1328f.org

Consumer Bankruptcy Research & Accountability

← All Reports

Measuring Bankruptcy Attorney Performance Using Public Data

Methodology Published March 2026

Summary

Attorney quality is the strongest predictor of Chapter 13 outcomes. This report describes a methodology for measuring attorney performance using publicly available PACER data, including control group construction, metric selection, and statistical validation. All tools are open-source and reproducible for any federal bankruptcy district.

1. Why Measure Attorney Performance

Chapter 13 success depends on sustained legal work over 3-5 years. Unlike Chapter 7 - where discharge is largely automatic for eligible debtors - Chapter 13 requires plan drafting, creditor negotiation, lien analysis, motion practice, compliance monitoring, and ongoing client communication. The quality of this work varies dramatically between practitioners.

Research confirms the variation. Professors Jimenez and Hynes (2025 JELS) documented significant attorney-effect differences in Chapter 13 outcomes. The Government Accountability Office has noted that Chapter 13 completion rates range from under 30% to over 60% depending on jurisdiction and practitioner. Yet no standard methodology exists for measuring individual attorney performance using publicly available data.

This report fills that gap. Using PACER Case Locator exports and straightforward statistical methods, any researcher, trustee, or bar authority can evaluate attorney performance against local benchmarks.

2. Data Source: PACER Case Locator

The PACER Case Locator (PCL) provides searchable case data for all federal bankruptcy courts. CSV exports contain the fields necessary for attorney-level performance analysis:

FieldUse
Case NumberUnique case identifier
Filing DateCase start date, cohort assignment
DispositionOutcome classification (discharged, dismissed, converted, open)
ChapterChapter 7 vs. 13 comparison
Attorney NamePractitioner identification and grouping
District/OfficeGeographic controls

Cost is minimal. PCL searches cost $0.10 per page, capped at $3.00 per search. A comprehensive attorney export for a single district typically costs $3.00-$9.00. Fee-exempt PACER accounts are available to researchers and qualifying organizations.

3. Key Metrics

3.1 Dismissal Rate (Resolved Cases)

The percentage of completed cases ending in dismissal rather than discharge. This is the primary outcome metric. Calculated as: dismissed / (dismissed + discharged), excluding open and converted cases. Resolving only completed cases avoids bias from recent cohorts that have high open-case percentages.

3.2 Discharge Rate (Resolved Cases)

The inverse of dismissal rate: discharged / (dismissed + discharged). A practitioner whose clients achieve discharge at significantly lower rates than peers in the same courts is underperforming on the most important metric.

3.3 Early Dismissal Rate

The percentage of cases dismissed within 90 days of filing. These are intake-stage failures - cases that never reached plan confirmation, often dismissed for nonpayment of filing fees, failure to file schedules, or failure to attend the 341(a) meeting. A high early-dismissal rate indicates systemic problems in case preparation and client screening.

3.4 Filing Day Distribution

The distribution of case filings across days of the week. Independent practitioners typically show a relatively even spread across weekdays, with perhaps modest concentration on certain days. Extreme concentration - more than 50% of filings on a single day - suggests batch processing rather than individualized case preparation.

3.5 Disposition Breakdown

What causes dismissals matters as much as the dismissal rate itself. PCL disposition codes and docket-level review can distinguish between:

A practitioner with a high rate of filing-fee and failure-to-file dismissals has a different problem than one whose clients fail at plan payment. The former suggests intake failures; the latter may reflect client-population effects.

3.6 Chapter Comparison

Comparing the same practitioner's Chapter 7 outcomes against Chapter 13 outcomes provides an internal control for client quality. If a practitioner's Chapter 7 discharge rate is comparable to peers but Chapter 13 outcomes are significantly worse, the divergence suggests a deficiency in the sustained legal work that Chapter 13 requires rather than a difficult client base.

4. Control Group Construction

4.1 Why Controls Matter

Raw metrics are meaningless without comparison. A 45% Chapter 13 dismissal rate could be excellent in a district where the baseline is 55%, or alarming in a district where the baseline is 30%. Comparing against national averages is insufficient because local factors - judges, trustees, local rules, economic conditions - produce significant district-to-district variation.

The valid comparison is against independent practitioners filing in the same courts during the same time periods.

4.2 Tiered Control Structure

Our methodology uses a two-tier control group:

TierCompositionPurpose
Primary (Tier 1)5 independent, high-volume practitioners with low dismissal ratesBest-practice benchmark
Secondary (Tier 2)10 additional independent practitioners with moderate-to-high volumeBroader baseline, reduces selection bias

4.3 Selection Criteria

Control attorneys must meet the following requirements:

4.4 Combined Control Set

In one analyzed portfolio, the combined control group included 15 attorneys and 16,000+ resolved Chapter 13 cases across 3 districts. This volume provides robust statistical power for detecting meaningful differences in performance.

5. Statistical Significance

With thousands of cases, small percentage-point differences can be highly statistically significant. We use chi-squared tests for independence to assess whether differences between the target practitioner and control group are likely due to chance.

Example

A practitioner with a 58% dismissal rate on 1,200 resolved cases vs. a control group dismissal rate of 43% on 16,000 cases produces a chi-squared statistic with p < 0.001. The 15-percentage-point gap is not plausibly attributable to random variation.

We report confidence intervals where appropriate and flag comparisons where sample sizes are too small for reliable inference (typically fewer than 50 resolved cases).

6. Controlling for Confounders

Several factors could produce outcome differences unrelated to attorney quality. The methodology addresses each:

ConfounderControl Method
Economic conditionsCompare same time periods. Year-over-year cohort analysis.
Judicial assignmentCompare same courts. Where case assignment is random, judge effects average out at sufficient volume.
Client demographicsChapter 7 as internal control for client quality (same attorney, same intake).
Case ageExclude recent cohorts with high open-case percentages. Use resolved-only metrics.
District variationWeight control-group metrics by district to match the target practitioner's filing distribution.

6.1 The Chapter 7 Internal Control

This is the most powerful confounder control available. If a practitioner's Chapter 7 clients achieve discharge at rates comparable to peers, the client base is not obviously different in quality. If Chapter 13 outcomes diverge sharply from Chapter 7 outcomes for the same practitioner - but not for controls - the divergence isolates the attorney's Chapter 13 work as the likely variable.

7. Interpreting Results

7.1 What Elevated Metrics Mean

A practitioner with metrics significantly worse than the control group is producing worse outcomes for their clients than other attorneys in the same courts. This is a factual statement about observable results. It does not, by itself, explain why.

7.2 What Elevated Metrics Do Not Mean

7.3 Multiple Metrics Are More Informative Than One

A single elevated metric has limited diagnostic value. Multiple elevated metrics - high dismissal rate, high early-dismissal rate, filing-day concentration, chapter divergence - collectively paint a more informative picture. See the bankruptcy mill definition report for a scoring framework using eight quantitative indicators.

8. Reproducibility

All tools used in this methodology are open-source and available on GitHub. The repository includes:

Any practitioner, researcher, bar authority, or journalist can replicate this methodology for their district using the same tools and data sources. The guide to reading PACER attorney records provides step-by-step instructions for obtaining the underlying data.

9. Related Reports

How to Cite

1328f.org, "Measuring Bankruptcy Attorney Performance Using Public Data," March 2026, https://1328f.org/reports/attorney-performance-methodology/

Not Legal Advice

This report describes a research methodology using publicly available court data. It does not constitute legal advice, does not evaluate any specific practitioner, and should not be used as the sole basis for any legal or disciplinary action. Researchers applying this methodology should exercise appropriate caution in interpreting results and consider all potential explanations for observed differences.

PACER cases made free through RECAP: 0 of 37.9 million

Every document we access becomes permanently free for the next researcher, attorney, or debtor. More added daily.

Support this mission