Research Brief 35

Research Brief

A series of reports summarizing current research from CJA

No. 35

September 2014 Executive Director, Jerome E. McElroy

Director, Research Dept., Richard R. Peterson, Ph.D. Research Brief Editor & Deputy Director, Research, Mary T. Phillips, Ph.D. Graphics & Production, Raymond P. Caligiure Administrative Associate, Annie Su

CJA is a not-for-profit corporation that provides a variety of criminal justice services under a contract with the City of New York.

CJA staff interview defendants arrested in New York City, make recommendations for pretrial release, and notify released defendants of upcoming court dates. CJA also operates two supervised release programs for nonviolent felony clients in Queens and Manhattan.

The Research Department conducts studies addressing a broad array of criminal justice policy concerns. The Research Brief series summarizes the results of some of these studies.

New York City Criminal Justice Agency, Inc. 52 Duane Street New York, NY 10007 PHONE: 646 213-2500 FAX: 646 213-2650 WEB: www.nycja.org

© 2014 NYC Criminal Justice Agency, Inc.

PAYING FOR BAIL ON CREDIT

By Mary T. Phillips, Ph.D. On March 26, 2012, the New up to a maximum of $2,500. A York County (Manhattan) Crimi- full refund is made when the case nal Court began a pilot program ends favorably for the defendant, for accepting bail payment by and a 3% fee is imposed when it credit card. The program was exends in conviction (same as for cash bail). The defenpanded to all five counties You of New York City starting dant must be physican use a January 28, 2013. credit card for cally present in the Prior to this procourthouse in order everything else — to pay bail by credit gram court cashiers did why not not accept credit cards card, after which reBAIL? for bail, even though the lease is immediate. law has authorized their use This research provides for decades. One hurdle was a preliminary assessment of the that the Office of Court Adminis- success of the program. We tration (OCA) had never set the examined the extent to which a amount of the administrative fee credit card option was set by the provided for in the law. OCA has courts and used by defendants now resolved this issue by ab- during the program’s first year of sorbing credit card companies’ Citywide implementation. Then charges without passing them we asked whether the program is along to clients. accomplishing its goals of helpUnder the program, the judge ing defendants gain release and must specify that a credit card is decreasing the time they spend in acceptable and the amount of bail detention, without raising failurethat may be paid by credit card, to-appear (FTA) rates.

This Research Brief is adapted from New York’s Credit Card Bail Experiment (2014)

by Mary T. Phillips, Ph.D., Deputy Director, Research Department The full report is available on CJA’s web site: www.nycja.org/library.php Research Assistance: Raymond P. Caligiure, Maria Annabel Mireles Systems Programming: Wayne Nehwadowich Address comments to the author at mphillips@nycja.org Please cite as follows, adapted to your citation style: Phillips, Mary T. 2014. “Paying For Bail On Credit.” Research Brief series, no. 35. New York: New York City Criminal Justice Agency, Inc. The mission of the New York City Criminal Justice Agency, Inc., is to assist the courts and the City in reducing unnecessary pretrial detention.

Utilization of the Credit Card Program by the Courts Figure 1 shows that in 2013, among all low-bail cases ($2,500 or less) in the combined boroughs, 3% had a credit card payment option set by the judge.

Brooklyn was the only borough with no credit card bail set throughout the study period, despite the largest volume of low-bail cases in the City.

Figure 1 Proportion of Low-Bail Cases ($2,500 or Less) With a Credit Card Option Arraigned 2013 Bronx

Manhattan

0

2% (130)

8% (644)

4% (239)

<1% (3)

3% (1,016)

8,481

6,228

8,035

5,836

1,471

30,051

N=

Queens

Staten Island

Combined Boroughs

Brooklyn

Arraigned March 27, 2012 â&#x20AC;&#x201C; December 31, 2012 Manhattan 3% (221)

N=

The program was used most frequently by the Court in Manhattan, where 8% of low-bail cases had a credit card option in 2013, up from 3% during the pilot phase of the project the previous year.

6,555

Samples Data Collection Cases in which a credit card payment option was set were identified manually by an examination of arraignment court calendars. Arraignment calendars were also the source of some data regarding the form in which bail was posted. Additional form-ofbail information was collected from the electronic database maintained by the Office of Court Administration (OCA).

Manually collected credit card information was linked to defendant and court processing data from the CJA database.

2

Credit Card Sample: All 1,237 cases in which a credit card option was set from March 26, 2012 (Manhattan only), through December 31, 2013 (Citywide) were included in the credit card sample. This included 1,016 cases with bail set in 2013 and 221 with bail set in 2012.

Non-Credit Card (Comparison) Sample: All cases with an arrest in 2013 with bail set at $2,500 or less with NO credit card option were included in the comparison sample.

Because there were no Brooklyn cases in the credit card sample, Brooklyn cases were also excluded from the comparison sample, leaving 20,554 comparison cases among the total of 30,051 low-bail cases in 2013.

Subgroup (Bail Paid By Credit Card): A credit card was used to post bail in at least 165 (13%) of the cases in the credit card sample. OCA may not have identified all credit card payments, so the actual number and percentage with credit card bail payment may be larger.

September 2014

Research Brief The Credit Card Sample During the study period, a total of 76 judges in the City set bail with a credit card option at least once (not shown). However, only a few did so with any frequency. No more than two or three judges were responsible for the majority of credit card options offered in each borough.

Figure 2 shows the proportion of cases in the credit card sample that had bail set by individual judges. In each borough the judge with the highest volume of credit card cases was coded “Judge A,” the judge with the second highest volume was coded “Judge B,” and so on.

In the Bronx and Manhattan, four judges accounted for 70% or more of the credit card cases. In Queens, only three judges accounted for a similar proportion of cases. In Staten Island, two judges accounted for all three credit card cases.

Moreover, there was a large gap between the top-volume judge in each borough and all the others.

In the Bronx, Judge A set three times as many credit card options as Judge B (36% vs. 12%). In Manhattan and Queens the difference was not as great, but in each borough Judge A led by a wide margin in setting bail with a credit card option.

Figure 2 Proportion of Credit Card Cases By Judge 100%

Judge A

80%

36%

60%

12%

40%

20%

0%

N=

28%

36%

Judge B

67%

17%

11% 11%

14%

22%

12%

15%

30%

29%

27%

33%

Bronx

Manhattan

Queens

Staten Island

130

865

239

3

The program authorizes amounts up to $2,500 for credit card payment. The mean credit card amount for the study sample was $1,260, and the median was $1,000.

Figure 3 shows that 60% of cases in the credit card sample had a credit card amount of

Judge C Judge D All Other

$1,000 or less. The credit card amount was exactly $1,000 in 18% of cases, and $2,500 in 19% of cases.

In nine cases, the amount was higher than the maximum authorized by OCA. These amounts ranged from $3,000 (4 cases) to $25,000 (one).

Figure 3 Credit Card Bail Amounts N=1,237 Mean = $1,260 $50 - $350 N = 61 5%

$450 - $500 N = 308 25% 0%

Research Brief #35

20%

$600 $750 N = 150 12% 40%

Median = $1,000 $1,250 $2,000 N = 241 19%

$1,000 N = 228 18% 60%

Above $2,500 N=9 1%

$2,500 N = 240 19% 80%

100%

3

The Credit Card Sample (continued) Judges tended to treat the credit card option as an alternative mechanism for posting cash bail, setting the same amounts for both. When a higher bond amount was set with a lower cash alternative, the credit card amount usually matched the cash alternative. When no cash alternative was set, the credit card amount usually matched the single amount set for bond and cash.

Figure 4 shows that in 87% of the credit card cases, the amount that could be posted by credit card was equal to the cash amount (the cash alternative, if there was one, or the single amount that could be posted as bond or cash). The credit card amount was lower than the cash amount 3% of the time, and higher than the cash amount 9% of the time.

In a very small number of cases (18, not shown) the credit card amount exceeded the bond amount.

Despite having been offered the option of using a credit card, many defendants paid cash or purchased a bail bond, or were unable to make bail in any form.

Figure 5 shows that only 13% of the defendants in credit card cases actually used a credit card.

The percentage was double that in the Bronx, where 26% did so.

In nearly half of the cases in the credit card sample, cash was apparently used to make bail.

Citywide, 45% posted cash — 54% in Queens. However, court records did not reliably distinguish between payments by credit card and by cash, so some credit card bail making could have been recorded as cash in error. It is possible that credit cards were used

5%

(123)

(1,077)

(165)

6% (75)

21%

45%

(261)

(552)

Combined Boroughs N = 1,233

6%

12% 19%

14% 41%

Bronx N = 130

23%

more often, and cash less often, than our research reflects.

One in five defendants in the credit card sample was detained to disposition (21%). This proportion was lowest in the Bronx (14%) — aside from Staten Island, where cash was posted in all three credit card cases — and highest in Manhattan (23%).

The defendant was released on recognizance (ROR) in 75 credit card cases (6% citywide), mostly to comply with the law requiring that detained defendants be released if the charges are not substantiated within five days.

In the remaining credit card cases, bail was posted by commercial bond (5%) or the form of bail was unknown (10%).

2% 6% 12%

6%

(117)

Credit Card = Cash 87%

10% 13%

(57)

Credit Card Higher Than Cash 9%

(43)

Credit Card Cash No Release ROR Bond Bail Posted (type unknown)

5% 26%

N = 1,237 Credit Card Lower Than Cash 3%

Figure 5 Pretrial Release Outcomes (Credit Card Sample)

11%

12% 4% 4%

Figure 4 Credit Card Amount Relative to Cash Bail

43%

Manhattan N = 862

100% 54%

Queens N = 238

Staten Island N=3

Four cases were excluded from bail-making analyses because of missing case processing data.

4

September 2014

Research Brief Comparisons With Non-Credit Card Cases Defendants who were offered the option of paying by credit card differed in significant ways from other defendants with similar bail. These differences were magnified in the subgroup of cases in which the defendant used a credit card to post bail.

Figure 6 shows that defendants in credit card cases were more likely to be white, em-

ployed, and at low risk of failure to appear, compared to defendants in the comparison sample. (The risk assessment is made in CJAâ&#x20AC;&#x2122;s pre-arraignment interview, using objectively validated criteria.)

The percentages of white and low-risk defendants in the subgroup (24% white, 38% low risk) were more than double the per-

centages in the comparison sample (10% white, 15% low risk).

The proportion with full-time employment in the subgroup (61%) was nearly double that of the comparison sample (35%).

The credit card sample as a whole was more similar to the comparison group, but still disproportionately white, employed, and at low risk of FTA.

Figure 6 Ethnicity, Employment, and Risk of Failure to Appear (FTA) Subgroup (Bail Paid by Credit card)

Credit Card Sample (Credit Card Bail Set)

Non-Credit Card (Comparison) Sample

N = 165

N = 1,237

N = 20,554

Percent Employed

24% (39)

61% (101)

14% (178)

46% (570)

21% (259)

35% (7,119)

15% (3,049)

Figure 7 Criminal History Percent With Prior Conviction 39% (62)

37% (60)

58% (698)

56% (669)

65% (13,113) Subgroup N= Credit Card N = Comparison N =

159 1,194 20,085

66% (13,288) Subgroup N= Credit Card N = Comparison N =

161 1,202 20,152

Cases with missing criminal history data were excluded from Figure 7, and cases with missing release data were excluded from Figure 8.

Figure 8 Pretrial Release Rates Percent Released 79% (967)

56% (11,566)

on pa ris m Co

Defendants who were offered a credit card option were much more likely to be released prior to disposition than defendants in the comparison sample who had similar bail but were not given the opportunity to pay by credit card. Figure 8 shows that the defendant was released in 79% of the credit card cases, compared to 56% of the non-credit card comparison sample. (The subgroup of cases with bail paid by credit card is not shown separately because by definition all were released.) Multivariate analyses showed that differences in defendant and case characteristics explained only a tiny part of the difference in release rates. The predicted probability of release, after accounting statistically for a wide range of other factors, was 75% for credit card cases and 58% for the comparison cases (not shown).

Research Brief #35

Percent With Prior Warrant

ar d

Defendants who paid bail by credit card (subgroup) were much less likely to have a prior criminal conviction (39%) than those in the comparison sample (65%, Figure 7). Results were similar for prior warrants: 37% in the subgroup, as opposed to to 66% in the comparison group, had a prior warrant. Percentages for the credit card sample as a whole were in between, but closer to the comparison group.

38% (62)

ed it C

10% (2,125)

Percent Low Risk of FTA

Cr

Percent White

N = 1,228

20,549

5

Comparisons With Non-Credit Card Cases (continued) Not only were release rates higher for defendants with a credit card option, but those who were detained spent less time in jail than their counterparts in the comparison group.

The effect was especially strong for the subgroup who paid by credit card. The mean detention length for the subgroup was 3 days (median zero), compared to 10 days (median 1 day) for the credit card sample as a whole,

and 18 days (median 5 days) for the comparison sample (Figure 9, left panel).

Accordingly, the research found large differences among sample groups in the percent released by specified intervals following arraignment. The differences diminished as time passed (Figure 9, right panel).

At arraignment, 77% of the subgroup were released, compared to 31% of the credit card

sample and 16% of the comparison group. After one day, the proportion who had been released (or had their cases disposed) was 85% for the subgroup, 51% for the credit card sample as a whole, and 26% for the comparison group.

After 30 days, most had been released or had their cases disposed in all groups: over 90% in both credit card groups and 84% in the comparison sample.

Figure 9 Length of Pretrial Detention Subgroup (Bail Paid by Credit card)

Credit Card Sample (Credit Card Bail Set)

Non-Credit Card (Comparison) Sample

N = 164

N = 1,223

N = 20,507

Mean and Median Detention Length (in days)

Percent Released (or case disposed) 77%

18

Mean

Leaving arraignment

31% 16% 85%

After 1 Day

51% 26%

10

92% After 5 Days

78%

5

61%

Median 3

Figure 10 Failure to Appear Rates (At-risk defendants only) 12% 10%

N = 164

mp

ari

so

n 967

Co

it C a rd ed Cr

rou

p

8% bg

Despite higher release rates and less time spent in detention, released defendants in the credit card groups were no more likely to fail to appear (FTA) than defendants in the comparison sample. In fact, Figure 10 shows that FTA rates were slightly lower among defendants who were offered payment by credit card (10%) than among the comparison sample of cases with similar bail (12%). Moreover, the lowest FTA rate was found among the subgroup of credit card cases with bail posted by credit card (8%).

Multivariate analyses showed that the predicted probability of FTA differed by only one percentage point between the credit card (11%) and comparison groups (12%), after accounting statistically for risk of FTA and other factors (not shown). This difference was not statistically significant.

6

95% 91% 84%

After 30 Days

Su

0

1

11,566

Only defendants who were released pre-disposition were included in the FTA analysis.

September 2014

Research Brief Summary and Conclusions The research showed that providing the option of posting bail by credit card resulted in higher pretrial release rates, less time spent in detention, and no increase in failure to appear rates. From that standpoint, the program must be judged a resounding success.

However, its overall impact is minimal because judges rarely use the credit card option in setting bail. Payment by credit card was offered to the defendant in only 3% of eligible cases during the program’s first full year of Citywide operation — and none in Brooklyn.

Why? For some judges, the decision may hinge on whether someone in the defendant’s family has a credit card. Poverty that results in an inability to make low bail also results in a low level of credit card ownership, and this factor may partially explain why credit card bail is not set more often. However, it doesn’t explain why no credit card bail was set

in Brooklyn, or why only one or two judges in each of the other boroughs accounted for most of the credit card bail that was set.

Moreover, only 13% of those who were offered the option actually used a credit card for bail. If most judges were screening defendants for credit card ownership, we would expect a much higher percentage of use, when offered.

There may be resistance to offering payment by credit card for fear that paying on credit will somehow encourage failure to appear. That fear is groundless, as this research has shown.

Another possible explanation is that it simply doesn’t occur to many judges to set credit card bail, accustomed as they are to setting bail in only two forms, cash and bond.

It was puzzling to find that the offer of a credit card option alone affected outcomes, even when no credit card was used. Payment by

credit card made the largest difference in every analysis, but the offer by itself also resulted in significantly higher likelihood of release and shorter detention. This finding, along with the surprisingly low proportion of cases with payment by credit card, and the confusing and unreliable bail-making records encountered in OCA, leads us to suspect that more defendants in the credit card sample paid by credit card than our data showed.

Finally, one might object that those who were offered credit card bail were those who would have made bail anyway. To some extent, this is true. However, regression analyses showed that even when compared statistically to others with the same charactistics, defendants in the credit card group were released more quickly and in greater numbers.

For the few who were given the chance to pay on credit, the program made a big difference.

Policy Implications The credit card bail program’s success in meeting its goals justifies its expansion to extend the benefits to a much larger group of defendants than the small number currently being served. Low credit card ownership among the defendant population creates built-in limits, but the clustering of credit card bail setting among a few judges in each borough, and the total absence of this form of bail in Brooklyn, suggest that those limits have not yet been reached. Three specific recommendations follow: • Raise Judicial Awareness. • Raise the Credit Card Cap • Improve Record Keeping.

Some of the courts’ apparent (while continuing to waive the The Office of Court Administrareluctance to set credit card bail fee). The current credit card tion (OCA), the official source could stem from unfamiliarity maximum of $2,500 is too low to of case processing data in New with this new bail form, as well benefit about 40% of defendants York, does not adequately distinas a tendency to adhere to old with bail. Raising the cap to guish between cash and credit habits. Administrative judges $3,500 would bring about 8,000 card in its coding of the form of in each borough (especially more defendants into the eligi- bail making. With credit cards in Brooklyn) could help raise ble bail range. However, if this making the transition into a perawareness. Defense attorneys comes at the price of instituting manent form of bail making, relishould also be alert to their cli- fees to offset higher costs, a bet- able record keeping by the courts ents’ resources and request ter strategy would be to focus on will be essential for assessing payment by credit card when- including more of the defendants the impact of this innovation in ever appropriate. who are already eligible. the future.

Research Brief #35

7

Research Brief from No. 35 (September 2014) Paying For Bail On Credit The New York City Criminal Justice Agency, Inc. 52 Duane Street New York, NY 10007

TO:

Research Brief from No. 35 (September 2014) Paying For Bail On Credit Forthcoming: No. 36 (January 2015) Queens Supervised Release: Legal Outcomes

by Freda F. Solomon, Ph.D.

Most recently published in this series: No. 34 (May 2014) Desk Appearance Tickets: Their Past, Present, and Possible Future (Phillips) No. 33 (January 2014) Case Processing in Brooklyn’s Integrated DV Court (Peterson) No. 32 (May 2013) Queens Supervised Release: A Brief Program Description (Curbelo et al.)

No. 31 (January 2013) The EVE Project (Peterson) No. 30 (September 2012) New York City’s Bail System — A World Apart (Phillips) No. 29 (May 2012) Evaluation of Brooklyn’s Video Statement Program For DV Cases (Peterson) No. 28 (January 2012) Implementing Brooklyn’s Video Statement Program For DV Cases (Peterson) No. 27 (September 2011) How Release Type Affects Failure To Appear (Phillips) No. 26 (May 2011) Commercial Bail Bonds In New York City (Phillips)

Reports available at www.nycja.org/library.php