Journal Of Personal Finance Vol. 10 Issue 1 by IARFC

Journal of Personal Finance

Volume 10, Issue 1 2011

The Official Journal of the International Association of Registered Financial Consultants

Volume 10, Issue 1

CONTENTS EDITORâ&#x20AC;&#x2122;S NOTES.......................................................................................... 9 RESEARCH & THEORY Public Awareness of Retirement Planning Rules of Thumb ................. 12 Robert N. Mayer, Ph.D., University of Utah Cathleen D. Zick, Ph.D., University of Utah Michelle Glaittli, University of Utah Retirement planning advice commonly takes the form of rules of thumb offered in self-help books, magazine articles, and Internet websites. The rules provide simple answers to questions about how much to save, how to allocate retirement investments, and how to safely draw down retirement savings. The accuracy of these rules is hotly debated among finance scholars, but little is known about the extent to which members of the public are aware of these rules. This study examines awareness of four widely-disseminated retirement rules of thumb among employees of a large university (N=3,095). Male respondents and those with higher levels of education are more aware of these rules than females and people with lower levels of education, but fewer than half of respondents are aware of even the best known of the four rules studied. Finally, we discuss the implications of the results for financial planning professionals. The Demand for Financial Planning Services ........................................ 36 Sherman D. Hanna, Ph.D., Ohio State University Based on 1998 to 2007 Survey of Consumer Finances datasets the proportion of households reporting use of a financial planner increased from 21% in 1998 to 25% in 2007, with an estimated increase of almost five million households between 2004 and 2007. Multivariate analysis shows that the likelihood of using a financial planner is strongly related to risk tolerance, with those with low risk tolerance the least likely, and those with above average risk tolerance the most likely to use a financial planner, controlling for income, net worth, age, and other factors. Those with substantial risk tolerance have significantly lower likelihood of using a financial planner than those with above average risk tolerance. Black households are more likely but Hispanic and Other/Asian households are less likely than comparable White households to use a financial planner. The likelihood of using a financial planner increases with net worth for ranges above zero, but also increases as net worth decreases below zero.

Journal of Personal Finance

Can Dual Beta Filtering Improve Investor Performance? .................... 63 James Chong, Ph.D., California State University, Northridge Shaun Pfeiffer, Ph.D. Candidate, Texas Tech University G. Michael Phillips, Ph.D., California State University, Northridge This study investigates the possibility that more efficient portfolios may be constructed by using the dual-beta model that screens out assets that exhibit more extreme downside risk sensitivity. Three portfolios were constructed, using the criteria of standard CAPM beta, downmarket beta, and a combination of up-market and down-market betas. Overall, the standard CAPM beta consistently lags the dual-betas. When compared to the Fama-French three-factor inspired DFEOX, the dual-betas also performed reasonably well, with the ability to contain the downside while participating in the upside. Safe Withdrawal Rates from Retirement Savings for Residents of Emerging Market Countries .................................................................... 87 Channarith Meng, Ph.D. Candidate, National Graduate Institute for Policy Studies (GRIPS) Wade Donald Pfau, Ph.D., National Graduate Institute for Policy Studies (GRIPS) Researchers have mostly focused on U.S. historical data to develop the 4 percent withdrawal rate rule. This rule suggests that retirees can safely sustain retirement withdrawals for at least 30 years by initially withdrawing 4 percent of their savings and adjusting this amount for inflation in subsequent years. But, the time period covered in these studies represents a particularly favorable one for U.S. asset returns that is unlikely to be broadly experienced. This poses a concern about whether safe withdrawal rate guidance from the U.S. can be applied to other countries. Particularly for emerging economies, definedcontribution pension plans have been introduced along with underdeveloped or non-existing annuity markets, making retirement withdrawal strategies an important concern. We study sustainable withdrawal rates for the 25 emerging countries included in the MSCI indices and find that the sustainability of a 4 percent withdrawal rate differs widely and can likely not be treated as safe.

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Financial Planning Literature Survey .................................................. 109 Benjamin E. Fagan, MSFE, PlusPlus Inc. Shawn Brayman, MES, PlusPlus Inc. This study is intended to provide an environmental scan of current research from Australia, Canada, United Kingdom and the United States, related to financial planning/services from 2003 to July 2010. The objective of this exercise is to try and highlight research areas where there may be gaps. This is not intended to review the research in any manner but rather to aggregate and document its existence in some broad based categories. The study was carried out in two parts. To begin with, research was collected, categorized and totalled to determine high and low volume areas. Finally, industry practitioners and academics were petitioned to provide their opinions. Based on our findings, Estate Distribution Analysis, Pension Alternatives and Tax Optimization were found to be the topics that require the most focus for further research. Modern Portfolio Theory, General Portfolio Management and Product Shelf were the categories that were determined to be the most overly researched areas.

Journal of Personal Finance

CALL FOR PAPERS JOURNAL OF PERSONAL FINANCE (www.JournalofPersonalFinance.com) OVERVIEW The new Journal of Personal Finance is seeking high quality manuscripts in topics related to household financial decision making. The journal is committed to providing high quality article reviews in a single-reviewer format within 45 days of submission. JFP encourages submission of manuscripts that advance the emerging literature in personal finance on topics that include: -

Household portfolio choice Retirement planning and income distribution Individual financial decision making Household risk management Life cycle consumption and asset allocation Investment research relevant to individual portfolios Household credit use Professional financial advice and its regulation Behavioral factors related to financial decisions Financial education and literacy

EDITORIAL BOARD The journal is also seeking editorial board members. Please send a current CV and sample review to the editor. JPF is committed to providing timely, high quality reviews in a single reviewer format. CONTACT Michael Finke, Editor Email: michael.finke@ttu.edu www.JournalofPersonalFinance.com ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

JOURNAL OF PERSONAL FINANCE VOLUME 10, ISSUE 1 2011 EDITOR Michael S. Finke, Texas Tech University ASSOCIATE EDITOR Wade Pfau, National Graduate Institute for Policy Studies (GRIPS) EDITORIAL ASSISTANT Benjamin Cummings, Texas Tech University EDITORIAL BOARD Steve Bailey, HB Financial Resources Joyce Cantrell, Kansas State University Monroe Friedman, Eastern Michigan University Joseph Goetz, University of Georgia Clinton Gudmunson, Iowa State University Sherman Hanna, The Ohio State University Karen Eilers Lahey, University of Akron Doug Lambin, University of Maryland, Baltimore County Jean Lown, Utah State University Angela Lyons, University of Illinois Ruth Lytton, Virginia Tech University Lewis Mandell, University of Washington and Aspen Institute Yoko Mimura, University of Georgia Robert Moreschi, Virginia Military Institute Edwin P. Morrow, Financial Planning Consultants David Nanigian, The American College Barbara O‘Neill, Rutgers Cooperative Extension Jing Xiao, University of Rhode Island Rui Yao, University of Missouri Tansel Yilmazer, University of Missouri Yoonkyung Yuh, Ewha Womans University Mailing Address:

IARFC Journal of Personal Finance The Financial Planning Building 2507 North Verity Parkway Middletown, OH 45042-0506

Journal of Personal Finance

Postmaster: Send address changes to IARFC, Journal of Personal Finance, The Financial Planning Building, 2507 North Verity Parkway, Middletown, OH 45042-0506 Permissions: Requests for permission to make copies or to obtain copyright permissions should be directed to the Editor. Certification Inquiries: Inquiries about or requests for information pertaining to the Registered Financial Consultant or Registered Financial Associate certifications should be made to IARFC, The Financial Planning Building, 2507 North Verity Parkway, Middletown, OH 45042-0506. Disclaimer: The Journal of Personal Finance is intended to present timely, accurate, and authoritative information. The editorial staff of the Journal is not engaged in providing investment, legal, accounting, financial, retirement, or other financial planning advice or service. Before implementing any recommendation presented in this Journal readers are encouraged to consult with a competent professional. While the information, data analysis methodology, and author recommendations have been reviewed through a peer evaluation process, some material presented in the Journal may be affected by changes in tax laws, court findings, or future interpretations of rules and regulations. As such, the accuracy and completeness of information, data, and opinions provided in the Journal are in no way guaranteed. The Editor, Editorial Advisory Board, the Institute of Personal Financial Planning, and the Board of the International Association of Registered Financial Consultants specifically disclaim any personal, joint, or corporate (profit or nonprofit) liability for loss or risk incurred as a consequence of the content of the Journal. General Editorial Policy: It is the editorial policy of this Journal to only publish content that is original, exclusive, and not previously copyrighted. Subscription Rates: Individual: Institution:

$55 U.S. $98 U.S.

$68 Non-U.S. $115 Non-U.S.

Single Issue:

$19 U.S.

$25 Non-U.S.

Send subscription requests with complete mailing address and payment to: IARFC Journal of Personal Finance The Financial Planning Building 2507 North Verity Parkway Middletown, OH 45042 ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

EDITORâ&#x20AC;&#x2122;S NOTES This first official 2011 issue of the Journal of Personal Finance is also the first of my tenure as full-time editor following in the footsteps of John Grable and Ruth Lytton, who have ably guided the Journal from its inception. This is my third issue as editor after having served as guest editor of two previous issues. In each of those issues I have relied on an extremely hard-working and capable group of reviewers who have committed to providing authors a high quality, timely manuscript evaluation. No journal can survive without the hard work of many scholars who volunteer to improve the quality of research in financial planning. I'd like to take this moment to thank the reviewers for this issue, and in particular the members of the editorial board who take on the bulk of reviewer responsibilities. This issue contains articles on a new approach to portfolio construction that has been used by institutional investors in the past, but is new to the field of individual portfolio management. Authors James Chong, Shaun Pfeiffer and G. Michael Phillips decompose Beta between upside and downside covariance with the market and seek to improve portfolio efficiency by looking for securities where Beta in bear markets is different from Beta in bull markets. Since the traditional Capital Asset Pricing Model assumes symmetry and prices assets based on both upside and downside risk, an investor could conceivably construct a portfolio of securities that have a relatively low total Beta (less than 0.7), but have a down-market Beta below 0.7 and an up-market Beta above 0.7. In other words, they perform better in an up-market without

Journal of Personal Finance

performing worse in a down market. The authors find evidence of superior performance using this portfolio technique, which may be particularly attractive for loss-averse investors seeking limited downside risk. I would also like to highlight the very interesting work of authors Channarith Meng and Wade Pfau. Dr. Pfau's recent work on retirement decumulation addresses the very real possibility that stock market returns used in previous decumulation shortfall studies, namely United States equity returns since 1926, may be overly optimistic. Extending the dataset into the 19th century, or simulating returns using a bootstrap method as in this article, provides more sober estimates of shortfall probabilities and of the optimal portfolio share held in equities during retirement. Since the U.S. had an unprecedented equity market run in the 20th century, Meng and Pfau ask how investors in other countries would have fared using the same decumulation methodology. In this issue they focus on sustainable withdrawal rates in developing nations and find substantial variation among countries and among strategies. I find this research particularly compelling since, as we are often reminded, past performance does not always predict the future - particularly in a world where the global capital market will have a strong influence on U.S. investors. In "The Demand for Financial Planning Services," Sherman Hanna finds that the use of financial planners climbed by five million between 2004 and 2007 and explores which Americans are more likely to use a planner. Among his more interesting findings are that, even independent of income and wealth, more educated households are more likely to hire a ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

professional to provide financial advice. It appears that many of those who are likely to be the most knowledgeable about personal finance also realize that they need an expert to help them make better financial decisions. Perhaps unsurprisingly, single women are also more likely than single men to hire a planner to help them with their finances. I am looking forward to the Winter issue of the Journal of Personal Finance and would again like to thank those who contribute to the Journal and to the readers and the IARFC for their support and interest in advancing the science of personal finance. ~Michael Finke

Journal of Personal Finance

PUBLIC AWARENESS OF RETIREMENT PLANNING RULES OF THUMB Robert N. Mayer, Ph.D. University of Utah Cathleen D. Zick, Ph.D. University of Utah Michelle Glaittli University of Utah Retirement planning advice commonly takes the form of rules of thumb offered in self-help books, magazine articles, and Internet websites. The rules provide simple answers to questions about how much to save, how to allocate retirement investments, and how to safely draw down retirement savings. The accuracy of these rules is hotly debated among finance scholars, but little is known about the extent to which members of the public are aware of these rules. This study examines awareness of four widelydisseminated retirement rules of thumb among employees of a large university (N=3,095). Male respondents and those with higher levels of education are more aware of these rules than females and people with lower levels of education, but fewer than half of respondents are aware of even the best known of the four rules studied. Finally, we discuss the implications of the results for financial planning professionals.

Retirement planning advice can never be simple, but it is often simplified in the form of ―rules of thumb.‖ These rules are offered in books and articles, websites, and television 

Robert N. Mayer, Department of Family and Consumer Studies, University of Utah, 225 South 1400 East, Salt Lake City, UT 84112-0080; (801) 581-5771; robert.mayer@fcs.utah.edu The research reported in this paper was supported by a grant from the Direct Selling Education Foundation. The authors also appreciate the help of Kara Glaubitz and Matt Argyle in the preparation of the revised manuscript. ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

programs. The rules address key questions of retirement planning: how much will I need to save, how can I reach my retirement savings goal, and how can I make my retirement nest egg last my entire lifetime? Clearly, using rules of thumb cannot replace planning that takes into account the specifics of an individualâ&#x20AC;&#x2DC;s situation, but these rules are the beginning of point of retirement planning for many individuals. Hence, regardless of the status of these rules among finance scholars and practitioners, it is important to understand public perception of these retirement guidelines. The article is organized as follows. First, we situate rules of thumb in the broader process of consumer decision making and financial planning for retirement. Second, we describe four common retirement rules of thumb, including the origins and evolution of these rules. Third, we describe a research study that examined public awareness of these four rules and present its major findings. The study should not be interpreted as an endorsement of these rules, only an acknowledgement of their ubiquity in the mass media. Finally, we comment on the implications of the studyâ&#x20AC;&#x2DC;s findings. Retirement Rules of Thumb in Context Individuals who wish to be deliberative about retirement planning can seek professional help, plan on their own, or employ some combination of the two approaches. Despite the availability of a variety of professional financial planners to assist individuals with their retirement planning, only a minority of people avail themselves of these professional services (Certified Financial Planning Board of Standards,

Journal of Personal Finance

2009; Elmerick, Montalto, & Fox, 2002). The majority of people are retirement planning do-it-yourselfers, and their planning activities may rely heavily on rules of thumb. Even people who avail themselves of professional financial help may use rules of thumb as a departure point for discussions with their advisors. The use of rules of thumb in retirement planning is relevant to three broad topics within consumer research: positive vs. normative, heuristics, and financial literacy. In a presidential address to the American Finance Association, John Y. Campbell (2006) highlighted the difference between observed (positive) and ideal (normative) financial behavior. Rules of thumb – however imperfect they may be – are normative statements about what people ought to do. These statements can be studied positively, however, by examining the extent to which people are aware of them, properly understand them, are aware of their limitations, and use them. The study reported here addresses the first of these research questions. Given the many complex choices that people face in their daily lives and the finite resources that can be devoted to these choices, people often use rules of thumb, shortcuts, and other ―heuristics‖ to facilitate these decisions (Kahneman, Slovic, & Tversky, 1982). Following these rules is designed to yield results that, while not perfect, are satisfactory (Simon, 1956; Schwartz, 2004). Despite the importance of rules of thumb, little attention has been devoted to their use by consumers in the retirement planning process.

Volume 10, Issue 1

The study of retirement rules of thumb can also be situated within the topic of financial literacy. Literacy is often discussed in terms of concepts that can be judged as correct or incorrect (Hung, Parker, & Yoong, 2009; Huston, 2010; Remund, 2010). Rules of thumb, in contrast, are only approximations of a correct response; they may need to be adjusted in light of individual circumstances. Nevertheless, awareness of rules of thumb is an important, if neglected, element of financial literacy. Common Retirement Rules of Thumb Retirement rules of thumb are appealing because they provide simple and concrete guidance for addressing the complex task of retirement planning. The centerpiece of retirement planning is calculating how much a person will need to fund a desired or â&#x20AC;&#x2022;comfortableâ&#x20AC;&#x2013; retirement lifestyle. Yet only a minority (42%) of people in the 2011 Retirement Confidence Survey conducted by the Employee Benefit Research Institute had made this basic calculation (Helman, Copeland, & VanDerhei, 2011). Advice dispensed by popular financial gurus such as Suze Orman and Dave Ramsey sidesteps this calculation by offering a simple rule of thumb: save a certain percent (typically, 10 or 15) of income for retirement. While easy to remember, this type of retirement planning rule fails to provide a retirement saving target, rationale, or method. The four rules discussed below, while simple as well, offer more specific guidance for retirement planning to those people wishing to follow them.

Journal of Personal Finance

Income Replacement Ratio Rule Retirement planning requires an estimate of how much income will be needed to cover anticipated expenses. In this regard, a rule of thumb is that a household needs roughly 7090% of its pre-retirement gross income to maintain its current standard of living. The fact that the ratio is less than 100% is based on the assumption that taxes decline in retirement as do many expenses (e.g., commuting expenses, clothing purchased for work). The idea of an income replacement ratio rule has a long history. An article published in July of 1965 told retirees that 50- 60% replacement income was the needed amount in retirement (Nuccio, 1965ab). Only a few months later, the same author revised this figure upward to between 50-75% (Nuccio, 1965). In 1981, a self-help book supported a replacement income of 75% (Schiller, 1981), a percent also found in a 1993 self-help book (Williamson, 1993). A 2009 article aimed at nurses recommended a replacement income between 60-80% (Strohfus & Schrader, 2009), and a self-help book for ―dummies‖ suggested that the ratio might be 100% (Benna, 2009). Like all rules of thumb, the income replacement ratio ignores individual differences in age and income. For example, some people may need far more than 80% of income in the early years of retirement as they try to catch up on the things they always wanted to do. Their income replacement needs could well be below 80% in their final years of life (assuming no large out-of-pocket heath care expenses). Similarly, income replacement needs vary by a person‘s ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

household income. Research by Aon Consulting and Georgia State University (Palmer, 2008) finds that a household earning only $30,000 per year, for example, might need to replace 90% of more of its pre-retirement income to maintain a given lifestyle, whereas a household earning $70,000 might only need to replace 77% (Palmer, 2008). Research by Scholz and Seshadri (2009) found a median optimal target replacement rate of 75% for married couples, but the authors urged caution in using rules of thumb due to enormous variability among households. 20 Times Income Rule Another rule of thumb that can be used to determine the total amount needed for retirement is multiplying an individual‘s current annual income or projected annual income requirement in retirement by a particular number. Robert Sheard popularized the number 20 in his book, Money for Life: The 20 Factor Plan for Accumulating Wealth While You’re Young (2000). Twenty-times-income is one of many rules that involve multiplying current income to derive a retirement savings goal. The author of a 1977 article in the Wall Street Journal recommended saving ten times one‘s annual income to produce a financially secure retirement (Moffitt, 1977). More recently, Stein and Demuth‘s (2009) self-help book promoted a factor of between twelve and sixteen when multiplying current salary. In August of 2009 Money magazine told readers that it was thirty times annual income (―Make Peace,‖ 2009). It is likely that the escalation of the factor used in this rule is driven, at least in part, by increasing retiree longevity.

Journal of Personal Finance

Charles Ferrell, author of Your Money Ratios (2009), suggests that the number by which current income is multiplied should increase as a person ages. When a person is 50 years old, multiplying by five may be sufficient. For someone who is 65 years old, multiplying by twelve is more appropriate (assuming additional income is available from Social Security). Regardless of the specific number that is used in multiplying current income (or required retirement income), the value of this rule of thumb is likely to be greatest when a person is close to retirement age. 110 Minus Age in Stocks In addition to guidance in setting retirement savings goals, individuals use rules of thumb to decide how to allocate their investments across asset classes. One such rule is to ―own your age in bonds.‖ This would mean, for instance, that a person who is 40 years old should allocate 40 percent of his or her investment portfolio to bonds, with the remainder going largely to stocks. This rule is based on the assumption that a person‘s holdings should be more conservatively invested as they age (Lozada, 2004). A rule that yields similar results to the own-your-age-inbonds rule is to subtract your age from a particular number to determine the percent of stocks in an investment portfolio. The most common form of this rule is 100-minus-age, although the origins of this formulation are not known. It would suggest that a 40-year old investor would have 60 percent of his or her holdings in stocks and 40 percent in bonds. Over time, though, the number used in the rule has migrated upward to 110-minusage or even 120-minus-age in response to increased longevity ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

(MarksJarvis, 2007). In August of 2008, Consumer Reports magazine told readers that the old rule of 100 was outdated and that 110 was now a more reasonable number to use in determining stock allocation; but even the number of 110 was questioned because of the increases in life expectancies (―Deep-Six,‖ 2008). A year later Money magazine also promoted 110 as the new number to use in stock allocation decisions (―Make Peace…,‖ 2009). William Bengen (1996) varies the number used in the rule to reflect differences in risk tolerance. He proposes 115 for people with low risk tolerance, 128 for those with moderate risk tolerance, and 140 for the aggressive investor. The moderate risk rate formula of 128 minus age is still more aggressive than the numbers typically publicized in the popular media. While the exact number may vary among formulations, the concept behind the various number-minus rules of thumb is embodied in target date and life-cycle mutual funds. These funds slowly and automatically decrease stock allocation as a person ages. Wang (2007) investigated the percent of stock allocation provided by various target date funds and compared these percentages to the rule of 120 minus age. His analysis showed that these funds varied from being too high by nearly nine percent to too low by over 21 percent. However, each fund did have an allocation option that was within two percent of using the rule at some point in the lifecycle. The various number-minus-age rules are meant to apply before retirement and beyond it. The rule encourages substantial stock ownership during the early years of

Journal of Personal Finance

retirement. Having too little invested in stock during retirement can jeopardize financial security, although having too much invested in stocks can have the same effect if there is a major market downturn. 4% Withdrawal Rate In addition to building a retirement nest egg, pre-retirees need a sense of how much money they will be able to withdraw annually during retirement without running a substantial risk of outliving their savings. A widely-cited rule of thumb addresses this issue: one can safely withdraw 4% per year adjusted for inflation. For example, if an individual has a nest egg totaling $1 million, withdrawing four percent the first year in retirement would make $40,000 available. The second year, $41,200 could be withdrawn if the inflation rate were three percent. The rule assumes that the unused portion of the retirement account is allocated in an age-appropriate fashion among asset classes. The 4% withdrawal rule during retirement is often associated with William P. Bengen. In 1994, Bengen wrote a seminal paper on the safe withdrawal rates from retirement portfolios. He concluded that a 4.1 percent withdrawal rate over thirty years is safe for a portfolio composed of 50% stocks and 50% intermediate government bonds (Bengen, 1994). Increasing the share of stock in the portfolio increases the funds that can be withdrawn but also increases the risk of exhausting the funds before the end of thirty years. The 4% withdrawal has generated a great deal of debate among academics and financial practitioners (Scott, Sharpe, & Watson, 2009). Nevertheless, Bengenâ&#x20AC;&#x2DC;s original formulation of ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

the rule appears to be reasonably accurate as a ballpark figure. For example, Cooley, Hubbard, and Walz (1999) concluded that a portfolio composed of 75% equities allowed for a 4-5% withdrawal rate over thirty years. For shorter payout periods, say, fifteen years or less, the withdrawal rate could be as high as eight or nine percent a year. Despite any shortcomings, the 4% withdrawal rule has worked its way into many textbooks and self-help sources of retirement advice (Armstrong & Doss, 2009; Eisenberg, 2006; Garman & Forgue, 2010). Summary Rules of thumb for retirement are common in the popular media and address some of the crucial aspects of retirement planning. Researchers do not know, however, the extent to which pre-retirees are aware of these rules; which types of people are most and least aware of these rules; and how these rules are used in the process of retirement planning. The research reported here addresses the first two of these three important questions. Study Design As part of National Consumer Protection Week 2011, the authors collaborated with the Division of Human Resources of a Mountain West university to create an educational event for the university‘s more than 20,000 full-time and part-time employees. The event took the form of an online quiz regarding retirement planning. The quiz, which centered on 12 knowledge questions and an additional 4 questions on retirement ―rules of thumb,‖ was available from Monday, March 7 through Friday, March 11. As an incentive to participate, 20 prizes were offered. In keeping with the theme

Journal of Personal Finance

of the event, these prizes consisted of a one-time $250 contribution to a new or existing supplemental retirement account administered for the employee by the University.1 The primary goal of our research was to examine public awareness of four retirement rules of thumb. To place our findings in a broader context, however, we interviewed four professional financial advisors to get their views on the value of retirement rules of thumb (Glaittli, 2011). The results of these interviews are reported in the Discussion section below. Measures As previously indicated, there is very little research on consumer awareness of retirement rules of thumb. An exception is a 2008 study conducted by Metropolitan Life Insurance Company. Among fifteen multiple-choice questions that were meant to measure ‗retirement IQ‖ were two regarding retirement rules of thumb. One referred to the income replacement ratio in retirement, the other to the 4% withdrawal rate rule. Both of these questions were used in this study, but with the answer categories modified to create equal numerical intervals between choices and to make one answer unambiguously reflective of the general presentation of these rules to the general public:

To increase the likelihood that participants took the quiz seriously and did not submit answers just for the sake of winning a prize, only those people who correctly answered 4 or more of the 12 knowledge questions in the quiz were eligible to win a prize. These knowledge questions covered a variety of retirement-related topics and were separate from the rule-ofthumb questions. In the case of the knowledge questions, respondents had an incentive to guess an answer rather than choose ―don‘t know/not sure.‖ This incentive did not exist, however, for the rule-of-thumb questions since answers to these questions did not affect prize eligibility. ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

What percent of pre-retirement income do financial experts think retirees will need in retirement? (2030%; 50-60%; 80-90%; 110-120%; Don‘t know/Not sure) To help ensure that an individual has enough money to make savings last his or her lifetime, experts recommend limiting the percent people withdraw from their savings principal each year to: (4%; 8%; 12%; 16%; Don‘t know/Not sure) Despite many references to the 20-times-income and 110minus-age rules in the popular press, we were unable to find survey questions covering these two rules of thumb. We therefore developed questions to gauge awareness of these two rules. After pretesting to ensure clarity, the two questions were finalized as follows: Financial experts suggest that individuals, in order to maintain their current standard of living during retirement, need to save an amount of money that equals their annual income multiplied by a certain number. What is the number that financial experts suggest using? (5; 10; 20; 30; Don‘t know/Not sure) Financial experts have a simple formula for recommending the percentage of stocks that people should have in their investment portfolios at different ages. This formula involves subtracting a person‘s current age from which of the following numbers? (50; 80; 110; 140; Don‘t know/Not sure)

Journal of Personal Finance

Answers to these four questions were coded into two categories: ―aware‖ and ―not aware‖ of the rule. Aware responses were those that reflect the general consensus among financial experts and commentators for each rule. All other answers (with the exception of omitted answers) were coded as not aware, as we had no way of comparing the remaining answers in terms of their proximity to the aware response. For example, is a person who responds that one can safely withdraw 12% of one‘s retirement savings each year during retirement more aware of the 4% withdrawal rule than someone who chose ―Don‘t know/Not sure‖ as an answer? In addition to the awareness measures of the four rules of thumb, respondents were asked to provide information about their basic socio-demographic characteristics. These characteristics included age, years of education, gender, marital status, household income, and percentage of household income accounted for by the individual respondent. In addition, respondents were grouped into four employment categories, each reflecting a different university retirement plan (or absence thereof). One group (―exempt‖) has a defined contribution retirement plan and represents roughly half of all respondents. A second group (―nonexempt‖) has a defined benefit retirement plan, with a small defined contribution component. This group comprises 38% of all respondents. The two remaining groups are both small, one consisting of part-time employees without a university-administered retirement plan (―non-benefited‖) and the other composed of respondents who could not classify themselves as exempt or non-exempt (―other‖). ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Sample Characteristics During the five-day event period, 3,180 employees took the quiz and submitted their answers. Of these respondents, 41 did not answer one or more of the rules of thumb questions and were subsequently dropped from the sample. Another 44 people were dropped because they did not provide an answer to one of the three socio-demographic questions that were coded as categorical variables (gender, marital status, and employment category). In the relatively small number of cases where there was a missing value for a continuous variable (age, education, household income, and percent of household income earned by the respondent), missing values were coded to the mean for that variable. Recoding missing responses to the mean value does not bias the coefficient estimates in the multivariate analyses but it does make the tests of statistical significance somewhat less conservative. Taken together, these adjustments resulted in a sample of 3095 people. The task of comparing the final sample with the overall population of university employees is complicated by the fact that university-wide data on socio-demographic characteristics are only available for full-time employees, that is, those drawing benefits. As it turned out, very few (259) nonbenefitted employees participated in the survey. (These employees do not have a retirement plan provided by the university, but they are eligible to establish a supplemental retirement account through the university.) The universityâ&#x20AC;&#x2DC;s Division of Human Resources estimated that approximately 10,000 benefitted employees received the email invitation to participate in the survey, yielding a cooperation rate among these employees of approximately 28%.

Journal of Personal Finance

Study Results Sample Characteristics A complete description of the sample characteristics is found in Table 1. Data are not available to compare the people who took the survey and those who declined to participate. Nevertheless, the benefitted employees who took the survey appear to be reasonably representative of the overall population of benefitted employees. Benefitted employees who participated matched university-wide data for age but were more likely than non-participants to be female and more likely to work for the health sciences units of the university than the non-health sciences units. Comparison data were not available for income or marital status. Note, however, that even if the sample were exactly representative of the university as a whole, the results of this study would not be generalizable to other populations. The results can only indicate trends among Table 1 Socio-Demographic Characteristics of Sample (N = 3,095) Variable

Definition

Std. Dev. 12.7 2.3 0.49 0.20 0.27

Mean

Age Education Benefited: Non-Exempt* Benefited: Other* Non-Benefited*

Age in Years 42.9 Years of Schooling 16.2 1=Non-Exempt, 0=Exempt 0.38 1=Other, 0=Exempt 0.04 1=Non-Benefited, 0.08 0=Exempt Gender: Female 1=Female, 0=Male 0.63 Household Income Income in $1,000s 82,223 Household Income Share Percent of Total Income 71.6 Marital Status: Married 1=Married/Cohabiting, 0.71 0=Otherwise *The omitted group in this sequence of dummy variables are those employees who are exempt and benefits eligible.

0.48 52,841 27.6 0.45

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

the respondents in this study, a shortcoming that is common in exploratory studies. The sample is fairly diverse in terms of socio-demographic characteristics (e.g., age, educational attainment, gender, and income) and therefore permits exploration of differences among individuals in awareness of rules of thumb. Awareness of Rules of Thumb The descriptive results for awareness of the four rules of thumb are shown in Table 2. Across the four questions, awareness of the rules of thumb was low with the modal Table 2 Responses for Four Retirement Rules of Thumb (N = 3,095) Response Income Replacement Ratio Rule 20-30% 50-60% 80-90% 110-120% Don‘t know/Not sure 20-Times-Income Rule 5 10 20 30 Don‘t know/Not sure 110-Minus-Age Rule 50 80 110 140 Don‘t know/Not sure 4% Withdrawal Rule 4% 8% 12% 16% Don‘t know/Not sure

Percent

Frequency

8.98 36.06 36.45 7.08 11.44

278 1,116 1,128 219 354

10.76 26.59 27.21 7.88 27.56

333 823 842 244 853

9.66 28.08 22.84 1.94 37.48

299 869 707 60 1,160

41.32 20.39 12.08 2.39 23.81

1,279 631 374 74 737

Journal of Personal Finance

number of rules selected that correspond to the advice of financial experts being 1-2. Only 108 (3.5%) respondents selected all four rules while 858 (28%) respondents identified none of them. In none of the four cases did a majority of respondents select the rule as it is formulated by financial experts and commentators. Awareness was highest for 4% withdrawal rule (41.32%) and lowest for the 110-minus-age rule (22.8%). Percentages of the sample ranging from 11.4 to 37.5 chose ―Don‘t know/Not sure,‖ but when eliminating these people, the number of people who misidentified a rule of thumb typically exceeded those who had correctly identified it. For example, 1,613 people chose an income replacement ratio below or above ―80-90%,‖ compared to 1,128 choosing this replacement interval. People who were aware of one rule of thumb were more likely to be aware of other rules of thumb, but only mildly so. Correlations among the four questions were consistently positive and statistically significant at the p<.0001 level, but they were also weak. The highest association was between awareness of the income replacement rule and awareness of the 110-minus age rule, but the correlation was only .16. Thus, it makes sense to analyze separately the socio-demographic predictors of awareness rather than try to create a measure that combines awareness across the four measures. Predictors of Awareness Given the dichotomous nature of the dependent variables, multinomial logit analyses were used to examine the connection between awareness and respondent characteristics. These characteristics were age, years of education, gender, ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Table 3 Estimated Odds Ratios for the Logistic Regressions (95% confidence interval in parentheses) Independent Variables

Replace. Ratio

20 X Income

110 - Age

4% Withdraw

Age

1.021 1.00 (1.02-1.03)** (1.00-1.00)

Education

1.05 1.06 1.14 1.03 (1.01-1.10)** (1.02-1.11)** (1.08-1.19)** (0.992-1.07)

Non-Exempt, Benefits Eligible1 Other, Benefits Eligible1 Non-benefited1 Gender Household Income (in $1000s) Household Income Share Marital Status: Married

1.00 1.01 (0.99-1.00) (1.01-1.02)**

0.83 (0.69-1.01)*

0.97 (0.79-1.19)

0.98 (0.78-1.22)

1.03 (0.86-1.24)

0.74 (0.47-1.17)

0.78 (0.47-1.28)

0.68 (0.38-1.23)

0.90 (0.58-1.38)

0.88 (0.64-1.21)

0.94 (0.68-1.31)

1.04 (0.73-1.47)

1.21 (0.90-1.63)

0.81 0.73 0.70 0.81 (0.69-0.96)** (0.62-0.87)** (0.58-0.85)** (0.69-0.95)** 1.00 (1.00-1.00)

1.00 1.01 1.01 (1.00-1.00) (1.00-1.01)** (1.00-1.01)**

1.00 (1.00-1.01)

1.00 (0.99-1.00)

1.00 (1.00-1.00)

1.00 (1.00-1.01)

1.34 0.99 (1.09-1.66)** (1.00-1.00)

0.92 (0.73-1.16)

1.07 (0.88-1.30)

Ď&#x2021;2 131.90** 31.08** 127.47** 46.64** The omitted group in this sequence of dummy variables are those employees who are exempt and benefits eligible. ** p<.05, *p<.10 1

marital status, household income, percentage of household income accounted for by the individual respondent, and employment category as it bears on type of retirement plan. The only characteristic that predicted awareness across all four awareness measures was the respondentâ&#x20AC;&#x2DC;s gender, with men being more aware than women. Higher levels of

Journal of Personal Finance

education predicted greater awareness for three of the awareness measures, the exception being the 4% withdrawal rule. Older respondents displayed greater awareness of the income replacement ratio rule and the 4% withdrawal rule than younger respondents, but there were no age differences for the other two rules. Interestingly, neither household income nor the percentage of the household income earned by the respondent predicted awareness. It might have been expected that people with relatively greater household incomes and who account for the majority of their householdâ&#x20AC;&#x2DC;s income would be more attuned to retirement planning information, including rules of thumb. Similarly, people with defined contribution plans have greater responsibility for guiding their retirement planning than those with defined benefit plans and therefore they might have been expected to show greater awareness of the rules of thumb. This was not the case, though. Discussion To put a human face on the research reported here, we spoke with four professional financial advisors to get their views on the value of retirement rules of thumb. Some advisors argued that awareness of retirement rules of thumb is an important element of financial literacy and can serve as a conversation starter in retirement planning. These advisors reported that clients who are aware of financial rules of thumb have little trouble understanding that these rules need to be customized to fit individual circumstances. One advisor felt that the sooner her clients learn these rules, the better. Another believed that these rules are most useful when clients are young, that is, just beginning their financial education. ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Other advisors with whom we spoke were less sanguine about the use of rules of thumb by their clients. These advisors reported having to spend time explaining the limitations of these rules. One advisor believed that clients can be lulled into a false sense of security just because they have met one or more of these rules. What about individuals who use retirement planning rules of thumb without the guidance of a financial professional? What is the likely impact of their reliance on these rules? First, it should be noted that overall awareness of the four rules of thumb studied here is fairly low, with only the 4% withdrawal rule exceeding a 40% awareness threshold for the sample. Thus, any help or harm that comes from the use of these rules is unlikely to be widespread. Second, people with higher levels of education tend to be more aware of the retirement rules of thumb, suggesting that these rules are a component of financial literacy rather than a substitute for it. Similarly, men are more aware of the rules than women. Given that research consistently reveals that men are more financially literate than women (Lusardi and Mitchell, 2008; Fonseca, Mullen, Zamarro, and Zissimopoulos, 2010), our finding again suggests that rules of thumb are used by those who are likely be able to assess the benefits and shortcomings of these rules. Conclusion Our study addresses only a by the existence indicated at the

was exploratory in nature and as such few of the research questions that are raised of retirement planning rules of thumb. As outset, our interest in documenting public

Journal of Personal Finance

awareness of these rules of thumb should not be interpreted as an endorsement of the accuracy or utility of these rules. Our results suggest, however, that awareness of these rules – especially among men and relatively well educated individuals – merits additional investigation. This research can examine awareness of these rules in a more nationally representative sample and, more important, any relationships among awareness, use of these rules, and retirement preparedness. Future research might also compare awareness, understanding, and use of retirement rules of thumb among clients of professional financial planners versus those who plan without professional help. To the extent that professional planners promote client awareness and use of rules of thumb, do planners favor conservative rules (to reduce the possibility of being viewed as ―failures‖) or aggressive ones (to increase commissions, fees, and other forms of remuneration)? Regardless of whether a person works with a financial planning professional or is a do-it-yourselfer, individuals need to be active participants in retirement planning. Rules of thumb, by virtue of their simplicity, may serve as steppingstones to more sophisticated retirement planning. As long as individuals understand the benefits and limitations of retirement rules of thumb, efforts to educate the public about these rules can have two types of benefits. First, awareness of these rules can serve as building blocks of financial literacy, especially when used in conjunction with professional financial assistance. Second, public education efforts can correct misperceptions about the content of common rules of thumb. We found that many people ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

inaccurately describe particular rules of thumb (e.g., subtracting their age from 80 rather than from 110 in the rule about asset allocation across the lifespan). If rules of thumb are to be useful at all, they need to be the rules of thumb that have achieved some rough consensus among scholars and practitioners, not some misunderstanding of these rules. References Armstrong, F., & Doss, J.R. (2009). The retirement challenge--Will you sink or swim? A complete, do-it-yourself toolkit to navigate your financial future. Upper Saddle River, NJ: FT Press. Bengen, W.P. (1994). Determining withdrawal rates using historical data. Journal of Financial Planning, 7(4), 14-24. Bengen, W.P. (1996). Asset allocation for a lifetime. Journal of Financial Planning, 9(4), 58-67. Benna, T. (2009). Managing your money all-in-one for dummies. Hoboken, NJ: Wiley. Campbell, J.Y. (2006). Household finance. Journal of Finance, 61(4), 1553-1604. Certified Financial Planning Board of Standards, Inc. (2009). 2009 National consumer survey on personal finance. Washington, D.C. Retrieved from http://www.cfp.net/downloads/CFP_Board_2009_National_ Consumer_Survey.pdf Cooley, P. L., Hubbard, C.M., & Walz, D.T. (1999). Sustainable withdrawal rates from your retirement portfolio. Financial Counseling and Advising 10(1), 40-51. Eisenberg, L. (2006). The number. New York: Free Press. Elmerick, S. A., Montalto, C. P., & Fox, J. J. (2002). Use of financial planners by U.S. households. Financial Services Review, 11(3), 217-213. Ferrell, C. (2009). Your Money Ratios. New York: Avery/Penguin. Fonseca, R., Mullen, K.J., Zamarro, G., & Zissimopoulos, J. (2010). What explains the gender gap in financial literacy? Rand Working Paper 762, June. Retrieved from http://www.rand.org/content/dam/rand/ pubs/working_papers/2010/RAND_WR762.pdf Garman, E.T., & Forgue, R.E. (2010). Personal finance. Boston, MA: Houghton Mifflin. Glaittli, M. (2011). Retirement rules of thumb: History and assessment by professionals, Senior Honorâ&#x20AC;&#x2DC;s Thesis, University of Utah, April 29. Helman, R., Copeland, C., & VanDerhei, J. (2011). The 2011 retirement confidence survey: Confidence drops to record lows, reflecting â&#x20AC;&#x2014;the

Journal of Personal Finance

new normal,‖ Issue Brief 355, March. Retrieved from http://www.ebri.org/pdf/briefspdf/EBRI_03-2011_No355_RCS2011.pdf Hung, A., Parker, A.M., & Yoong, J. (2009). Defining and Measuring Financial Literacy. RAND Working Paper 708. September. Retrieved from http://www.rand.org/content/dam/rand/pubs/working_papers/ 2009/ RAND_WR708.pdf Huston, S.J. (2010). Measuring financial literacy. Journal of Consumer Affairs, 44(2), 296-316. Kahneman, D., Slovic, P., & Tversky, A. (1982). Judgment under uncertainty: Heuristics and biases. New York: Cambridge University Press. Lozada, G.A. (2004). Constructing age-dependent portfolios, Journal of Personal Finance, 3(4), 59-75. Lusardi A. and O. Mitchell (2008). Planning and financial literacy: How do women fare? American Economic Review, 98(2), 413-417. Make Peace With Your Money (2009). Money, August, p.6. MarksJarvis, G. (2007). Saving for retirement without living like a pauper or winning the lottery. Upper Saddle River, NJ: Financial Times. Moffitt, D. (1977, January 24). How some professional pension-planning ideas can show how much to save for retirement. Wall Street Journal, p. 34. Nuccio, S. (1965a, July 8). Personal finance: Planning pensions. New York Times, p. 39. Nuccio, S. (1965b, December 9). Personal finance: Facing retirement. New York Times, p. 73. Palmer, B.A. (2008). 2008 GSU/Aon RETIRE project report. Georgia State University. J. Mack Robinson College of Business. Retrieved from http://rmictr.gsu.edu/Papers/RR08-1.pdf Deep-six the ‗rule of 110‘ (2008, August). Consumer Reports Online. Retrieved from http://www.consumerreports.org/cro/money/personalinvesting/recession-financial-advice/deep-six-the-rule-of-110/ recession-financial-advice-deep-six-the-rule-of-110.htm ?loginMethod=auto Remund, D.L. (2010). Financial literacy explicated: The case for a clearer definition in an increasingly complex economy. Journal of Consumer Affairs 44(2), 279-295. Schiller, M. K. (1981). Personal and family finance: Principles and applications. Boston, MA: Allyn & Bacon. Scholz, J.K., & Seshadri. A. (2009). What Replacement Rates Should Households Use? University of Michigan Retirement Research Center, Working Paper WP 2009-214. http://www.mrrc.isr.umich.edu/ publications/papers/pdf/wp214.pdf Schwartz, Barry (2004). The Paradox of Choice. New York, Ecco. ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Scott, J.S., Sharpe, W.F., & Watson, J.G. (2009). The 4% rule at what price? Journal of Investment Management, 7(3), 31-48. Sheard, R. (2000). Money for life: The 20 factor plan for accumulating wealth while you're young. New York: HarperBusiness. Simon, H. A. (1956). Rational choice and the structure of the environment. Psychological Review, 63, 129â&#x20AC;&#x201C;138. Stein, B., & DeMuth, P. (2005). Yes, you can still retire comfortably! The baby-boom retirement crisis and how to beat it. Carlsbad, CA: New Beginnings. Strohfus, P.K., & Schrader, V. (2009). Assisting nursing students to plan for retirement. Nurse Educator, 34(2), 54-55. Wang, J. (2007). Stock allocation rule: 120 minus age. Bargaineering Personal Finance Blog. January 7. Retrieved from http://www.bargaineering.com/articles/stock-allocation-rule-120minus-age.html Williamson, G.K. (1993). Sooner than you think: Mapping a course for a comfortable retirement. Homewood, IL: Business One Irwin.

Journal of Personal Finance

THE DEMAND FOR FINANCIAL PLANNING SERVICES Sherman D. Hanna, Ph.D.ď&#x20AC;Ş Ohio State University Based on 1998 to 2007 Survey of Consumer Finances datasets the proportion of households reporting use of a financial planner increased from 21% in 1998 to 25% in 2007, with an estimated increase of almost five million households between 2004 and 2007. Multivariate analysis shows that the likelihood of using a financial planner is strongly related to risk tolerance, with those with low risk tolerance the least likely, and those with above average risk tolerance the most likely to use a financial planner, controlling for income, net worth, age, and other factors. Those with substantial risk tolerance have significantly lower likelihood of using a financial planner than those with above average risk tolerance. Black households are more likely but Hispanic and Other/Asian households are less likely than comparable White households to use a financial planner. The likelihood of using a financial planner increases with net worth for ranges above zero, but also increases as net worth decreases below zero.

The proportion of households using financial planners has increased, but is at a relatively low level, even at high levels of income and net worth. What factors are related to the use of financial planners? Which types of households seem to be underserved by financial planners? This paper uses a combination of the 1998 to 2007 Survey of Consumer Finances datasets to analyze the effects of household characteristics and risk tolerance on the use of financial planners. The empirical results are discussed in the context of normative analyses of the

ď&#x20AC;Ş

Sherman D. Hanna, Consumer Sciences Department, Ohio State University, 1787 Neil Avenue, Columbus, OH 43210; (614) 292-4584; hanna.1@osu.edu ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

value of financial planning advice, thus providing context for the identification of underserved segments of the population. Literature Review Studies analyzing empirical patterns on use of financial planners have not focused on the theoretical relationship between the value of financial planning advice and risk tolerance. Bae and Sandager (1997) reported results from convenience samples, and found that respondents were most interested in advice on retirement funding, investment/asset growth, and reducing tax burden. Elmerick, Montalto, and Fox (2002) used the 1998 Survey of Consumer Finances (SCF) dataset to analyze the types of households that reported using a financial planner for comprehensive advice, advice on savings and investment, or advice on credit. They did not provide a theoretical framework other than a brief mention of modern portfolio theory, but noted that many households sought comprehensive financial planning advice. They reported that 21% of households used a financial planner for some type of advice. In their multivariate analysis, those under 35 were more likely to use a financial planner than those 35 and older, use of financial planners increased with education, Blacks were more likely and Hispanics less likely than Whites to use financial planners, unmarried female households were more likely than married households to use financial planners, use of financial planners increased with income to the $50,000 to $74,999 range and then was roughly the same above that level, and the use of financial planners increased with net worth and also with the level of financial assets.

Journal of Personal Finance

Chang (2005) discussed some sociological theoretical aspects of the decision to seek help with savings and investment decisions from a social network versus professional help, and noted ―…although socioeconomic status should be positively related to access … those with the most resourcerich networks may be least likely to use them to search for financial information because they have greater access to alternative sources of information, such as financial professionals.‖ Chang used the 1998 SCF to analyze the likelihood of seeking financial advice from paid financial professionals: financial planners, accountants, brokers and lawyers. Chang reported that the most common source of advice was friends or relatives, mentioned by 41% of those who reported saving or investing, compared to about 36% who consulted some type of paid financial professional. Chang‘s multivariate analysis showed that use of paid financial professionals increased with education and liquid asset level but decreased with income, was higher for single female head households than for married couples, higher for Black households than for White households, but lower for Other (Hispanic and Other/Asian combined) than for White households, and increased with risk tolerance. Peterson (2006) suggested that a household‘s need for financial planning services should be related to the complexity of its financial situation, which he stated should depend on the number of goals, the number of financial accounts, the number of dependents, and the level of financial resources. He noted that the need for financial planning services must be balanced against the cost of the services. He analyzed the 2004 SCF dataset, and found that the use of a financial planner was ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

positively related to resources and the number of goals and of accounts, but not related to the number of dependents. Hanna and Lindamood (2010) discussed the theoretical benefits of using a financial planner based on expected utility analysis, and estimated the monetary value of hypothetical ideal advice to a na誰ve consumer. Assuming plausible values of risk aversion, advice that is likely to increase wealth in the future is not valued as much as the expected wealth increase, and those with high risk aversion (low risk tolerance) would not place much value on such advice. However, advice that reduces the risk of large wealth losses has very high value, even if the probability of the loss is very low, and the value of such advice increases substantially with risk aversion. Consumers with very high risk aversion (very low risk tolerance) might value such advice very highly. In one example for a household with total wealth of $2,500,000, they demonstrated that advice that eliminates the risk of one in a thousand chance of a loss of 80% of household wealth would have a value of $1,620 if relative risk aversion is very low, but $932,709 if relative risk aversion is very high. Therefore, those with very high risk aversion (very low risk tolerance) should place high values on risk reducing advice. This article analyzes a combination of the 1998 to 2007 Surveys of Consumer Finances in use of financial planners by households, and therefore represents an advance over previous research in testing for changes over time in the use of financial planners. This article is also the first to test separately for the effect of negative net worth on the use of financial planners. By discussing the results in terms of a normative model for the

Journal of Personal Finance

benefits of financial planning services, this article also provides more insights into underserved segments of the U.S. population. Theory Given that financial planners are paid by commissions or fees or a combination of methods, it makes sense that a householdâ&#x20AC;&#x2DC;s resources, including income and assets, would affect its demand for financial planning services. I will focus on the use of financial planners, and ignore the interaction between the demand for financial planning services and the demand for financial advice from others such as bankers and brokers, but including the demand for other types of advice would be an obvious extension to this research. All other things equal, those with low risk tolerance should place a much higher value on financial planning advice that reduces risk than those with high risk tolerance (Hanna & Lindamood, 2010). The reverse is true for advice that increases the expected value of wealth, but the difference in value of such advice for low and high risk tolerance households is much smaller than the difference in value for risk reduction advice. Therefore, households with low risk tolerance should have a higher demand for financial planning services than households with high risk tolerance. The need for financial planning services may be related to the ability of the household to do its own planning, which is presumably related to the complexity of its financial situation as well as its knowledge and cognitive ability. Warschauer (2008) discussed some major issues in financial planning, and ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

obviously the simpler types of households, e.g., a young one person household with no savings or discretionary income, might have low need for financial planning services, whereas an older household with higher income and assets might have higher needs. The ability of a household, in terms of experience with its financial management, might be related to age and cognitive ability, as well as formal learning. Education is related to cognitive ability (Berry, Gruys, & Sackett, 2006). However, even though a person with high cognitive ability may be more likely to be able to manage his or her own financial planning tasks, such a person might also be more likely to recognize the need. As Yuh and Hanna (2010) discussed, education might be related to being more future-oriented, and therefore more educated households might place a higher value on the future benefits of financial planning services. For a particular level of net worth, complexity, and ability of the household to manage its own finances, age may be related to the perceived value of financial planning services in terms of the value of future benefits, based on remaining life expectancy. Discounting future benefits at some rate, e.g., 3% per year, would mean that the present value of the benefits of financial planning services would be much lower for somebody with a 20 year remaining life expectancy than for a 30 year remaining life expectancy, though for younger households there would not be a large difference between a 30 year and a 40 year life expectancy. Therefore, in terms of age, there would not be much difference between a 30 year old and a 40 year old with otherwise similar situations in terms of the present value of future benefits of financial planning, but there might be a substantial difference between a 60 year old and a

Journal of Personal Finance

70 year old, as remaining life expectancies might decrease substantially. As for single head versus couple households, with equal abilities, the remaining life expectancies might imply greater benefit for couple households, but task specialization (Lindamood & Hanna, 2005) would imply that for a given level of resources and complexity, couples would have less need for paid financial planning services than single head households. Methods Data and Variables I use a combination of the 1998, 2001, 2004, and 2007 SCF datasets to study the demand for financial planners. For more information about the SCF datasets and methodological issues, see Bucks, Kennickell, Mach, and Moore (2009), Lindamood, Hanna, and Bi (2007), and Hanna, Lindamood and Huston (2009). The SCF dataset contains five implicates. I use the repeated-imputation inference (RII) method to correct for underestimation of variances due to imputation of missing data (Montalto & Sung, 1996). The descriptive results are weighted to represent the population proportions of households, with the SCF population weights adjusted so that the apparent sample size was equal to the actual sample size. In general, I follow methods suggested by Lindamood, et al. (2007). The dependent variable is whether a household reported using a financial planner for information on savings or investment decisions, and/or for borrowing or credit decisions. One of the questions was: â&#x20AC;&#x2022;What sources of information do you use to make decisions about saving and investments?â&#x20AC;&#x2013; That question, and a similar question about borrowing or credit ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

decisions, presented some alternatives such as magazines/newspapers, and open-ended responses were also coded (see discussions in Elmerick, et al., 2002 and in Chang, 2005). The explanatory variables included in the study are age of the head, education of the household, job status of the household, risk tolerance, household income, presence of children aged under 19, homeownership, and household type, as well as the racial/ethnic self-identification of the respondent. The racial/ethnic categories are those available in the public datasets of the SCF, White, Black, Hispanic, and a combined Other category which is likely to be mostly Asian/Pacific Islander (Hanna & Lindamood, 2008). The possibility of nonlinear effects for age makes it reasonable to include both age and age squared to account for non-linear effects of age in our multivariate analysis, but in the descriptive analyses (Table 2) I classify age using six categories: under age 30, age 30-39, age 40-49, age 50-59, age 60-69, and age 70 and over. Education may have an impact on the financial knowledge of the household, and therefore its choices. For non-couple households, education is based on the highest education attained by the head, but for couple households, it is based on the partner with the higher level of education. For instance, if a husbandâ&#x20AC;&#x2DC;s highest education is a high school diploma and the wife has a bachelor degree, the education of the household is coded as bachelor degree. Job status is based on the head for non-couple households, and for couple households I use the status of both the head and the partner/spouse based on the following: if one or both are self-employed I count the household status as self-employed, if neither is self-employed

Journal of Personal Finance

but at least one is an employee I count the household status as employee, if neither is employed or self-employed but neither is of retirement age I count the household status as no work, and if neither is employed or self-employed but at least one is of retirement age I count the household status as retired. Couples may make different choices and generally may have more potential resources than single people. Having a dependent child under the age of 19 may increase the number of goals but also reduce the amount available for investing (Yuh & Hanna, 2010), so it is unclear whether it will have a positive or negative effect on the use of financial planning services. The income and wealth-related factors include household income, net worth, and homeownership. Household income and net worth are measured using natural logs to capture the possible non-linearity of the relationship, although for our descriptive results in Table 2, I present results using categories of income and net worth. For values of income and net worth equal to zero, the log of 0.01 is used. Net worth is specified as a piecewise (Suits, Mason, & Chan, 1978) log variable to allow for different effects for positive and for negative net worth. Households with negative net worth are different from households with low net worth (Chen & Finke, 1996) so I allow for separate effects of net worth in the negative range versus in the positive range of net worth.1 1

For positive values of net worth, the log of net worth is used, and otherwise that variable is computed as the log of 0.01. A separate variable is computed for negative values of net worth, the log of the absolute value of net worth, and for non-negative values of net worth, that variable is computed as the log of 0.01. ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Statistical Analysis For descriptive results, statistical tests for differences in the proportions of households using financial planners are calculated allowing for the implicate structure of the SCF datasets (SAS code is shown in Chen, 2007), with differences calculated relative to the same reference categories used in the multivariate analyses, e.g., the mean for Hispanic households is compared to the mean for White households. Logistic regression (Logit) is an appropriate technique for a multivariate analysis of a dependent variable with a small number of levels (Allison, 1999). As suggested by Montalto and Sung (1996), this study uses the repeated-imputation inference (RII) method to correct for underestimation of variances due to imputation of missing data.2 I also created graphs to illustrate selected logit results, along with descriptive results.3 Results Descriptive Results The proportion of households using a financial planner increased from 21% in 1998 to 25% in 2007 (Table 1). Based on the SCF sampling weight, in 2007 over 29 million households reported using a financial planner, an increase of almost five million households over 2001. Table 2 contains means tests of using a financial planner by categories of 2

Deaton (1997) suggested that weighting regression procedures using endogenous weights might result in biased estimates, so I did not weight the logistic regression. 3 For the logit results, I used a transformation (Allison, 1999, p. 14) of the estimated coefficients, e.g., for age and age squared, but applied them at the mean levels of each corresponding descriptive category, and adjusted the calculated likelihood so that the mean of the patterns corresponded to the overall sample mean.

Journal of Personal Finance

Table 1 Number of Households Using a Financial planner, and Percent of All Households, by Survey Year Year

Number of Households Using a Financial Planner

Percent of Households Using a Financial Planner

1998

21,670,000

21.1%

2001

21,300,000

22.0%

2004

24,350,000

21.7%

2007

29,300,000

25.2%

Calculated by author, weighted projections from 1998, 2001, 2004, and 2007 Surveys of Consumer Finances

independent variables. (For income and net worth, I used mean rates by categories for the descriptive table, even though I use continuous variables in the logistic regression.) There are significant differences in the likelihood of using a financial planner by most of characteristics used in this study. The likelihood of using a financial planner was roughly the same for 1998, 2001, and 2004, and then it increased significantly in 2007 to 25%. Only 11% of those who said they were unwilling to take any risks with investments used a financial planner, with the other levels of risk tolerance having higher rates, with the peak rate of 33% being for ―above average,‖ and the ―substantial‖ level having a significantly lower rate (29%) than the rate for ―above average.‖ The proportion using a financial planner increased, then decreased with age, from 18% for the less than 30 category to 27% for the 50 to 59 category, then decreased to 16% for the 70 and older category. Married households were the most

Volume 10, Issue 1

Table 2 Using a Financial Planner by Various Characteristics, Bivariate Analysis, Combined 1998-2007 Datasets (Means Test) Variable

Category

% in category (n=17,684)

Survey year

Risk tolerance

Age

Marital status

Racial/ethnic status of respondent

Household education {for couples, maximum level of either partner}

year 1998 year 2001 year 2004 year 2007 no risk average above average substantial Less than 30 30-39 40-49 50-59 60-69 70 and over married single male single female partner White Black Hispanic Other/Asian < high school high school some college bachelor degree post-bachelor degree

24.3 25.1 25.6 25.0 40.7 38.1 17.2 4.0 13.5 19.1 22.2 17.8 12.0 15.5 50.0 14.5 27.2 7.3 75.4 12.8 8.4 3.4 11.1 28.2 26.5 20.1 14.1

Using a Financial Planner Persig. cent level1 21.1 20.0 .004 21.7 .131 25.2 .000 10.8 .000 28.2 .000 33.2 28.7 .000 17.6 23.7 .000 22.9 .000 26.9 .000 23.8 .000 15.7 .000 24.6 .000 19.4 19.5 .842 18.8 .261 23.6 20.8 .000 11.6 .000 17.7 .000 7.2 15.2 .000 23.5 .000 28.7 .000 35.1 .000

Journal of Personal Finance

Table 2 Using a Financial Planner by Various Characteristics, Bivariate Analysis, Combined 1998-2007 Datasets (Means Test) Using a Financial Variable Category Planner Persig. (n=17,684) cent level1 Child<age 19 yes 43.6 22.4 .006 56.4 21.6 no Employment status 61.4 23.1 employee {of household} self-employed 14.0 28.0 .000 retired 21.4 16.3 .000 not employed 3.2 12.9 .000 Homeowner yes 67.9 15.5 .000 32.1 25.1 No Household income 24.9 11.3 0-23,654 23,654-46,250 25.3 17.4 .000 46,251-82,966 24.8 24.6 .000 82,967-135,242 15.0 31.7 .000 >135,242 10.0 39.2 .000 Household net worth <0 7.4 16.2 .000 17.6 10.7 0-14,000 14,001-102,753 25.0 17.3 .000 102,754-333,200 25.0 23.1 .000 333,201-822,716 14.9 32.6 .000 > 822,716 10.1 39.5 .000 All households 100.0 22.0 1 Significance test is for mean difference from reference category for each variable. Bold is the reference category; weighted data; RII technique is used. % in category

likely to use a financial planner (25%), while other types of households had rates roughly five percentage points lower. Only 12% of households with Hispanic respondents used a financial planner, compared to 24% of those with White respondents, 21% of those with Black respondents, and 18% of ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

households with respondents choosing an â&#x20AC;&#x2022;otherâ&#x20AC;&#x2013; racial/ethnic category. The likelihood of using a financial planner increased steadily with education, from 7% for those with less than a high school degree to 35% of those with a post-bachelor degree. Having a child under 19 in the household was related to a slightly higher rate of using a financial planner. Households with a self-employed head or spouse had the highest rate of using a financial planner, 28%, compared to 23% for households with an employee, 16% for retired households, and 13% for those otherwise not employed. Homeowners were more likely to use a financial planner than renters. The likelihood of using a financial planner increased with income, from 23% of households with annual incomes under $23,654 to 39% of households with incomes over $135,242. Over 7% of households had negative net worth. The likelihood of using a financial planner was higher for those households (16%) than was the likelihood for households with net worth of zero to $14,000 (11%) and about the same as the rate for households with net worth of $14,001 to $102,753. The rate steadily increased net worth increases, with almost 40% of those with net worth over $822,717 using a financial planner. Multivariate Results The logistic regression shows the effects of independent variables on the likelihood of using a financial planner (Table 3). Most of the effects are similar to the descriptive patterns shown in Table 2. Figure 1 shows the actual and calculated likelihoods of using a financial planner by survey year. The calculated results are based on the logit coefficients for survey

Journal of Personal Finance

Table 3 Using a Financial Planner, Multivariate Logistic Analysis (n=17,684) Variable1 Intercept Survey year (1998) year 2001 year 2004 year 2007 Risk tolerance (above average) no risk average substantial Age Age squared Marital status (married) single male single female partner Racial/ethnic status (White) Black Hispanic Other/Asian Education (< high school) high school degree some college bachelor's degree post-bachelor degree Presence of a child < 19 Employment status (employee) self employed no work but not retired retired

Using a Financial Planner Coeff.2 p-val.3 s.e. Odds ratio -3.0562

.000

0.2459

-0.0742 0.0532 0.3506

.155 .302 .000

0.0522 0.0515 0.0505

0.928 1.055 1.420

-0.8803 -0.0767 -0.2624 0.0278 -0.0003

.000 .078 .001 .000 .000

0.0611 0.0436 0.0797 0.0077 0.0001

0.467 1.009 0.899 1.028 1.000

-0.1841 0.2107 -0.0879

.004 .000 .292

0.0635 0.0555 0.0835

0.832 1.234 0.916

0.3013 -0.2360 -0.3529

.000 .014 .001

0.0695 0.0958 0.1068

1.352 0.790 0.703

0.3383 0.6038 0.7017 0.8276 -0.1564

.003 .000 .000 .000 .000

0.1149 0.1152 0.1177 0.1196 0.0421

1.403 1.829 2.017 2.288 0.855

-0.0649 -0.2429 0.0487

.159 .082 .471

0.0461 0.1396 0.0676

0.937 0.784 1.050

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Table 3 Using a Financial Planner, Multivariate Logistic Analysis (n=17,684) Using a Financial Planner Coeff.2 p-val.3 s.e. Odds ratio

Variable1

Income (log) [if ≤ 0, log(.01)] 0.0261 .033 0.1225 Net worth (log) [if ≤ 0, log(.01)] 0.1015 .000 0.0111 -Net worth (log) [if ≥ 0, log(.01)] 0.0950 .000 0.0135 Homeowner 0.1004 .089 0.0590 Concordance (mean) 70.2% 1 Reference category in parentheses. 2 Unweighted analysis combining all five implicates. 3 Significance level and standard error based on RII technique.

1.026 1.107 1.100 1.106

Figure 1 Rate of Use of Financial Planner by Survey Year, and Rate Calculated Based on Mean Values of Other Variables 27% 25%

23% 21% 19% 1998 Mean by Year

2001

2004

2007

Calculated at Mean Values of Other Variables

Actual rates based on results shown in Table 2. Calculated rates based on logit results shown in Table 3, at mean values of other variables.

Journal of Personal Finance

Figure 2 Rate of Use of Financial Planner by Risk Tolerance, and Rate Calculated Based on Mean Values of Other Variables 35% 30% 25% 20% 15% 10% no risk

average

above average

substantial

mean by risk tolerance level calculated at mean values of other variables

Actual rates based on results shown in Table 2. Calculated rates based on logit results shown in Table 3, at mean values of other variables.

year in Table 3, with all other variables set at the overall sample means. In other words, the calculated results show what the financial planner use would have been if characteristics such as risk tolerance, income, net worth, and household composition had not changed during the period. For both the actual and calculated results, there was not much change for 1998, 2001, and 2004, but there was a substantial increase between 2004 and 2007 for both the actual and calculated likelihoods. Figure 2 shows the actual and calculated likelihoods of using a financial planner by risk tolerance. As with the descriptive results, the highest likelihood of using a financial planner is for those with above average risk tolerance, although those with average risk tolerance were not significantly ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Figure 3 Rate of Use of Financial Planner by Category of Age of Head, and Rate Calculated Based on Mean Values of Other Variables 27% 25% 23% 21% 19% 17% 15% 25

45 Mean of Age Group

Calculated Based on Logit

Actual rates based on results shown in Table 2 (age of head <30, 30-39, 40-49, 50-59,60-69, 70 and over.) Calculated rates based on logit results shown in Table 3, at mean values of other variables.

different from those with above average risk tolerance based on the logit. Those having substantial risk tolerance were significantly less likely to use a financial planner than those with above average risk tolerance. As with the actual pattern (Table 2), the logit implies that those unwilling to take any risk were much less likely to use a financial planner than those willing to take average or above average risk, though the differences were somewhat reduced because of the setting of income, net worth, and other household characteristics at the overall sample mean.

Journal of Personal Finance

Figure 4 Rate of Use of Financial Planner by Racial/Ethnic Category of Respondent, and Rate Calculated Based on Mean Values of Other Variables 27% 25% 23% 21% 19% 17% 15% 13% 11% White mean by groupl

Black

Hispanic

Asian/other

calculated at mean values of other variables

Actual rates based on results shown in Table 2. Calculated rates based on logit results shown in Table 3, at mean values of other variables.

Figure 3 shows the actual and calculated likelihoods of using a financial planner by the age of the head. The combined effect of age and age squared implies that the likelihood of using a financial planner increases until age 42, then decreases, so the peak is lower than the peak age range in the descriptive results, 50 to 59. Note that the calculated likelihood of using a financial planner for those under 30 is almost as high as for those age 30 to 39 or 40 to 49, which is because of the assumption that the younger households had the same net worth and other characteristics as the sample means. Both the actual and calculated likelihoods decreased substantially from about age 55 to age 80.

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Figure 5 Rate of Use of Financial Planner by Household Education, and Rate Calculated Based on Mean Values of Other Variables 35% 30% 25% 20% 15% 10% 5% <HS

Some college

Post-BS

mean by education level calculated at mean values of other variables

Actual rates based on results shown in Table 2. Calculated rates based on logit results shown in Table 3, at mean values of other variables. Household education is defined by the highest education level of the head for single head households, and the maximum education level of either partner for couple households.

Figure 4 shows the actual and calculated likelihoods of using a financial planner by the racial/ethnic identification of the respondent. Unlike the actual patterns, the calculated patterns show that households with a Black respondent would be more likely than households with a White, Hispanic, or Other/Asian respondent to use a financial planner, if each group had the overall sample mean levels of net worth and other household characteristics. Households with Hispanic respondents and households with Other/Asian respondents would be less likely than households with White respondents to use a financial planner, given equal household characteristics.

Journal of Personal Finance

Figure 6 Rate of Use of Financial Planner by Household Income Category, and Rate Calculated Based on Mean Values of Other Variables 40% 35% 30% 25% 20% 15% 10% ă&#x20AC;&#x20AC;0-23,654

23,65446,250

46,25182,966

82,967135,242

>135,242

Income Category mean by income category calculated at mean values of other variables

Actual rates based on results shown in Table 2. Calculated rates based on logit results shown in Table 3, at mean values of other variables.

Figure 5 shows the actual and calculated likelihoods of using a financial planner by household education. Both patterns show rates substantially increasing with education, though the calculated pattern is less steep than the actual pattern, because of the assumption that each group has the overall sample means of net worth and other characteristics. Figure 6 shows the actual and calculated likelihoods of using a financial planner by household income. The effect of income in Table 3 is statistically significant, but the magnitude of the effect shown in the graph is small, with most of the effect for increases from very low income to the mean of the

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Figure 7 Rate of Use of Financial Planner by Household Net Worth Category, and Rate Calculated Based on Mean Values of Other Variables 40% 35% 30% 25% 20% 15% 10% <0

0-14,000

14,001102,753

102,754333,200

333,201- >822,716 822,716

Net Worth Category mean by net worth category calculated at mean values of other variables

Actual rates based on results shown in Table 2. Calculated rates based on logit results shown in Table 3, at mean values of other variables.

lowest categoryď&#x20AC;´ (not shown in Figure 6). A household with an annual income of $0.01 would have a calculated likelihood of using a financial planner of 16%, assuming mean values of net worth and other characteristics, while one with an income of $343,455 (mean of top decile) would have a calculated likelihood of 23%, and one with an income of $25,000,000 would have a calculated likelihood of 25%. Figure 7 shows the actual and calculated likelihoods of using a financial planner by household net worth. The ď&#x20AC;´

For the age, income, and net worth graphs (Figures 3, 6, and 7) the logit results were used to calculate likelihoods at the mean levels for the descriptive categories, e.g., the calculated likelihood for the lowest income category in Figure 6 is for the mean income in that category, $13,437.

Journal of Personal Finance

calculated likelihood of using a financial planner increases strongly with net worth as net worth increases from zero, but it also increases strongly as net worth becomes more and more negative. The calculated likelihood shown for the negative net worth category is for the mean level of net worth for that category, -$16,032. The logit coefficient for Ln(-net worth) implies that a household with a negative net worth of $300,000, would be as likely to use a financial planner as an otherwise similar household with positive net worth of $99,980. Controlling for net worth, income, and other characteristics, single headed female households are significantly more likely than married couple and single headed male households to use a financial planner, unlike the actual pattern of married couple households being more likely than single female households to use a financial planner. There is a substantial difference between single female and single male households in the calculated likelihood of using a financial planner, presumably because of the greater self-confidence of males and their reluctance to seek help. Households with a child under 19 are less likely to use a financial planner than otherwise similar households without a child under 19, although the difference is small. Homeowners are not significantly different from otherwise similar renters in the likelihood of using a financial planner. Controlling for other characteristics, households with employee job status are not significantly different from those categorized as selfemployed, retired, or not working.

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Implications The substantial increase in the use of financial planners between 2004 and 2007 may provide optimism for the financial planning industry, but some of the patterns suggest underserved market segments. The result that even after controlling for income, age, and net worth, those with low risk tolerance (unwilling to take investment risks) are less likely to use a financial planner than those with higher risk tolerance may seem reasonable in terms of the idea of financial planning as portfolio management, but in terms of the theoretical results demonstrated by Hanna and Lindamood (2010), those with low risk tolerance but high net worth or income should place substantially higher value on the risk management aspects of comprehensive financial planning than should households with high risk tolerance. Those under 30 are unlikely to use financial planners, but the logit results suggest that level is appropriate relative to the low net worth of young households, especially to the extent that benefits of financial planning are more related to protecting assets than increasing assets. The decrease in use of financial planners by elderly households seems reasonable in terms of decreasing future benefits because of more limited remaining life expectancies, but for those with substantial assets, the value of reducing risks should still be substantial (Hanna & Lindamood, 2010). Single male headed households also seem to be an underserved segment. As Elmerick, et al. (2002) showed, other things equal, households with a Black respondent are much more likely than similar households with a White respondent to use a financial planner for credit or borrowing decisions, and somewhat more

Journal of Personal Finance

likely to use a financial planner for savings or savings or investment decisions, so credit problems of Black households might be part of the differences in terms of Black-White differences in overall use of financial planners. The result that households with Hispanic and with Other/Asian respondents are significantly less likely to use financial planners than those with White or Black respondents suggests that populations with substantial proportions of immigrants are underserved by financial planners. Almost 40% of Hispanics in the U.S. are immigrants, and 67% of Asians are immigrants (U.S. Census Bureau, 2010). Immigrants who lack familiarity with financial planning in the United States may be a factor, but increased marketing to these segments may be beneficial. Chatterjee (2009) found that immigrants have lower participation in U.S. financial markets than native-born Americans, so that difference may also help explain the lower use of financial planners by Hispanics and the Asian/other group in the Survey of Consumer Finances. The strong effect of education on the likelihood of using a financial planner after controlling for net worth and other characteristics suggests that less educated affluent households may be underserved by financial planners. To the extent that low education is related to being more present-oriented, it is possible that those households might not value the future benefits of financial planning highly, but presumably those households might find financial planning by themselves to be more challenging than more educated households. The small negative effect of having a dependent child under the age of 19 suggests that even though the number of ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

goals may be higher for those with one or more children, the reduction in the amount available for investing outweighs that effect. Net worth seems to be much more important than household income in the likelihood of using a financial planner. Further research should consider the impact of different components of net worth on the likelihood of using a financial planner. However, the result that being a homeowner does not have a significant effect in the logit suggests that the most important variation in typical household net worth does not matter much in whether households use a financial planner. References Allison, P. D. (1999). Logistic regression Using SAS: Theory and application. Cary, NC: SAS Institute, Inc. Bae, S. C. & Sandager, J. P. (1997). What consumers look for in financial planners. Financial Counseling and Planning, 8(2), 9-16. Berry, C. M., Gruys, M. L., & Sackett, P. R. (2006). Educational attainment as a proxy for cognitive ability in selection: Effects on levels of cognitive ability and adverse impact. Journal of Applied Psychology, 91(3), 696-705. Bucks, B. K., Kennickell, A. B., Mach, T. L., & Moore, K. B. (2009). Changes in U.S. family finances from 2004 to 2007: Evidence from the Survey of Consumer Finances. Federal Reserve Bulletin, 95, A1-A55. Chang, M. L. (2005). With a little help from my friends (and my financial planner). Social Forces, 83(4), 1469-1498. Chatterjee, S. (2009). Do immigrants have lower participation rates in U.S. financial markets? International Journal of Business and Finance Research, 3(2), 1-13. Chen, C.C. (2007). Changes in retirement adequacy, 1995-2004: Accounting for retirement stages. Dissertation, The Ohio State University. Chen, P. & Finke, M. S. (1996). Negative net worth and the life cycle hypothesis. Financial Counseling and Planning, 7, 87-95. Deaton, A. (1997). The analysis of household surveys: A microeconometric approach to development policy. Baltimore, MD: Johns Hopkins University Press.

Journal of Personal Finance

Elmerick, S. A., Montalto, C. P., & Fox, J. J. (2002). Use of financial planners by U.S. households. Financial Services Review, 11(3), 217213. Hanna, S. D. & Lindamood, S. (2008). The decrease in stock ownership by minority households. Journal of Financial Counseling and Planning, 19(2), 46-58. Hanna, S. D., & Lindamood, S. (2010). Quantifying the economic benefits of personal financial planning. Financial Services Review, 19(2), 111127. Hanna, S. D. Lindamood, S., & Huston, S. J. (2009). National household datasets for financial research: Survey of Consumer Finances. Proceedings of the Academy of Financial Services. Retrieved from http://www.academyfinancial.org/09Conference/09Proceedings/(2E) Hanna, Lindamood, Huston.pdf Lindamood, S., & Hanna, S. D. (2005). Determinants of the wife being the financially knowledgeable spouse. Proceedings of the Academy of Financial Services. Lindamood, S., Hanna, S.D., & Bi, L. (2007). Using the Survey of Consumer Finances: Methodological considerations and issues. Journal of Consumer Affairs, 41(2), 195–214. Montalto, C. P. & Sung, J. (1996). Multiple imputation in the 1992 Survey of Consumer Finances. Financial Counseling and Planning, 7, 133– 146. Peterson, B. (2006). Are households with complex financial management issues more likely to use a financial planner? Thesis, University of Wisconsin - Madison. Suits, D. B., Mason, A.. & Chan, L. (1978). Spline functions fitted by standard regression methods. Review of Economics and Statistics, 60, 132-139. U.S. Census Bureau (2010, January). Race and Hispanic Origin of the foreign-born population in the United States: 2007. American Community Survey Reports. Warschauer, T. (2008). The economic benefits of personal financial planning. Proceedings of the Academy of Financial Services. Yuh, Y. & Hanna, S. D. (2010). Which households think they save? Journal of Consumer Affairs, 44(1), 70-97.

Volume 10, Issue 1

CAN DUAL BETA FILTERING IMPROVE INVESTOR PERFORMANCE? James Chong, Ph.D.ď&#x20AC;Ş California State University, Northridge Shaun Pfeiffer, Ph.D. Candidate Texas Tech University G. Michael Phillips, Ph.D. California State University, Northridge This study investigates the possibility that more efficient portfolios may be constructed by using the dual-beta model that screens out assets that exhibit more extreme downside risk sensitivity. Three portfolios were constructed, using the criteria of standard CAPM beta, down-market beta, and a combination of up-market and down-market betas. Overall, the standard CAPM beta consistently lags the dual-betas. When compared to the Fama-French threefactor inspired DFEOX, the dual-betas also performed reasonably well, with the ability to contain the downside while participating in the upside.

Introduction and Literature Review Individual investors appear more sensitive to investment losses than would be predicted by neoclassical economic preferences (Tversky & Kahneman, 1991; Ang, Chen, & Xing, 2006). This sensitivity compromises realized portfolio performance by inducing extreme rebalancing toward safety ď&#x20AC;Ş

James Chong, Department of Finance, Real Estate, and Insurance, California State University, Northridge, 18111 Nordhoff Street, Northridge, CA 91330-8379; (818) 677-4613; jchong@macrorisk.com The authors are grateful to the editor, Michael Finke, and an anonymous referee for useful comments on a previous version of this paper. The usual disclaimer applies.

Journal of Personal Finance

following market declines (Barberis, Huang & Santos, 2001) or by limiting exposure to equities (Siegel, 2005). Consequences from loss aversion can significantly reduce future financial well-being when the portfolio does not reflect this type of risk preference. Prior research also notes that asset volatility may not be symmetric between gains and losses (Estrada, 2007). If certain assets exhibit more extreme declines in performance during a market decline, it may be possible to identify these securities ex ante in order to construct more optimal consumer portfolios that are more attractive to individual investors. This study investigates the possibility that more efficient portfolios may be constructed by using a portfolio selection technique that screens out assets that exhibit more extreme downside risk sensitivity. Loss Aversion Loss aversion is defined as higher sensitivity to investment losses than gains. Research estimates that the pain of loss for a typical investor is roughly twice the pleasure from an equivalent gain (Kahneman, Knetsch, & Thaler, 1990; Tversky & Kahneman 1991). This type of risk preference makes risky assets less appealing to the investor. Loss aversion can be magnified by behavioral biases such as mental accounting. For example, sensitivity to losses is shown to increase with frequency of account evaluation (Benartzi & Thaler, 1995). Findings from Barber and Odean (2000) suggest that average investors turn over roughly 75% of their portfolios each year, which supports the notion of frequent account evaluation. Higher levels of loss aversion are associated with less equity exposure, a desire for portfolio insurance, or some combination of protective strategies and a reduction in equity ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

exposure (Berkelaar, Kouwenberg & Post, 2004). According to Siegel (2005), loss-averse investors reduce stock holdings and forego a substantial equity risk premium over longer investment horizons. Additionally, Barber, Odean, and Zheng (2000) find loss aversion, or the propensity to hold losers and sell winners, leads to an annual reduction in portfolio returns of roughly 3.5%. In turn, research suggests that loss aversion may be more important than risk aversion when constructing a portfolio (Basu, Raj & Tchalian, 2008). Portfolio performance of a loss-averse investor can suffer due to rebalancing into safer asset classes after poor market returns. For example, Barberis, Huang and Santos (2001) suggest that investors become more risk seeking following gains and more risk averse following losses. This leads to a buy high and sell low strategy where the investor realizes lower than average returns over the investment horizon. Specifically, investors fail to benefit from mean reversion that is associated with asset prices over longer investment horizons (Debondt & Thaler, 1985). Additionally, portfolio transactions triggered by behavioral biases and loss aversion have been shown to reduce returns by 1% to 5% per year versus a buy and hold strategy (Barber et al., 2000; Barber & Odean, 2000). Cochrane (1999) suggests that this type of portfolio rebalancing represents a shift in investor risk preference, which sacrifices returns in order to reduce risk. Portfolio Construction Investors may delegate portfolio decisions to a financial planner in an attempt to reduce the negative effects of loss aversion and other behavioral biases. Hanna and Lindamood

Journal of Personal Finance

(2010) note loss prevention as one of the primary benefits financial planners offer to their clients. The planner attempts to construct an optimal portfolio based on the goals and unique risk preferences of each client (Eyssell, 2003) in addition to the individual investment attributes relative to the overall portfolio (Markowitz, 1952). Mean-variance optimization is a process that many planners use to derive a long-term asset allocation. Farrelly (2006) notes that optimization involves a great deal of art for many practitioners. In other words, practitioners frequently constrain the optimization output in order to account for errors in the optimization assumptions, behavioral biases and risk preferences of the client. Many practitioners rely on the tenets of Modern Portfolio Theory (MPT) and Capital Asset Pricing Theory (CAPM) when constructing investment portfolios. The notion that investors should consider risk in portfolio decisions is central to seminal studies in finance (Markowitz, 1952; Sharpe, 1964). MPT defines risk as the variance of investment returns. Beta represents risk in the CAPM framework. Beta is systematic risk and is seen as the covariance of returns between an investment and the market portfolio relative to the variance of returns of the market portfolio. In short, CAPM states that the expected return on an investment is solely a function of beta, the investor is not compensated for bearing unsystematic risk, and high beta stocks are expected to outperform low beta stocks in periods of positive market returns. Additionally, CAPM suggests that more risk-averse investors should increase the amount of riskfree securities while maintaining the value-weighted exposure to risky assets (Canner, Mankiw, & Weil, 1997). However, practitioners rely on risk metrics such as beta and variance of ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

sub-asset classes to make asset allocation decisions. Beta is widely used by investment advisors to measure portfolio risk (Levy, 2010; Chan & Lakonishok, 1993). Research estimates that 70% of practitioners use the CAPM beta as a measure of systematic risk (Graham & Harvey, 2001). Risk Measurements Empirical evidence suggests that beta is an imperfect measure of investment risk. Potential estimation error associated with beta due to lower r-squared statistics (Eyssell, 2003) has led many researchers to estimate beta on portfolios rather than individual securities (Blume, 1975; Fama & French, 1992). Aggrawal and Waggle (2010) find that beta varies significantly across different financial websites. The authors note that the deviation is due to the use of different proxies for the market. In addition to the errors in beta, there is mixed empirical evidence on CAPM. Early empirical evidence supports the claims of CAPM (Jensen, 1969; Downs & Ingram, 2000). Subsequent studies, however, find many empirical contradictions in relation to the claims of CAPM (Fama & Macbeth, 1973; Black, Jensen, & Scholes, 1972). Research finds a size (Banz, 1981) and value effect (Stattman, 1980; Rosenberg, Reid, & Lanstein, 1985) are important in explaining investment returns. Together these findings suggest that beta is not the only factor explaining returns and eventually lead to the formation of the three-factor model (Fama & French, 1992). Findings that suggest beta is not positively related to average returns over varying periods of analysis are even more troubling to the predictions of CAPM (Fama & French, 1992). In other words, the relationship

Journal of Personal Finance

between beta and average returns is not reliable and the CAPM beta may not be the best way to define risk. Beta can remain useful to practitioners in spite of the empirical evidence against it. The findings that contradict CAPM suggest that the relationship between beta and average returns is flatter than predicted by CAPM. Black (1993) notes that beta remains useful to investors and planners as a risk measure. Specifically, low beta investments offer higher riskadjusted returns than high beta investments. Other studies suggest that beta is a useful indicator of downside risk exposure in declining markets. Grundy and Malkiel (1996) find that higher beta stocks consistently underperform lower beta stocks during periods where the S&P drops by more than 10%. Together, these findings suggest that the strength of empirical support for beta is weaker than CAPM would suggest; however, this does not mean beta is a useless measure of risk. Research also provides many alternative measures of risk for practitioners to use in portfolio design. The concern for downside risk measures is captured in many early studies. For example, Markowitz (1959) notes that semi-variance, or downside deviation, is a better measure of risk than variance. The author adds that the use of semi-variance, rather than variance, in the optimization process can lead to better portfolios. These suggestions are based on the idea that investors are typically loss averse. Collectively, the concerns associated with the traditional CAPM beta and the concept of loss aversion has led to the suggestion that there may be better measures of risk than beta (Chan & Lakonishok, 1993). Using the idea of semi-variance, Estrada (2007) constructs a ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

downside beta. This risk measure captures the sensitivity to the market when market returns are negative or below some threshold such as average market returns. Galagedera (2009) suggests that the downside beta is a better measure of systematic risk than the CAPM beta and that the difference between these risk measures is greatest for low volatility portfolios. Additional research suggests that the use of downside risk measures is better able to estimate an appropriate amount to allocate to risky assets (Berkelaar, Kouwenberg, & Post, 2004). Loss Prevention Strategies Loss avoidance is of key importance and the dual-beta model is found to be of value in capturing the downside risk. The same can be said for many financial planners and their clients who are concerned with capital preservation and loss avoidance (Bajtelsmit, 2005). However, there are many strategies that a financial planner can employ to mitigate portfolio losses. Asset allocation, rebalancing, the use of derivatives, and reducing exposure to stocks are a few ways to mitigate portfolio losses. First, note that total risk is a function of systematic risk and unsystematic risk (Xiong, Ibbotson, Idzorek, & Chen, 2010). Asset allocation, which includes diversification within and across asset classes, is an approach to reduce unsystematic risk and the potential for significant portfolio losses (Markowitz, 1952). Correlation tightening during market declines (Bauer, Haerden, & Molenaar, 2004) and positive correlation between stocks and bonds over longer investment horizons (Campbell & Ammer, 1993) are limitations of asset allocation as a tool to mitigate portfolio losses. Research estimates that rebalancing can reduce

Journal of Personal Finance

volatility by roughly 1% and increase portfolio returns by approximately 50 basis points per year (Plaxco & Arnott, 2002). Although rebalancing incurs transaction costs and taxes the benefit of maintaining a certain risk exposure is important to the planning profession (Daryanani, 2008). Portfolio insurance, or rolling put options, is another loss prevention strategy that may be considered by planners. However, Arnott (1998) notes that the cost of this strategy can be as high as 5% per year. Eliminating exposure to equities can reduce portfolio shocks; however, the client forfeits the benefit of an equity risk premium. Our study focuses on the use of downside beta in an attempt to construct more attractive portfolios for clients. We acknowledge that portfolio strategies based on downside beta should be used alongside proper asset allocation and rebalancing criteria. Recent research shows that low beta portfolios can provide higher returns and lower volatility than high beta portfolios (Baker, Bradley & Wurgler, 2011). Our study attempts to identify more efficient portfolios based on the downside beta used in Estrada (2006) by screening out assets that exhibit greater downside sensitivity. Our work is closely related to that of Pettengill et al. (1995), who provide contrary evidence to that of Fama and French (1992), in that there is a significant relationship between beta and returns so long as one segregates beta into its up-market and down-market components (henceforth, referred to as the dual-beta model).1 Our empirical findings are clearâ&#x20AC;&#x201D; 1

Further literature on up- and down-market betas can be found in Moelli (2007) and the references therein. ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

the standard CAPM beta consistently lags the dual-betas, in terms of average daily returns and return-to-standard deviation ratio. As such, the dual-beta model is superior to the standard CAPM model. When compared to the Fama-French threefactor inspired portfolio, the dual-betas also performed reasonably well, with the ability to contain the downside while participating in the upside. The relatively poor performance of the traditional beta portfolio suggests the need for financial planners to explore alternative forms of stock selectionâ&#x20AC;&#x201D;the Fama-French three-factor model is such an alternative but a more tractable solution can be found in the dual-beta model. The findings of this study advocate to financial planners, if they have not done so already, the use of the dual-beta model for stock selection and portfolio construction. The paper is structured as follows. We begin by providing a brief overview of our sample data, followed by a description of our methodology. We then proceed to present some results from our findings. Finally, we end with our conclusions. Data and Methodology DFA Core Equity 1 Portfolio (DFEOX) The efficacy of the Fama-French three-factor model2 has led many to conclude that â&#x20AC;&#x2022;alpha can be elusive when measured against the three-factor modelâ&#x20AC;&#x2013; (Pollock, 2007). 2

In addition to beta, Fama and French found size (i.e., the return on small stocks minus the return on big stocks, SMB) and value (i.e., the return on high book-to-market stocks minus the return on low book-to-market stocks, HML) to be of significance in explaining average returns and therefore, valid proxies for risk.

Journal of Personal Finance

Subsequently, a new benchmark was proposed—―the new face of indexing‖ (Fama, 2000)—where ―the goal of indexing switches from diversification across the available stocks to diversification across the available risk-return dimensions,‖ resulting in the creation of the Dimensional Fund Advisors (DFA) Core Equity 1 Portfolio (DFEOX), which seeks ―to buy the total U.S. market in proportions that provide higher exposure to the risk premiums associated with size and value identified by Fama and French.‖3 The DFEOX is categorized under ―Large Blend‖ by Morningstar and therefore is deemed an appropriate benchmark for large cap portfolio performance.

Standard CAPM Model Although beta has been shown by Fama and French (1992) to be an imperfect measure of investment risk, the standard CAPM model, where beta is derived from, is still popular among financial planners and investment professionals and can be expressed as (

where

)

(1)

is the risk-free rate (we use the overnight U.S. Federal

funds rate as proxy),

is the return on asset j, (

observed excess return on asset j, intercept, called alpha,

) is the

is the estimated regression is the estimated excess return

on the market index (here, the S&P 500 index, SPX), and is the unexplained portion of the model. In our paper, we estimate the standard CAPM beta using one-year daily returns. 3

Volume 10, Issue 1

Dual-Beta Model The dual-beta model is an extension of the standard CAPM model. It estimates the parameters separately for upmarket, when the daily return for the market index is nonnegative, and down-market, when the daily return for the market index is negative. The dual-beta model can thus be described as (

)

(

) ,

where

, and

(2)

are the estimated parameters for up-

market and down-market days respectively; on days the market index did not decline and on days it did; D is a dummy variable, which takes the value of 1 when the market index daily return is non-negative, and zero otherwise. If there is no asymmetry in beta, then and , i.e., equations (1) and (2) are identical. As with the standard CAPM beta, we estimate the up-market and down-market betas using one-year daily returns. Portfolio Construction with Standard CAPM and Dual-Beta Models For our analysis, we employ daily data, from January 1, 2006 to March 4, 2011, for a total of 1,350 data points.4 DFEOX was established on November 1, 2005, and for convenience, we used January 1, 2006 as the start date. We construct three separate portfolios for comparison to DFEOX. The portfolio construction and rebalancing processes

These data were provided by MacroRisk Analytics from their database.

Journal of Personal Finance

are initiated at the beginning of each quarter,5 using a buy-list of stocks in the S&P 500 index.6 The criteria we impose on the choice of stocks are a standard CAPM beta of less than 0.7,7 down-market beta of less than 0.7, and the combination of down-market beta of less than 0.7 and up-market beta of greater than 0.7. The portfolio is then constructed with equal weighting on the stock components. The rationale for these various beta models is an attempt to capture various characteristics of the market. The standard CAPM beta (henceforth referred as Beta) is one of the most popular measures of investment risk, and as such, we also employ it here. We are taking a conservative approach and impose a filter of Beta that is less than 0.7. The down-market beta (referred to as Dbeta) criterion of less than 0.7 is taking on a risk-averse stance only when the market goes down. The combination beta of Dbeta of less than 0.7 and up-market beta (Ubeta) of greater than 0.7 (referred to as Combination) is to ensure conservatism on down-market days but acquire more risk on up-market days. Lastly, DFEOX is our large cap performance benchmark, which utilizes the Fama-French threefactor model.

As the process is initiated at the beginning of each quarter, there is no lookahead bias. 6 The Fama-French three-factor model is applied to the total market, which is defined as companies listed on the NYSE, AMEX, and NASDAQ Global Market System. By restricting ourselves to only S&P 500 stocks, we are limiting the effectiveness of our portfolio. 7 In theory, market beta equals 1. However, Chong and Phillips (2009) found that the median beta of stocks listed on the New York Stock Exchange is 0.7. ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Conditional Volatilities and Correlation Estimating conditional volatilities and correlations, with the GARCH (1,1) and DCC (1,1) models respectively, has become almost standard practice in finance. With the GARCH (1,1) model (Bollerslev, 1986), an asset‘s conditional variance ( ) can be expressed as (3)

subject to With the DCC (1,1) model (Engle, 2002), the time-varying covariance matrix is expressed as , where is the diagonal matrix of GARCH (1,1) volatilities, is the timevarying correlation matrix, is a diagonal matrix comprising the square root of the diagonal elements of , and is ̅

(  t 1t 1 )

(4)

where ̅ is the unconditional covariance and a and b are scalars. The coefficients of (3) and (4) are estimated by the maximum likelihood procedure using the BFGS algorithm. Results A graphical illustration of how the various beta models performed in relation to DFEOX is presented by Figure 1. We begin at $1 on January 1, 2006 and end on March 4, 2011. For our sample period, the Combination beta had the highest cumulative wealth of $1.3974. This was followed by DFEOX ($1.2465), Dbeta ($1.2298), and Beta ($1.1401). Prior to the financial crisis, the various portfolios tracked each other closely, with separation between the portfolios occurring at approximately December 2008. Further, we note that none of

Journal of Personal Finance

Figure 1 Cumulative Wealth

the beta-generated portfolios plummeted as did DFEOX during the financial crisis. DFEOX reached its trough on March 9, 2009 ($0.5560) but made a surge in the remainder of our sample period, eventually surpassing Beta and Dbeta. This would suggest that DFEOX has higher volatility than the betagenerated portfolios. In Table 1, Panel A, we report summary statistics of returns and risk for the whole sample period. The results (for mean daily return, standard deviation) confirmed somewhat our analysis of Figure 1. Although the Combination beta has the second highest volatility, this is offset by its returns, resulting in the highest return-to-standard deviation ratio (0.0255) among the portfolios. Coming in second is Dbeta (0.0193), whose ratio exceeded that of DFEOX (0.0182). ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Table 1 Summary Statistics of Returns and Risk Beta < 0.7

Dbeta < 0.7

Dbeta < 0.7, Ubeta > 0.7

DFEOX

Panel A: Whole Period (1/1/06 – 3/4/11) Mean 0.0156% 0.0216%

0.0334%

0.0288%

Median

0.0439%

0.0285%

0.0263%

0.0835%

Standard Deviation

1.0812%

1.1222%

1.3115%

1.5787%

Maximum

10.0472%

9.3606%

8.6777%

11.0165%

Minimum

-7.0931%

-7.0606%

-8.0649%

-9.3727%

Ratio*

0.0144

0.0193

0.0255

0.0182

Correlation

0.8984

0.9263

0.9312

1.0000

1,350

Panel B: First Period (1/1/06 – 3/8/09) Mean -0.0321% -0.0278%

Sample

-0.0212%

-0.0550%

Median

0.0293%

0.0000%

0.0011%

0.0000%

Standard Deviation

1.2255%

1.2758%

1.4621%

1.6778%

Maximum

10.0472%

9.3606%

8.6777%

11.0165%

Minimum

-7.0931%

-7.0606%

-8.0649%

-9.3727%

-0.0262

-0.0218

-0.0145

-0.0328

0.9134

0.9425

0.9467

1.0000

830

Panel C: Second Period (3/9/09 – 3/4/11) Mean 0.0917% 0.1005%

0.1205%

0.1626%

Median

0.0804%

0.0794%

0.0561%

0.0988%

Standard Deviation

0.7937%

0.8145%

1.0215%

1.3975%

Maximum

3.8441%

4.3498%

5.0254%

7.3028%

Minimum

Ratio* Correlation Sample

-3.0903%

-3.0619%

-3.4971%

-4.8949%

Ratio*

0.1155

0.1234

0.1180

0.1164

Correlation

0.8741

0.9011

0.8985

1.0000

520

Sample *

Ratio = Mean return divided by standard deviation.

Journal of Personal Finance

Figure 2 Conditional Correlation with DFEOX, using the DCC (1,1) Model

Further analysis is undertaken by separating the sample period in two, at the point when DFEOX was at its lowest (see Figure 1, when DFEOX was at $0.5560 on March 9, 2009). This allows us to assess the portfolio structure prior to and during the financial crisis (Panel B of Table 1) and subsequent recovery (Panel C of Table 1), while also ensuring the robustness of our findings. Prior to March 9, 2009, all portfolios experienced loss (Panel B of Table 1). Even though the various beta-generated portfolios were highly correlated with DFEOX, with correlation coefficients in excess of 0.9, their average daily returns differed—the Combination beta registered average daily returns of -0.0212% while DFEOX‘s average daily returns was -0.0550%. On closer examination of their ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Figure 3 Conditional Volatility with the GARCH (1,1) Model

conditional correlations (Figure 2), sizeable fluctuations in correlation are evident, which explains the phenomenon of high correlation accompanied by vastly differing returns. It is also apparent from Figure 3 (and corroborated by Table 1, Panel B) that DFEOX has consistently higher volatility, and consequently inferior return-to-standard deviation ratio, than beta-generated portfolios. Post-March 2009 witnessed a surge by DFEOX with an average daily return of 0.1626%. However, associated with higher return was higher volatility relative to other portfolios; unlike pre-March 2009, the difference in volatility between DFEOX and the other portfolios is much greater (Figure 3). Accordingly, the return-to-standard deviation ratio of DFEOX lagged those of Dbeta and Combination beta. Dbeta, with

Journal of Personal Finance

relative low volatility, had a superior ratio over Combination beta (0.1234 vs. 0.1180) despite a lower average daily return (0.1005% vs. 0.1205%). Overall, the standard CAPM beta consistently lags the dual-betas, in terms of average daily returns and return-tostandard deviation ratio. As such, the dual-beta model, which segregates the traditional beta into its up- and down-market components, is superior to the standard CAPM model. When compared to the Fama-French three-factor inspired DFEOX, the dual-betas also performed reasonably well, with the ability to contain the downside while participating in the upside. This augurs well for the dual-beta model, which is considerably simpler to implement and explain to clients than the FamaFrench three-factor model. Limitations This studyâ&#x20AC;&#x2DC;s main objective is to investigate the possibility that more efficient portfolios may be constructed by using the dual-beta model that screens out assets that exhibit more extreme downside risk sensitivity. However, there are questions left unanswered. For instance, accounting for transaction costs in establishing and maintaining a stock-only portfolio may result in reduced efficacy of the dual-beta model. In Table 2, we provide summary statistics of average transactions (and transaction costs) per quarter. The Combination beta has the lowest average number of transactions per quarter. This is intuitive since there are fewer stocks that meet the criteria imposed by this model. Assuming an investor executes stock trades via a discount brokerage ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Table 2 Summary Statistics of Average Transactions (Costs) per Quarter Beta < 0.7

Dbeta < 0.7

Portfolio Components

Dbeta< 0.7, Ubeta > 0.7 29

No. of Transactions

101

Transaction Costs ($) Cost as % of NAV

574

707

308

0.06%

0.07%

0.03%

(e.g., Scottrade at $7 per stock trade), the transaction costs incurred by the Combination model, in absolute terms and as a percentage of net asset value (NAV), are the lowest of the (dual-)beta models. On the other hand, the Dbeta model suffers from the highest average transaction costs. While the transaction costs mentioned above may be reasonable for a financial planner, they may be exorbitant for an individual investor, in which case, employing exchange traded funds (ETFs) would be an alternative strategy, given that commission-free ETFs are being offered by brokers (e.g., Scottrade, Charles Schwab). Of course, for more institutional activities, prime brokerage operations allow for extremely inexpensive trades. Summary and Conclusion This article has sought to provide a review of the standard CAPM model and the dual-beta model available for stock 

The transaction costs as a percent of net asset value is dependent on the amount under management, which we assumed to be $1 million. 

A separate analysis using ETFs (not shown, but available from the authors on request) showed improved performance for all (dual-)beta models over their stock-only counterparts.

Journal of Personal Finance

selection and their effectiveness. Three portfolios were constructed, using the criteria of standard beta less than 0.7, down-market beta less than 0.7, and the combination of downmarket less than 0.7 and up-market greater than 0.7. The criteria used are especially appropriate for a loss-averse investor, who has a higher sensitivity to investment losses than gains. Thus, this article could be viewed as having provided evidence on the effectiveness of loss prevention strategies in stock selection. Further, in the quest for wealth enhancement and loss prevention, a strategy of combining up- and downmarket betas was employed with success. In addition to standard and dual-betas, we chose a performance benchmark inspired by the Fama-French three factor model, the DFA Core Equity 1 Portfolio (DFEOX). Recall that Fama and French (1992) found, in addition to beta, size (i.e., the return on small stocks minus the return on big stocks) and value (i.e., the return on high book-to-market stocks minus the return on low book-to-market stocks) to be of significance in explaining average returns and therefore valid proxies for risk. It is therefore a worthwhile exercise to compare portfolios formed via a (dual-)beta filter with DFEOX. The relatively poor performance of the traditional beta portfolio suggests the need for financial planners to explore alternative forms of stock selection. While the Fama-French three-factor model is such an alternative, a more tractable solution is the dual-beta model. The findings of this study advocate to financial planners, if they have not done so already,

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

the use of the dual-beta model for stock selection and portfolio construction. Interestingly, this study finds that a simple down-market beta scheme produces a portfolio with relatively low variance while generating positive returns. Such models are simple and can be estimated using a spreadsheet (the combination of upand down-market betas is only slightly more involved), thus potentially rendering considerably more complex and cumbersome models, such as the Fama-French three-factor model, hardly worth the additional effort. References Aggrawal, P., & Waggle, D. (2010). The Dispersion of ETF Betas on Financial Websites, The Journal of Investing, 19(1), 13-24. Ang, A., Chen, J., & Xing, Y. (2006). Downside Risk, Review of Financial Studies, 21(4), 1767-1794. Arnott, R. (1998). Options and Protective Strategies, The Journal of Investing, 7(2), 16-22. Baker, M., Bradley, B., & Wurgler, J. (2011). Benchmarks as Limits to Arbitrage: Understanding the Low-Volatility Anomaly, Financial Analysts Journal, 67(1), 40-56. Banz, R. (1981). The Relationship between Return and Market Value of Common Stocks, Journal of Financial Economics, 9, 3-18. Barber, B. & Odean, T. (2000). Trading is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors, The Journal of Finance, 55(2), 773-806. Barber, B., Odean, T., & Zheng, L. (2000). The Behavior of Mutual Fund Investors, Working paper. Barberis, N., Huang, M., & Santos, T. (2001). Prospect Theory and Asset Prices, Quarterly Journal of Economics, 116(1), 1-53. Basu, S., Raj, M., & Tchalian, H. (2008). A Comprehensive Study of Behavioral Finance, Journal of Financial Service Professionals, 62(4), 51-62. Bajtelsmit, V.L. (2005). Personal Finance: Skills for Life. Hoboken, NJ: Wiley. Bauer, R., Haerden, R., & Molenaar, R. (2004). Asset Allocation in Stable and Unstable Times, The Journal of Investing, 13(3), 72-80.

Journal of Personal Finance

Benartzi, S., & Thaler, R. (1995). Myopic Loss Aversion and the Equity Risk Premium Puzzle, The Quarterly Journal of Economics, 110(1), 73-92. Berkelaar, A., Kouwenberg, R., & Post, T. (2004). Optimal Portfolio Choice under Loss Aversion, The Review of Economics and Statistics, 86, 973-987. Black, F. (1993). Beta and Return, The Journal of Portfolio Management, 20, 8-18. Black, F., Jensen, M., & Scholes, M. (1972). The Capital Asset Pricing Model: Some Empirical Tests. Studies in the Theory of Capital Markets, 79-122. Blume, M. (1975). Betas and Their Regression Tendencies, The Journal of Finance, 30(3), 785-795. Bollerslev, T. (1986). Generalized Autoregressive Conditional Heteroscedasticity, Journal of Econometrics, 31, 307-327. Campbell, J., & Ammer, J. (1993). What Moves the Stock and Bond Markets? A Variance Decomposition for Long-term Asset Returns, The Journal of Finance, 48(1), 3-37. Canner, N., Mankiw, G., & Weil, D. (1997). An Asset Allocation Puzzle, The American Economic Review, 87(1), 181-191. Chan, L. & Lakonishok, J. (1993). Are the Reports of Beta‘s Death Premature? The Journal of Portfolio Management, 19(4), 51-62. Chong, J., & Phillips, G.M. (2009). The Performance of Safe Portfolios during Stressful Times: How Beta Failed, Why New Measures are Needed. Financial Planning and Analysis Corporate Round Table, Financial Management Association, October 21. Cochrane, J. (1999). Portfolio Advice for a Multifactor World, Economic Perspectives, 23(3), 59-79. Daryanani, G. (2008). Opportunistic Rebalancing: A New Paradigm for Wealth Managers, Journal of Financial Planning, 21(1), 48-61. Debondt, W., & Thaler, R. (1985). Does the Stock Market Overreact? The Journal of Finance, 40(3), 793-805. Downs, T., & Ingram, R. (2000). Beta, Size, Risk, and Return. The Journal of Financial Research, 23(3), 245-260. Engle, R. (2002). Dynamic Conditional Correlation: A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroscedasticity Models, Journal of Business & Economic Statistics, 20, 339-350. Estrada, J. (2007). Mean-Semivariance Behavior: Downside Risk and Capital Asset Pricing, International Review of Economics and Finance, 16, 169-185. Eyssell, T. (2003) What‘s the Proper Beta? Financial Advisors and the ‗Two-Beta Trap‘, Journal of Financial Planning, 16(9), 54-57. Fama, E.F. (2000). The New Indexing. Retrieved from http://www.dfaus.com/2009/05/the-new-indexing.htm ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

Fama, E., & French, K. (1992). The Cross-Section of Expected Stock Returns, The Journal of Finance, 47(2), 427-465. Fama, E., & MacBeth, J. (1973). Risk, Return, and Equilibrium: Empirical Tests, The Journal of Political Economy, 81(3), 607-636. Farrelly, T. (2006) Asset Allocation for Robust Portfolios, Journal of Investing, 15(4), 53-63. Galagedera, D. (2009). An Analytical Framework for Explaining Relative Performance of CAPM Beta and Downside Beta, International Journal of Theoretical and Applied Finance, 12(3), 341-358. Graham, J. & Harvey, C. (2001). The Theory and Practice of Corporate Finance: Evidence from the Field. Journal of Financial Economics, 60, 187-243. Grundy, K., & Malkiel, B. (1996). Reports of Betaâ&#x20AC;&#x2DC;s Death Have Been Greatly Exaggerated, The Journal of Portfolio Management, 22(3), 36-44. Hanna, S., & Lindamood, S. (2010). Quantifying the Economic Benefits of Personal Financial Planning, Financial Services Review, 19(2), 111-127. Jensen, M. (1969). The Pricing of Capital Assets, and the Evaluation of Investment Portfolios, The Journal of Business, 42(2), 167-247. Kahneman, D., Knetsch, J., & Thaler, R. (1990). Experimental Tests of the Endowment Effect and the Coase Theorem, Journal of Political Economy, 98(6), 1325-1348. Levy, H. (2010). The CAPM is Alive and Well: A Review and Synthesis, European Financial Management, 16(1), 43-71. Markowitz, H. (1952). Portfolio Selection, The Journal of Finance, 7, 77-91. Markowitz, H. (1959). Portfolio Selection: Efficient Diversification of Investments (Wiley, Yale University Press, 1970, Basil Blackwell, 1991). Morelli, D. (2007). Beta, Size, Book-to-Market Equity and Returns: A Study Based on UK Data, Journal of Multinational Financial Management, 17, 257-272. Pettengill, G., Sundaram, S., & Mathur, I. (1995). The Conditional Relation between Beta and Returns, Journal of Financial and Quantitative Analysis, 30, 101-116. Plaxco, L., & Arnott, R. (2002). Rebalancing a Global Policy Benchmark, The Journal of Portfolio Management, 28(2), 9-22. Pollock, S.L. (2007). Using Three-Factor Theory to Assess Investment Performance, Journal of Financial Planning, 20(6), 66-75. Rosenberg, B., Reid, K., & Lanstein, R. (1985). Persuasive Evidence of Market Inefficiency, Journal of Portfolio Management, 9, 18-28. Sharpe, W. (1964). A Theory of Market Equilibrium under Conditions of Risk, The Journal of Finance, 19(3), 425-442.

Journal of Personal Finance

Siegel, J. (2005). Perspectives on the Equity Risk Premium, Financial Analysts Journal, 61(6), 61-73. Stattman, D. (1980). Book values and stock returns, The Chicago MBA: A Journal of Selected Papers, 4, 25-45. Tversky, A. & Kahneman, D. (1991). Loss Aversion and Riskless Choice: A Reference Dependent Model, Quarterly Journal of Economics, 107, 1039-1061. Xiong, J., Ibbotson, R., Idzorek, T., & Chen, P. (2010). The Equal Importance of Asset Allocation and Active Management, Financial Analysts Journal, 66(2), 1-9.

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

SAFE WITHDRAWAL RATES FROM RETIREMENT SAVINGS FOR RESIDENTS OF EMERGING MARKET COUNTRIES Channarith Meng, Ph.D. Candidateď&#x20AC;Ş National Graduate Institute for Policy Studies (GRIPS) Wade Donald Pfau, Ph.D. National Graduate Institute for Policy Studies (GRIPS) Researchers have mostly focused on U.S. historical data to develop the 4 percent withdrawal rate rule. This rule suggests that retirees can safely sustain retirement withdrawals for at least 30 years by initially withdrawing 4 percent of their savings and adjusting this amount for inflation in subsequent years. But, the time period covered in these studies represents a particularly favorable one for U.S. asset returns that is unlikely to be broadly experienced. This poses a concern about whether safe withdrawal rate guidance from the U.S. can be applied to other countries. Particularly for emerging economies, defined-contribution pension plans have been introduced along with under-developed or non-existing annuity markets, making retirement withdrawal strategies an important concern. We study sustainable withdrawal rates for the 25 emerging countries included in the MSCI indices and find that the sustainability of a 4 percent withdrawal rate differs widely and can likely not be treated as safe.

Introduction What is the safe withdrawal rate from un-annuitized retirement savings that will provide the most retirement income for retirees without exhausting their savings? Potential retirees must answer this question to know if their expected spending ď&#x20AC;Ş

Channarith Meng, National Graduate Institute for Policy Studies (GRIPS), 7-22-1 Roppongi, Minato-ku, Tokyo 106-8677, Japan; Phone: 81-3-64396225; channarithmeng@yahoo.com

Journal of Personal Finance

needs can be reasonably supported from their savings. When the withdrawal rate is too high, retirees are vulnerable to the risk of income shortfalls and poverty at later ages. A low withdrawal rate, on the other hand, may lead retirees to sacrifice the opportunity of a higher sustainable living standard. Recent interest in addressing this issue has resulted in a growing literature. Using various simulation techniques including historical overlapping, bootstrapping, and Monte Carlo simulations, researchers have developed a variety of rules and strategies in the hope of giving retirees appropriate guidelines for their retirement planning. A range of withdrawal rates have been recommended along with asset allocation strategies to safely sustain retirees for a required number of years. Among numerous studies, the 4 percent withdrawal rule has been widely accepted as a safe sustainable withdrawal rate, and it has become an established baseline for testing other approaches. In the pioneering study for this field, Bengen (1994) suggests that an initial withdrawal rate of 4 percent adjusted for inflation in subsequent years should be safe and sustainable for at least 30 years. He further recommends a starting allocation to stocks between 50 and 75 percent. In subsequent research, Bengen (1996) indicates that a 4 percent withdrawal rate is sustainable even when the proportion of stocks in the portfolio is gradually reduced over time. Bengen (1997) includes small capitalization stocks into the portfolio mix and finds a notable increase in the sustainable withdrawal rate. In his latest

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

research, Bengen (2006) indicates that even 5 percent can be safely sustainable under certain conditions. Other studies also give support for the sustainability of 4 percent or higher withdrawal rates. Cooley, Hubbard, and Walz (2011), updates earlier findings to show that when using historical simulations, a 50/50 portfolio for stocks and bonds provides a 96 percent historical success rate for a 4 percent real withdrawal rate over 30 years. The success rate increases to 100 percent when increasing the share of stocks to 75 percent. Monte Carlo simulations by Ameriks, Veres, and Warshawsky (2001) indicate that a 4.5 percent real withdrawal rate is possible with an 8.3 percent chance of exhausting money in 30 years. Tezel (2004), using historical simulations, finds that 4.5, 5.5, and 6.5 percent real withdrawal rates work for time horizons of 30, 20, and 10 years, respectively, with the chance of exhausting money during retirement below 8 percent. Spitzer, Strieter, and Singh (2007) also find that a 4.4 percent real withdrawal rate with 50 percent stocks can be used with a 10 percent chance for failure within 30 years. These studies also find importance for allocating a high proportion to stocks in the portfolio mix. Terry (2003), on the other hand, suggests a negative relationship exists between stock allocations and withdrawal rates. Studies by Pye (2000), Guyton (2004), Guyton and Klinger (2006), Robinson (2007), Spitzer, Strieter, and Singh (2007), Spitzer (2008), and Stout (2008) also explore various decision rules for variable withdrawal strategies to achieve higher initial withdrawal rates without harming the overall chances for success. Scott, Sharpe, and Watson (2009) suggest that using financial derivatives could

Journal of Personal Finance

support a higher spending rate than offered by the 4 percent rule. While much of the existing literature supports the safety of the 4 percent withdrawal rate or even higher rates, the conclusions are usually based on the data for U.S. asset returns since 1926. This covers a particularly fortuitous time period for the U.S. that is unlikely to be attained over a regular basis by any country. Blanchett and Blanchett (2008) acknowledge that past market conditions may not suitably represent what will happen in the future. They note that, based on the average expected forecast for future stock returns from a variety of sources, the future real returns for a 60/40 portfolio of stocks and bonds in the U.S. can be expected to be between 1 and 2 percentage points less than historical averages. Dimson, Marsh, and Staunton (2004) also argue that looking at the past U.S. data for future predictions will lead to â&#x20AC;&#x2022;success bias.â&#x20AC;&#x2013; This expectation of lower future stock returns in the U.S. is also noted by Bogle (2009) and Krugman (2005). Overall, conclusions reached by previous studies may provide overly optimistic recommendations about future sustainable withdrawal rates, which could therefore jeopardize retirement spending at later ages. Very few studies about safe withdrawal rates consider countries other than the U.S. Pfau (2010) is one exception that includes 17 developed market economies. The study shows that the U.S. enjoyed consistently low inflation, and high returns and low volatility on stocks and bonds, relative to other countries. With historical simulations, his results show that only 4 countries including Canada, Sweden, Denmark, and the ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

U.S. could attain a maximum worst-case withdrawal rate exceeding 4 percent for a 30-year retirement duration. This calculation does not include account fees and assumes that retirees in each year had the perfect foresight to choose the best performing asset allocation. He also finds that the best worstcase maximum withdrawal rates occur with stock allocations of at least 48 percent for all countries except Switzerland. These findings, in addition to the potentially weaker performance of future market returns, pose a concern about the wide applicability of the 4 percent rule. Estimating sustainable withdrawal rates is of particular importance for less developed economies with limited annuity markets and growing reliance on defined-contribution pension plans. In many of these countries, existing defined-benefit pension funds provide limited coverage for the population. As well, worldwide trends of decreasing fertility and increasing lifespans are leading to increasingly aging populations. Table 1 summarizes these demographic trends for the countries included in this study, showing how the percentage of the population aged 60 and over is rapidly growing from an average of 7.6 percent in 1970, to 11.5 percent in 2010, to a projected 25.7 percent in 2050. Related to this, life expectancy at birth has grown from an average of 61.3 in the early 1970s to a projected 79.5 by the 2050s. At the same time that populations are aging, the traditional network of having extended families support their elderly members, which is so important in emerging market countries that Holzmann and Hinz (2005) include it as the fourth pillar of the old-age support network in the World Bankâ&#x20AC;&#x2DC;s revised pension framework, is being threatened by reduced family sizes and

Journal of Personal Finance

Table 1 Population Aged 60 and over and Life Expectancy at Birth Population Aged 60+ (% of Total Population)

Life Expectancy at Birth (Years) Country 1970- 2010- 20501970 2010 2050 1975 2015 2055 Argentina 10.7 14.6 25.0 67.2 76.1 81.4 Brazil 5.6 10.3 29.0 59.5 74.0 79.9 Chile 7.8 13.2 30.3 63.7 79.3 83.3 China 6.6 12.3 33.9 64.6 73.8 79.7 Colombia 5.5 8.6 23.7 61.7 74.0 79.7 Czech Republic 18.0 21.8 34.2 70.3 77.9 82.9 Egypt 5.4 8.0 20.2 51.7 73.5 79.7 Hungary 17.2 22.5 32.2 69.4 74.7 80.5 India 5.5 7.6 19.1 50.8 66.0 74.4 Indonesia 5.5 8.2 25.5 53.4 70.0 78.2 Israel 10.4 14.8 22.5 72.6 82.0 86.8 Jordan 5.2 5.8 18.2 62.6 73.6 79.1 Korea 5.4 15.7 38.9 63.2 80.7 85.1 Malaysia 5.4 7.7 20.4 64.9 74.6 80.3 Mexico 5.6 9.0 25.8 62.6 77.2 82.3 Morocco 6.2 8.2 24.2 53.0 72.5 79.3 Pakistan 5.8 6.4 15.8 54.6 65.8 72.7 Peru 5.6 8.8 22.7 55.5 74.3 79.9 Philippines 4.9 5.7 15.3 61.4 69.2 77.1 Poland 12.8 19.2 35.3 70.6 76.4 81.4 Russia 11.9 17.8 31.2 69.0 69.2 76.3 South Africa 5.5 7.4 14.8 53.7 53.8 65.8 Sri Lanka 5.9 12.3 27.4 64.1 75.2 80.7 Thailand 5.3 12.9 31.8 61.0 74.4 80.1 Turkey 6.1 9.0 26.0 51.3 74.3 80.0 Average 7.6 11.5 25.7 61.3 73.3 79.5 The data is based on the medium-variant projection. Source: Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat, World Population Prospects: The 2010 Revision

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

increased labor mobility. Increasingly, elderly will be left to fend for themselves. Pfau and Atisophon (2009) provide as case study about these demographic trends for Thailand, a country which is working to create a defined-contribution National Pension Fund to supplement the existing rudimentary defined-benefit system. Whitehouse (2007) provides details about pension reform in many different countries, indicating how defined-contribution pensions have now become commonplace in Latin America, the Caribbean, Eastern Europe, and Central Asia. As such, the issue of sustainable retirement spending in emerging market countries is quite important. To the best of our knowledge, we are providing the first attempt to address this issue for emerging market economies in a stochastic framework that incorporates volatility and probabilities of success for retirement withdrawal strategies. We investigate both the applicability of the widely accepted 4 percent withdrawal rule, as well as the issue of asset allocation during retirement. Data and Methodology This study uses data from a variety of sources available through the end of 2009. Returns on domestic stocks for the 25 countries are obtained from the MSCI Stock Indices. They are calculated as the annual percentage change at year end for the MSCI Standard Core Gross Indices. We also use domestic currency deposit rates, taken from the International Monetary Fundâ&#x20AC;&#x2DC;s International Financial Statistics (IFS), to represent the local fixed income returns. Two exceptions are that we use the central bank discount rate for India and Jordan in 1988-89 and the call money rate for Pakistan. Also, for Poland, we made

Journal of Personal Finance

adjustments to match recent and earlier deposit rates after a change in the methodology of reporting deposit rates in 2002. Inflation rates are also taken from the IFS. We use the longest available time period of data for each country, except that we drop the periods of extreme hyperinflation in Argentina and Brazil. Analysis is based on the real returns for stocks and deposit rates. Even though we would also like to consider short-term and long-term government debt, such data is not available for many of the emerging countries. Unlike Pfau (2010), which could consider historical simulations with 109 years of data for each developed market country, we use a bootstrapping Monte Carlo approach with the limited historical data for emerging markets. Annual in-sample returns are randomly selected with replacement to form hypothetical multi-year simulation periods for asset returns. We simulate 10,000 hypothetical asset return paths for retirees in each country. For each simulation, we optimize across the two domestic assets, finding the fixed asset allocation that provides the highest sustainable withdrawal rate for 30 years. This is called the perfect foresight assumption, and it provides an overly optimistic assessment for sustainable withdrawal rates. To correct for this, we also investigate how sustainable withdrawal rates vary by asset allocation. We consider 21 possibilities for fixed asset allocations, ranging in 5 percentage point increments from 0 to 100 percent stocks, with the remainder allocated to bank deposits. We assume a fixed retirement duration of 30 years to be analogous with previous studies. Modifying this assumption is simple, and most studies find that sustainable withdrawal rates decrease, but at a decreasing rate, as the retirement duration increases. Other ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

assumptions include no deductions for administrative fees, annual rebalancing to the targeted asset allocations, and no taxes. We assume that the annual account withdrawal is set as a percentage of the accumulated portfolio at the retirement date. Since we have adjusted our data to eliminate the impact of inflation, our resulting withdrawal rates are expressed in terms of real purchasing power. Constant withdrawals are made at the start of each year. The remaining account balance, divided among the two assets, then grows or shrinks by that yearâ&#x20AC;&#x2DC;s asset returns, and at the end of the year the portfolio is rebalanced to the target asset allocation. If the withdrawal pushes the account balance to zero, the withdrawal rate was too high and the portfolio failed to be sustainable for 30 years. We calculate the maximum sustainable withdrawal rate for each simulation. Results Table 2 provides summary statistics for asset returns and inflation for the available time periods in 25 emerging market economies. Asset returns are provided in real terms after removing the effects of inflation. The returns for stocks and fixed income assets vary across countries. Stocks provide double-digit average returns for all countries except China, Israel, Jordan, Morocco, and Poland. However, stock volatility, as measured by standard deviation, tends also to be very high. Standard deviations for real stock returns were under 30 percent in only 4 of the 25 countries. On the other hand, fixed income assets tend to provide lower average returns and risks

Journal of Personal Finance

Table 2 Summary Statistics Real Fixed Income Returns

Argentina

Corr. Between Stocks & Period Fixed Std. Std. Std. Income Mean Mean Mean Dev. Dev. Dev. Assets 1992-2009 11.5 37.8 3.6 6.4 7.2 8.1 -0.15

Brazil

1995-2009

19.1

47.8

Chile

1988-2009

18.0

China

1993-2009

4.7

Columbia Czech Republic Egypt

1993-2009

Hungary

Real Stock Returns

Country

Inflation

9.5

7.3

11.0 15.6

0.30

29.5

3.4

8.4

6.9

-0.09

45.9

-0.2

3.8

4.9

7.3

0.31

18.7

41.3

4.4

3.4

11.6

7.2

-0.59

1995-2009

11.7

30.4

-1.0

1.6

4.5

3.4

0.56

1995-2009

30.0

62.6

1.3

5.3

7.3

5.0

0.09

1995-2009

18.4

47.6

0.8

2.7

10.4

7.6

-0.23

India

1993-2009

13.9

39.8

1.2

2.6

6.8

3.0

0.04

Indonesia

1988-2009

23.9

67.3

4.6

5.9

11.2 11.1

0.09

Israel

1993-2009

8.9

30.2

2.8

5.0

4.3

0.34

Jordan

1988-2009

6.7

29.6

1.0

5.2

5.5

6.1

0.20

Korea

1988-2009

10.7

37.4

2.8

1.9

4.6

2.2

0.04

Malaysia

1988-2009

12.0

35.1

1.8

1.5

2.9

1.3

0.06

Mexico

1988-2009

18.6

34.6

-1.2

7.2

17.7 23.7

0.26

Morocco

1998-2009

7.9

22.8

2.6

1.6

1.9

1.1

-0.30

Pakistan

1993-2009

16.5

53.6

0.3

3.3

8.6

4.6

0.16

Peru

1993-2009

21.0

38.0

-0.4

7.0

8.3 11.9

0.04

Philippines 1988-2009

10.8

44.1

1.7

2.4

7.4

3.6

-0.08

2.1

2.2

9.4

9.9

-0.14

34.2 49.4

0.19

Poland

1994-2009

2.0

34.3

Russia South Africa Sri Lanka

1995-2009

14.4

60.0

1993-2009

10.4

22.8

3.7

2.4

6.9

2.5

-0.06

1993-2009

12.7

55.8

-0.1

4.1

10.3

4.7

0.45

Thailand

1988-2009

15.1

51.0

2.5

2.9

3.8

2.3

0.07

-9.9 11.5

Turkey 1988-2009 39.1 120.6 2.0 8.4 52.1 31.2 0.04 Source: Own calculations using data described in Data and Methodology section. ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

in these countries. Fixed income assets provide real returns under 5 percent in all countries except Brazil. Real average returns are even negative for some countries. At the same time, Russia is the only country which experienced fixed income asset volatility above 10 percent. For inflation, average rates were above 10 percent in Brazil, Columbia, Hungary, Indonesia, Mexico, Russia, Sri Lanka, and Turkey. Table 2 also includes the correlations between stocks and fixed income assets. The correlation coefficients are small and even negative in eight cases, implying potential diversification benefits. Table 3 provides simulation results for sustainable withdrawal rates over 30 years at various distribution percentiles. The distributions are based on whichever asset allocation provides the highest withdrawal rate over 30 years in each simulation. In the worst-case scenario, only retirees in Brazil, Colombia, South Africa, Chile, Morocco, and Korea could sustain a 4 percent withdrawal rate, and retirees in 12 countries could not sustain a 3 percent withdrawal rate. In Egypt, Peru, Jordan, China, Sri Lanka, Turkey, Mexico, and Russia, the highest withdrawal rate for the worst-case scenario is lower than 2 percent. Focusing on the worst-case scenario from 10,000 simulations may be criticized as overly pessimistic or risk averse, and the table also provides withdrawal rates at the 1st, 5th, and 10th percentiles. These percentiles provide the withdrawal rates which can be sustained for 30 years with a 1 percent, 5 percent, and 10 percent chance of failure, respectively. However, Terry (2003) argues that when dealing with irreplaceable assets and uncertainties, even a 1 percent

Journal of Personal Finance

Table 3 Sustainable Withdrawal Rates % Failure % Failure Within 30 Within 30 1st 5th 10th Country Min. years at 4% years at 5% %ile %ile %ile Withdrawal Withdrawal Rate Rate Brazil 5.00 6.23 7.08 7.59 0 0 Columbia 4.95 5.53 5.91 6.18 0 0 South Africa 4.40 4.81 5.10 5.30 0 3.2 Chile 4.34 5.05 5.93 6.66 0 0.9 Morocco 4.09 4.40 4.55 4.65 0 26.6 Korea 4.08 4.32 4.51 4.62 0 29.5 Israel 3.64 4.03 4.27 4.41 0.8 32.5 Poland 3.60 3.86 4.00 4.09 5.0 74.5 Malaysia 3.53 3.87 4.08 4.23 2.7 27.1 Thailand 3.35 3.92 4.22 4.39 1.7 26.7 Indonesia 3.26 4.19 4.90 5.36 0.6 5.9 Philippines 3.14 3.59 3.84 3.98 11.3 43.4 Argentina 3.06 3.78 4.29 4.60 2.1 18.8 Hungary 2.92 3.40 3.74 4.06 9.0 22.7 India 2.91 3.38 3.68 3.89 12.5 30.1 Pakistan 2.41 2.79 3.09 3.33 24.0 39.3 Czech Republic 2.38 2.58 2.83 3.21 19.4 30.0 Egypt 1.85 3.14 4.06 4.82 4.8 11.4 Peru 1.84 3.09 4.19 5.17 4.0 9.3 Jordan 1.81 2.53 2.93 3.18 34.5 54.9 China 1.80 2.37 2.62 2.75 65.1 78.1 Sri Lanka 1.73 2.34 2.65 2.82 40.5 55.5 Turkey 1.62 2.55 3.23 3.73 13.4 25.8 Mexico 1.39 2.62 3.91 5.02 5.4 9.9 Russia 0.06 0.17 0.30 0.41 74.1 78.8 Assumptions include perfect foresight, a 30-year retirement duration, no administrative fees, annual inflation adjustments, and annual rebalancing. Results are based on 10,000 simulations using bootstrapping with replacement. Source: Same as Table 2. ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

probability of failure is excessively high. Fullmer (2008) also argues that downside risk is a painful aspect of risk and is more unbearable after retirement when options such as continuing to work have declined. At the 1st percentile (i.e. 99 percent chance of success), sustainable withdrawal rates exceed 4 percent in 8 out of 25 countries: Brazil, Columbia, South Africa, Chile, Morocco, Korea, Israel, and Indonesia. If a 5 percent failure rate is accepted, a 4 percent withdrawal rate is sustainable in 14 countries. With a 5 percent failure rate, a withdrawal rate of 7 percent is possible in Brazil, and it is almost 6 percent in Columbia and Chile, and 5 percent in South Africa. However, in 5 countries even a 3 percent withdrawal rate was not sustainable. The number of countries with withdrawal rates exceeding 4 percent increases to 16 with a 10 percent failure rate, but this leaves 9 countries with sustainable rates below 4 percent even with a 10 percent chance of failure. The last two columns of Table 3 show the percentage of failures with fixed withdrawal rates of 4 and 5 percent. With the 4 percent withdrawal rate, 4 countries experience failures in more than 25 percent of cases, while 15 countries experience this outcome with a 5 percent withdrawal rate. Table 4 shows the number of years for which 4 and 5 percent withdrawal rates are sustainable at various percentiles. In the worst case, all countries except Russia find 4 percent and 5 percent to be sustainable for at least 10 years. The number of sustainable years increases when a higher chance for failure is accepted. As well, there tends to be a large drop in the number of sustainable years when the withdrawal rate increases from 4

100

Journal of Personal Finance

Table 4 Number of Sustainable Years for Various Withdrawal Rates 4% Withdrawal Rate 5% Withdrawal Rate 1st 5th 10th 1st 5th 10th Min. Min. %ile %ile %ile %ile %ile %ile Brazil >50 >50 >50 >50 29 >50 >50 >50 Columbia >50 >50 >50 >50 29 44 >50 >50 Chile 36 >50 >50 >50 24 33 >50 >50 South Africa 35 47 >50 >50 24 29 33 37 Korea 31 35 38 40 23 25 26 27 Morocco 31 36 38 40 23 25 27 27 Israel 27 31 34 36 20 23 24 26 Poland 26 29 30 31 20 22 23 23 Malaysia 25 29 32 34 20 22 23 24 Thailand 24 30 34 37 19 22 24 26 Indonesia 23 38 >50 >50 18 24 33 46 Philippines 23 27 29 31 18 20 22 23 Argentina 22 29 38 46 17 21 26 29 Hungary 22 25 28 31 18 20 21 23 India 21 25 28 30 17 20 21 22 Czech Republic 19 20 21 23 16 17 17 18 Pakistan 19 21 23 25 15 17 18 20 Peru 16 23 37 >50 13 17 23 38 Sri Lanka 16 19 21 22 13 15 17 18 China 15 19 20 21 13 15 17 17 Egypt 15 23 33 >50 13 17 22 29 Jordan 15 20 23 24 13 16 18 19 Turkey 14 19 24 29 12 15 18 21 Mexico 12 18 31 >50 10 14 20 32 Russia 6 9 10 11 6 8 9 10 Assumptions include perfect foresight, no administrative fees, annual inflation adjustments for withdrawals, and annual rebalancing. >50 means at least 50 years of sustainability. Source: Same as Table 2. Country

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

101

Figure 1 Asset Allocation Providing Maximum Sustainable Withdrawal Rate for 100 10th percentile

5th percentile 1st percentile

Percentage Allocation to Stocks

Minimum

70 60 50 40 30 20 10

RUS

TUR

MEX

LKA

JOR

CHN

PER

EGY

CZE.

IND

PAK

HUN

ARG

IDN

PHL

THA

MYS

ISR

POL

KOR

MAR

ZAF

CHL

COL

BRA

Various Failure Probabilities

percent to 5 percent, especially for Brazil, Colombia, Chile, and South Africa. Figure 1 shows the asset allocations that achieved the perfect foresight maximum sustainable withdrawal rates shown for Table 3. Interestingly, for most countries the optimums occur with a low proportion of stocks. This contrasts with the Pfau (2010) study for developed markets, which found that the stock allocation which provided the highest withdrawal rate was at least 50 percent in 16 of 17 countries. For the more volatile emerging market countries, from the minimums to the 10th percentiles of the simulations, the optimums occur with

102

Journal of Personal Finance

Figure 2 Sustainable Withdrawal Rates across Distribution of Stock Allocations with 5 Percent Failure Probability 8

BRA

Sustainable Withdrawal Rate

CHL

COL

ZAF

IDN

4POL

MAR KOR ISR ARG THA MY S PHL IND

EGY

MEX

TUR PAK JOR

3 CHN

PER

HUN

CZE.

LKA

RUS

40 50 60 Percentage Allocation to Stocks

100

stock allocations below 30 percent for all countries except Chile, the Czech Republic, Egypt, Peru, and Mexico. Figure 2 illustrates the distribution of sustainable withdrawal rates across stock allocations for each country with a 5 percent probability of failure. For each country‘s distribution, the highest withdrawal rate attained is labeled with the country‘s name code. In the case of ties, the smallest stock allocation is labeled. The highest withdrawal rates are achieved with 30 percent or less stock allocations for all countries except Chile, Peru and Mexico, where the highest withdrawal rates occur with 50, 55, and 80 percent stock allocations, respectively. Strikingly, 19 out of the 25 countries achieve the highest sustainable withdrawal rates with stock allocations of ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

103

Figure 3 Probability of Failure for 4% Withdrawal Rate by Stock Allocation 100

80 RUS

Probability of Failure

CHN

LKA

40 JOR

30 PAK

CZE. TUR

PHL

IND HUN

POL ISR

BRA COL ZAF 0KOR 0

MY S THA IDN CHL MAR

EGY

MEX

PER

ARG

40 50 60 Percentage Allocation to Stocks

100

15 percent or less. The distribution of stock allocations has a downward sloping trend for many countries, noticeably when stock allocations rise above 20 percent. Allocating a high proportion to stocks does more harm than good for sustainable withdrawal rates in these emerging market countries. Finally, Figures 3 and 4 show the probability of failures with 4 and 5 percent withdrawal rates, respectively, across the range of stock allocations. Again, for each countryâ&#x20AC;&#x2DC;s distribution, the lowest probability of failure is labeled by the countryâ&#x20AC;&#x2DC;s name code. The distributions of failure probabilities exhibit a convex shape (or roughly U-shaped) for many countries. This pattern is more apparent when the withdrawal rate is 5 percent. There is a large drop in the probability of

104

Journal of Personal Finance

Figure 4 Probability of Failure for 5% Withdrawal Rate by Stock Allocation 100

RUS

CHN POL

Probability of Failure

60 LKA

JOR

50 PHL

PAK

ISR

IND

KOR THA TUR

MAR

CZE.

MY S HUN

ARG

EGY

MEX

PER

IDN ZAF

0BRA 0

CHL

COL

40 50 60 Percentage Allocation to Stocks

100

failure when stocks are initially introduced, but the marginal drop decreases to a minimum. Then the failure probabilities increase for higher stock allocations. Moreover, the minimum probability of failure for a 4 percent withdrawal rate occurs at points where stock allocations are less than 50 percent for most countries, except Czech Republic, Mexico, Peru, and Russia. It is not surprising that when the withdrawal rate increases to 5 percent, minimum probabilities of failure move to higher stock allocations, since more risk is needed to fund higher withdrawals. Even though more stocks are needed to increase withdrawals to 5 percent, still only 6 countries experience optimal stock allocations of more than 50 percent. These results improve the robustness of ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

105

our previous findings that, differently from developed countries, high stock allocations are not appropriate for maintaining sustainable withdrawals with emerging market assets. Conclusion Numerous studies based on U.S. data exist to help retirees plan safe and sustainable withdrawals from their retirement savings. In the existing literature, the well-known finding is that an annual 4 percent inflation-adjusted withdrawal rate over 30 years is considered safe for retirees with a stock allocation above 50 percent. However, this widely-accepted rule-of-thumb is not necessarily applicable for the situation in other countries. For emerging market economies, this issue is quite important, as annuity markets are not developed and recent pension reforms are moving toward defined-contribution pension plans in which retirement income management is handled individually by retirees. Therefore, guidelines about sustainable withdrawal rates are needed. Our study, based on the 25 emerging market economies included in the MSCI indices, finds that the sustainability of the 4 percent withdrawal rule is questionable in many cases. Using the bootstrapping approach, our results show that, in the worst-case scenario, only retirees in 6 out of 25 countries could sustain their 30 years of withdrawals with 4 percent. Even with a 5 percent chance of failure, 4 percent is not sustainable in 11 countries, and even 3 percent is not sustainable in 5 countries. Moreover, our study indicates that the optimal asset allocation for providing the highest withdrawal rates with a low

106

Journal of Personal Finance

chance for failure occur at low stock allocations for most emerging market economies, in contrast to previous studies on developed economies. Though a higher proportion of stocks increases the chance of success at higher withdrawal rates, higher withdrawal rates are also accompanied by increased failure probabilities. To attain a 4 or 5 percent withdrawal rate, a portfolio mix composed of less than 50 percent stock is needed for most of the countries in our sample. The bootstrapping approach used here provides a way to incorporate volatility into the issue of retirement planning, which gives a more realistic picture than using fixed rates of returns for these financial assets. However, the approach is far from perfect. This study makes an implicit assumption that past patterns in financial markets are reflective of the type of situation these countries will face in the future. Further developments may help to reduce the financial market volatility in these countries, but that is not yet clearly going to be the case. Given these uncertainties, the findings of this research suggest that retirement saving will be very important in emerging markets as the 4 percent rule is not reliable. For the most part, asset allocations should be lower as well for these retirees, suggesting that appropriate asset allocations for developed and emerging market countries may be quite different. This issue is deserving of greater research focus in the future, as citizens of emerging market countries cannot rely on the results of retirement planning studies conducted for the U.S. case.

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

107

References Ameriks, J., Veres, R., & Warshawsky, M. J. (2001). Making retirement income last a lifetime. Journal of Financial Planning, 14, 12, 60-76. Bengen, W. P. (1994). Determining withdrawal rates using historical data. Journal of Financial Planning, 7(1), 171-180. Bengen, W. P. (1996). Asset allocation for a lifetime. Journal of Financial Planning, 9(4), 58-67. Bengen, W. P. (1997). Conserving client portfolios during retirement, part III. Journal of Financial Planning, 10(6), 84-97. Bengen, W. P. (2006). Baking a withdrawal plan ‗Layer Cake‘ for your retirement clients. Journal of Financial Planning, 19(8), 44-51. Blanchett, D. M. & Blanchett, B. C. (2008). Data dependence and sustainable real withdrawal rates. Journal of Financial Planning, 21(9), 70-85. Bogle, J. C. (2009). Enough: True measure of money, business, and life. New Jersey: John Wiley and Sons. Cooley, P. L., Hubbard, C. M., & Walz, D. T. (2011). Portfolio Success Rates: Where to Draw the Line. Journal of Financial Planning, 24(4), 48-58. Dimson, E., Marsh, P., & Staunton, M. (2004). Irrational optimism. Financial Analysts Journal, 60(1), 15-25. Fullmer, R. K. (2008). The fundamental differences in accumulation and decumulation. Journal of Investment Consulting, 9(1), 36-40. Guyton, J. T. (2004). Decision rules and portfolio management for retirees: Is the ‗safe‘ initial withdrawal rate too safe? Journal of Financial Planning, 17(10), 54-61. Guyton, J. T. & Klinger, W. J. (2006). Decision rules and maximum initial withdrawal rates. Journal of Financial Planning, 19(3), 49-57. Holzmann, R. & Hinz, R. (2005). Old Age Income Support in the 21st Century: An International Perspective on Pension Systems and Reform. World Bank: Washington, DC. Krugman, P. (2005). Confusions about social security. The Economist’s Voice, 2(1), 1-8. Pfau, W. D. (2010). An international perspective on safe withdrawal rates from retirement savings: The demise of the 4 percent rule? Journal of Financial Planning, 23(12), 52-61. Pfau, W. D., & Atisophon, V. (2009). The impact of the National Pension Fund on the suitability of elderly pensions in Thailand. Asian Economic Journal, 23(1), 41-63. Pye, G. B. (2000). Sustainable investment withdrawals. Journal of Portfolio Management, 26(4), 73-83. Robinson, C. D. (2007). A phased-income approach to retirement withdrawals: A new paradigm for a more affluent retirement. Journal of Financial Planning, 20(3), 44-56.

108

Journal of Personal Finance

Scott, J.S., Sharpe, W. F., & Watson, J. G. (2009). The 4% rule – At what price? Journal of Investment Management, 7(3). Spitzer, J. J. (2008). Retirement withdrawals: an analysis of the benefits of periodic ―midcourse‖ adjustments. Financial Services Review, 17(1), 17-29. Spitzer, J. J., Strieter, J. C., & Singh, S. (2007). Guidelines for withdrawal rates and portfolio safety during retirement. Journal of Financial Planning, 20(10), 52-59. Spitzer, J. J., Strieter, J. C., & Singh, S. (2008). Adaptive withdrawals. Journal of Investing, 17(2), 104-113. Stout, R. G. (2008). Stochastic optimization of retirement portfolio asset allocations and withdrawals. Financial Services Review, 17(1), 1-15. Terry, R. L. (2003). The relation between portfolio composition and sustainable withdrawal rates. Journal of Financial Planning, 16(5), 64-72. Tezel, A. (2004). Sustainable retirement withdrawals. Journal of Financial Planning, 17(7), 52-57. Whitehouse, E. (2007). Pensions Panorama: Retirement-Income Systems in 53 Countries. World Bank: Washington, DC.

Volume 10, Issue 1

109

FINANCIAL PLANNING LITERATURE SURVEY Benjamin E. Fagan, MSFE PlusPlus Inc. Shawn Brayman, MES ď&#x20AC;Ş PlusPlus Inc. This study is intended to provide an environmental scan of current research from Australia, Canada, United Kingdom and the United States, related to financial planning/services from 2003 to July 2010. The objective of this exercise is to try and highlight research areas where there may be gaps. This is not intended to review the research in any manner but rather to aggregate and document its existence in some broad based categories. The study was carried out in two parts. To begin with, research was collected, categorized and totalled to determine high and low volume areas. Finally, industry practitioners and academics were petitioned to provide their opinions. Based on our findings, Estate Distribution Analysis, Pension Alternatives and Tax Optimization were found to be the topics that require the most focus for further research. Modern Portfolio Theory, General Portfolio Management and Product Shelf were the categories that were determined to be the most overly researched areas.

Introduction To provide an unbiased review of financial planning research, different search methods from several sources were employed, followed by an assessment from industry professionals. The two main search methods used were to ď&#x20AC;ŞShawn Brayman, PlusPlus Inc.,55 Mary St. Suite #200, Lindsay, Ontario, Canada K9V 5Z6; (705) 324-8001 ext. 306; shawn@planplus.com The authors wish to thank the FPSC Foundation for sponsoring the research that is highlighted in this report. The FPSC Foundation is a charitable organization that promotes and disseminates research for the benefit of the public, financial planners, academia and industry. www.fpscfoundation.ca

110

Journal of Personal Finance

contact stakeholders in the financial planning industry directly and have them submit their research and to conduct an environmental scan using academic databases and search engines. Six hundred and two (602) institutions were directly contacted and asked to submit research. This number included 572 universities, 3 journals, 19 financial planning associations and 12 other industry related institutions from Canada, the United States, the United Kingdom, New Zealand and Australia. Description of Review The environmental scan was accomplished by searching for different keywords in academic databases over the study‘s timeframe. ProQuest was used, at the University of Waterloo, to perform the search across a collection of databases (for a complete listing of the databases used, see Appendix A). The results of this methodology could be potentially biased due to the selection of keywords. To reduce this bias each category, subcategory, and the general term ―financial planning‖ were used as keywords. Once collected, the research was grouped into specific categories and sub-categories that had been previously selected by PlanPlus and the FPSC Foundation (see Appendix B). This was done based on the titles of the research if they contained certain keywords. Also the abstract was reviewed if further scrutiny was necessary. Because of the nature of the search methodology, duplicates were common and were removed. From 2003 through 2010 the literature search returned 1,978 papers and articles from 379 different sources. ©2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

111

It is difficult to make any assessment of whether or not there are research gaps based solely on article volume. To provide some cross-validation, industry professionals were asked to provide their feedback of areas in financial planning that currently require the most attention and those that have been overly researched (in their opinion). This survey was directed electronically to the same 572 universities mentioned above, distributed in paper form at the FPSC Professional Development Day in Canada as well as limited distribution at the Financial Planning Association Conference in Denver. Both events where held in October 2010. Individual emails were also directed to several hundred planning professionals and executives and a mailing was carried out by the FPSC to all registered CFP速 certificate holders in Canada. Based on the feedback from 743 industry professionals an importance ranking was established for each research category. The importance ranking, in conjunction with article volume, was then used to determine specifically what areas of research are in need of focus or potentially receive too much focus. The ranking methodology can be found in Appendix H.1. Literature Scan Total Published Articles by Topic/Category In total, 1,978 literary articles were sourced in the literature survey (as presented in Table 1). The breakdown of each category can be seen in the following table. The most active categories were Retirement Planning (419 articles), Portfolio Management (317 articles) and Behavioural Finance (246 articles). The least active areas of research were Regulatory & Compliance (48 articles), Tax Planning (58

112

Journal of Personal Finance

articles), and Holistic Planning (75 articles). See Appendix C for a complete listing of articles per category. Table 1 Articles per Main Category Category Retirement Planning Portfolio Management Behavioural Finance Business Practices Investment Planning Other Planning Estate Planning Insurance Planning and Risk Management Cash Flow & Liability Mangement Holistic Planning Tax Planning Regulatory & Compliance Total

Articles 419 317 246 157 148 144 132 126 108 75 58 48 1,978

Publication Trends over Time Over the past several years we have seen substantial growth in the production of financial planning literature. In 2003 less than 200 articles were published. This has grown each year to an estimated nearly 400 articles in 2010 (as shown in Figure 1). Please note that the 2010 estimate is based on a linear extrapolation from the articles collected through July of that year. The vast majority of journals searched included articles from 2003 or earlier so we feel this trend is not significantly biased by the availability of the data.

ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

113

Articles Produced

Figure 1 Articles produced from 2003-2010 related to financial planning 450 400 350 300 250 200 150 100 50 0

2010 estimated

2003 2004 2005 2006 2007 2008 2009 2010 Year

Sources for Financial Planning Research The literature search encompassed articles from 379 scholarly journals, magazines and other publication sources. Figure 2 shows, as a percentage of the total 1,982 articles, the publication sources that have published the most research related to financial planning since 2003. The majority of articles have primarily come from, Journal of Financial Planning (20%), Financial Analysts Journal (13%), Journal of Financial Service Professionals (11%), Journal of Family and Economic Issues (11%) and the Financial Services Review (10%). Geographic Source of Research The source of the research is largely driven by the location of the journal which published it, as it was not possible to determine the domicile of authors. As is evident in Table 2, the United States clearly dominates the source of research. The Netherlands was an unexpected surprise largely as a result of

Publication Source

0.1

0.15

0.2

0.25 0.3 Percent of Total

0.35

0.4

0.45

0.5

0.05

20.13% 47.18%

1.11% 1.26% 1.34% 1.63% 2.01% 2.30% 2.60% 2.82% 3.49% 3.71% 6.17% 6.91% 9.58% 10.85% 11.22% 12.85%

Risk Management Journal of Risk and Insurance Canadian Tax Journal Benefits Quarterly The Journal of Wealth… Journal of Business Ethics The CPA Journal Journal of Accountancy Insurance: Mathematic &… The Journal of Portfolio … Journal of Personal Finance Journal of Financial Planning … Financial Services Review Journal of Family and … Journal of Financial Service … Financial Analysts Journal Journal of Financial Planning Other (361 others)

Figure 2 Publication source of financial planning related articles from 2003-2010

114 Journal of Personal Finance

Volume 10, Issue 1

115

publications from ‗Insurance: Mathematics & Economics‘ and the ‗Journal of Business Ethics.‘ Table 2 Geographic Source of Research Country

Articles

% Total

1,584

80.1%

United Kingdom

168

8.5%

Netherlands

United States

114

5.8%

Canada

3.1%

Australia

1.0%

Switzerland

0.5%

0.4%

0.8%

Germany Other

Authors of Research Although not central to the scope of this research, and in some cases difficult to consolidate as a result of different variations of an author‘s name, we felt that the distribution of authorship was an interesting by-product of this scan. Based on the total of 1,978 papers there were a total of 4,220 authors or co-authors. After some attempt to clean up the data for consistent naming, we arrived at 2,658 unique authors. As can be seen in Table 3, the vast majority of authors published only one paper that appeared in our scan with just under 1% of authors publishing 7 or more papers. Also compiled is a list of researchers who have authored 8 or more articles in Table 4. John E. Grable and Moshe A. Milevsky were found to have the most contributions, each with 17, followed by J. Timothy Lynch at 15. For a more extensive list, please see Appendix E.

116

Journal of Personal Finance

Table 3 Frequency of Authorship # of Papers Authored 1 2 3 4 5 6 7 8 9 10 or more Total

Authors 2,192 286 80 40 20 14 9 4 6 7 2,658

% 82.47% 10.76% 3.01% 1.50% 0.75% 0.53% 0.34% 0.15% 0.23% 0.26%

Table 4 Most Articles by Author Articles

Researcher

Articles

Researcher

John E. Grable

Angela C. Lyons

Moshe A. Milevsky

Willi Semmler

J.Timothy Lynch

David Blake

William Reichenstein

Amin Mawani

Sherman D. Hanna

Barbara O'Neill

Sharon A. DeVaney

Michael S. Finke

John J. Spitzer

Michael J. Roszkowski

Deanna L. Sharpe

Dennis C. Reardon

Neal E. Cutler

Opinion Survey on the Need for Additional Research The second stage of the research was to try and develop a weighting for the perceived need of additional research in the various topic areas which could then be combined with the ÂŠ2011, IARFC. All rights of reproduction in any form reserved.

Volume 10, Issue 1

117

basic inventory of published research. The approach was to develop a simple two-question survey (Appendix D) using the same research categories used to group the papers from the research scan. The approach was to ask respondents to list the three topics they felt were most in need of additional research and the three topics they felt were already overly researched. From the results of the survey, the categories and subcategories were separately ranked. The survey was distributed to academics and industry professionals as outlined in the initial description. This resulted in 743 responses of which 120 were partial answers where respondents selected items they felt required more research but did not feel anything was ―over researched.‖ We found that although the instructions indicated to select the 3 more important items most respondents selected far more. Of the 84 options the 743 respondents selected 5,547 categories or an average of 7.47 items they felt needed more research. The 623 respondents who chose an option about ―overly researched‖ selected 2,679 or an average of 4.3 topics (see Appendix F for results). As a result of the much stronger expression of topics requiring additional research, the overall weighting results in a much longer list of topics where more research is desired. We feel this is reflective of the actual belief and intent of the respondents. We felt that the design of the opinion survey would provide a number of interesting perspectives on the various topics presented:

118

Journal of Personal Finance



Magnitude of interest in the topic, either positive or negative.



Specific responses on each topic.

Like the literature scan itself, the raw results are difficult to use. As an example, both Post-Modern Portfolio Theory and Job Change/Loss had 40 respondents that listed them as topics requiring more research, but there were 24 people who felt that Post-Modern Portfolio Theory was over-researched and only 9 who felt Job Change/Loss was over-researched. We therefore created a ―weighting factor‖ (see Appendix G that outlines the methodology) that looked at the overall interest in the topic based on total responses, the net difference on the responses as either positive or negative, and the degree of consensus on the topic as well as the volume of articles that were tracked during the literature survey. We felt that topic areas with high consensus of opinion provided a stronger reading of a topic‘s appropriateness (Importance Weight). In Table 5 – Importance Weight, you can see the weighting of specific topics based on the combined factors of the consensus level. The Importance Weight only considers the opinions of the respondents, while in the following section Importance Rank further considers the volume of articles collected in each category. A 100% Consensus would mean all votes on that topic were consistent for More or Over Researched and the net score for the topic which indicated the magnitude of the opinion on that topic area. The categories towards the top of the table indicate that ‗More Research‘ is needed. The categories towards the bottom indicate ‗Less Research‘ is needed. If the Importance Weight is near zero, no ©2011, IARFC. All rights of reproduction in any form reserved.

610 539 502 404 347 708 203 559 450 663 209 353

Behavioural Finance

Other Planning

Cash Flow & Liability Management

Tax Planning

Retirement Planning

Holistic Planning

Insurance Planning & Risk Management

Business Practices

Investment Planning

Regulatory & Compliance

Portfolio Management

Estate Planning