Financial Consequences of Pretrial Detention: Slutzky & Xu (2025)

What this is. The paper’s core results, identification design, and estimating equations: enough to understand what was found and how, without reading all 45 pages. To replicate or extend, read the original at https://doi.org/10.1093/rfs/hhaf009.

TL;DR

This paper asks whether pretrial detention, which holds individuals in jail before trial because they cannot afford bail, causes subsequent household financial distress. Using Maryland criminal court data (2000-2016) matched to bankruptcy filings from Gross, Notowidigdo, and Wang (2014), foreclosure, and judgment lien records, it exploits the quasi-random assignment of court commissioners to cases as an instrument for detention decisions. A more lenient commissioner reduces the probability of detention, and this variation is unrelated to defendant characteristics. The main finding is that pretrial detention causally increases household insolvency, raising chapter 7 bankruptcy rates by 0.79 percentage points (30% of the mean) and judgment lien rates by 0.56 percentage points (35% of the mean) within three years. Foreclosures increase significantly (2.9 pp, 23% of the mean) only in ZIP codes with declining house prices, consistent with home equity acting as a liquidity buffer. The financial burden falls primarily on family members, not defendants themselves, consistent with cohabiting relatives bearing bail-related costs. Commercial bail bonds appear to contribute alongside other channels, but the authors caution that the evidence does not establish that bonds cause defaults. Insolvency effects also persist when commercial-bond cases are excluded and in samples that largely eliminate income-loss and criminal-record channels.

Core results

Magnitudes and significance are as reported; \*\* = 5%, \*\*\* = 1%. All regressions instrument detention with the residualized leave-out mean commissioner leniency measure (first-stage F-stat exceeds 6,000 unless noted).

#	Result	Locator	Magnitude
R1	Pretrial detention raises chapter 7 bankruptcy at the 1-3 year horizon; null for chapter 13	Table 5, p. 3353	+0.44 pp at 1 year**, +0.76 pp at 2 years**, +0.79 pp at 3 years** (30% of 2.65% mean); chapter 13 insignificant at all horizons
R2	Pretrial detention raises judgment lien rates at the 3-year horizon	Table 6, p. 3355	+0.56 pp at 3 years** (35% increase relative to 1.6% mean); insignificant at shorter horizons
R3	Foreclosure effects are null overall but large in areas with declining house prices	Table 7, p. 3356	Full sample: +0.92 pp at 3 years (insignificant); negative-HPI subsample: +2.9 pp** at 3 years (23% of mean); null in positive-HPI subsample
R4	Overall household insolvency (bankruptcy + lien + foreclosure) rises by 2.4 pp at 3 years	Table 8, p. 3359	+1.7 pp** at 2 years, +2.4 pp*** at 3 years (16% of 14.8% mean); clean zero in backward-looking placebo tests (Figure 8)
R5	Insolvency burden falls on family members, not defendants	Table 9, p. 3362	InsolventDef: +0.0027 (insignificant, mean 1.6%); InsolventFamily: +0.0075** (22% of 3.4% mean); InsolventUnrelated: +0.0123 (insignificant)
R6	Commercial bail bonds contribute but do not fully explain the insolvency effect	Table 10, p. 3365	ROR/commercial-bond sample: +4.22 pp***; excluding commercial bonds: +1.96 pp*; +6.39 pp* in the sample also restricted to timely release and dismissed cases
R7	Insolvency effects are stronger for younger defendants and longer-maturity mortgages	Table 11, p. 3367	Young defendants (age < 30): +2.78 pp*; mortgage-financed properties: +3.86 pp**; long-maturity mortgages: +5.48 pp**; short-maturity estimate +2.36 pp is insignificant; consistent with older relative bearing bail costs
R8	Failure to appear magnifies the insolvency effect for defendants and family members via bond forfeiture	Table 13, p. 3369	FailToAppear x Detained interaction: +0.0308*** (InsolventDef, col 2); +0.0088*** (InsolventFamily, col 3); insignificant for InsolventUnrelated (col 5); OLS result, causal interpretation limited
R9	Last-name-matched insolvency also rises	Table 9, p. 3362	InsolventName: +0.0119** (col 3), mean 0.0491; 2SLS
R10	No statistically significant spillover to unrelated people at the defendant’s address	Table 9, p. 3362	InsolventUnrelated: +0.0123 (SE 0.0082), insignificant, 12% of mean 0.0987; table is positive but the paper’s discussion calls the estimate negative; 2SLS
R11	Residualized commissioner leniency is balanced on observed case and prior-financial characteristics	Table 2, p. 3347	Joint F-test p = 0.4993 for 10 covariates in column 3; court-year and ZIP-code-year FE; commissioner-clustered SE
R12	First-stage estimates are large in magnitude across subsamples; the table’s negative sign conflicts with the text’s release interpretation	Table 3, p. 3349	Table coefficient -0.9457*** (SE 0.0081), F = 13,526.11; subgroup estimates -0.8980 to -1.0819. Table labels the instrument ReleasedRIV, but prose describes a positive ReleasedRIV-release relationship; paper defines DetainedRIV = 1 - ReleasedRIV for detention specifications
R13	Detention reduces timely release and increases detention duration	Table 4B, p. 3351	2SLS: timely release -0.5441*** (SE 0.0163); log detention duration +1.3162*** (SE 0.0514)
R14	Detention raises the likelihood of a guilty disposition	Table 4B, p. 3351	2SLS coefficient +0.0356*** (SE 0.0086)
R15	Detention reduces failures to appear in court	Table 4B, p. 3351	2SLS coefficient -0.1301*** (SE 0.0094)
R16	Detention raises recidivism	Table 4B, p. 3351	2SLS coefficient +0.0191*** (SE 0.0032)
R17	Pre-hearing placebo outcomes show no systematic insolvency differences	Figures 4, 5, and 8, pp. 3354, 3355, 3360	Backward-looking chapter 7/13 bankruptcy, judgment lien, and overall-insolvency coefficients are small and statistically indistinguishable from zero
R18	Demographic heterogeneity is limited, while insolvency is higher for young defendants at property purchase	Table 11A, p. 3367	Young +0.0278* vs old +0.0211*; Black +0.0215** vs non-Black +0.0396**; young at purchase +0.0286** vs old at purchase +0.0193 (insignificant)
R19	Housing-linked insolvency is concentrated in higher-exposure subsamples	Table 11B, p. 3367	High-price +0.0532*** vs low-price +0.0022; mortgage +0.0386** vs no mortgage +0.0158; long maturity +0.0548** vs short +0.0236; negative HPI +0.0370** vs positive +0.0162
R20	Bond use responds to home-equity conditions, with commercial surety use rising across all splits	Table 12, p. 3368	Property bond: +0.0525*** in positive-HPI areas and +0.0409*** for short maturity; commercial surety: +0.1403 to +0.2434*** across HPI and maturity splits

Overall (paper’s conclusion). Pretrial detention imposes significant household financial costs that extend beyond the defendant to cohabiting family members, consistent with relatives bearing bail-related costs. The 3-year bankruptcy effect (0.79 pp) is comparable in magnitude to Dobkin et al. (2018), who find that hospital admissions raise bankruptcy rates by 0.4-1.4 pp. Home equity cushions households from insolvency: the foreclosure effect is concentrated in areas with declining house prices, where households cannot tap equity to meet liquidity shocks. Commercial bail bonds appear to contribute alongside other channels, though the evidence does not establish that bonds cause defaults. The findings add to the literature on the collateral damage of the criminal justice system and are relevant to ongoing debates about bail reform.

Theory / model

The paper has no formal economic model. It tests whether pretrial detention creates household financial distress through immediate bail costs, lost income during detention, or longer-run income losses following conviction and reduced employment (pp. 3330-3332). Gupta, Hansman, and Frenchman (2016) document that pretrial detention raises conviction rates and reduces formal employment, motivating these potential income channels. The authors argue that liquidity pressure should affect chapter 7 bankruptcy and liens, while foreclosure should be most likely when negative home equity prevents households from using property as a buffer. Foote, Gerardi, and Willen (2008) describe this double-trigger foreclosure channel. The family-name results test whether distress falls on relatives who post bail or cosign bonds.

Identification. Maryland court commissioners vary in release leniency. The design assumes conditional case assignment is unrelated to potential financial outcomes, and that commissioner leniency affects those outcomes only through detention. The balance test does not reject joint orthogonality of the residualized leave-out measure to observed covariates (Table 2, p. 3347); first-stage estimates remain sizeable across the race, sex, age, and court subsamples (Table 3, p. 3349). Monotonicity is not directly testable, so the authors use the consistent first-stage direction across subsamples as supporting evidence (p. 3348).

Method

The authors adapt the commissioner leniency instrument used by Dahl, Kostøl, and Mogstad (2014) and Dobbie, Goldin, and Yang (2018). They first residualize release decisions on defendant and case characteristics, then calculate each commissioner’s leave-out average residual release decision for the year. The focal defendant’s own cases are removed from that average (pp. 3344-3345).

The paper’s first numbered equation gives the baseline estimating relationship (Equation 1, p. 3344):

$$Y_{ict} = \beta_0 + \delta \, \text{Released}_{ic} + X_{ict}\beta + \epsilon_{ict} \tag{1}$$

The residual release decision and the leave-out commissioner average are defined in Equations 2 and 3 (p. 3345):

$$\text{Released}^{*}_{ict} = \text{Released}_{ic} - \gamma X_{ict} = \text{ReleasedRIV}_{ctj} + \epsilon_{ict} \tag{2}$$ $$\text{ReleasedRIV}_{ctj} = \frac{1}{n_{tj} - n_{itj}}\left(\sum_{k=0}^{n_{tj}} \text{Released}^{*}_{ikt} - \sum_{c=0}^{n_{itj}} \text{Released}^{*}_{ict}\right) \tag{3}$$

Here, $X_{ict}$ contains defendant and case characteristics, including age and fixed effects described in Section 2.5; $j$ indexes the commissioner, and $t$ the year. The paper complements the release leniency instrument with its detention counterpart, $\text{DetainedRIV}=1-\text{ReleasedRIV}$. The first-stage F-statistic is 13,526.11 in the main sample (Table 3, p. 3349); the table’s negative coefficient sign is inconsistent with the narrative’s positive release relationship. Standard errors are clustered by commissioner.

Empirical specifications

For the main 2SLS analyses, the paper instruments the binary detention decision with the leave-out commissioner measure and estimates Equation 1 with the paper’s listed fixed effects. The outcomes include indicators for case results and cumulative household insolvency within horizons of three months, six months, one year, two years, and three years. The baseline fixed effects are court-by-year, ZIP-code-by-year, month, day-of-week, sex, race, and charge; standard errors are clustered by commissioner (pp. 3344, 3347). Table 3 reports a coefficient of -0.9457 (SE 0.0081), with F = 13,526.11, although its negative sign is inconsistent with the text’s stated positive ReleasedRIV-release relationship (p. 3349). The bankruptcy sample has N = 306,722 and the principal ZTRAX-matched insolvency sample has N = 275,325 (Tables 5 and 8, pp. 3353 and 3359).

For binary or cumulative outcome $Y^{\tau}_{ict}$, the estimation uses the 2SLS version of Equation 1, replacing the endogenous treatment with detention and instrumenting it with DetainedRIV. The reported coefficient is the effect of detention; each observation is a criminal case matched to an individual or household financial outcome. The specifications below report the major samples and outcomes; all use the fixed effects and clustered standard errors stated above unless noted.

Pretrial case outcomes (Table 4B, p. 3351). The 2SLS specifications estimate timely release, log days detained, guilty disposition, failure to appear, and recidivism. The five coefficients are -0.5441 (SE 0.0163), +1.3162 (0.0514), +0.0356 (0.0086), -0.1301 (0.0094), and +0.0191 (0.0032), respectively; N ranges from 438,736 to 502,546.

Household financial outcomes (Tables 5-8, pp. 3353-3359). Bankruptcy specifications use N = 306,722; lien specifications use N = 502,546; foreclosure and combined insolvency specifications use N = 275,325. The estimates are horizon-specific cumulative indicators. The 3-year chapter 7 coefficient is +0.0079 (SE 0.0037), while chapter 13 estimates are insignificant at all horizons. The 3-year lien coefficient is +0.0056 (0.0023). For foreclosure, the 3-year coefficient is +0.0092 (0.0074) in the full sample, +0.0291 (0.0123) in negative-HPI ZIP codes, and -0.0033 (0.0089) in positive-HPI ZIP codes. Combined insolvency at three years is +0.0242 (0.0090).

Household members and mechanism specifications (Tables 9-13, pp. 3362-3369). Table 9 uses 2SLS with address and name matches for defendant, family, last-name, and unrelated insolvency outcomes, N = 275,325. Table 10 estimates three-year insolvency for bail-bond subsamples, with samples from N = 69,437 to 229,728; its first-stage F-statistics range from 144.60 to 3,220.57. Table 11 estimates three-year insolvency across defendant and housing subsamples; Table 12 estimates property-bond and commercial surety-bond use for ZTRAX-matched cases. These specifications retain the baseline fixed effects and commissioner-clustered standard errors. Table 13 is explicitly OLS because failure to appear is endogenous and is not instrumented. Its specification is:

$$Y_{ict} = \beta_0 + \beta_1\text{Detained}_{ic} + \beta_2\text{FailToAppear}_{ic} + \beta_3(\text{FailToAppear}_{ic} \times \text{Detained}_{ic}) + X_{ict}\beta + \epsilon_{ict}$$

The Table 13 sample has N = 275,820 and includes insolvency indicators for the defendant’s address matched to the defendant, family names, broader last names, or unrelated names. It uses the baseline fixed effects and commissioner-clustered standard errors. The positive interaction for defendant and family outcomes is suggestive only because failure to appear is endogenous (Table 13, pp. 3369-3370).

Placebos and robustness. Figures 4, 5, and 8 report no systematic pre-hearing insolvency effects (pp. 3354, 3355, 3360). The paper also reports robustness checks with a lagged instrument, a non-residualized instrument, apartment-inclusive samples, and absorbing-event exclusions in Internet Appendix Tables A.III-A.VI (pp. 3360, 3363-3364). The text describes these estimates as qualitatively unchanged; exact appendix magnitudes are not transcribed here.

Datasets used

Dataset	Role in paper	Wiki page
Maryland Judiciary public access database	Criminal case records: 1.08 million cases 2000-2016; commissioner ID, release decisions, bail types, charge categories	Maryland Judiciary
PACER (Public Access to Court Electronic Records)	Consumer bankruptcy filings 2000-2011: 318,000 filings in Maryland; chapter 7 and chapter 13 type, filing date, address	PACER
ZTRAX (Zillow Transaction and Assessment Database)	Real estate transactions 1993-2020: 9 million Maryland transactions; foreclosure events post-2007, property-level matching	ZTRAX (licensed)
Maryland Judiciary civil court records	Judgment lien filings 2000-2020: 386,938 lien filings; plaintiff/defendant address, filing date	Maryland Judiciary
Federal Housing Finance Agency HPI	ZIP-code-level annual house price index; used to split sample into negative/positive HPI growth subsamples	No page yet

Sample: over 500,000 criminal cases in Baltimore City, Montgomery County, and Prince George’s County District Courts. 78% from Baltimore City. 81% Black defendants, 83% male, median age 30.

When to read the full paper

Read the original if you are: studying the economics of the bail system or pretrial detention reform; working on household insolvency and liquidity shocks more broadly; replicating or extending the leave-out commissioner leniency instrument to other outcomes or jurisdictions; or testing the double-trigger hypothesis for foreclosures with a new source of liquidity shocks. Tables 5-9 contain the headline regressions; the Internet Appendix (available on the RFS website) contains robustness tests including a lagged IV, non-residualized IV, apartment-inclusive sample, and absorbing-state tests.

Attribution and rights

Source: peer-reviewed, The Review of Financial Studies 38(11), November 2025. This distillation was extracted by an LLM on 2026-06-06 and is not human-verified or independently reproduced. The paper is paywalled (OUP standard publication reuse rights); no verbatim reproduction. Access the original at https://doi.org/10.1093/rfs/hhaf009.

Citation: Slutzky, Pablo, and Sheng-Jun Xu. “The Financial Consequences of Pretrial Detention.” The Review of Financial Studies 38, no. 11 (2025): 3329-3373. DOI: 10.1093/rfs/hhaf009.