be regarded as representative for the total population and allowing for generalizations, the response sample distribution is largely similar to that of the entire population with regard to industry classification and parent nationality [see GUS 2013]. Thus the collected data make it possible to conduct a detailed exploration of the sectoral, geographic, modal and organizational structure of Polish OFDI (see Tables 1a–c). Table 1a. Sectoral distribution of major FDI of the sample firms (N = 60) Sector

Total manufacturing

Total services

Other

48.0

45.0

7.0

Number of FDI Percentage Source: Survey data.

Table 1b. Geographic distribution of major FDI of the sample firms (N = 60) Country

Ger many

Ukraine

Roma nia

Czech Repu blic

Slovakia

Russia

Other

Number of FDI

20.0

17.0

12.0

10.0

7.0

5.0

30.0

Percentage

Source: Survey data.

Table 1c. Firm size distribution of FDI in the sample (N = 60) Size (employment) Number of firms Percentage

0–49

50–99

100– 249

250– 499

500– 999

1000– 1999

>2000

8.0

5.0

17.0

18.0

15.0

10.0

26.0

Source: Survey data.

Whereas in order to qualify for the study the firms had to be registered in Poland their ultimate owners might be located abroad. Therefore firms with more than 10% of foreign capital constituted 47% of the sample. With regard to the FDI forms used, 58% of the firms had previous experience with wholly-owned green-field subsidiaries, whilst 20% had established joint ventures abroad. Notably 48% of the sample firms had made foreign acquisitions, out of which 12% could be classified as brownfield investments [Meyer and Estrin 2011]. The firms studied located their major FDI projects mostly in Germany

Economics and Business Review, Vol. 1(15), No. 4, 2015

(20%), Ukraine (17%), Romania (12%) and the Czech Republic (10%) (see Table 1b). This was based on a request to respondents to refer to subsidiaries engaged only in manufacturing and distribution, thus omitting special purpose entities/vehicles and other similar elements of corporate financial structure, and as a consequence, diminishing the significant role of such locations as Luxembourg, Switzerland or the Netherlands [Zimny 2011]. The current limited scope of foreign operations was reflected in the fact that 70% of the firms had foreign subsidiaries in up to 3 countries only, where sales and marketing activities were predominant. This finding remains in line with earlier studies in the CEE context outlined in Section 1.

3.3. Data operationalization The FDI establishment mode is that of the largest foreign affiliate in terms of total assets in the last fiscal year. It assumes a value of 0 if it is a greenfield subsidiary and 1 if it is an acquisition [Dikova and van Witteloostuijn 2007; Slangen and Hennart 2008]. Whilst there are studies jointly comparing joint ventures, green-fields and acquisitions [e.g. Kogut and Singh 1988; Anand and Delios 1997], establishment modes can be perceived as a separate decision problem in international expansion. Thus ownership choices (joint ventures vs. wholly-owned subsidiaries) were not analyzed here [see also Gorynia, Nowak, and Wolniak 2007]. The motives for establishing the largest foreign affiliate of the outward investor were measured on a five-point Likert scale for each of the four motive categories [Dunning and Lundan 2008]: market-seeking (foreign market share increase and expansion to further markets, Cronbach’s α = 0.67), efficiencyseeking (lower production costs, economies of scale and access to low cost labour, Cronbach’s α = 0.55), strategic asset-seeking (new brands, new distribution channels, human resources and new technology, Cronbach’s α = 0.62) and resource-seeking (access to raw materials and components; no Cronbach value since only one item was used for measurement). Table 2 summarizes the data operationalization. Whilst different studies have used secondary data to measure host-country and firm-level variables, Brouthers [2013] suggests that antecedents of managerial decisions such as entry modes need to be measured as perceptions of the decision-makers as opposed to objective variables which affect post-entry performance, as they refer to the actual environment for foreign operations. Obviously perceptual measures are burdened with the limitation of respondent subjectivity. However, in the context of FDI decisions, prior research suggested that the use of subjective measures is particularly appropriate in exploring the behaviour of firms from emerging economies and that such measures significantly correlate with objective indicators [Luo and Peng 1999]. Moreover, in the context of this research, reference to survey questions addressed to repre-

M. Gorynia, J. Nowak, P. Trąpczyński, R. Wolniak, Does outward FDI…

Table 2. Operationalization of motives Question

Scale

Motives for establishing the largest foreign subsidiary: 1. Market-seeking (foreign market share increase, expansion to further markets) 2. Efficiency-seeking (lower production costs, economies of scale, access to labour) 3. Strategic asset-seeking (new brands, distribution channels, personnel and technology) 4. Resource-seeking (access to raw materials and components)

Five-point Likert scale (1-no importance, 5-key importance)

Cronbach’s Alpha 0.67 0.55 0.62

Source: Survey data.

sentatives directly responsible for FDI decisions in firms allows an evaluation of the relevant decision determinants instead of merely inferring them.

4. Findings In order to test to Hypothesis 1 the data on firms’ foreign expansion were aggregated and summarized (Tables 3a and 3b). As many as 48% of the sample firms preceded their first FDI project with exports and 7% with contract manTable 3a. Host-country experience prior to the first FDI project (N = 60) Operating mode

Exports

Contract manufacturing

Licensing

No experience

Number of firms

48.0

7.0

2.0

48.0

Percentage Source: Survey data.

Table 3b. Host-country experience prior to subsequent FDI projects (N = 41) Opera ting mode

Exports

Contract manufac- Licensing turing

Fran chising

Another FDI

No experience

Number of firms

Percen tage

59.0

5.0

2.0

29.0

Source: Survey data.

Economics and Business Review, Vol. 1(15), No. 4, 2015

ufacturing. Moreover from the 41 firms which also made further investments in other countries, 59% previously operated in these host-countries through exports, 5% by contract manufacturing and another 5% through licensing. Therefore support could be found for Hypothesis 1. At the same time it is worth noticing that 48% of firms did not have any host-country experience prior to their initial FDI project, nor did 29% of those investing in more than one country. In order to address Hypothesis 2 mean scores of the motive categories were computed (Table 4). In line with this hypothesis market-seeking motives appeared as the dominant driver of Polish OFDI, followed by efficiency-seeking and strategic asset-seeking motives, with only a marginal role left to resourceseeking motives. Table 4. The significance (1 – low, 5 – high) of FDI motives (N = 60) Market-seeking

Efficiencyseeking

Strategic assetseeking

Resourceseeking

Mean score

3.55

2.72

2.05

1.57

1.25

0.94

0.87

1.14

FDI motive category

Source: Survey data.

Additionally, in order to address Hypotheses 3a and 3b, a chi-square test was run for the relationship between the motives and FDI establishment modes (Table 5). No statistically significant relationship was found between these variables thus providing no support for Hypotheses 3a and 3b. Table 5. Chi-square tests for the FDI motive-mode relationship Motive

FDI Mode

Marketseeking

Efficiencyseeking

Strategic assetseeking

Resourceseeking

Total

Green-field

Acquisition

Total Pearson Chi-Square

5.408

Likelihood Ratio

5.464

Linear-by-Linear Association

2.543

Source: Survey data.

M. Gorynia, J. Nowak, P. Trąpczyński, R. Wolniak, Does outward FDI…

In conformity with existing theory and research on the internationalization of firms from emerging and transition economies most Polish companies expanded gradually by preceding equity entry modes with exports to target markets. This corresponds with previous findings in the CEE context [see e.g. Antalóczy and Éltető 2003] but also to some of those related to advanced economy environments [see e.g. Holmlund, Kock, and Vanyushyn 2007]. However, in the case of firms originating from CEE countries, the evolutionary behaviour in the internationalization process can be interpreted as a phenomenon driven by the exploitation of previous business connections frequently established before the beginning of the transition process in the early 1990s. Previous experience with doing business in the CEE region can be regarded as a key advantage in embarking on capital expansion there [Gorynia et al. 2013a]. At the same time a significant number of the largest FDI projects undertaken by the sample firms were made without any prior experience in the target country or elsewhere. Given the predominance of market-seeking motives in the said investments it cannot be concluded that these investments were driven by the need to acquire deficient resources quickly, which characterized OFDI generated by Asian firms [Luo and Tung 2007]. Rather the non-linear character of Polish firms’ internationalization, which becomes more apparent with a careful scrutiny of case-by-case evidence, can arguably be attributed to the fact that most CEE countries share a similar, historically shaped institutional background, which tends to facilitate foreign expansion and omit certain stages of the gradual expansion sequence [Del Sol and Kogan 2007]. Nonetheless this aspect of CEE firm internationalization still requires further investigation in order to understand the relevance of the advantages stemming from a similar institutional context. Moreover this accelerated internationalization is a known phenomenon in the case of developed country multinationals also which can be traced back to a number of explanatory variables, such as home market size or firm size [Dimitratos and Jones 2005]. Thus it is yet to be determined on a larger sample of companies as to what extent the phenomenon of OFDI from emerging markets is specifically related to their origin rather than other factors which are independent of this country context. As for the role of particular FDI motives most foreign investment projects belong to the market seeking category thus confirming the argument of Dunning that this category usually appears in strategies of firms embarking on a move into foreign markets in the early, extensive stages of the internationalization process and also corroborates some earlier studies of FDI from CEE [Varblane, Reitjan, and Roolaht 2003; Czaplewski and Wiśniewska 2007; Karpińska-Mizielińska and Smuga 2007]. However no support could be found for the argument that the predominance of efficiency-seeking motives increases the propensity to choose green-field investment as the FDI mode, nor for the impact of the strategic asset-seeking motive. Both motive categories are normally indicative of strategies followed by mature MNEs, reflecting their

Economics and Business Review, Vol. 1(15), No. 4, 2015

intensive approach to continued internationalization, hence the limited scope of Polish firms’ internationalization does not yet permit any firm conclusions. However the argument about the link between acquisition and the need to capture new intangible assets abroad remains consistent with the evidence of other emerging multinationals, especially those from China, where OFDI is capable of fulfilling the strategic goals of both the investors and government [Cui and Jiang 2010; Yamakawa, Peng, and Deeds 2008]. A case-by-case analysis of the present sample provides indications that this relationship holds true also for Polish firms to some extent, particularly the investment in more advanced economies.

Conclusions The findings of the quantitative analysis on a sample of outward investors based in Poland presented in this paper can be summarized as follows. Whilst foreign expansion of Polish firms is mostly of a gradual character, i.e. FDI projects are preceded with other modes of foreign expansion and many investments also skip the earlier, non-equity stages of internationalization. The study points to the fact that market-seeking motives explain most of the FDI undertaken by these latecomer firms. No clear support was found for the relationship between the motives for FDI and the choice of FDI establishment modes. Nonetheless the analysis provided some preliminary evidence that strategic asset-seeking motives tend to be at the heart of foreign acquisitions, although this relationship received little support here. The study is clearly subject to certain limitations, the key one being its limited sample size. Thus the present analysis cannot properly account for factors such as sector of activity, firm size, or host-country characteristics, which seem to be relevant from the point of view of specific FDI motives. Nonetheless it is part of a broader research project and therefore these initial results should first of all be regarded as the exploration of the international expansion patterns of Polish firms. Another limitation pertains to the unique use of subjective measures in order to probe the concepts investigated. Although the intention was to capture managerial perceptions which are crucial to expansion decisions, the use of secondary data might have improved the reliability and robustness of the research results obtained. Further studies on the expansion of firms from CEE countries could investigate the role which different FDI motives have in different locations, notably economically and institutionally less or more advanced than the home country. These motives might affect both the entry mode itself as well as location choice. Sub-sample analysis would be an appropriate research project, requiring however a larger total sample. Another relevant research problem is the multitude of motivations underlying FDI decisions. Hence the adopted categorization of

M. Gorynia, J. Nowak, P. Trąpczyński, R. Wolniak, Does outward FDI…

motives inevitably leads to certain oversimplification and does not fully reflect, from the perspective of the parent firm, the complexity of roles attributed to outward FDI. Thus a more nuanced approach to identifying a higher number of motives behind the investment can provide more meaningful insights. In the same vein the role of subsidiaries in the value chain of the parent firm should be taken into account in future investigations. The present study did not look into the effects of the industry sector on foreign expansion decisions, which also constitutes a promising avenue for future studies.

References Anand, J., Delios, A., 1997, Location Specificity and the Transferability of Downstream Assets to Foreign Subsidiaries, Journal of International Business Studies, vol. 28, no. 3: 579–603. Antalóczy, K., Éltető, A., 2003, Outward Foreign Direct Investment in Hungary. Motivation and Effects, in: Svetličič, M., Rojec, M. (eds.), Facilitating Transition by Internationalization: Outward Direct Investment from Central European Economies in Transition, Ashgate, Aldershot: 155–174. Barkema, H.G., Vermeulen, F., 1998, International Expansion Through Start-up or Acquisition: A Learning Perspective, Academy of Management Journal, vol. 41, no. 1: 7–26. Bartosik-Purgat, M., Schroeder, J., 2014, Polish Exporters’ Use and Perception of Various Sources of Market Information, Poznań University of Economics Review, vol. 14, no. 2: 97–117. Bohata, M., Zeplinerová, A., 2003, Internationalization of Czech Companies via Outward Investment, in: Svetličič, M., Rojec, M. (eds.), Facilitating Transition by Internationalization: Outward Direct Investment from Central European Economies in Transition, Ashgate, Aldershot: 111–132. Brouthers, K.D., 2013, Retrospective: Institutional, Cultural and Transaction Cost Influences on Entry Mode Choice and Performance, Journal of International Business Studies, vol. 44, no. 1: 14–22. Buckley, P.J., Casson, M.C., 1998, Analyzing Foreign Market Entry Strategies: Extending the Internalization Approach, Journal of International Business Studies, vol. 29, no. 3: 539–562. Buckley, J., Clegg, J. L., Cross, A. R., Voss, H., 2010, What Can Emerging Markets Learn from the Outward Direct Investment Policies of Advanced Countries?, in: Sauvant, K.P., McAllister, G., Maschek, W.A. (eds.), Foreign Direct Investments from Emerging Markets. The Challenges Ahead, Palgrave Macmillan, New York: 243–276. Busse, M., Hefeker, C., 2007, Political Risk, Institutions and Foreign Direct Investment, European Journal of Political Economy, vol. 23, no. 2: 397–415. Chen, S.-F.S., Hennart, J.-F., 2002, Japanese Investors’ Choice of Joint Ventures versus Wholly-Owned Subsidiaries in the US: The Role of Market Barriers and Firm Capabilities, Journal of International Business Studies, vol. 33, no. 1: 1–18. Cui, L., Jiang, F., 2010, Behind Ownership Decision of Chinese Outward FDI: Resources and Institutions, Asia Pacific Journal of Management, vol. 27: 751–774.

Economics and Business Review, Vol. 1(15), No. 4, 2015

Cuervo-Cazurra, A., Genc, M., 2008, Transforming Disadvantages into Advantages: Developing-country MNEs in the Least Developed Countries, Journal of International Business Studies, vol. 39: 957–979. Czaplewski, L., Wiśniewska, E., 2007, Zamierzenia i motywy podejmowania bezpośrednich inwestycji zagranicznych przez polskie przedsiębiorstwa, in: Karaszewski, W. (red.), Bezpośrednie inwestycje zagraniczne w budowaniu potencjału konkurencyjności polskich przedsiębiorstw i regionów, Wydawnictwo Uniwersytetu Mikołaja Kopernika, Toruń: 229–244. Del Sol, P., Kogan, J. 2007, Regional Competitive Advantage Based on Pioneering Economic Reforms: The Case of Chilean FDI, Journal of International Business Studies, vol. 38: 901–927. Deng, P., 2007, Investing for Strategic Resources and Its Rationale: The Case of Outward FDI from Chinese Companies, Business Horizons, vol. 50: 71–81. Dikova, D., Van Witteloostuijn, A., 2007, Foreign Direct Investment Mode Choice: Entry and Establishment Modes in Transition Economies, Journal of International Business Studies, vol. 38, no. 6: 1013–1033. Dimitratos, P., Jones, M., 2005, Future Directions for International Entrepreneurship Research, International Business Review, vol. 14: 119–128. Dunning, J.H., 1998, Location and the Multinational Enterprise: A Neglected Factor?, Journal of International Business Studies, vol. 29: 45–66. Dunning, J.H., Lundan, S., 2008, Multinational Enterprises and the Global Economy, Edward Elgar Cheltenham, U.K. Fonfara, K., 2011, A Typology of Company Behaviour in the Internationalisation Process (a Network Approach), Poznań University of Economics Review, vol. 11, no. 2: 5–25. Gorg, H., 2000, Analysing Foreign Market Entry – The Choice between Greenfield Investment and Acquisitions, Journal of Economic Studies, vol. 27, no. 3: 165–181. Gorynia, M., Nowak, J., Trąpczyński, P., Wolniak, R., 2013a, Internationalization of Polish Firms via Foreign Direct Investment: A Multiple-Case Study Approach, in: Marinov, M.A., Marinova, S.T. (eds.), Successes and Challenges of Emerging Economy Multinationals, Houndmills, Palgrave Macmillan, New York: 184–216. Gorynia, M., Nowak, J., Trąpczyński, P., Wolniak, R., 2013b, The Internationalization of Polish Firms: Evidence from a Qualitative Study of FDI Behavior, in: Turkina, E., Thai, M.T.T. (eds.), Internationalization of Firms from Economies-in-transition: The Effects of Politico-economic Paradigm Shift, Edward Elgar Publishing: 39–66. Gorynia, M., Nowak, J., Trąpczyński, P., Wolniak, R., 2014, Should Government Support Outward FDI? The Case of Poland, in: Marinova, S. (ed.), Institutional Impacts on Firm Internationalization, Palgrave Macmillan, pp. 120–145. Gorynia, M., Nowak, J., Wolniak, R., 2007, Motives and Modes of FDI in Poland: An Exploratory Qualitative Study, Journal for East European Management Studies, vol. 12, no. 2: 132–151. Gorynia, M., Nowak, J., Wolniak, R., 2008, Poland’s Evolving Net Outward Investment Position: A Cross-industry Analysis, Poznań University of Economics Review, vol. 8, no. 2: 5–25. Gorynia, M., Nowak, J., Wolniak, R., 2010, Foreign Direct Investment in Central and Eastern Europe: The IDP Trajectories of Selected Countries, Poznań University of Economics Review, vol. 10, no. 1: 5–26.

M. Gorynia, J. Nowak, P. Trąpczyński, R. Wolniak, Does outward FDI…

Gorynia, M., Nowak, J., Wolniak, R., 2011, Outward Foreign Direct Investment of Poland: Exploring Geographic and Industry Trends and Patterns, paper presented at the Annual Conference of the European International Business Academy, 8–10.12.2011, Bucharest, Romania. Gorynia, M., Nowak, J., Wolniak, R., 2012, Emerging Profiles of Polish Outward Foreign Direct Investment, Journal of East-West Business, vol. 18, no. 2: 132–156. GUS, 2013, Polskie inwestycje bezpośrednie w 2011 roku, Departament Strategii i Analiz, Warszawa: 5–47. Hennart, J.-F., Reddy, S., 1997, The Choice between Mergers/Acquisitions and Joint Ventures: The Case of Japanese Investors in the United States, Strategic Management Journal, vol. 18: 1–12. Holmlund, M., Kock, S., Vanyushyn, V., 2007, Small and Medium-sized Enterprises’ Internationalization and the Influence of Importing on Exporting, International Small Business Journal, vol. 35: 459–475. Jaklič, A., Svetličič, M., 2009, Emerging Markets Global, Players EMGP Project, Vale Columbia Center on Sustainable Investment, Columbia University, New York. Jaworek, M., Szałucka, M., Szóstek, A., 2009, Knowledge and Skills as Competitiveness Factors of Polish Direct Investment Companies, in: Karaszewski, W. (ed.), Foreign Direct Investment of Polish Companies: Its Scale, Structure, Determinants, Influence on the Competitiveness, Wydawnictwo Naukowe UMK, Toruń: 41–62. Johanson, J., Vahlne, J.E., 1977, The Internationalization Process of the Firm – A Model of Knowledge Development and Increasing Foreign Market Commitments, Journal of International Business Studies, vol. 8, no, 1: 23–32. Johanson, J., Vahlne, J.E., 1990, The Mechanism of Internationalisation, International Marketing Review, vol. 7, no. 4: 11–24. Johanson, J., Vahlne, J.E., 2009, The Uppsala Internationalization Process Model Revisited: From Liability of Foreignness to Liability of Outsidership, Journal of International Business Studies, vol. 40: 1411–1431. Johanson, J., Wiedersheim-Paul, F., 1975, The Internationalization of the Firm – Four Swedish Cases, Journal of Management Studies, vol. 12, no. 3: 305–323. Kaliszuk, E., Błaszczuk-Zawiła, M., Wancio, A., 2012., Polish Multinationals Go Beyond Europe, Vale Columbia Center on Sustainable Investment, Columbia University, New York, Institute for Market, Consumption and Business Cycles Research, Warsaw. Kaliszuk, E., Wancio, A., 2013, Polish Multinationals: Expanding and Seeking Innovation Abroad, Vale Columbia Center on Sustainable Investment, Columbia University, New York, Institute for Market, Consumption and Business Cycles Research, Warsaw. Kalotay, K., 2004, Outward FDI from Central European Countries, Economics of Planning, vol. 37: 141–172. Karpińska-Mizielińska, W., Smuga, T., 2007, Determinanty bezpośrednich inwestycji polskich przedsiębiorstw na rynkach zagranicznych, Gospodarka Narodowa, vol. 9: 31–53. Kępka, H., 2011, Struktura geograficzna polskich bezpośrednich inwestycji, in: Pałys, L. (red.): Polskie inwestycje za granicą, Studia i Materiały, no. 92, Instytut Badań Rynku, Konsumpcji i Koniunktur, Warszawa: 18–47. Kim, Y., Gray, S.J., 2008, The Impact of Entry Mode Choice on Foreign Affiliate Performance: The Case of Foreign MNEs in South Korea, Management International Review, vol. 48, no. 2: 165–188.

100

Economics and Business Review, Vol. 1(15), No. 4, 2015

Kogut, B. Singh, H., 1988, The Effect of National Culture on the Choice of Entry Mode, Journal of International Business Studies, vol. 19, no. 3: 411–432. Kuznetsov, A., 2013, Global Expansion of Russian Multinationals after the Crisis: Results of 2011, Vale Columbia Center on Sustainable Investment, Columbia University, New York, Institute of World Economy and International Relations MEMO of Russian Academy of Sciences, Moscow. Larimo, J., 1993, Foreign Direct Investment Behaviour and Performance: An Analysis of Finnish Direct Manufacturing Investments in OECD Countries, Acta Wasaensia, 32, University of Vaasa, Vaasa. Larimo, J., 2003, Form of Investment by Nordic Firms in World Markets, Journal of Business Research, vol. 56, no. 10: 791–803. Luo, Y., Peng, M.W., 1999, Learning to Compete in a Transition Economy: Experience, Environment and Performance, Journal of International Business Studies, vol. 302: 269–295. Luo, Y., Tung, R., 2007, International Expansion of Emerging Market Enterprises: A Springboard Perspective, Journal of International Business Studies, vol. 38: 481–498. Makino, S., Lau, C.-M, Yeh, R.-S., 2002, Asset-Expoitation versus Asset-Seeking: Implications for Location Choice of Foreign Direct Investment from Newly Industrialized Economies, Journal of International Business Studies, vol. 38: 481–498. Meyer, K., Estrin, S., 2011, Brownfield Acquisitions. A Reconceptualization and Extension, Management International Review, vol. 51: 483–509. Mathews, J.A., 2006, Dragon Multinationals: New Players in 21st Century Globalization, Asia Pacific Journal of Management, vol. 23: 5–27. Moon, H.C., Roehl, T.W., 2001, Unconventional Foreign Direct Investment and the Imbalance, International Business Review, vol. 10: 197–215. Narula, R., 2006, Globalization, New Ecologies, New Zoologies, and the Purported Death of the Eclectic Paradigm, Asia Pacific Journal of Management, vol. 23, no. 2: 143–151. Obłój, K., Wąsowska, A., 2012, Strategiczne wybory polskich firm – motywy i kierunki budowania powiązań zewnętrznych drogą kapitałowego umiędzynarodowienia działalności, in: Morawski, W. (red.), Powiązania zewnętrzne. Modernizacja Polski, Wolters Kluwer, Warszawa: 102–120. Padmanabhan, P., Cho, K.R., 1995, Methodological Issues in International Business Studies: The Case of Foreign Establishment Mode Decisions by Multinational Firms, International Business Review, vol. 4, no. 1: 55–73. Padmanabhan, P., Cho, K.R., 1996, Ownership Strategy for a Foreign Affiliate: an Empirical Investigation of Japanese Firms, Management International Review, vol. 36, no. 1: 45–65. Peng, M.W., 2012, The Global Strategy of Emerging Multinational from China, Global Strategy Journal, vol. 2: 97–107. Radło, M.-J., 2012, Emerging Multinationals and Outward FDI Development: the Case of Poland, Eastern European Economics, vol. 50, no. 2: 59–84. Radło, M.-J., Sass, M., 2012, Outward Foreign Direct Investments and Emerging Multinational Companies from Central and Eastern Europe. The Case of Visegrád Countries, Eastern European Economics, vol. 50, no. 2: 5–21. Rosati, D., Wilinski, W., 2003, Outward Foreign Direct Investments from Poland, in: Svetličič, M., Rojec, M. (eds.), Facilitating Transition by Internationalization:

M. Gorynia, J. Nowak, P. Trąpczyński, R. Wolniak, Does outward FDI…

101

Outward Direct Investment from Central European Economies in Transition, Ashgate, Aldershot: 175–204. Rugraff, E., 2010, Strengths and Weaknesses of the Outward FDI Paths of the Central European Countries, Post-Communist Economies, vol. 22, no. 1: 1–17. Slangen, A.H.L., 2005, Studies on the Determinants of Foreign Entry Mode Choices and Performance, Doctoral Dissertation, University of Tilburg, Tilburg. Slangen, A.H.L., Hennart, J.-F., 2008, Do Multinationals Really Prefer to Enter Culturally Distant Countries Countries through Greenfields Rather than through Acquisitions?, Journal of International Business Studies, vol. 39: 472–490. Svetličič, M., Jaklič, A., 2003, Outward FDI by Transition Economies: Basic Features, Trends and Development Implications, in: Svetličič, M., Rojec, M. (eds.), Facilitating Transition by Internationalization: Outward Direct Investment from Central European Economies in Transition, Ashgate, Aldershot: 49–76. Svetličič, M., Rojec, M., Trtnik, A, 2000, The Restructuring Role of Outward Foreign Direct Investment by Central European Firms: The Case of Slovenia, Advances in International Marketing, vol. 10: 53–88. Śliwiński, R., 2012, Internationalization Strategies of Polish Fast Growing Enterprises, Poznań University of Economics Review, vol. 12, no. 1: 17–39. Trąpczyński, P., 2013, Does Foreign Direct Investment Theory Explain Subsidiary Performance? A Critical Literature Review, Poznań University of Economics Review, vol. 13, no. 2: 47–64. Trąpczyński, P., Wrona, T., 2013, From Going International to Being International – Strategies for International Competitiveness, Poznań University of Economics Review, vol. 13, no. 1: 89–114. Turnbull, W., 1987, A Challenge to the Stages Theory of the Internationalization Process, in: Reid, S., Rosson, P. (eds.), Managing Export Entry and Expansion, Praeger, New York. Varblane, U., Reiljan, E., Roolaht, T., 2003, The Role of Outward Foreign Investments in the Internationalization of Estonian Firms, in: Svetličič, M., Rojec, M. (eds.), Facilitating Transition by Internationalization: Outward Direct Investment from Central European Economies in Transition, Ashgate, Aldershot: 133–154. Vissak, T., Ibeh, K., Paliwoda, S., 2007, Internationalising from the European Periphery: Triggers, Processes, and Trajectories, Journal of Euromarketing, vol. 17, no. 1: 35–48. Visser, H., 2006, Outward Foreign Direct Investment: Is It a Good Thing?, in: Meijert, G., Heijman, W.J.M., van Ophem, J.A.C., Verstegen, B.H.J. (eds.), Heterodox Views on Economics and the Economy of the Global Society, Wageningen Academic Publishers, Wageningen,. Yamakawa, Y., Peng, M.W., Deeds, D.L., 2008, What Drives New Ventures to Internationalize from Emerging to Developed Economies?, Entrepreneurship Theory and Practice vol. 32: 59–82. Zemplinerová, A., 2012, Czech OFDI: Investment Strategies and Motivation to Invest Abroad, Eastern European Economics, vol. 50, no. 2: 22–40. Zimny, Z., 2011, Outward FDI from Poland and its Policy Context, Columbia FDI Profiles, Vale Columbia Center on Sustainable International Investment, New York.

Economics and Business Review, Vol. 1 (15), No. 4, 2015: 102–125 DOI: 10.18559/ebr.2015.4.7

The complex relationship between intrinsic and extrinsic rewards1 Orni Gov2

Abstract : This study aimed to determine the relationship between the two main components of the organizational rewards system; intrinsic and extrinsic rewards. Pay satisfaction was chosen to represent the extrinsic rewards whilst the Thomas et al. four intrinsic rewards represented the intrinsic rewards. A total of one hundred and fifty six self-administered questionnaires were collected from Israeli employees. Analysis using Pearson correlation indicated that there is no relationship between the two types of rewards. However, a regression analysis demonstrated that employees with temporary status, and outsourcers’ employees, are experiencing less pay satisfaction than organizations’ directly-employed staff. Another regression analysis established that a sense of choice predicted a positive correlation with pay satisfaction whilst a sense of competence predicted a negative correlation. Based on the findings it can be concluded that there is a downside to an organization’s decision to reward employees by empowering and giving them autonomy. The empowerment process might increase the employees’ sense of competence, which may lead to a decrease in pay satisfaction. Keywords : employees’ motivation, extrinsic rewards, intrinsic rewards, organizational rewards, performance, pay satisfaction. JEL codes : J30, J31, J32, J33, M12, M50, M52, M54.

Introduction Organizations spend significant parts of their budgets offering intrinsic and extrinsic rewards to members aiming to improve human resources outcomes [Mahaney and Lederer 2006]. Work rewards, meaning all the benefits that workers receive from their jobs [Bratton and Gold 2012], are considered one of the most important factors that encourage workers to invest extra effort and work more efficiently. When employees are rewarded, they get work done [Njanjal et al. 2013]. Due to its significant implications to the success of an 1

Article received 10 April 2014, accepted 15 September 2015. Hadassah Academic College, Department of Management of Service Organizations, Hneviim St., 39, Jerusalem, Israel, 9101001, e-mail: ornigov@gmail.com. 2

O. Gov, The complex relationship between intrinsic and extrinsic rewards

103

organization, the reward system has been studied by numerous researchers. Yet there is some inconsistency in the literature regarding the effects of the different incentives on employees. The main debate has focused on the effect of extrinsic and intrinsic rewards on organizational outcomes and the interrelation between both types of rewards. Alfandi and Alkahsawnen [2014] discuss suitable incentives and claim that the absence of them may negatively affect employees’ performance by weakening their motivation to attain the organization’s goals. Therefore it is useful to understand the dynamic of the reward system and to identify the types of work rewards and their effect on employees’ willingness to contribute to an organization’s effectiveness and performance. Hertzberg [1962] distinguishes between two groups of motivational factors: hygiene and motivation. The hygiene factors include rewards such as salary, company compensation policy, working conditions and security. The absence of hygiene rewards can de-motivate or cause dissatisfaction, however they do not necessarily create satisfaction when they are present. On the other hand, the motivation factors do motivate or create satisfaction and are rarely the cause of dissatisfaction. The Motivation rewards are based on an individual’s need for personal growth. They include intrinsic rewards such as status, responsibility, gaining recognition and performing challenging work. Following Hertzberg a number of scholars noted the importance of intrinsic rewards and their advantage as motivation drivers [Manz 1991; Senge 1990]. Extrinsic rewards played a dominant role in traditional organizations, where work was generally technical and bureaucratic and where obeying the rules and procedures was vital. Since this type of work offered employees few intrinsic rewards, extrinsic rewards were often the only motivational tools available to organizations [Thomas 2009]. Extrinsic rewards remain significant for workers, especially when making decisions regarding the acceptance of a new job, but according to Thomas extrinsic rewards are less important now, whilst day-today motivation is more strongly driven by intrinsic rewards. Other researchers such as Frey [2012] and Zhou, Zhang, and Sanchez [2009] go even further in their perceptions of the relationship between both groups of rewards. Zhou and his colleagues found that motivation is a result of the combination of the two groups of factors; some amount of extrinsic reward is necessary yet too much of it will decrease employees’ motivation. Frey’s findings are more decisive; according to him, tangible rewards generally have negative effects on intrinsic motivation. The importance of further understanding the relationship between the two types of rewards is obvious when we are taking into account the importance of the reward system to an organization’s success and the enormous cost of an inefficient one. If we can learn more about the effects of both types of rewards it will allow organizations to invest their resources in the right direction to maximize their effectiveness and reach the best results.

104

Economics and Business Review, Vol. 1(15), No. 4, 2015

This paper is divided into four sections. The first section reviews the literature that deals with the elements at focus of this article: intrinsic rewards and pay satisfaction. The research also highlights the mixed approaches regarding the relationship between them. The second section describes the research methodology including the data collection procedure, the participants and the measures used by the researcher. The third section presents the results. Finally, the last section provides a discussion of the implications of the resaults and the contribution to the body of knowledge regarding employee rewards.

1. Literature 1.1. Intrinsic rewards Intrinsic motivation is the human need to be competent and self-determining in relation to the environment, to engage one’s interests and exercise one’s capacities [Deci and Ryan 1985]. Intrinsic motivation is about passion and positive feelings that people get from their work. These feelings reinforce and energize employees’ self-management efforts and make work personally fulfilling. Building intrinsic motivation helps to create an upward spiral of positive feelings and experiences [Thomas 2009]. Intrinsic motivation is experienced when a person is moved to act for the ‘fun’ or challenge entailed rather than because of external pressures or rewards. Employees are thought to be motivated to work hard to produce quality results when they have pride in their work, they believe their efforts are important to the success of the team and their jobs are fun, challenging, and rewarding [Mahaney and Lederer 2006]. Having intrinsically motivated employees can be the greatest asset for any organization [Masvaure, Ruggunan, and Prowse 2014]. Two of the most established intrinsic rewards’ models are Hackman and Oldham’s [1976] Job characteristic model and Deci and Ryan’s Self-determination theory [1991]. Both made significant contributions to the study of motivation yet they also had some limitations which should be taken into consideration. Hackman and Oldham [1976] developed the job characteristics model, which has largely replaced Herzberg’s theory on organizational settings. Their model describes the effects of five dimensions of job design: skill variety, task identity, task significance, autonomy and feedback. Research reviewed by Fried and Ferris [1987] shows that these five job dimensions generally have an impact on job satisfaction and other favourable outcomes. However there is less support for the motivational core of the model. That part of the model says that intrinsic motivation occurs when three “psychological states” are present: experienced meaningfulness of the work, experienced responsibility for outcomes of the work, and knowledge of the actual results of the work. Thomas [2009] claims that this list has two shortcomings. First, it focuses only on task outcomes (sim-

O. Gov, The complex relationship between intrinsic and extrinsic rewards

105

ilar to purposes) as a factor in intrinsic motivation, leaving out activity-related rewards. Second, only meaningfulness is clearly an intrinsic reward. Deci and Ryan [1991] developed the Self-determination theory, the most widely known theory for intrinsic motivation. According to them motivation is intrinsic if activity is undertaken for the immediate satisfaction of a person’s needs. The Self-determination theory is rooted in a set of assumptions about human nature and motivation [Ryan and Deci 2000]. Humans are inherently motivated to grow and achieve and will fully commit to and engage in even uninteresting tasks when their meaning and value is understood. Passive behaviour of employees is a learned behaviour, which has an undermining effect on inherent motivation [Stone, Deci, and Ryan 2008]. The Self-determination theory of motivation differentiates between autonomous or truly willing actions and imposed actions that are controlled by forces experienced as external to the self. According to Deci and Ryan’s view task activities are intrinsically rewarding when people experience a sense of self-determination (choice) and of competence, and thus they become motivated. The shortcoming of their view is that the model focuses only on task activities as sources of intrinsic rewards so that purposes are excluded [Wiersma 1992]. Due to the shortcoming of both classic theories, Thomas and Velthouse [1990] suggest the four intrinsic rewards’ model, based on reducing Hackman and Oldham’s three psychological states into two intrinsically rewarding states, a sense of meaningfulness and a sense of progress and then to add the missing activity-related rewards based on the Self-determination theory, a sense of choice and a sense of competence. The Thomas, Jansen, and Tymon [1997] model emphasizes the relationship between self-management and intrinsic rewards. According to their view intrinsic motivation is derived from the self-management process and, in parallel, it reinforces and energizes continued self-management. According to the four intrinsic rewards model, each of the stages in the self-management process affects one’s intrinsic rewards, and together and separately, they lead to employee engagement. A person who is allowed to self-manage can gain motivation and energy when he/she has a meaningful objective, has the freedom to choose the strategies to achieve that objective, feels competent whilst doing it and has a sense of progress. The four Intrinsic Rewards are: –– Sense of meaningfulness – this reward involves the meaningfulness or importance of the purpose the employee is trying to fulfill. The individual feels that he has an opportunity to accomplish something of actual value, something that matters in the larger scheme of things. He feels that he is on a path that is worth his time and energy, giving him a strong sense of purpose or direction. The feeling of meaningfulness occurs when an individual is progressing on a path that they believe is worth their time and energy. –– Sense of choice – the employee feels free to choose how to accomplish his work, to use his best judgment to select those work activities that make the

106

Economics and Business Review, Vol. 1(15), No. 4, 2015

most sense to him and to perform them in ways that seem appropriate. The person feels ownership of his work, believes in the approach he is taking, and feels responsible for making it work. The feeling of choice occurs when an individual feels free to choose activities that make sense to him/her and is able to perform them in ways that seem appropriate. –– Sense of competence – the employee feels that he is handling his work activities well, that his performance of these activities meets or exceeds his personal standards and that he is doing good, high-quality work. The employee feels a sense of satisfaction, pride, or even artistry in how well he handles these activities. The feeling of competence, involves whether the individual feels skillful in performing the task activities that they have chosen. According to Bandura’s social cognitive theory [1986, 1997] individuals are inherently motivated to perform if they feel confident in their ability. –– Sense of progress – the employee is encouraged by the feeling that his efforts are really accomplishing something. He feels that his work is on track and moving in the right direction. The individual sees convincing signs that things are working out, giving him confidence in the choices he has made and confidence in the future. The feeling of progress involves a person’s sense that a task is moving forward and that their activities are really accomplishing something. Stumpf et al. [2013] suggest a virtuous cycle of intrinsic rewards supporting the employees’ positive work-related feelings. These feelings lead to work experiences that foster a greater sense of intrinsic rewards: the experience of work as meaningful, the ability to exercise some degree of choice, the experience of progress, and the development of a greater sense of competence. The senses of competence and progress result from the degree of accomplishment, and the qualities of the performance. These intrinsic rewards are based on monitoring events that occur late in the self-management process, following the task behaviour. Choice and meaningfulness in contrast are rewards based on events that take place prior to the occurrence of the task behaviour. They are derived from the task opportunities; the ability to use one’s own judgment and to pursue a valuable purpose [Thomas 2009]. The literature suggests significant support for the four intrinsic rewards model. In this review some studies are found to be based on research which examined the model directly [Thomas and Velthouse 1990; Tymon, Stumpf, and Doa 2010]. Others test factors such as autonomy or competence which are considered to be intrinsic rewards [Conchie 2013; Harell and Daim 2010; Mundhra and Jacobs 2011]. The model was shown to be robust when considered from either perspective”. Thomas [2009] reports that employees with high levels of intrinsic rewards become informal recruiters and marketeers for their organization. Such employees experience more positive feelings and fewer negative ones on the job. Their job satisfaction is higher, they report fewer stress symptoms and are more

O. Gov, The complex relationship between intrinsic and extrinsic rewards

107

likely to feel that they are developing professionally. A later study by Masvaure, Ruggunan, and Maharaj [2014] found similar results. Harell and Daim [2010] identified autonomy, responsibility, variety of task and sense of accomplishment as significant motivators. The studies of Mundhra and Jacob [2011] and Davis and Wilson [2000] show similar results. In this study the four intrinsic reward model has been adopted since it combines the two most popular models, and it fits the characteristics of the current employees. The view has been taken here that the Thomas et al. four intrinsic rewards model has not been given enough recognition in motivation and reward management theory and practice. As has been discussed reward systems have usually included extrinsic rewards in addition to intrinsic rewards, therefore a complete review should include the extrinsic and intrinsic reward literature and a summary of the main finding regarding the relationships between them. Extrinsic rewards are rewards that result from external, non-job-related factors [Malhorta, Budhwar, and Prowse 2007]. Extrinsic rewards include elements such as pay, fringe benefits, job security, promotions, private office space, social climate [Ozutku 2012]. Mahaney and Lederer [2006] claim that of all the extrinsic rewards pay is the most significant to employees. They give some examples of types of pay as extrinsic rewards such as competitive salaries, pay rises, merit bonuses, and indirect forms of payment as compensatory time off. In the current research Mahaney and Lederer’s view was adopted. Therefore the focus is on pay as the representative of extrinsic rewards.

1.2. Pay and pay satisfaction The word pay refers to all forms of compensation, such as direct cash payments, indirect remuneration, non-cash payments and the amount of pay rises and the process by which the compensation system is administered [Williams, McDaniel, and Nguen 2006]. Harell and Daim [2010] identified pay and bonuses as the top tangible’ motivators of employees. Heneman and Judge [2000] claim that Williams, McDaniel, and Nguen’s [2006] definition of pay is too broad and recommended its replacement with a multi-dimensional conceptualization of pay, including the dimension of pay satisfaction, focusing on each dimension separately. They base this suggestion on their findings which, show a weak positive relationship between an employee’s objective pay level and his/her subjective pay satisfaction. Miceli and Lane’s [1991] definition fits Heneman’s and Judge’s suggestion. According to them, pay satisfaction is the amount of overall positive or negative sentiments that individuals have towards their pay. Employee pay dissatisfaction was found to be associated with undesirable outcomes such as turnover, absenteeism, reduced performance, decreased commitment and theft [Currall et al. 2005; Heneman and Judge 2000; Miceli and Mulvey 2000]. On the other hand employees’ pay satisfaction was found to be related to favorable behaviour such as organizational citizenship [Nwankwo

108

Economics and Business Review, Vol. 1(15), No. 4, 2015

et al. 2013] and successful adaptation to change [Smits and Bowden 2015]. Hence, rather than focus on level of pay, it may be useful to gain a deeper understanding of pay satisfaction. Equity Theory [Adams 1963] is often used to show the relationship between pay and the employee’s pay satisfaction. Adams claims that individuals want to be compensated fairly for their contribution; therefore employees seek the balance between what they invest in their jobs in terms of effort, knowledge, skills, and performance and what they get as an outcome through extrinsic and intrinsic compensation. Equity is achieved when the input-output ratio of the individual is perceived by that individual to be equivalent to that of relevant other employees in the organization or external to it [Adams 1963; Singh and Loncar 2010]. The empirical data regarding the Equity theory is inconclusive, mostly with regard to the overpayment condition [Gray 2000]. Some Scholars support the notion that employees evaluate their pay through social comparisons [Van Buren 2008; Williams, McDaniel, and Nguen 2006] whilst other found results that reject it. Ahmad et al. [2010] in a recent study found that employees are likely to feel ‘rewarded’, ‘motivated’ and satisfied when they know that they are able to get fair pay which reflects the amount of work that they do. They found that employees are particularly concerned with pay discrimination and that this may lessen their level of motivation to do their job well. Harell and Daim [2010] identified pay and bonuses as the top tangible motivators of employees. According to Deci’s self-determination theory; people generally see monetary incentives as a controller of their behaviour and this frustrates the need for autonomy and leads to less autonomous motivation and a decrease in their performance [Falk and Kosfeld 2006; Weibel, Rost, and Osterloh 2010]. Weibel, Rost, and Osterloh [2010] finding support this notion. They demonstrate that monetary compensation generally has a negative effect on performance with the exception of uninteresting tasks or ones that do not allow autonomy. When the task is boring and when the self-control is low, money will have a positive effect on the employees’ motivation and performance. Ledford, Gerhart, and Fang [2013] accept Deci’s data yet suggest a different explanation. According to Ledford et al. the employee interpretation of the extrinsic rewards shape his reaction to them. When the employee believes that the rewards provide positive information about their own competence and self-control over results, intrinsic motivation will increase. If the employee interprets the results as indicating external control, decreasing their feelings of self-control and competence, intrinsic motivation decreases. Gray [2000] suggests that performance based bonus payments are often perceived by the employee more as a risk to lose money, which he expects to receive, and not as an incentive. Therefore employees may perform less well in a cash-bonus condition. Hertzberg [1962] examines another aspect of pay satisfaction, by suggesting that salary is mainly a hygienic factor but may also be a motivational factor when

O. Gov, The complex relationship between intrinsic and extrinsic rewards

109

the employee perceives it as an expression of his achievements. Gray [2000] in his summary of Hertzberg’s views of monetary rewards, states “it meant more than money; it meant a job well done; it meant that the individual was progressing in his job” [p. 16]. Ozutku [2012] accepts this notion. He claims that money is not always a motivational factor – today’s workers wish to find internal satisfaction, rather than external compensation. These employees usually already receive better wages, therefore managers need to use inner motivators for the job in order to motivate them. Ozutku also suggests that even though it is generally thought that money plays an important role in the motivation of unqualified individuals with low wages, money may not always provide the expected motivation. For these individuals practices like the employee of the month, successful employee award ceremonies and sincere celebration of success can be more effective than monetary rewards. Lawler [2003] suggests two factors which determine the reward’s effect on employees: the first is the amount of reward that is given and the second is the importance attributed by an employee to each reward. Others focus on the employee’s attitude to his rewards and claim that the importance of his rewards in general, and his pay specifically, is affected by national differences [Hofstede 2001; Adler and Gundersen 2008; Van de Vliert, Van Yperen, and Thierry 2008]. Van de Vliert, Van Yperen, and Thierry [2008] demonstrate that wages are perceived as more important in poor countries with cold climates, like Azerbaijan, than in rich countries with cold climates like Canada and in poor and rich countries with temperate climates like Uganda and Singapore. His explanation of the results regarding cold countries is that “increasingly cold residential areas result in increased needs for thermal comfort, nutrition, and health, as well as an increased need for money to satisfy these basic biological needs” [Van de Vliert, Van Yperen, and Thierry 2008: 81]. They rationalize their results using the Vroom [1964] expectancy theory, “money would be more important in poor countries where homeostatic goods are more significant and desirable (valence), and where pay enables employees to buy these goods (instrumentality)” [Van de Vliert, Van Yperen, and Thierry 2008: 81]. Based on this literature review is possible to summarize that pay satisfaction is a subjective attitude that develops based on both cultural and demographical aspects. Pay satisfaction has significant implications for both employees and organizations. Therefore there is a need to expand our understanding of pay satisfaction and its relationship to intrinsic rewards.

1.3. The connection between intrinsic rewards and pay satisfaction The literature review regarding the connection between the two variables indicates a significant relationship but the direction of it is inconclusive. Williams et al. [2006] based on a meta-analysis of 28 correlations of pay satisfaction from 203 studies suggested the existence of a positive relation between perceived job

110

Economics and Business Review, Vol. 1(15), No. 4, 2015

characteristics known to be intrinsic rewards such as autonomy, task feedback and skill variety and pay satisfaction. Khan et al. [2013] in a recent study on the relationship between rewards and performance of banking employees in Pakistan show a mix of extrinsic (monetary compensation, recognition, company policies, relationship with supervisor and co-worker and working conditions) and intrinsic rewards (ability utilization, variety, moral value, activity, authority, social service and security). Ledford, Gerhart, and Fang [2013] also claim, based on a large meta-analysis, that there is no trade-off between intrinsic and extrinsic rewards. They demonstrate that motivation is the sum of the two types of rewards. Extrinsic rewards such as pay could increase or decrease intrinsic rewards based on a variety of elements including the individual’s relevant perceptions. Ruiz-Palominoet, Saez-Martinez, and Martinez-Canas [2013] found a positive relationship between job characteristics which are known as intrinsic rewards and pay satisfaction. According to them employees perceive an enriched job as a reward in itself and a source of self-realization therefore they are less concerned with pay issues and feel content with their pay. Nevertheless “The recurring theme in popular management literature is that extrinsic rewards diminish intrinsic motivation and this problem is so serious that it can render extrinsic incentives for performance of any kind as ineffective or even counterproductive” [Ledford, Gerhart, and Fang 2013: 17]. Kohn [1993] was one of the first to indicate extrinsic rewards as actually reducing intrinsic motivation. According to him the more closely we tie compensation (or other rewards) to performance, the more damage we do. Experiments have shown that, after inducing temporary states of extrinsic motivation, participants exhibit poorer concept attainment and impaired problem solving and lower creative output [Baer, Oldham, and Cummings 2003]. Thomas further claims that, when organizations wanted only compliance from workers, they bought it with money and other extrinsic rewards, whilst intrinsic motivation was found to drive employees to self-determination and job satisfaction and to try to make a difference, being innovative and more passionate towards the job [Thomas, Jansen, and Tymon 2009]. Zhou, Zhang, and Sanchez [2009] tested the impact of extrinsic and intrinsic rewards on employee creativity. They conclude that basic amounts of tangible rewards are necessary to encourage employee creativity yet too many rewards could decrease the employee’s selfmotivation and, as a result, his creativity. In addition they show that intrinsic rewards such as providing learning resources, job rotation and tutoring have a more significant effect on employees creativity performance than the extrinsic rewards. Their third and most important conclusion is that extrinsic and intrinsic rewards have an interference interaction effect on innovative behaviour. Frey [2012] based upon a literature review of 128 experiments discusses the “crowding out” effect of tangible rewards on intrinsic motivation. He claims that intrinsic motivation occurs when people choose to act in a certain way just for the pleasurable feeling that the act evokes in them or because they

O. Gov, The complex relationship between intrinsic and extrinsic rewards

111

internalized social norms. Providing monetary rewards for specific activities may crowd out both the self-esteem and the feeling of responsibility of the individual for his own fate. As a result a decrease in creativity, innovation and effort would be seen. Baer, Oldham, and Cummings [2003] in their study of the relationship between the types of rewards and creative performance describe a complex picture which is effected by the task and its level of difficulty. Extrinsic rewards will ‘crowd-out’ the intrinsic motivation and reduce the creative performance of employees. Yet, when intrinsic motivation is low since the job is simple, extrinsic rewards will give a certain degree of control to the employees which will lead to higher levels of creative performance. Based on the different data the current research aims to explore and determine the relationship between intrinsic rewards and pay satisfaction.

2. Research methodology 2.1. Data collection procedure The population, for the purpose of the study, is composed of Israeli employees in general, with no specific focus on any professional or demographic group. The sampling used a convenience technique. An invitation to participate in an online survey was sent by Facebook and in addition the invitation included a request to share the survey with others. The data were collected during three months. One hundred and sixty responses were received but five were disqualified due to mistakes in their completion.

2.2. Participants The total sample was 156 individuals from different organizations in Israel. Charts No. 1 and No. 2 demonstrate the sub-samples distributed according to demographic characteristics: The majority of the participants were women (62.8%) and employed (83.3%). As shown in table 1, the average age of the participants is 32.9 (sd. 11.5). The age range is 20–73. It is clear that the percentage of young participants (aged 30 and under) is very high (59%) whilst the percentage of participants aged 50 and above is relatively low (11.5%). The distribution of educational level demonstrates that the sample is heterogeneous: 21% of the participants have a high school education, 22% have a professional education, 39% have a Bachelor’s degree and 18% have other higher education (Table 1). Table 2 demonstrates that a little less than a quarter of the sample report that they are managers, as compared to 77% that have no management role. For the 32 managers the number of subordinates is 1–100, with an average of 18.4 and sd. of 24.0. The average time of the participant within their organization is 4.6

112

Economics and Business Review, Vol. 1(15), No. 4, 2015

Table 1. The samples distribution according to demographic characteristics Variables

Values

Gender

male

37.2

female

62.8

30 and under

59.0

31 to 40

19.2

41 to 50

10.3

51 and up

11.5

high school

21.2

professional

21.8

bachelor’s degree

39.1

master’s degree and above

17.9

employed

130

83.3

self-employed

16.7

Age

Education

Type of employment

Table 2. The sample’s distribution by the occupational related variables Variables

Values

Employment terms (N – 141)

outsourcer’s and temporary employee

12.8

company’s employee

123

87.2

service

17.3

sales

12.2

traditional manufacturing

8.3

Field of occupation

Number of employees in the organization

Hierarchy (N = 156)

high-tech

12.8

education

10.9

administration

17.9

others

20.5

10 and under

12.8

10 to 50

23.7

51 to 100

16.7

101–500

19.9

500 and up

26.9

not in a managerial position

120

76.9

manager

23.1

113

O. Gov, The complex relationship between intrinsic and extrinsic rewards

cont. Table 2 Number of subordinates (N – 32)

Time with the organization (N – 140)

Time in the position (N = 149)

5 and under

34.4

6 to 10

28.1

11 to 50

25.0

50 and up

12.5

a year and under

29.3

1.5 to 3 years

35.0

3.5 to 5 years

13.6

5.5 to 10 years

10.7

10.5 and more years

11.4

a year and under

30.9

1.5 to 3 years

36.9

3.5 to 5 years

10.1

5.5 to 10 years

6.7

10.5 and more years

15.4

years (sd. 5.8) and in their position or a similar position in another organization 5 years (sd 7.4). The range of time in the organization is 1–30 years. The range of time in the role is between 1 to 50 years.

2.3. Measures A self-administered questionnaire was used in order to achieve the objectives of the study. The questionnaire was chosen based on a comprehensive review of extant literature. The questionnaire is divided into two parts. Part I includes questions that aim at obtaining details regarding the level of intrinsic reward (and its four dimensions: meaningfulness, choice, competence, and progress) and pay satisfaction. Part II consists of questions seeking information about demographics (such as age, gender, educational qualifications, occupation). All the data were collected within the same time frame. The questionnaire includes the following tools: 2.3.1. The Work Engagement Profile (WEP) Intrinsic reward was measured using the Thomas Work Engagement Profile (WEP) [Thomas, Jansen, and Tymon 2009]. The WEP is a 24-item questionnaire that measures intrinsic reward which individuals may receive directly from their work. The WEP tests four dimensions of intrinsic reward using a seven-point Likert response scale ranging from one (strongly disagree) to seven (strongly agree):

114

Economics and Business Review, Vol. 1(15), No. 4, 2015

–– Sense of meaningfulness is measured by items No. 3, 7, 10, 13 18, 21 of the questionnaire. It consists of items such as ‘I care about what I’m doing’ and ‘what I am trying to accomplish is meaningful to me’. –– Sense of choice is measured by items No. 4, 8, 12, 16, 19, 22 of the questionnaire. It consists of items such as ‘I feel free to select different paths or approaches in my work’ and ‘I exercise a lot of choice in what I do’. –– Sense of competence is measured by items No. 2, 5, 11, 14, 17, 23 of the questionnaire. It consists of items such as ‘I am good at my work’ and ‘I am doing my work capably’. –– Sense of progress is measured by items No. 1, 6, 9, 15, 20, 24 of the questionnaire. It consists of items such as ‘I am making good progress on my projects’ and ‘My tasks are moving forward’. The WEP is considered accurate with a reliability coefficient greater than 90 [Thomas and Tymon 1994]. Statistical relationships between the WEP scores and other variables provide strong evidence that the instrument is measuring what it was designed to measure [Thomas, Jansen, and Tymon 2009]. 2.3.2. Pay Satisfaction – PSQ Pay satisfaction was measured by the 4 – items (No. 36–39 in the questionnaire of the study) which consist of the “pay level satisfaction” dimension based on the PSQ – Pay Satisfaction Questionnaire of Heneman and Schwab [1985]. Participants are asked to indicate their satisfaction regarding items such as ‘Your take-home pay’ or ‘Size of your current salary’ on a 5-point Likert scale from 1 (very dissatisfied) to 5 (very satisfied). The efficient alphas for the satisfaction from the pay dimension exceed 0.90.

3. Results 3.1. Descriptive statistics Table 3 demonstrates the measures of central tendency and variability of the independent variable, intrinsic rewards and its four dimensions; sense of meaningfulness, sense of choice, sense of competence and sense of progress – varied from 1 to 7. The mean values show a strong tendency toward the high scores of both the variable and its four dimensions (between 4.7 for sense of choice to 5.6 for sense of competence). This tendency is consistent with the distribution of WEP scores as published by Thomas, Jansen, and Tymon [2009]. The standard deviation ranged from 1.08 for sense of competence to 1.59 for sense of meaningfulness. The findings regarding pay satisfaction display average values which are not extreme in either direction were: mean 3.1, standard deviation 1.14.

115

O. Gov, The complex relationship between intrinsic and extrinsic rewards

Table 3. Measures of central tendency and variability of intrinsic rewards and its dimensions and pay satisfaction Mean

Median

Mode

S.D.

Min.

Max.

Intrinsic rewards

5.1

5.4

5.5

1.15

1.6

6.9

Sense of meaningfulness

5.1

5.5

7.0

1.59

1.0

7.0

Sense of choice

4.7

5.0

1.48

1.0

7.0

Sense of competence

5.6

5.8

6.7

1.08

2.5

7.0

Sense of progress

5.1

5.3

6.0

1.26

1.8

7.0

Pay satisfaction

3.1

3.0

4.0

1.14

1.0

5.0

3.2. Reliability Cronbach’s alpha coefficient is calculated to verify the internal reliability of the results for all the variables. High values of alpha indicate that the items are highly correlated with true scores [Collis and Hussey 2003]. Table 4, below shows the Cronbach’s alpha coefficient of all the variables. Table 4. The independent and the dependent variables: scales, items, serial number and Cronbach’s alpha values Variable no.

Variable form

Items

Scale

Cron bach’s alpha

Intrinsic rewards

Var1

Int_Rew

1–24

1–7

0.96

Sense of meaningfulness

Var2

Int_Rew1

3, 7, 10, 13, 18, 21

1–7

0.94

Sense of choice

Var3

Int_Rew2

4, 8, 12, 16, 19, 22

1–7

0.93

Sense of competence

Var4

Int_Rew3

2, 5, 11, 14, 17, 23

1–7

0.92

Sense of progress

Var5

Int_Rew4

1, 6, 9, 15, 20, 24

1–7

0.92

Pay satisfaction

Var6

Pay Sat

36–39

1–5

0.98

Variables/dimensions

Nunnally [1994] suggests that instruments used in exploratory research should have a reliability of 0.7 or better. As can be seen in Table 4 the Cronbach’s alpha estimated for the current study shows acceptable levels of reliability of 0.80 to 0.98.

3.3. Pearson’s correlation between the variables A Pearson’s Correlation Coefficient was calculated between the intrinsic rewards and their four dimensions and pay satisfaction. Table 5, below, shows the relationship between pay satisfaction and the dimensions of intrinsic rewards.

116

Economics and Business Review, Vol. 1(15), No. 4, 2015

Table 5. Pearson’s correlation coefficient between the independent variable and its dimensions and the dependent variable Independent variable Dependent variable

Intrinsic rewards

Sense of meaning fulness

Sense of choice

Sense of competence

Sense of progress

0.114

0.100

0.257**

–0.141

0.107

Pay satisfaction * p < 0.05. ** p < 0.01.

As shown in Table 5, no relationship was found between intrinsic rewards and three of their dimensions and pay satisfaction. Yet sense of choice has a positive significant relationship to pay satisfaction (r = 0.257; p < 0.01).

3.4. The two regression tests for the dependent variable Following the testing of the Pearson coefficient correlations two prediction models were developed in order to test the relationship between the two kinds of reward using a multiple linear regression for the dependent variable. The first prediction formula includes the independent variable, intrinsic rewards and some of the control demographic variables. The model was developed in two stages: in the first stage, five control demographic variables were included in the regression: age, seniority in the employment position, type of employment (salaried/self-employed), employment terms (outsourcer’s and temporary employee /organization’s employee), hierarchy (manager/non-manager). Table 6. First linear regression for pay satisfaction Variable

The whole sample Step 1

Step 2

Age

0.216

0.195

Type of employment

0.072

0.044

Employment terms

–0.152

–0.199*

Hierarchy

0.096

0.083

Seniority in the employment position

–0.131

–0.144

Intrinsic rewards 2

0.169

0.06

0.08

F(5,130) = 1.66ns

F(6,129) = 1.96ns

* p < 0.05. ** p < 0.01.

O. Gov, The complex relationship between intrinsic and extrinsic rewards

117

Table 7. Second linear regression of pay satisfaction Intrinsic rewards dimensions

Beta

Sense of meaningfulness

–0.027

Sense of choice

0.297**

Sense of competence

–0.436**

Sense of progress 2

0.252

0.16

F(4,151) = 7.29** * p < 0.05. ** p < 0.01.

Those variables were selected based on their relevance to the dependent variable according to the Pearson correlations. In the second stage the independent variable, intrinsic rewards added to the regression. The findings, as shown on Table 6, indicate that the whole model is not significant and it predicts less than 10% of the variance of the pay satisfaction. Hence, Pay Satisfaction is not predicted by Intrinsic Rewards factors. However the control demographic variable, Employment Status, has a significant contribution to the prediction of Pay Satisfaction F(6,129) = 1.99; p < 0.05. The second prediction formula includes the four independent dimensions; sense of meaningfulness, sense of choice, sense of competence. Table 7 indicates that the regression model is significant [F(4,151) = 7.29; p < 0.01] and explains 16% of the pay satisfaction variance. The two dimensions that are responsible for most of the explained variance are: sense of choice (β = 0.30; p < 0.01) with positive beta and sense of competence (β = –0.44; p < 0.01) with negative beta. As the sense of choice rises, the level of pay satisfaction increases. On the other hand, as the sense of competence rises the pay satisfaction decreases.

4. Discussion 4.1. The relationship between intrinsic rewards and pay satisfaction In the current study, a recurring theme has been chosen and with it in mind, the expectation was that the relationship between intrinsic rewards and pay satisfaction is negative. Consistent with Hertzberg’s [1962] view, the results of this research show no significant relationship between intrinsic rewards and pay satisfaction. The following paragraphs present, some explanations for the results. In this research pay satisfaction has been treated as a one-dimensional construct. Yet there is a growing body of research to support the notion that

118

Economics and Business Review, Vol. 1(15), No. 4, 2015

pay satisfaction is a multi-dimensional construct [Heneman and Judge 2000; Williams, McDaniel, and Nguyen 2006; Ruiz-Palomino, Saez-Martinez, and Martinez-Canas 2013]. Ruiz-Palomino, Saez-Martinez, and Martinez-Canas [2013] found that the “supervisor Ethical Leadership” serves as a moderator variable which strengthens the positive influence of “job motivation potential” (autonomy, feedback, task significant etc.) on pay satisfaction. Their findings demonstrate that a supervisor who practises higher levels of ethical leadership makes a difference in terms of pay satisfaction and its effect is more significant to employees with high intrinsic rewards than to employees with low intrinsic rewards. Williams, McDaniel, and Nguyen [2006] suggest a more complicated model regarding pay satisfaction based on equity theory [Adams 1965] and discrepancy theory [Lawler 2003]. According to them pay satisfaction is a complex variable which been influenced by a variety of elements such as job characteristics, equity, organizational policies, actual payment and the employees perceptions. A study of pay satisfaction should regard as a multi-dimensional phenomenon. Hence we could assume that the one-dimensional attitude in the present research could be a limitation which could account for the insignificant results. Future research regarding the relationship between Intrinsic Rewards and Pay Satisfaction is recommended to measure Pay Satisfaction as a multidimensional construct. Another possible explanation for these results could be derived from the focus on pay satisfaction and not on pay amount. Pay satisfaction entails the feeling of contentment one has with one’s pay [Nwankwo et al. 2013]. The rationale for measuring pay satisfaction instead of pay amount was that regardless of the actual amount an individual is earning his/her work-related attitude will be affected by satisfaction level. Employees react differently to the same pay amount and compensation according to their perceptions of equity, their pay expectations, prior pay experiences, nationality and culture and even gender [Nwankwo et al. 2013]. Hence focusing on the subject of pay satisfaction was thought to be an appropriate decision. However considering the insignificant finding a different approach would be recommended for future studies that includes testing the amount of pay as a dependent variable instead or in combination with pay satisfaction. The interpretation of the these research results should take into consideration the age distribution of the sample. Fifty nine percent of the subjects were thirty years old or less. Another nineteen percent were between the ages of thirty-one to forty. This is a major deviation toward generation Y that could lead to some doubts regarding the validity of the current research results for older employees. A fourth interpretation of the results would be that that they are accurate and reflect actual reality for Israeli employees. The literature review regarding the relationship between intrinsic and extrinsic rewards in general, and more specifically between the four intrinsic rewards and pay satisfaction, is based

O. Gov, The complex relationship between intrinsic and extrinsic rewards

119

mainly on research conducted in different countries such as the United States, Pakistan and Italy. None was conducted in Israel. Israel’s culture, according to Hofstede’s dimensions [2001] has a low long time orientation (38), very low power of distance (13), middle individualism (54) and masculinity (47) and a high long-term orientation (81). According to Hofstede Israeli society is characterized by a strong belief in the ideal of self-actualization which is strongly connected to the self-management process and the intrinsic rewards model. The low power of distance score of the Israeli culture also supports the importance of intrinsic rewards to its employees: they expect to have autonomy, to be consulted, and empowered. On the other hand Israelis masculinity characteristics give a significant importance to extrinsic rewards such as salary, working conditions and status symbols. Based on the current research results we can conclude that according to Hofstede’s culture model and dimensions both types of rewards are relevant to Israeli employees yet intrinsic rewards have a stronger effect and that there are no significant relationships between the two type of rewards. The current research demonstrates that there is no significant relationship between intrinsic rewards and pay satisfaction. Nevertheless the results identified other important relationships of both theoretical and practical relevance regarding employee reward. The implication of these results will be discussed in the next sections.

4.2. Pay satisfaction and employment terms The research demonstrates that pay satisfaction is significantly predicted by the variable employment terms. The results of the study demonstrate that outsourcer’s and temporary workers are less satisfied with their pay in comparison with the organization’s employees. Business has massively reduced costs over the last twenty years by hiring temporary employees and contracting out its workforce [Hubbard 2012]. Foote [2004] describes the attitude of many of the temporary employees as “second class employees”. Foote’s findings demonstrate a higher tendency for voluntary turnover, less job satisfaction and a higher level of stress in comparison to in-house employees. In addition he shows that the outsourcers and temporary employees work fewer hours and earn lower wages than permanent workers. Foote’s findings could help to explain the findings of this research that the employment terms significantly predict pay satisfaction.

4.3. The contrasted effect of sense of choice and sense of competence on pay satisfaction The results of the prediction regression model indicate that the two intrinsic rewards which explained 16% of the pay satisfaction variance are sense of choice (β = 0.30; p < 0.01) and sense of competence (β = –0.44; p < 0.01). Accordingly,

120

Economics and Business Review, Vol. 1(15), No. 4, 2015

employees with high autonomy in their jobs will feel more satisfied with their pay. On the other hand, employees with a high sense of competence will feel less satisfied with their pay regardless of the actual amount of their salary. This finding could be explained by the equity theory [Adams 1963]; a worker with a sense of competence feels that his input to the organization is high, therefore he expects higher pay (output) and tend to feels unsatisfied with his actual monetary compensation. Nonetheless an employee whose sense of choice and autonomy in his job is high will tend to perceive his job as a place that empowers him and enables self-actualization and the development of skills. Hence such an employee will consider the equity equation as balanced and will tend to feel satisfied with his actual pay since he receives so much non-monetary output for his efforts at work. Practitioners who base their compensation policies on those findings have to invest more resources in empowering employees and develop their sense of choice. Unfortunately, following this direction could lead to a paradox; the empowerment process might increase the employees’ sense of competence which will lead to a decrease of pay satisfaction. Stumpf et al. [2013] findings support this notion. They demonstrated that employees who are empowered with choice might begin to assume their capabilities exceed their current position’s expectations, and as a result, their engagement level decreases and they show a higher tendency to leave.

4.4. Generation Y and rewards The interpretation of the current search results should take into consideration the age distribution of its sample; Fifty nine percent of the subjects were thirty years old or less. Another nineteen percent were between the ages of thirtyone to forty. This is a major deviation toward generation Y that could explain the results in general and the finding regarding Sense of Progress particularly. Today’s workforce is known to include four generations that work together side by side. A generation is a group of people born during the same period, usually about two decades. Because they live through the same experiences, such as economic recession, war, crises and upheavals, they tend to share similar values and behaviour [Schullery 2013]. Today’s four generations are; Traditionalists (1925–1945); Baby Boomers (1946–1964); Generation X (1965–1981); and the Generation Y or the Millennials (1982–1999). Generation Y tends to want an intellectual challenge, needs to succeed, strives to make a difference, and seeks employers who will further their professional development [Meier and Crocker 2010]. Accordingly to the current study’s findings, setting and achieving personal goals matters to employees from Generation Y, as does performing meaningful work that has the potential to contribute to a better world [Allen 2004]. Furthermore Generation Y’s employees are accustomed to being involved in family decisions, consequently they expect

O. Gov, The complex relationship between intrinsic and extrinsic rewards

121

a similar amount of authority and choice in their work [Hurst and Good 2009]. As we can see Thomas’ two dimensions; sense of meaningfulness and sense of choice are a fundamental component of Generation Y’s nature. However The Millennial has other attributes, as people who grew-up playing computer video games they expect work to be fun [Tapscott 2009]. An expectation that could explain the results of this research regarding sense of progress. Progress is fulfilling but it also usually involves a delay of gratification. Delay of gratification is known to be frustrating since it is usually a long process, which requires patience and willingness to defer the needs and pleasures of today for the success and advancement of tomorrow [Frostmeier, Drobetz, and Maercker 2011]. It is often the opposite of enjoyable activities and, as such, it could be unrewarding especially to Generation Y. It is important to remember that generation Y’s characteristics are broad trends of this specific population and that individuals sharing the same generation are shown to have a large diversity in their values, characteristics and behaviours.

4.5. Limitation and further research There is no single, best research strategy and limitations always exist. The results of this study should be analyzed in the light of its main limitations, such as the use of self-reporting measures. Furthermore, participation in the research was voluntary; self-selection may have contributed to a response bias. The integration of objective measures, such as absenteeism and real turnover, would increase confidence in the validity of the resulting data [Goffin and Gellatly 2001]. Another limitation is the lack of experimental and longitudinal design. As a result, it is impossible to draw inferences about causality [Mathieu and Taylor 2006]. Future studies should consider experimental methods, and test longterm effects on the relationships examined in the present study and on outcomes such as actual turnover behavior and absenteeism, peer or supervisor review, employees’ performance and organizational membership. Further research is required in order to learn more about the influence of the two intrinsic rewards types on pay satisfaction in order to build a rewards system which will encourage employees to be both less dependent and more skillful without diminishing their motivation and satisfaction from their salaries. It may be suggested that additional value can be derived from focusing on this subject in the context of virtual teams which Kauffmann [2015] suggests are becoming one of the main solutions of organizations addressing the competitive demands of the twenty first century business reality. In addition, as discussed, there is a need to further explore the relationship between intrinsic rewards and pay satisfaction in order to conclude that there are no significant relationships between those two important variables.

122

Economics and Business Review, Vol. 1(15), No. 4, 2015

Conclusions The results imply that the subject of rewards is a complicated and that the relationship between intrinsic rewards and pay satisfaction is complex and therefore there is a need for further exploration. Meanwhile managers may need to tailor a motivational strategy whenever possible. They need to make sure that they are in tune with their employees’ motivators and not just blindly following a motivational theory. Asking employees what motivates them, listening and acting based on their responses is very important for shaping organizational reward policies and achieving the right balance between extrinsic and intrinsic rewards.

References Adams, J.S., 1963, Toward an Understanding of Inequity, Journal of Abnormal and Social Psychology, vol. 67: 422‑436. Adler, N.J., Gundersen, A., 2008, International Dimensions of Organizational Behavior, 5th ed., Thomson South-Western Educational, Mason, OH. Ahmad, S., Ali, I., Rehman, K., Khan, M.A., Waseemullah, A., 2010, Insecure Job and Low Pay Leads to Job Dissatisfaction Interdisciplinary, Journal of Contemporary Research in Business, vol. 11: 222–263. Alfandi, A.M., Alkahsawneh, M.S., 2014, The Role of the Incentives and Reward System in Enhancing Employee’s Performance – A Case of Jordanian Travel and Tourism Institutions, International Journal of Academic Research in Business and Social and sciences, vol. 4, no. 4: 326–341. Allen, P., 2004, Welcoming Y, Benefits Canada, vol. 28 no. 9: 51–53. Baer, M., Oldham, G.R., Cummings, A., 2003, Rewarding Creativity: When Does It Really Matter? The Leadership Quarterly, vol. 14, no. 45: 569–586. Bandura, A., 1986, Social Foundations of Thought and Action: a Social Cognitive Theory, Englewood Cliffs, NJ: Prentice-Hall. Bandura, A., 1997, Self-efficacy: the Exercise of Control, W.H. Freeman, New York, NY. Bratton, J., Gold, J., 2012, Human Resource Management Theory and Practice, Macmillan, Basingstoke, NY. Collis, J., Hussey, R., 2003, Business Research, 2th ed., Palgrave MacMillan, NY. Conchie, S.M., 2013, Transformational Leadership, Intrinsic Motivation, and Trust: A Moderated-mediated Model of Workplace Safety, Journal of Occupational Health Psychology, vol. 18, no. 2: 198–210. Currall, S.C., Towler, A.J., Judge, T.A., Kohn. L., 2005, Pay Satisfaction and Organizational Outcomes, Personnel Psychology, vol. 58, no. 3: 613‑640. David, W., 2006, Mental Pressure Caused by Employees’ Duty, Management, vol. 15, no. 3: 28–35. Davis, J., Wilson, S.M., 2000, Principals and Efforts to Empower Teacher: Effects on Teacher Motivation and Job Satisfaction and Stress, The Clearing House, vol. 73, no. 6: 349–353.

O. Gov, The complex relationship between intrinsic and extrinsic rewards

123

Deci, E.L., 1971, Effects of Externally Mediated Rewards on Intrinsic Motivation, Journal of Personality and Social Psychology, vol. 18, no. 1: 105–115. Deci, E.L., Ryan, R.M., 1985, Intrinsic Motivation and Self-determination in Human Behavior, Plenum, NY. Deci, E.L., Ryan, R.M., 1991, A Motivational Approach to Self: Integration in Personality. in: Dienstbier, R. (ed.), Nebraska Symposium on Motivation. Perspectives on Motivation, University of Nebraska, Lincoln, NE, vol. 38: 37–288. Falk, A., Kosfeld, M., 2006, The Hidden Costs of Control, The American Economic Review, vol. 96, no. 5: 1611–1630. Foote, D.A., 2004, Temporary Workers: Managing the Problem of Unscheduled Turnover, Management Decision, vol. 42, no. 8: 963–973. Frey, B.S., 2012, Crowding Effects on Intrinsic Motivation, Journal of Labor Politics, vol. 20, no. 2/3: 91–98. Fried, Y., Ferris, G.R., 1987, The Validity of the Job Characteristics Model: A Review and Meta-analysis, Personnel Psychology, vol. 40, no. 2: 287–322. Frostmeier, S., Drobetz, R., Maercker, A., 2011, The Delay of Gratification Test for Adults: Validating a Behavioral Measure of Self-motivation in a Sample of Older People, Motiv Emot, vol. 35: 118–134. Goffin, R.D., Gellatly, I.R., 2001, A Multi-rater Assessment of Organizational Commitment: are Self-report Measures Biased?, Journal of Organizational Behavior, vol. 22: 437–451. Gray, R., 2000, Workplace Motivation, www.rodericgray.com/workplacemotivation. pdf retrieved 28/04/2015. Hackman, J.R., Oldham, G.R., 1976, Motivation Through the Design of Work: Test of a Theory, Organizational Behavior and Human Performance, vol. 16: 250–279. Harell, G., Daim, T.U., 2010, HDM Modeling as a Tool to Assist Management with Employee Motivation: The Case of Silicon Forest, Engineering Management Journal, vol. 22, no. 1: 23–33. Heneman, H.G., Judge, T.J., 2000, Incentives and Motivation. in: Rynes, S., Gerhart, B. (eds.), Compensation in Organizations: Progress and Prospects, New Lexington, San Francisco, CA: 61–103. Heneman, H.G., Schwab, D.P., 1985, Pay Satisfaction: Its Multidimensional Nature and Measurement, International Journal of Psychology, vol. 20: 129–141. Herzberg, F., 1962, Work and the Nature of Man, Thomas Y., Crowell, New York, NY. Hofstede, G., 2001, Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organizations Across Nations, 2th ed., Sage, Thousand Oaks, CA. Hubbard, L., 2012, Industrial, Australian Nursing Journal, vol. 20, no. 12: 27–28. Hurst, J.L., Good L.K., 2009, Generation Y and Career Choice – The Impact of Retail Career Perceptions, Expectations and Entitlement Perceptions, Career Development International, vol. 14, no. 6: 570–594. Kauffmann, D., 2015, How Team Leaders Can Improve Virtual Team Collaboration through Trust and ICT: A Conceptual Model Proposition, Economics and Business Review, vol. 1: 52–75. Khan, I., Shahid, M., Nawab, S., Wali, S.S., 2013, Influence of Intrinsic and Extrinsic Rewards on Employee Performance: The Banking Sector of Pakistan, Academic Research International, vol. 4, no. 1: 282–291. Kohn, A., 1993, Punished by Rewards, Houghton Mifflin, Boston, MA.

124

Economics and Business Review, Vol. 1(15), No. 4, 2015

Lawler, E.E., 2003, Treat People Right, Jossey-Bass Inc, San Francisco, CA. Ledford, G.E., Gerhart, B., Fang, M., 2013, Negative Effects of Extrinsic Rewards on Intrinsic Motivation: More Smoke than Fire, World at Work Journal, vol. 91: 877–951. Mahaney, R.C., Lederer, A.L., 2006, The Effect of Intrinsic and Extrinsic Rewards for Developers on Information Systems Project Success, Project Management Journal, vol. 37, no. 4: 42–54. Malhorta, N., Budhwar, P., Prowse, P., 2007, Linking Rewards to Commitment: An Empirical Investigation of Four UK Call Centers, International Journal of Human Resource Management, vol. 18, no. 12: 2095–2127. Manz, C.C., 1991, Leading Employees to Be Self-managing and Beyond toward Establish of Self-leading in Organizations, Journal of Management Systems, vol. 3, no. 3: 15–24. Masvaure, P., Ruggunana, S., Maharaj, A., 2014, Work Engagement, Intrinsic Motivation and Job Satisfaction among Employees of a Diamond Mining Company in Zimbabwe, Journal of Economics and Behavioral Studies, vol. 6, no. 6: 488–499. Mathieu, J.E., Taylor, S.R., 2006, Clarifying Conditions and Decision Points for Meditational Type Inferences in Organizational Behavior, Journal of Organizational Behavior, vol. 27, no. 8: 1031–056. Meier, J., Crocker, M., 2010, Generation Y in the Workforce: Managerial Challenges, The Journal of Human Resource and Adult Learning, vol. 6, no. 1: 68–78. Miceli, M.P., Lane, M.C., 1991, Antecedents of Pay Satisfaction: A Review and Extension, in: Rowland, K., Ferris, K. (eds.), Research in Personnel and Human Resources Management, Greenwich, CT: JAI Press: 235–309. Miceli, M.P., Mulvey, P.W., 2000, Consequences of Satisfaction with Pay System: Two Field Studies, Industrial Relations, vol. 39: 62‑87. Milkovich, G., Newaman, J.M., 2008, Compensation, 10th ed., McGraw-Hill, New Delhi, India. Mundhra, D.D., Jacob, W., 2011, Intrinsic Motivators in the Indian Manufacturing Sector: An Empirical Study, Journal of Organizational Behavior, vol. 10, no. 2: 21–40. Njanja, W.L., Maina, R.N., Kibet, L.K., Kageni, N., 2013, Effect of Reward on Employee Performance: A Case of Kenya Power and Lighting Company Ltd., Nakuru, Kenya, International Journal of Business and Management, vol. 8, no. 2: 41–49. Nunnally, J.C., 1994, Psychometric Theory, 4th. ed, McGraw-Hill, New York, NY. Nwankwo, B.E., Sydney-Agbor, N., Agu, S.A., Aboh, J.U., 2013, Influence of Pay Satisfaction and Length of Service on Organization Citizenship Behavior of Bankers, International Journal of Academic Research in Business and Social Sciences, vol. 3, no. 9: 238–245. Ozutku, H., 2012, The Influence of Intrinsic and Extrinsic Rewards on Employee Results: An Empirical Analysis in Turkish Manufacturing Industry, Business and Economics Research Journal, vol. 3, no. 3: 29–48. Ruiz-Palomino, P., Saez-Martinez, F.J., Martinez-Canas, R., 2013, Understanding Pay Satisfaction: Effects of Supervisor Ethical Leadership on Job Motivating Potential Influence, Journal Bus Ethics, vol. 118: 31–43. Ryan, R.M., Deci, E.L., 2000, Self-determination Theory and the Facilitation of Intrinsic Motivation, Social Development, and Wellbeing, American Psychologist, vol. 55, no. 1: 68–78.

O. Gov, The complex relationship between intrinsic and extrinsic rewards

125

Schullery, N.M., 2013, Workplace Engagement and Generational Differences in Values, Business and Professional Communication Quarterly, vol. 76, no. 2: 252–265. Senge, P., 1990, The Fifth Discipline: the Art and Practice of the Learning Organization, Doubledday Currency, New York, NY. Singh, P., Loncar, N., 2010, Pay Satisfaction, Job Satisfaction and Turnover Intent, Industrial Relations, vol. 65, no. 3: 470–490. Smits, S.J., Bowden, B.E., 2015, A Perspective on Leading and Managing Organizational Change, Economics and Business Review, vol. 1, no. 2: 3–21. Stone, D.N., Deci, E.L., Ryan, R.M., 2008, Beyond talk: Creating Autonomous Motivation through Self-determination Theory, Journal of General Management, vol. 34: 75–91. Stumpf, S.A., Tymon, W.G., Favorito, F., Smith, R.R., 2013, Employees and Change Initiatives: Intrinsic Reward and Feeling Valued, Journal of Business Strategic, vol. 34, no. 2: 21–29. Tapscott, D., 2009, Grown up Digital: How the Net Generation is Changing Your World, McGraw-Hill, New York, NY. Thomas, K.W., 2009, Intrinsic Motivation at Work: What Really Drives Employee Engagement, Berrett-Koehler, San Francisco, CA. Thomas, K.W., Jansen, E., Tymon, W.G. Jr., 1997, Navigating in the Realm of Theory: An Empowering View of Construct Development, Research in Organizational Change and Development, vol. 10: 1–30. Thomas, K.W., Jansen, E., Tymon, W.G. Jr., 2009, Work Management Profile, Cpp, Mountain View, Ca. Thomas, K.W., Tymon, W.G. Jr., 1994, Does Empowerment Always Work: Understanding the Role of Intrinsic Motivation and Personal Interpretation, Journal of Management Systems, vol. 6, no. 2: 1–13. Thomas, K.W., Velthouse., B.A., 1990, Cognitive Elements of Empowerment: An Interruptive Model of Intrinsic Motivation, Academy of Management Review, vol. 15, no. 4: 666–681. Tymon, W.J., Stumpf, S.A., Doa, J.P., 2010, Exploring Talent Management in India: The Neglected Role of Intrinsic Rewards, Journal of World Business, vol. 45, no. 2: 109–121. Van Buren, H., 2008, Fairness and the Main Management Theories of the Twentieth Century, Journal of Business Ethics, vol. 82: 633‑644. Van de Vliert, E., Van Yperen, N.W., Thierry, H., 2008, Are Wages More Important for Employees in Poorer Countries with Harsher Climates?, Journal of Organizational Behavior, vol. 79: 79–94. Weibel, A., Rost, K., Osterloh, M., 2010, Pay for Performance in the Public Sector: Benefits and (Hidden) Costs, Journal of Public Administration Research and Theory, vol. 20: 387–412. Wiersma, U.J., 1992, The Effects of Extrinsic Rewards in Intrinsic Motivation: A Metaanalysis, Journal of Occupational and Organizational Psychology, vol. 65: 101–114. Williams, M.L., McDaniel, M.A., Nguyen, N.T., 2006, Meta-analysis of the Antecedents and Consequences of Pay Level Satisfaction, Journal of Applied Psychology, vol. 91, no. 2: 392–413. Zhou, Y., Zhang, Y., Sanchez, A.M., 2009, How do the Reward Approaches Affect Employees’ Innovative Behavior? – An Empirical Study in Chinese Enterprises, Academy of Management Proceeding, vol. 8: 1–6.

Economics and Business Review, Vol. 1 (15), No. 4, 2015: 126–153 DOI: 10.18559/ebr.2015.4.8

Improvement of the communication between teachers and students in the coaching programme and in a process of action research1 Michal Lory2

Abstract : The paper seeks to explain the process of the coaching programme for teachers in elementary3 schools in Israel in an action research, for the improvement of the communication between the students and teachers. The objective of the present research is to launch a group coaching programme amongst fellow teachers in the school, including to design and examine this experience in the format of an action research with emphasis on the potential influence of the nature of the communication between the teachers and students and as a result on the academic achievement of the students. The research presented in this paper is based on the qualitative and quantitative research approaches. The validation of the research findings is performed primarily on the basis of triangulation between the interviews held with a group of teachers and the personal questionnaires4 distributed to the teachers and their analysis [Ellis, Adams, and Bochner 2011]. Keywords : group coaching, action research, research programme, reflection. JEL codes : H75, H52, J24, K12.

Introduction Coaching is defined according to the International Federation of Coaches in research studies dedicated to the understanding of the concept of coaching, as a process of ongoing relations between the coach and the coachee, relations that focus on the achievement of a personal vision or goal and the realization of a desire. Coaching includes concepts, models, and results-oriented tools from the fields of philosophy, thinking, professional didactics, strategic planning, supportive guidance, sociology, psychology, management and market 1

Article received 17April 2015, accepted 15 September 2015. Shalhevet, 1 HaCramim st., Shoham, Israel, e-mail: Loryjm@gmail.com. 3 In Israel the elementary school is from first grade to sixth grade and is for children six to twelve years of age. 4 The questionnaire is available from the author upon request. 2

M. Lory, Improvement of the communication between teachers and students

127

ing. The extensive integration creates a synergy effect that enables a high level of achievements to be reached in a relatively short period of time. Coaching provides proven results, increased productivity and improved achievements. Group coaching as action research is, according to Tsellermeir and Tabac [2004], important in the building of a professional learning community. To create dialogue and openness amongst teachers who participate in action research it is necessary to build a discussion community that will enable systems of democratic and respectful relations, which will lead the teachers to see their work and behaviour from the perspective of others. It is necessary to create a space for discussion amongst the teachers coached that is open, authentic and caring and although it is critical, that will create willingness to listen, whilst increasing the teachers’ confidence in themselves and belief in their knowledge and rehabilitating their ability to handle criticism in parallel to their subjective and professional educational perceptions [Tsellermeir and Tabac 2004; Wilson et al. 2012]. The system of relations between the teachers of an open discussion and dialogue on their work in the class will be the reflection that will be performed on the basis of the experiences of thinking and will be used to examine the teachers’ knowledge and to improve the educational and scholastic processes in the class. The reflection will enable the drawing of conclusions, insights and learning from the guidance and from the study and in-service training days about professionalization [International Coaching Federation 2010]. Coaching helps the teachers avoid the engagement in teaching in a procedural or technical manner, through reflection, which in turn allows the analysis of the behaviour and approaches and in-depth examination of the needs of both the teacher and the learners and creates the ability to meet these needs [Hayes 2010]. Taking into consideration the research objective the following research questions were posited: –– What are the characteristics and products of experience in the group coaching programme amongst the fellow teachers? –– Does the group coaching programme amongst fellow teachers contribute to the improvement of the academic achievement of the students from the perspective of its beneficial influence on the nature of the communication between teachers and students? Coaching is intended to improve the achievements and the reduction of student failures in tests and in their studies, through the improvement of the quality of teaching and the creation of interpersonal communication between teachers and students. For this purpose the educational system must be mindful of the field of human resources. The field of human resources includes engagement in the disclosure and development of knowledge, in the identification of the workers’ characteristics and in their adjustment to their roles, in a way that the maximum quality of functioning will be obtained. In addition

128

Economics and Business Review, Vol. 1(15), No. 4, 2015

the field of human resources is intended to implement instruction sessions for the workers, development and adoption of ways of instruction relevant to the areas of occupation, their professional promotion and effective and efficient handling of human resource factors in the organization. The strategy today of human resources assumes that those who engage in the field need to know to manage the organizational processes on three levels: TLC – technology, learning, and counselling. The teachers’ subjective perceptions are an outcome of their professional training and experience in actual teaching. Action research that addresses them, examines them, and disputes them may erode their confidence and inspire objection instead of leading to cooperation [Wilson et al. 2012]. It should be noted that the format of the action research is modern, effective, and especially useful in the field of education, specifically as an effective aid in the promotion of the teachers and consequently also for the promotion of the organizational product at the level of the school and of the educational system [Carr and Kemmis 2003; Stringer 2004; Levi 2006]. As a member of the educational staff in the school and as the social education coordinator in the school, I became aware of the issue of the problem of communication between teachers and students. In particular I was forced to deal with the problem by myself in the framework of my position and I was aware of the existence of the problem from conversations with my colleagues who frequently complained about the difficulties in the communication and in the transfer of knowledge to their students. As a part of the group of teachers who participated in the programme, I am the researcher and initiator of the plan. I work as the homeroom teacher of a first grade class and social coordinator in the school. I have professional training in the instruction of groups and coaching from Bar-Ilan University. Taking into consideration the unique characteristics of the teaching profession and educational activity, the action researchers in education encounter the need to bridge the gap between the theory that developed in academic institutions in which the teachers acquired their professional training and the everyday experience of the teachers in the field that continuously offers a broad range of challenges and issues that require the constant development of skills beyond the closed set of professional tools [Stringer 2004; Allpert 2006]. In the framework of the action research, the educational system in general and the teachers in particular are given the possibility of acting as a knowledge community, where not only its practical but also its abstract products are disconnected from the area of activity. In this the action researchers use the potential of a beneficial transformation both on the level of the individual and in the broad social circles of the school, the educational system and society, in light of the fact that they rely on self-reflection of field factors in relation to the perceptions, patterns of behaviour, ways of action and work methods as topics of methodical critical analysis [Carr and Kemmis 2003; Levi 2006].

M. Lory, Improvement of the communication between teachers and students

129

In terms of the research method within the present research the qualitative research method was employed to examine the experience in the group coaching programme amongst fellow teachers, as well as a quantitative research component as an aid in the evaluation of the influence of the programme on the students’ academic achievements. Regarding the qualitative research paradigm this is a method that enables the research and understanding of the human experience with emphasis on the outlooks and subjective interpretations of the given social reality. The starting point of the qualitative research paradigm is that it is not possible to quantify subjective experience but it can be conceptualized on the basis of the data that can be collected using qualitative instruments such as observations and interviews [Sabar Ben-Yehoshua 2001; Shkedi 2003]. The qualitative research method was selected in the case of the present research study because the present research study focuses on the subjective experience of teachers in a group work process, when this is a new experience for the teachers and thus it cannot be clarified using a closed constellation of defined variables. This is the research method that suits and is useful for the study of complex social phenomenon, including the examination of topics that have not yet been studied and investigated in-depth and thus an extensive corpus of knowledge on them does not exist [Shkedi 2003]. In addition, in light of the fact that the qualitative research method provides an answer in cases of the research of complex social phenomena on the basis of components of experience and subjective experience of those who act in the given social context, it is customary to use it precisely in the framework of action researches in general and in action researches in the field of education in particular [Stringer 2004; Levi 2006]. In respect of the integration of the quantitative research component, this is a commonly accepted procedure in social research and its aim is to utilize the advantages of the two research paradigms [Bryman 2012; Beyt-Marom 2014]. Since the present research study has the goal of examining the potential of the influence of the group coaching programme amongst fellow teachers on the academic achievement of their students, it was decided to use the quantitative research method and to examine the differences in academic achievement whilst comparing the semi-annual test scores in the core subjects amongst the students of the teachers who had participated in the coaching programme and those of the students of the teachers who did not participate in the programme in terms of the degree of improvement in the scores as a measure of the amount of improvement in the academic achievement of the students. The action research is intended to bring about a professional change amongst the teachers and improve their abilities to cooperate and reflect on their behaviour in a way that will increase the teachers’ desire to improve the communication between them and their students. For this purpose I intend to present descriptions of the instructional behaviour in the school, to enable interpretation of occurrences and to illustrate to the teachers the importance of the reflection

130

Economics and Business Review, Vol. 1(15), No. 4, 2015

and the action research in the change of fixed patterns and in the adoption of additional instruction methods. These are vital to the improvement of teaching, good communication, the professional and personal encouragement of the teachers and the improvement of the learning amongst the students. Listening to others and being exposed to different experiences of teachers who participate in the programme, along with the development of a recognition of the likelihood of new ways of action additional to those they adopt, may lead to the establishment of a belief in teaching and in a reinforcement of the teachers’ community which will provide moral support for its members [Gallucci et al. 2010]. The article presents the theory of the coaching programme and the importance of the action research for the participating teachers. Section 1 describes the concept of peer coaching and its influence on the coachees. Section 1.1 explains the definitions of the concept of peer coaching and its influence on the coachees. Section 1.2 presents the process of implementation of the coaching programme for the teaching staff. Section 1.3 explains the effectiveness of the peer coaching programme. Section 2 explains the importance of circle of action research in the coaching sessions with the teaching staff participating in the programme. Section 2.1 explains the importance of reflection in action research in the peer coaching programme. Section 2.2 describes the methods of teaching in the coaching process for the improvement of communication between the teaching staff and the students. Section 2.3 describes the programme structure, the programme sessions with the participating teaching staff and the goal of every session and every stage of the programme. Section 3 ends with the summary of the indices built from the participation of the teaching staff in the programme.

1. The peer coaching programme and action research 1.1. The concept of peer coaching and its influence on the coachees The concept of coaching defines a professional process, where the basis of its mission is to achieve a change that will lead to optimal self-fulfilment and significant changes in the coachee’s life, in a way that will reflect his aspirations, desires, and self-defined goals. Coaching reflects a professional relationship between the coach and the coachee, a relationship that incorporates the use of tools and models for the improvement of the coachee’s conduct. Coaching is used to define the coachee’s vision and mission and define his personal values, on which basis an applicative action programme is designed and built, enabling this realization and implementation. The main objective of coaching is to create in the coachee’s life a sustainable change that leads to optimal self-fulfilment and to exceptional achievements in his life, his career paths, and the different frameworks in which he acts. The coaching process is based on an ongoing

M. Lory, Improvement of the communication between teachers and students

131

system of relations with the coach, in a way that enables the coachee to define a vision, mission, and scale of values in the areas of his life that require, in his opinion, definition and change [Israel Coaching Association 2004]. Coaching is intended to enable the coaches to cope with a crisis in process, to cope with the objections that the coachee presents in his unwillingness to deal with changes and to relinquish previous perceptions and in a refusal to adopt approaches that will create opportunities for the formation of a personal vision and processes for the change of existing conduct and situations. The coach must involve the coaching participants in the processes of making decisions and clarify the importance of reflection to the process. They must discuss with the participants the difficulties, provide encouragement and support and enable them to create and express different interpretations of occurrences and to base this upon data collected during the action research between the teachers who participate in the programme [Israel Coaching Association 2004]. Levin [as cited by Burnes 2004] addressed the increased depth of the factors of human resistance to the creation of changes and the ways in which it is possible to overcome the resistance, so as to enable processes of change. McNiff [2010] maintained that coaching processes are based on learning and assume that although the coachee already knows a lot, he must find answers to his reflections and therefore he needs a coach, who will focus the processes of the search for answers. The coach in essence is a supporter or critic who listens to the coachee’s ideas and helps him find alternatives of behaviour. Thus the coach also experiences the reciprocal learning process along with the coachee’s experiences [McNiff 2010]. Like Levin [as cited by Burnes 2004] and McNiff [2010], Tsellermeir and Tabac [2004] addressed the teachers’ perception and asserted that the engagement in coaching and action research contradicts the messages that the teachers receive as students during their studies. They form a perception of the knowledge that they absorb as cohesive knowledge that cannot be disputed, whilst in action research they are supposed to create knowledge and theories for the situations that they experienced in the past. To cope with the perceptual fixation, Tsellermeir and Tabac [2004] maintained that it is necessary to present the coaching as a long-term multi-stage process, which is expressed in ups and downs and is characterized by conflicts.

1.2. Process of implementation of the coaching programme The peer coaching programme is a unique concept that has barely been investigated amongst teachers in schools in Israel. The model is only beginning to be introduced in Israel and is not established and institutionalized. Cooperation amongst peers is accepted in many schools and is expressed in the visits of teachers to the classes of fellow teachers, in observation of the way in which they teach, in their exposure to other approaches and in shared consultations.

132

Economics and Business Review, Vol. 1(15), No. 4, 2015

Peer coaching enables teachers to be involved in class discussion and provides support in the learning processes of their students. Peer coaching enables teachers to exchange the role of coach and coachee. During peer coaching the teachers work in the programme collaboratively and share the instruction, support and learning [Murray, Ma, and Mazur 2009]. Peer coaching is based on the examination of coaching as a professional method that has been established in research since 1983, when Joyce and Showers [1983, as cited by DeFord 2007; Hayes 2010] examined how it could be implemented in teaching, through workshops, learning sessions for the teachers, feedback, professional research and coaching practiced actively, whilst leading to change and growth of the teaching in the schools. Peer coaching is most suitable and relevant to the group of peer teachers in the school, is effective in comparison with other types of coaching, suits work within a team and meets the budget limitations. The method enables intensive coaching because all the participants are to be found in one place and because it has low operational costs and there is no need for an external coach [Grant 2012b]. The instructor in this model is a partner in the professional interaction at the clinical and academic focus, which complement one another. The academic focus addresses the field of human resources and addresses the development of an instructional staff that will participate in in-service training courses during its work, in “instructional skills, setting instructional strategies, and planning systems and tools for work seminars with the teachers” [Du Toit and Reissner 2012]. The enrichment of the teachers in the school through a coaching programme and the instructor’s participation in instruction may provide a basis for a different approach to teaching, both personal and school, which does not require special organization or great financial resources, but may cause a significant change in the teaching-learning processes [Burley and Pomphrey 2011]. The educational system focuses the control over knowledge in education, over the knowledge disciplines and over the planning of the teaching in a way that will facilitate the socialization of the learners and their preparation for life. In the past knowledge in education was emphasized as the embodiment of the main objectives of education so as to inculcate the culture and social order, through the teacher, who was the main source of knowledge. Today, to prepare the learners for the challenges of the future, the purpose of education is not the formation but the liberation of awareness, the reinforcement of critical skills and the inculcation and development of the ability of self-inquiry, curiosity, cognitive flexibility and moral and intellectual choices. The educational system is required to cope with the learners’ manner of education and their transformation into moral, caring, enlightened adults who are sensitive to others and who take responsibility for their actions [Karnieli 2010]. It appears that schools, the teaching body and the educational system do not dedicate sufficient time and resources to meet the aforementioned challenges

M. Lory, Improvement of the communication between teachers and students

133

of education. The constant race to teach knowledge and to measure student achievements does not leave time and resources in relation to the learners’ social and emotional needs, which are not the top priorities and they do not produce a scholastic atmosphere and educational climate relevant to these challenges. Teachers learn in the process of their professional training primarily in one area of knowledge and there is almost complete avoidance of the intention of oral, social, and educational engagement. There is a gap between the educational policy that should be implemented and the ability to translate it into pedagogical activity. The need to cultivate value-oriented and moral learners with sensitivity to others obligates the search for a different method of teaching that will bring the teachers to gain experience in new ways [Kfir 2011; Arnon, Frankel, and Rubin 2012; Jimenez, King, and Tan 2012]. The use of coaching in teaching helps link the teachers’ learning as a practical tool and their ability to teach interactively with their students. The coaching helps the professional development and the adjustment and appropriation of means that support the making of decisions that create changes in teaching methods and aids the shaping and improvement of the teachers’ professional experiences. The personal improvement and empowerment through coaching are today perceived as a vital element in the teachers’ professional improvement and they reflect thinking in one-directional terms that offer focus on teaching in the class over time [Grossman and McDonald 2008; Burley and Pomphrey 2011; Du Toit and Reissner 2012].

1.3. Importance of reflection in action research and its effectiveness in the peer coaching programme Reflection is the higher-order thinking about an action, before, during, or after it, intended to explain the accumulated knowledge, practical experience, planning, performance and evaluation of the action and to improve performances in the future. Reflective thinking on processes addresses the reconstruction of stages of action, deliberations, judgment in making decisions, attempts to explain and understand successes and mistakes and how to improve the process and results in the future [Hayes 2010]. Reflection is an action of self-observation (understanding and assessment) on processes of thinking and emotional and social processes and not only on learning itself. Reflection represents intentional, voluntary but not random thinking, which is expressed in questions on the activity of the past, so as to improve and promote the activity in the future. In the peer coaching programme, the reflection is re-evaluated through the teacher’s experience, the insights he reached, the intuition that guided his conduct, the experiences he experienced, and the conclusions that were reached. Reflection represents the ability of analysis, synthesis of situations, evaluation of knowledge, skills, thinking strategies, value-oriented perceptions and social positions, which are established and develop during the teaching and learning.

134

Economics and Business Review, Vol. 1(15), No. 4, 2015

There are two types of reflection: reflection during the action, which appears when a person reflects his behaviour during its occurrence with the goal of developing awareness of its existence, and reflection without action, which is the action of self-feedback or feedback on the part of somebody who was a witness of the behaviour after the occurrence and it enables the person to repeat the conduct with the goal of understanding it, describing it, analysing it and evaluating the processes of behaviour with the goal of obtaining insights and profiting from the ability to create behavioural change in the future. Reflection is a cognitive process of inquiry and helps the teachers focus on the change of processes of teaching and the adjustment of these processes to the learners’ needs. Coaching helps teachers avoid the engagement in teaching in a procedural or technical manner, through reflection, which enables analysis of behaviour and approaches, an in-depth examination of the teacher’s and learners’ needs, and produces the ability to respond to them [Hayes 2010]. The coaching programme in action research is based on reflectiveness, as an inseparable part of the teaching of qualitative research. The teachers participating in the process are active participants who are involved in the research process throughout its entire course and engage in reflective observation of the learned or researched phenomena. Therefore it is recommended to use reflective observation in the planning and performance of the process. It is important to perform similar activity in the process of the teaching of the qualitative research, with the observation of four dimensions: (1) the personal background, (2) prejudices and previous assumptions, (3) attitude to the research topic and contents and (4) influences of the behaviour of the staff participating in the research process [King and Horrocks 2010].

2. The importance of the action research circle in the peer coaching sessions 2.1. The importance of reflection in action research in the peer coaching programme Action research is a practical-applicable component that enables teachers to further the depth of their understanding of the teaching-learning processes, to improve them and to propose alternative ways of coping, which may suit a broad sector of teachers in the school. Action research is implemented for the most part in collaborations that combine teachers who have a shared interest and it constitutes a basis of action that guides the investigation of their teaching activity and professional improvement under the guidance of external academic faculty members, senior teachers, or experienced researchers [Stringer 2004]. The approach enables the presentation of different viewpoints of teachers and of people from outside their immediate system of interaction. The goal is to

M. Lory, Improvement of the communication between teachers and students

135

present a broad perspective, diverse approaches and different points of reference that will help in the construction of a methodical, trustworthy and participative research between guided teacher peers and their instructors. Action researches are intended to prevent the teaching from becoming a technical routine and they reflect the teachers’ professional knowledge, incorporate it in relevant theories and examine the complexity of the teaching work. Action researches present the processes of change required in teaching and thus promote the interests, both those of the teachers and of the schools, while they produce processes of change in the quality of the teaching [Norton 2009]. Action research consists of two processes. The first process is to identify the situation and analyse it, whilst the second process is to identify alternative and possible solutions. Action research is intended to emphasize the required change of action and recognizes that successful and effective action is based on the correct analysis of the situation, through the detection and identification of all the possible alternative solutions and the choice of the most appropriate solution. For this purpose, the person’s tangible need for change and recognition of the essential urgency of the change is necessary. The theoretical elements of action research are found in Gestalt psychology, which emphasized the need for reflection and the acquisition of new insights as a condition of the successful feasibility of the changes in situations [Burley and Pomphrey 2011]. Action research reflects a circular process of action, planned, based on the identification of the problem, the understanding of the need for change, the planning of activity for the achievement of change, the collection of data,

Figure 1. The model of three stages in action research

Source: http://creativeeducator.tech4learning.com/vo7/articles/Embracing_Action_Research

136

Economics and Business Review, Vol. 1(15), No. 4, 2015

the analysis of data, and the assessment of data, the drawing of conclusions, and the learning of lessons that will facilitate the formation of new behaviour [Burnes 2004]. The beginning of action research was in the 1940s in the United States, following the research study of the social sociologist Kurt Lewin, who combined his three stage model with action research, a term he coined in 1946, so as to create a rational and guided approach to the planning of changes [as cited by Burnes 2004]. The development of action research in education derives from the criticism that arose regarding the gap created between academic research and problems and topics that occupy the teachers in their everyday teaching work in the schools. The academic professionalization of teaching requires teachers to broaden and deepen their knowledge and personal and professional autonomy, which actually require a bridging of the gap between academic research and activity. Action researches connect between scientific academic knowledge in the field of education and the teachers’ experience, when the practical aspect has importance and validity similar to academic theoretical information [Allpert 2006]. In recent years the understanding that there is a need for the development of learning amongst teachers and for the professional development amongst teachers as well as the fact that teaching is performed through self-critical examination of the teachers’ work has meant the development of planned programmes for teacher training by the academic training institutions based on the perception of teaching as a reflective field, which goes into and studies the teaching work in depth. The study of teaching as a reflective action enables teachers to identify problems in their work and to understand and cope with them, through observation and collection of data, for the analysis of the teaching, processes of learning and learning products, in a way that will improve the quality of the teaching. For this action research is required [McNiff 2010; Tsellermeir 2011]. Tsellermeir and Tabac [2004] see the academic training of teachers to be one culture and the field of teaching another culture. In their opinion, on the basis of the opinions of Wenger [1998, as cited by Tsellermeir and Tabac 2004], the third culture is that which develops as a result of the joint learning of teachers, managers, teacher trainees and teacher trainers. The establishment of this culture requires the establishment of professional learning communities in organizations and education institutions in which the organization’s educational vision will be shaped. Through them the main goal of the organization will be achieved as a part of democratic pluralistic society, whose members have individual perceptions. The approach of society to these perceptions as an asset and a resource will facilitate the creation of institutional processes of change in this society and will constitute a part of the processes of the formation of educational leadership.

M. Lory, Improvement of the communication between teachers and students

137

This activity facilitates the formation, extension, and constant examination of temporary interpretations of knowledge [Tsellermeir and Tabac 2004; Gallucci et al. 2010].

2.2. Methods of teaching in the coaching process for the improvement of communication between teachers and their students In peer instruction there is teaching and learning. The participants show to their peers programmes that they have learned and experienced, exchange information about sources of knowledge and accept from the group emotional support that enables the clarification of emotional processes and values related to an event that the group addresses. The structure of the relations and the interactions amongst peers is at the heart of peer coaching and is based on reciprocal respect and trust [Lu 2010; Grant 2012b]. Coaching has the potential to improve the effectiveness of teaching through the maintenance of a non-judgmental system of relations based on trust and reciprocal learning and that empowers the ability for independent work with others [Gallucci et al. 2010]. Shidler [2009] addressed the sense of self-efficacy of teachers. She saw important elements in the field of teaching which can be fulfilled through coaching for teachers and which influence their students’ quality of learning. She based her work on the research of Ross [1992, as cited by Shidler 2009], who maintained that the students’ achievements increase as the teachers undergo coaching more intensively and that the topics of teaching and learning addressed in coaching influence directly the rise in the students’ achievements. Shidler [2009] maintained that the teachers’ experience in coaching improves their approach to teaching and consequently the students’ achievements. However, Shidler asserted that the sessions need to focus on specific contents, certain techniques and instructional practices. The teachers participating in the coaching need to observe other teachers, dedicate hours for counselling and joint work with other teachers and improve their reflective ability. The coaching needs to focus on cognitive activities and on the channelling of the thinking and conduct of the teachers to a defined objective and clear goal. Coaches need to be present in lessons that teachers give, to support them and to simplify the teaching techniques and dialogue ability of teachers with their students. The use of coaching to increase the teachers’ self-efficacy and improve the teaching effectiveness is a part of the educational reforms implemented around the world, such as “No Child Left Behind” in the United States. The group peer coaching enables teachers to be involved in the learning processes of the students, to be found in constant discussion with them and to be supportive of them. It enables teachers to attempt teaching using alternative methods, to create reflection regarding the new methods and to hold reciprocal observations, which are effective both for teachers and for students. Peer coaching is intended to

138

Economics and Business Review, Vol. 1(15), No. 4, 2015

make the teacher have the ability to analyse and provide reflection on the educational and professional activity and in the way that its conduct will be intentional and not random [Van-Eekelen, Boshuizen, and Vermunt 2005; Lu 2010].

2.3. Structure of the peer coaching programme The structure of the programme presented in the peer coaching process is within a cooperative and unique paradigm, in that peer coaching has a commitment to an ongoing process through exchanges in the roles of counsellor and counselled. The peers constitute a figure for imitation by one another, although the assessment does not exist perfectly, in a structured and methodical manner, despite the feedback given to the participants on their functioning. During the coaching for teachers, the effectiveness of four components is examined: teaching specific information, shaping instructional techniques and approaches, observation and examination of teaching methods, consultation and reflection. In the first year, coaching emphasized the teachers’ self-efficacy and teaching effectiveness, whilst focusing on specific contents in teaching and methods that were implemented and supported directly by the coaches in the classes [Wageman 2001; de Hann et al. 2010]. The structure of the programme suits the teaching methods that are in accord with the teaching of qualitative research in general and the teaching of action research in particular. The learning is undertaken in a computerized learning environment with the help of a variety of technological instruments. The significance of the performance of action research with peers and their place in the processes of involvement, writing and documenting enable the corpus of knowledge that exists in the team of learners to be revealed beyond the researcher himself [Glesne 2006]. The main issue that the process raises addresses significant learning processes, through adjustment between the methods and values of the qualitative research approach learned openly and the values conveyed to the learners in the method of teaching chosen for their teaching [Ellis, Adams, and Bochner 2011]. An additional issue of teaching that derives from the process addresses the essential characteristics of the interaction between the instructor and the teaching team that lead to the creation of knowledge for the existence of optimal social action. These include gradual and spacious teaching which enables dialogue with the teachers in a variety of issues. Dialogue is essential in the academic in-service training courses that accompany the teachers’ social action [Hughes 2008]. The programme was about one year in one elementary school and in an action research in which I was a partner. The population of teachers participating in the coaching programme consisted of fifteen teachers, five men and ten women. The teachers who participated in the programme included three science teachers, three mathematics teachers, three English teachers, two language

M. Lory, Improvement of the communication between teachers and students

139

teachers, two history teachers, one art teacher and one sports teacher. In terms of the professional experience as teachers, about 10% of the teachers who participated in the programme were relative newcomers, without considerable experience in education and teaching, whilst the rest were experienced teachers. The sessions were once every two weeks, for a total of twenty sessions every year. Every session lasted about an hour and a half (about two lessons). The sessions were in the teachers’ room, once a week, about two lessons in the school schedule. For a detailed description of the process, in each session, with the reasons and problems, please see the Appendix. The coaching programme operates in two main ways: 1. Work of the school staff participating in the research programme, the professionalism of each teacher participating, the skills and communication ability of the teacher in his class with his students and the teacher’s expectations from the innovative process in which he participates. 2. The work of the instructor (coach) in the process of action research who leads the process in the school and stimulates the perception of improvement of the communication between the teachers and students. He encourages the involvement of the teaching staff in educational dialogue, to listen and to find solutions through an in-service training programme, as well as to improve the students’ achievements. The coach is required to have knowledge in the field of coaching, to have proven coaching ability, to have knowledge in relevant fields – in child development, in education and in teaching. The main research tools in the programme are observation and discussions over time in the field researched, so as to collect and interpret data. In ethnographic research reciprocal relations and systems of relations develop between the researchers and the research participants. For the most part the observation of the teachers is a ‘participative observation’ and the research includes interviews. The research is intended to understand phenomena, to challenge inadequate positivist approaches and to provide appropriate answers to educational and professional questions [Allpert 2006]. The teachers who participated in the programme performed reciprocal peer criticism and experienced a reflective process in the framework of which actual difficult cases of teacher-student communication were analysed and discussed, as they arose from observation of the work of the participants themselves. The process enabled the transfer of knowledge and learning from failures and successes in a positive and supportive atmosphere, which entails the potential for an experience that encourages growth and empowers. In the framework of group discussion in the observations of the lesson, two main summative insights arose. First, the teachers defined the need for differential and relevant individualized reference according to the needs of the different students as an essential condition for teacher-positive communication. Second, the teachers emphasized the existence of a structural limitation related to the

140

Economics and Business Review, Vol. 1(15), No. 4, 2015

ability to perform differential teaching adjusted to needs, when the learning processes occur in a heterogeneous class with a rather large number of students, so that the level of crowdedness produces excessive functional load and denies the teachers the possibility of meeting the unique needs of each one of the students in the class. The evaluation of the results of the programme was undertaken in two main dimensions. In the first dimension, during the last session a summative reflective discussion was held in relation to the group process and the joint work during the programme and in relation to the results of the process against the participants’ expectations, as they were raised. In addition the participants were asked to write in their free time voluntary feedback regarding their impression of the programme and its outcomes. In the second dimension, a quantitative evaluation of the students’ academic achievements was performed through the examination of the differences between the achievements of students, of teachers who participated in the group coaching programme amongst fellow teachers and the achievements of students of teachers who did not participate.

Conclusions The topic of the research was formed, on the one hand, in light of my awareness of the issue of teacher-student communication, as a result, of both my professional and personal experience and my exposure to similar experiences of my professional colleagues, and on the other hand, in light of my training in the instruction of groups and coaching. The framework of the action research attributes importance to the researcher being part of the research field itself, with the processes that occur in it, unlike research approaches in which the researcher is an external factor looking from the sidelines. In the framework of the action research the researcher establishes the reference to the action she seeks to research. I realized that there are relatively few research works on the topic of group coaching for teachers and peer coaching amongst teachers, although some of them provide empirical support of the positive influence of coaching on the teachers’ functioning and the students’ achievements. I further realized that research studies of this type have not yet been performed in elementary schools in Israel. In parallel I performed an exploratory consultation amongst peers with whom I shared the idea of the research and I received responses of interest and support. The development of action researches in education derives from the criticism that arose in relation to the gap created between academic research and problems and topics that engage teachers in their everyday teaching work in the schools. The academic professionalization of teaching requires of the teachers a broadening and extension of their knowledge and personal and professional

M. Lory, Improvement of the communication between teachers and students

141

autonomy, which requires a bridge between the academic research and the action in practice. The action researches connect scientific academic knowledge and theoretical academic knowledge [Allpert 2006]. The ethical influences of the programme were that the teachers participating chose to make suggestions for organizational change in two main dimensions. First, the participating teachers focused on the promotion of the inclusionary education reforms that fully integrate the model of inclusion. Inclusionary education is perceived by the teachers as an educational arena that promotes communication between the teachers and the students through the strengthening of the students’ sense of belonging, promotion of equality from differences, and the creation of possibilities for achievement devoid of obstacles. Second, the teachers saw the goal of the promotion of attentive and sensitive dialogue communication as doable in the context educational activity as an activity of education for values. The understanding that it is necessary to develop learning amongst teachers and in their professional development and the fact that teaching is performed in the self-critical examination of the teachers’ work have obliged in recent years the academic training institutions to plan programmes for teacher training in recent years, which are based on the perception of teaching as a reflective field, which looks in depth at the planning of teaching and the research of same. The research of teaching as a reflective action enables teachers to identify problems in their work and to understand and cope with them, through observation and collection of data and for the purpose of the analysis of teaching, the learning process, and the learning product, in a way that will improve the quality of the teaching. For this purpose, action research is required [McNiff 2010; Tsellermeier 2011]. In group peer coaching the rise in openness towards the experiences and in the externalization of experiences was measured. In group peer coaching the great benefit that the coaching participants derived was measured. The research results were analysed through the examination of repeat elements. In the peer coaching group high achievements of commitment to achieve goals were found in the different measures that were examined, such as control, striving to achieve goals, positive reference, personal growth, environmental control, positive relations with others, goals in life, self-acceptance, autonomy, sense of mission, and activity for a goal, perseverance in the path, sense of general hope (optimism), and the reduction of negativity. Peer coaching is an important strategy and management and leadership programmes need to take it into account and include it in the frameworks intended for improvement, despite the discussion on its effectiveness although it has not been accepted as a coaching method unanimously accepted. Coaching produces flexibility in collaborative professional development (CPD) amongst teachers and helps develop inquiry-based knowledge, creates the ability to extrapolate and draw conclusions, synthesizes knowledge, avoids relying on

142

Economics and Business Review, Vol. 1(15), No. 4, 2015

fixed processes and creates continuous learning, essential for teachers today. Thus coaching has the potential to improve teaching effectiveness and maintain a non-judgmental system of relationships between teachers and students, based on reciprocal trust, mutual learning, and empowerment of independent work between students and teachers, amongst teachers and students. The basis for peer coaching is the cooperation that exists in any event between peer teachers in many schools, in the visits of teachers to the classrooms of their fellow teachers, in the observation of the teaching processes of others, in the exposure to different teaching approaches and in shared consultations. Peer coaching necessitates full communication amongst all the participants, coaches and coachees. The coaches are required to have knowledge and abilities in the field of coaching, proven coaching ability and knowledge in the fields of child development, education, and teaching. The implementation of coaching requires enough time that will be sufficient for the performance of the coaching processes that will reflect the coachees’ progress. The participants need time, privacy and a sympathetic atmosphere so as to implement the coaching process. These are also the basic themes for peer coaching, and therefore it constitutes an effective tool for the improvement of the quality of teaching and learning, promotion of the teachers’ and students’ success, through joint teamwork, inter-staff discussions, for the improvement of the teaching [Burnes 2004; Lloyd and Modlin 2012]. Peer coaching enables teachers to be involved in their students’ learning processes and teaches the teachers to be in constant discussion and in a position of support of the students. Peer coaching enables teachers to experience alternative teaching methods and to create a self-reflection and reflection with their peers in a way that will improve the teachers’ efficiency and their students’ achievements, so that the coaches and coachees benefit. The peer coaching process relies on psychological principles, such as the increase of the level of mutual awareness of the peers in the process and promotes the personal goals of the participants. It upholds the counselling of equals, sharing of responsibility, guidance and coaching, interaction on the basis of similarity or difference in the profession, status, or training and on the basis of the peers’ experience. In peer coaching there are group processes, equal and symmetrical relationships amongst the participants. The dimension of the dependence on the instructor or coach is limited, since all the peers are also coaches in their turn. The simultaneous give and take empowers the participants, creates a sense of belonging and social cohesion and reduces the sense of loneliness. Peer coaching is characterized by teaching and learning, exposure to experiences, exchange of information, sharing of knowledge sources, brainstorming, emotional support, clarification of emotional and value-oriented processes and reflection. Peer coaching enables diverse perspectives on events, proposes alternatives of action and produces commitment to an ongoing process and imitation [Murray, Ma, and Mazur 2009; Burns and Gillon 2011].

[143]

Session 1 October, about a month after the start of the studies

Number of sessions

Recruitment of people on the teaching staff, approximately 15 teachers in one school, for the continuation of the process and activation in the continuation of the programme. The instructor explains the importance of teamwork in the school as an inseparable part of professionalization

Information about concepts and their importance to the continuation of the writing of the research (Combined research, qualitative, with a quantitative component, Quan/Qual). The process is investigated to clarify whether improvement of the communication between the teachers and the students improves the students’ scholastic achievements. This approach offers diverse options for the extension of the knowledge and skills from another perspective for teachers and in reference to the improvement of the communication between them and the students [Allpert 2006]

Before the beginning of the programme. In-depth reading and review of research studies and articles on action research and on important concepts. Writing of the Review of the Literature

At the beginning of the programme for the school staff, explanation and disclosure of the topic, the failures in the students’ achievements, the innovative research of the peer coaching programme, and its implementation as goal-focused in the action research [Williams 2008]

Reason for the process

Description of the process

Research project process

Appendix: Research project in detail

The problem – Financing and resources for the programme. The solution – My participation in all stages of the programme, as a female educator, knowing the teaching staff, guiding groups, and engaging in coaching. Allotment of time and space – the teachers’ room, once every two weeks. As part

The problem with the collection of the material is that not much research information on the topic, which addresses the research goal, was found. The solution is the analysis and conclusion of the findings from my participation in the action research. The process is important in the construction of a learning team, in the teachers’ actual experience in, and in collaboration that creates dialogue and openness among the participant teachers [Tsellermeir and Tabac 2004]

Problems in all stages of the research process

[144]

Number of sessions

Description of the process

Problems in all stages of the research process

The importance of the educational diaof the school in-service training sessions logue between the staff and their students held every year in the school helps the continuation of the process and encourages and extends the involvement of all school staff for the conceptualization of the programme’s goals until its success, in two aspects: From the organizational aspect: To create observation and shared learning. From the practical aspect: To reveal information relevant to the implementation of the staff who run the programme. The Hypothesis: As the programme includes a process of evaluation and documentation, it can ensure its improvement and effectiveness, through the presentation of the findings obtained and the raising of proposals for the continuation of its activity based on the data in the field. In addition, the instructor reminds the teaching staff that it is necessary to keep a documentary journal of the entire process that occurs during the programme with the staff and instructor. Encouragement and empowerment of the teaching staff for the improvement of

Reason for the process

cont. Table

[145]

Sessions 3–4 Before the session – every teacher will Stage 2 in the process receive homework (independent work): to write on his abilities as a teacher, how he can help, from his role, the success of the process.

Session 2 Stage 1 – After the recruitment of the Stage 1 in the process educational staff in the school for the process, distribution of the open questionnaire (5–8 questions) for the teaching staff in the school, after the explanation and disclosure to them. The questionnaire is individual and every teacher participating in the process responds personally and freely

Empowerment – Giving expression to each member of the team, involvement of the team in the learning of attention, importance of belonging to the group. Williams [2008] investigated the peer

The questionnaire content will be based on the data of the scholastic achievements in the teacher’s class, the level of communication between the teacher and the student, the actual level and the desired level. The teacher must write the reason he has joined and the reason for his desire to be a partner in the innovative process

communication and success of their students’ achievements

The problem – Lack of desire to share difficulties and to change professional thinking on the part of the teachers in their training.

The problem – Reliability and validity of the respondents to the questionnaire. The solution – Research studies showed that one of the methods for examination in a qualitative research is a questionnaire at the start, middle, and end of a programme. The research programme is goal-focused and specific to the action research [Murray, Ma, and Mazur 2009; Burns and Gillon 2011]. The combination of the methods that I will present has a number of goals. The most prominent is the validation of phenomena through the cross-checking of data (questionnaires, observations, and interviews), and this is the basis for the production and presentation of a variety of data for the purpose of the interpretation, research needs, questions, extension and increased depth of accumulated knowledge, and contributes to the validity and reliability of the research [Journal of Mixed Methods Research – JMMR 2007]

[146]

Number of sessions

Reason for the process coaching model and improvement of the personal results and found that this model has advantages over other types of coaching. The level of the group achievements was significantly higher than that of another group of teachers. There was a positive influence on the group members who participated in the process. The sense of unity in the group is strengthened

Description of the process

What are his abilities in his class? What is his knowledge of the subject he teaches? Every teacher who participates in the process will write down his expectations from the coach and from the programme at the end of the process. Afterwards, he will present his responses before every member of staff participating in the process. A session to create an educational dialogue between the teacher and the team participating in the research, the abilities and talents of each teacher. A session to get to know people and coordinate expectations amongst the participants

The solution – Explanation of the importance and positive influence of the programme in each of the stages. Reading and study of research studies that showed the relation between improvement of the communication and improvement of the scholastic achievements. Shidler [2009] found an essential relationship between the effectiveness of peer coaching for teachers and the teacher’s ability to bring about an increase in the level of the students and in their achievements. His argument was that every teacher who participates in the peer coaching process over time and focuses on specific goals helps the accumulation of knowledge. The quality of learning of the students rises, there is effective teaching and theories are put into practice. There is an improvement in the teacher’s relationship with the student, and thus student achievement is improved. Creation of the dialogue between the participants as an important tool in action research. [Shidler 2009] Increase of the motivation in the creation of the dialogue as a research group for one aim – improvement of student achievement

Problems in all stages of the research process

cont. Table

[147]

The learning experience through observation facilitates the accumulation of information for the achievement of goals for the teacher, improvement of the quality of teaching, examination of the teaching methods, consultation, and reflection – to increase the depth of the familiarity with one another and to follow up after their behaviour in new situations [Wallace Foundation 2008]

Session 9–12

Session 12

Another distribution of questionnaires and collection of data, the same questions as in the questionnaire at the start of the process.

Observation of the teaching staff in three lessons recorded in the stages of the proStage 4 in the process cess. (The recording of the lessons occurs in each one of sessions 5–8.) Each time the observation is of a different teacher in his class. The lessons are analysed together with the teaching staff (sessions 9–12) and in the continuation of the programme. Analysis of the lessons will focus on the communication between the teacher and his students. Reflection is performed – to improve the doing and action [Gur-Zeev 2001]

The reading of the sources enriches and increases the depth of the teacher and can help explain extensively and find additional solutions in each one of the programme stages. Motivation to find solutions and ways to cope for better scholastic and social functioning

Sessions 5–8 Involvement in the communication diffiStage 3 in the process culties between the teacher and the student in class. Disclosure and discussion of each one of the participants. Each teacher needs to have a journal of sessions and document what is done in at every stage in the process. In every journal, the teacher will write bibliographic sources suitable for reading and study for that stage

The problem – Systemic organization of the teaching staff to participate in the observation of the teacher in one class and at the same time. Therefore – The solution – Early assessments of the researcher in the recording of the lessons and analysis of them afterwards with the group of teachers participating in the process. The coaching process is effective also in the dedication of hours to the instruction of teachers without their students’ presence and for reflection [self-observation, understanding, and assessment; Hayes 2010]

The problem – Lack of desire to share the difficulties of the teacher in his class. Erosion of the teacher’s self-confidence, as a criticism of a lack of professionalism and the effective functioning in class. The solution – Explanation of the importance of involvement and exposure of each one of the participants. Ensure success of the process by providing help through practical tools (positive thinking, listening to every student, calm conversation between the teacher and the student, and so on)

[148]

Sessions 13–16 Stage 5

Number of sessions

Discussion of problems of communication in class, from the teaching staff participating in the coaching process and discussion of the findings and of tools for coping

Additional distribution of questionnaires and collection of data on the students’ achievements

Description of the process

The peer coaching process is intended to make the teacher participating in the process have the ability to analyse and provide the reflection of his educational and professional activity in a goal-oriented way and to suggest alternatives of action [Grant 2012a]. Research studies showed that peer coaching directly influences the students’ achievements [Glazer and Hannafin 2006; Shidler 2009]. As the teachers participate in more intensive coaching, the teachers’ functioning is stronger, with an essential insight into change in the educational approach, transfer of knowledge, and reference to the emphasis of the inculcation of knowledge amongst the learners. Thus coaching facilitates the teachers’ learning process and creates opportunities for improvement of teaching and learning in the class. As the students are required to develop, advance,and acquire knowledge and skills to improve their future, the teachers are committed to improving

Reason for the process

The problem – Cooperation and reciprocity in the participating teaching staff requires the development of new perceptions and more complicated thinking than that which the teacher has by himself in the classroom. The solution – One of the significant abilities for the formation of solutions in peer coaching is establishing reciprocal relations. The personal and professional improvement of the teacher will enrich his class and the entire school. Teachers learn [Murray, Ma, and Mazur 2009; Gallucci et al. 2010]

Problems in all stages of the research process

cont. Table

[149]

The entire process of the programme, in all its stages, will be documented throughout with photographs Documentation of the lessons that were recorded.

Distribution of the last questionnaire, to the same group of teachers participating in the programme.

Session 20

Session 19

Summary of the coaching process in the action research performed and built with the team. Identification of all the possible alternative solutions and choice of the most appropriate solution during the change.

Sessions 17–20 Summary Stage 6 Action research ensures the effectiveness of the change, only on the level of setting it as a group value and makes all the participants a part of the process [Burnes 2004] Comparison of differences and indicators amongst the questionnaire answers at the start and middle of the process. Cross-checking data between the students’ achievements before and after the coaching process is performed using a quantitative research tool that combines qualitative action research but complements the research needs, explains the relation between the group coaching process and the improvement of the students’ learning achievements, and facilitates the analysis of the methodology and the presentation of the findings [Creswell and Plano Clark 2007]. Help teachers who teach in the coming years. The research will address the management of the staff in one elementary school

their abilities and skills in ways that will lead them to a different way of thinking and responding to the students’ needs and to bridging the gap between the learning experiences and the goals of the curricula. In this way the students’ achievements will be improved [Fullan 2002; Murray 2012] The problem – The development of action research in education derives from the criticism raised in relation to the gap created between the research process and problems of communication in the class and topics that engage teachers in the everyday teaching work in the school that necessitate bridging the gap between the research process and actual work. Development in learning is necessary amongst teachers, as is professional development [McNiff 2010]. The solution – Construction of an educational vision of the organization (the school), when the main goal is to enable shared learning of teachers, with pluralistic perceptions, as an asset and resource that will help in the processes of change and will constitute a part of the processes of the formation of educational leadership [Galluci et al. 2010]

[150]

Number of sessions

Description of the process in the process of group coaching. The research is conducted by the coach who explains, guides, observes and directs the participants to a feeling of belonging and to success of the programme. In addition the school principal needs to build a management system for the group coach. The principal joins the programme stages, listens, is interested, observes the programmeâ&#x20AC;&#x2122;s participants, and organizes methodically for the coming years a system for the teaching staff with goals of empowerment for the improvement of the communication between the teachers and students. (Morning talks, active recesses with talks, etc.). The management system will enable the processing of data between the teachers who participated in the coaching process and teaching staff that did not participate. There is the creation of a dialogue and discussions on the programme, and thus all the teachers who learn in the school will learn from the programme and will experience the coaching process for the improvement of the communication between themselves and the students, which leads to the improvement of achievements

Reason for the process

Problems in all stages of the research process

cont. Table

M. Lory, Improvement of the communication between teachers and students

151

References Allpert, B., 2006, Two Ways of Participation in Qualitative Research in the Schools: The Researcher as Ethnographer and the Researcher in the Action Team, in: Levi, D. (ed.), Action Research: Theory and Practice, Philosophical Connections and Methodologies between Action Research and the Qualitative Research Paradigm, Mofet Institute: 197–213. Arnon, R., Frankel, P., Rubin, E., 2012, To Be or Not to Be (a Teacher)? The Image of the Teaching Profession as an Attractive Occupation, Research Paths, no. 18, Mofet Institute: 33–44. Beyt-Marom, R., 2014, Research Methods in the Social Sciences: Research Principles and Styles, The Open University, Tel Aviv. Bryman, A., 2012, Social Research Methods, Oxford University Press. Burley, S., Pomphrey, C., 2011, Mentoring and Coaching in Schools: Professional Learning through Collaborative Inquiry, Routledge, Taylor and Francis Group. Burnes, B., 2004, Kurt Lewin and the Planned Approach to Change: A Re-appraisal, Journal of Management Studies, vol. 41, no. 6: 977–1002. Burns, L., Gillon, E., 2011, Developing a Teaching Agenda for Coaching Psychology in Undergraduate Programs, The Coaching Psychologist, vol. 7, no. 2: 90–106. Carr, W., Kemmis, S., 2003, Becoming Critical: Education Knowledge and Action Research, Routledge. Creswell, J.W., Plano Clark, V.L., 2007, Designing and Conducting Mixed Methods Research, Sage, Thousand Oaks, CA. De Hann, E., Bertie, C., Day, A., Sills, C., 2010, Critical Moments of Clients and Coaches: A Direct-Comparison Study, International Coaching Psychology Review, vol. 5, no. 2: 109–128. DeFord, D.E., 2007, Learning through Coaching: Ongoing Professional Development for Literacy Coaches, The Primer, Massachusetts Reading Association, vol. 36, no. 2: 1–24. Du Toit, A., Reissner, S., 2012, Experiences of Coaching in Team Learning, International Journal of Mentoring and Coaching in Education, vol. 1, no. 3: 177–190. Ellis, C., Adams, T.E., Bochner, A.P., 2011, Autoethnography: an Overview, Qualitative Social Research, vol. 12, no. 1. Fullan, M., 2002, The Role of Leadership in the Promotion of Knowledge Management in Schools, OECD Conference, March: 18–19. Gallucci, C., DeVoogt, M., Van Lare, Yoon, I.H., Boatright, B., 2010, Instructional Coaching: Building Theory about the Role and Organizational Support for Professional Learning, American Educational Research Journal, December, vol. 47, no. 4: 919–963. Glazer, E.M., Hannafin, M.J., 2006, The Collaborative Apprenticeship Model: Situated Professional Development within School Settings, Teaching and Teacher Education, vol. 22: 179–193. Glesne, C., 2006, Becoming Qualitative Researchers: an Introduction, 3rd ed., Pearson Education, Boston. Grant, A.M., 2012a, An Integrated Model of Goal-Focused Coaching: an Evidence-Based Framework for Teaching and Practice, International Coaching Psychology Review, vol. 7, no. 2: 146–165.

152

Economics and Business Review, Vol. 1(15), No. 4, 2015

Grant, A.M., 2012b, Australian Coaches’ Views on Coaching Supervision: A Study with Implications for Australian Coach Education, Training and Practice, International Journal of Evidence Based Coaching and Mentoring, vol. 10, no. 2: 17–33. Grossman, P., McDonald, M., 2008, Back to the Future: Directions in Research in Teaching and Teacher Education, American Educational Research Journal, vol. 45: 184–205. Gur-Ze’ev, I., 2001, Normalizing Education, Counter Education, and Reflection, Studies in Education, vol. 5, no. 1: 133–158. Hayes, R.S., 2010, The Study of Literacy Coaching Observations and Interviews with Elementary Teachers, Ph.D. Dissertation, Georgia State University, Atlanta. Hughes, S.A., 2008, Maggie & Me: A Black Professor and a White Urban School Teacher Connect Autoethnography to Critical Race Pedagogy, Educational Foundations, vol. 22, no. 3–4: 73–95. International Coaching Federation – ICF, 2010, http://www.coachfederation.org. Israel Coaching Association, 2004, Curriculum in the Profession of Human Resources Management, The Ministry of Education, Culture, and Sport, The Administration for Science and Technology, http://www.ilcc.org.il/home/doc.aspx?mCatID=9951. Jimenez, E., King, E.M., Tan, J.P., 2012, Making the Grade, Finance and Development, vol. 49, no. 1: 1–4. Karnieli, M., 2010, Curiosity and Investigativeness: Foundation Stones in the Empowerment, Ramot, Ra’anana. Kfir, D., 2011, A Fateful Search: Society in Israel Is Looking for Good Teachers – A Collection of Articles, Mofet Institute. King, N., Horrocks, C., 2010, Interviews in Qualitative Research, Sage, London. Levi, D. (ed.) 2006, Action Research: Theory and Practice, Philosophical Connections and Methodologies between Action Research and the Qualitative Research Paradigm, Mofet Institute. Lloyd, C.M., Modlin, E.L., 2012, Coaching as a Key Component in Teachers’ Professional Development Improving Classroom Practices in Head Start Settings, OPRE Report, New York. Lu, H.L., 2010, Research on Peer Coaching in Preservice Teacher Education – A Review of Literature, Teaching and Teacher Education, vol. 26: 748–753. McNiff, J., 2010, Action Research for Professional Development: Concise Advice for New and Experienced Action Researchers, 3rd ed., September Books, Dorset. Murray, J., 2012, Professional Learning Opportunities in U.S. Independent Schools, Lambert, London, England. Murray, S., Ma, X., Mazur, J., 2009, Effects of Peer Coaching on Teachers’ Collaborative Interactions and Students Mathematics Achievement, Journal of Education Research, vol. 102, no. 3: 203–212. National Education Technology Plan, http://www.ed.gov/technology/netp-2010. Norton, L.S., 2009, Action Research in Teaching and Learning: A Practical Guide to Conducting Pedagogical Research in Universities, Routledge, New York and London. Sabar Ben-Yehoshua, N., 2001, Trends and Currents in Qualitative Research, Dvir, Tel Aviv. Sabar Ben-Yehoshua, N., Dushnik, L., 2006, Informed Consent – a Principle that Requires Change and Adjustment to Action Research, in: Levi, D. (ed.), Action Research: Theory

M. Lory, Improvement of the communication between teachers and students

153

and Practice, Philosophical Connections and Methodologies between Action Research and the Qualitative Research Paradigm, Mofet Institute. Shidler, L., 2009, The Impact of Time Spent Coaching for Teacher Efficacy on Student Achievement, Early Childhood Education Journal, vol. 36: 453–460. Shkedi, A., 2003, Words That Try to Touch: Qualitative Research – Theory and Implementation, Ramot, Tel Aviv. Stringer, E., 2004, Action Research in Education, Pearson/Merrill/Prentice Hall, Upper Saddle River, NJ. Tsellermeir, M. 2011, Analysis of Data in Action Research: Story of a Journey for SelfDiscovery, Research Paths, no. 17: 42–52. Tsellermeir, M., Kozulin, A. (eds.), 2004, Learning in the Social Context: Development of the Higher Psychological Processes of Lev Vygotsky, HaKibbutz HaMeuchad Press, Tel Aviv. Tsellermeir, M., Tabac, I., 2004, We will Do It between Us! Or How Action Research Builds a Professional Learning Community in the Framework of Collegiality in a Field College, Mofet Institute. Van Eekelen, I.M., Boshuizen, H.P.A., Vermunt, J.D., 2005, Self-Regulation in Higher Education Teacher Learning, Higher Education, vol. 50: 447–471. Wageman, R., 2001, How Leaders Foster Self-Managing Team Effectiveness: Design Choices versus Hands-on Coaching, Organization Science, vol. 12, no. 5: 559–577. Wallace Foundation, 2008, Becoming a Leader: Preparing School Principals for Today’s Schools. Williams, P., 2008, The Emergence of Coaching as a New Profession and Its Global Influence, in: Kinder, A., Hughes, R., Cooper, C.L. (eds.), Employee Well-being Support: a Workplace Resource, Chichester: John Wiley & Sons: 285–296. Wilson, K.P., Dykstra, J.R., Watson, L.R., Body, B.A., Crais, E.R., 2012, Coaching in Early Education Classrooms Serving Children with Autism: A Pilot Study, Early Childhood Education Journal, vol. 40: 97–105.

Economics and Business Review, Vol. 1 (15), No. 4, 2015: 154–156 DOI: 10.18559/ebr.2015.4.9

BOOK REVIEWS

Barney G. Glaser, Choosing Classic Grounded Theory: a Grounded Theory Reader of Expert Advice, CA: Sociology Press, Mill Valley 2014 This book is designed for the researcher seeking a methodology to aid in their development of substantive theories. Grounded theory methodology can be utilized in either qualitative or quantitative research. A difficulty researcher’s face is deciding which grounded theory methodology to select. Unfortunately the majority of texts available do not even identify the version of grounded theory that is being put forth and the novice researcher is left unaware of the stark differences between methodologies. This book takes us back to the roots of grounded theory and why it is important to stay true to the original methodology. Dr. Glaser as cofounder of grounded theory explains the added value the researcher gains by using Classical Grounded Theory (CGT). The book outlines how and why different versions of grounded theory developed and the strengths of CGT as the only true grounded theory methodology. Dr. Glaser puts forward that all other grounded theory methodologies are variations of qualitative data analysis (QDA). This book provides the much needed clarity on what is and is not grounded theory. The book takes a pragmatic approach to answering key questions for both the experienced and novice researcher. Will CGT fit my data and area of research? Can I use CGT for my PhD? What questions should I ask to determine if CGT is suited for my research? Do I need a mentor? Can the

methodology fit the time frame of my research? Dr. Glaser’s new book allows both the novice and expert researcher to better understand the role of CGT as a robust research methodology. This fifteen chapter 439 page book takes the reader through the process from selection to application of the methodology. The challenges and myths around the methodology are clearly outlined and the book identifies what is and is not CGT. In the world of technology a google of “grounded theory” will produce a range of books offering the researcher an abundance of grounded theory methodologies and techniques to utilize. What is not apparent to the novice researcher is the epistemological issues that separate the various versions of grounded theory. Dr. Glaser points out that while similarities appear on the surface other forms of grounded theory put forth are variations of QDA and break the governing rules of true grounded theory. This is not to say they don’t have value but should be seen as a form of QDA. This book provides the reader with a clear understanding of grounded theory and why researchers should take care not to blend or slur together what are totally different methodologies. Joining Dr. Glaser in this book are contributions by four outstanding experts in the field of grounded theory (Dr. Christiansen, Dr. Holton, Dr. Lowe and Dr. Simmons). Each expert has chapters

Book reviews

dedicated to the selection and use of CGT. For the PhD researcher a wide range of examples and supporting documentation is provided by those who have successfully used CGT methodology for their PhD dissertation. Chapters one and two clarify grounded theory myths and the basic rules in selecting CGT as a methodology. Dr. Glaser points out that CGT is based on conceptualization and therefore it produces “enduring concepts for years compared to the stale dated descriptions” (Glaser 2014: 49) associated with QDA. The PhD researcher is given particular focus in these two chapters citing excellent dissertation examples and a candid approach to some of the challenges that the PhD researcher may face in selecting CGT as a methodology. For researchers new to grounded theory Dr. Glaser strips away much of the confusion around the different versions of grounded theory and provides sound advice on the use of mentors. Researchers who have the opportunity to attend workshops organised by Dr. Glaser cite the benefits they received and why it is strongly recommended. Dr. Glaser advocates that such workshops will benefit the new researcher, however, the methodology can be learned with or without a mentor. In fact mentors who do not understand CTG can potentially do more harm than good. Chapter three is a contribution by Dr. Christiansen taking the reader through the research procedures, stages and unique terminology of CGT. Warnings are given for the novice CGT researcher of some of the common errors in the use of CGT. The use of technology for example “may lead to a built-in pre-framing” a problem that exists with QDA based methodologies. Dr Glaser in Chapter four addresses the hallmark issue of conceptualization contrasted with descriptive generalization. The importance of being conceptual

155 is a core of CGT. The “GT researcher can start with the details of an existing case and constantly compare, generate a core category and its properties on an issue and start a substantive theory” (Glaser 2014: 144). This chapter brings out the importance of conceptualization. An excellent outline is given on how descriptive cases can be utilized as input data for CGT. Chapters five to ten provide CGT background that can be used by the experienced and novice researcher. The desire by researchers in the social sciences to emulate the scientific world of research draw them to QDA methodologies that profess to be grounded theory but whose roots are firmly in QDA. Chapter five is focused on the novice researcher and in particular the PhD student and the importance for the supervisor to provide encouragement and not to act as a block to their ability to be creative and original. Dr. Glaser puts forward “Make no mistake about it, the best GT is done in the hands of beginners” (Glaser 2014: 157). Dr. Christiansen in chapter six discusses ontology and epistemology and the debate on theory building. In chapters seven and eight Dr. Glaser discusses in detail the differences found in constructivist grounded theory and Straussian grounded theory and explains why these variations of QDA fail the researcher whose goal is to break through to new conceptualizations and new theories. These chapters provide the reader a clear separation between the different methodologies. The question of how and when to select CGT is covered with examples from researchers who were searching for a methodology that would allow them to work directly with the data to develop a theory beyond the limitations offer by descriptive research methods. Dr. Simmons in chapter nine provides the reader the historical perspective on the Constructivist and Straussian methodol-

156

Economics and Business Review, Vol. 1(15), No. 4, 2015

ogies and why these methodologies have gone down the QDA path. In chapter ten Dr. Lowe asks the PhD students four simple questions to help them decide if CGT is the methodology for them. If they answer “yes” to the four questions he encourages them to read on and discusses some of the potential challenges faced by PhD students who select CGT as a methodology. Dr. Glaser in chapter eleven takes on the ongoing issue of credibility in qualitative research and puts forth in his book:“credibility is not the question. The question for GT is being applicable to explaining how a main concern is continually resolved in a substantive area and its general, conceptual applicability… The theory has fit, relevance and works and is modifiable when compared, conceptually, with new data” (Glaser 2014: 317). The due diligence of the methodology and constant comparative analysis is what makes CGT work. The fact that it is based on conceptualization versus descriptive allows CGT to generate theories that go beyond the time constraints of descriptive based methodologies. This is highlighted in Chapter twelve by Dr. Holton in her discussion of CGT as a general research methodology. The reader is re-

minded again that the four pillars of CGT are; fit, work, relevance and modifiability Dr. Glaser and Dr. Holton in chapter thirteen warn researchers of the dangers of mixing methodologies, and the importance of following well-defined methodology procedures. Chapter fourteen highlights the success of CGT in development of PhD dissertations. The final chapter fifteen reviews letters Dr. Glaser wrote defending the CGT approach. In summary this is a well written book and displays the level of attention to detail that Dr. Glaser has become known for. The book is an excellent reference manual for anyone considering grounded theory as a methodology. It clearly portrays the strengths of CGT whilst being honest on some of the challenges that users of the methodology will face. Even advocates of Constructionist or Straussian grounded theory will not debate the importance of not blending these very different methodologies. I would highly recommend this book to both PhD students and their supervisors. If you are beginning to explore grounded theory this is a “must read”. Gary Evans

University of Prince Edward Island

Aims and Scope Aims and Scope Economics and Business Economics Review andisBusiness the successor Review toisthe thePoznań successor University to the Poznań of Economics University Review of Economics which wasReview which was published by the Poznań published University by the Poznań of Economics University Press ofin Economics 2001–2014. Press TheinEconomics 2001–2014. and ThBusiness e Economics Review and Business Review is a quarterly journal is a quarterly focusing on journal theoretical focusing andon applied theoretical research andwork applied in the research fields work of economics, in the fields manof economics, management and finance. agement The Review and finance. welcomes The Review the submission welcomes of the articles submission for publication of articles dealing for publication with micro, dealing with micro, mezzo and macro mezzo issues. and All macro texts are issues. double-blind All textsassessed are double-blind by independent assessedreviewers by independent prior to reviewers acceptance. prior to acceptance.

Notes for Contributors Notes for Contributors 1. Articles submitted 1. Articles for publication submittedinfor thepublication Economicsinand theBusiness Economics Review and Business should contain Revieworiginal, should contain original, unpublished workunpublished not submitted work for not publication submitted elsewhere. for publication elsewhere. 2. Manuscripts intended 2. Manuscripts for publication intended should for publication be writtenshould in English be written and edited in English in Word andand edited sentin to:Word and sent to: review@ue.poznan.pl. review@ue.poznan.pl. Authors should upload Authors twoshould versions upload of their twomanuscript. versions of One theirshould manuscript. be a comOne should be a complete text, while inplete the second text, while all document in the second information all document identifying information the author(s) identifying should thebe author(s) removed should be removed from files to allowfrom themfito lesbe tosent allowtothem anonymous to be sent referees. to anonymous referees. 3. The manuscripts 3. are Thetomanuscripts be typewritten arein to12’ be typewritten font in A4 paper in 12’ format font inand A4 be paper left-aligned. format and Pages be left should -aligned. Pages should be numbered. be numbered. 4. The papers submitted 4. The should papers submitted have an abstract shouldofhave not an more abstract than 100 of not words, morekeywords than 100and words, the keywords Journal and the Journal of Economic Literature of Economic classification Literature code.classification code. 5. Acknowledgements 5. Acknowledgements and references to grants, and references affiliation, to postal grants,and affiliation, e-mail addresses, postal andetc. e-mail should addresses, appearetc. should appear b, etc a, b, etc as a separate footnote as a to separate the author’s footnote name toa,the author’s name and should not be and included should innot the be main included list of footnotes. in the main list of footnotes. 6. Footnotes should 6. Footnotes be listed consecutively should be listed throughout consecutively the text throughout in Arabicthe numerals. text in Arabic Cross-references numerals. Cross-references should refer to particular should refer section to particular numbers: e.g.: section Seenumbers: Section 1.4. e.g.: See Section 1.4. 7. Quoted texts of7.more Quoted thantexts 40 words of more should thanbe 40separated words should frombe theseparated main body from by the a four-spaced main bodyindenby a four-spaced indentation of the margin tation as aof block. the margin as a block. 8. Mathematical notations 8. Mathematical should meet notations the following should meet guidelines: the following guidelines: – symbols representing – symbols variables representing should be variables italicized, should be italicized, – avoid symbols above – avoid letters symbols and use above acceptable letters and alternatives use acceptable (Y*) where alternatives possible, (Y*) where possible, – where mathematical – where formulae mathematical are set out formulae and numbered are set out these andnumbers numbered should thesebe numbers placed against should be placed against the right margin as... the (1), right margin as... (1), – before submitting – before the final submitting manuscript, the ficheck nal manuscript, the layout of check all mathematical the layout of formulae all mathematical carefullyformulae carefully (including alignments, (including centring alignments, length of centring fraction lines lengthand of fraction type, sizelines and and closure type,ofsize brackets, and closure etc.), of brackets, etc.), – where it would –assist where referees it would authors assistshould refereesprovide authorssupplementary should provide mathematical supplementary notes mathematical on the notes on the derivation of equations. derivation of equations. 9. References in the 9. text References should be in the indicated text should by thebeauthor’s indicated name, by the date author’s of publication name, date andofthe publication page num-and the page number where appropriate, ber where e.g. Acemoglu appropriate, ande.g. Robinson Acemoglu [2012], and Robinson Hicks [1965a, [2012], 1965b]. HicksReferences [1965a, 1965b]. shouldReferences should be listed at the endbeoflisted the article at the in end theofstyle the article of the in following the styleexamples: of the following examples: Acemoglu, D., Robinson, Acemoglu, J.A.,D., 2012, Robinson, Why Nations J.A., 2012, Fail.Why The Origins Nations of Fail. Power, The Prosperity Origins of Power, and Poverty, Prosperity and Poverty, Profile Books, London. Profile Books, London. Kalecki, M., 1943, Kalecki, Political M., Aspects 1943, ofPolitical Full Employment, Aspects of Th Full e Political Employment, Quarterly, The Political vol. XIV,Quarterly, no. 4: 322–331. vol. XIV, no. 4: 322–331. Simon, H.A., 1976,Simon, From Substantive H.A., 1976,toFrom Procedural Substantive Rationality, to Procedural in: Latsis, Rationality, S.J. (ed.),in: Method Latsis,and S.J. Appraisal (ed.), Method and Appraisal in Economics, Cambridge in Economics, University Cambridge Press, Cambridge: University Press, 15–30. Cambridge: 15–30. 10. Copyrights will 10.beCopyrights establishedwill in the be established name of theinE&BR the name publisher, of the namely E&BR publisher, the Poznań namely University the Poznań of University of Economics Press. Economics Press. More informationMore and advice information on the and suitability advice and on the formats suitability of manuscripts and formats canofbe manuscripts obtained from: can be obtained from: Economics and Business Economics Review and Business Review al. Niepodległościal. 10Niepodległości 10 61-875 Poznań 61-875 Poznań Poland Poland e-mail: review@ue.poznan.pl e-mail: review@ue.poznan.pl www.puereview.ue.poznan.pl www.puereview.ue.poznan.pl

Subscription Economics and Business Review (E&BR) is published quarterly and is the successor to the Poznań University of Economics Review. The E&BR is published by the Poznań University of Economics Press. E&BR is listed in ProQuest, EBSCO, and BazEkon. Subscription rates for the print version of the E&BR: institutions: 1 year – €50.00; individuals: 1 year – €25.00. Single copies: institutions – €15.00; individuals – €10.00. The E&BR on-line edition is free of charge. Correspondence with regard to subscriptions should be addressed to: Księgarnia Uniwersytetu Ekonomicznego w Poznaniu, ul. Powstańców Wielkopolskich 16, 61-895 Poznań, Poland, fax: +48 61 8543147; e-mail: info@ksiegarnia-ue.pl. Payments for subscriptions or single copies should be made in Euros to Księgarnia Uniwersytetu Ekonomicznego w Poznaniu by bank transfer to account No.: 96 1090 1476 0000 0000 4703 1245.