Evidence from Senegal shows that the length of legal pre-trials can be reduced whilst not impacting quality of decisions
The speed of justice matters for development. Court efficiency affects transaction costs in enforcing contracts and allowing gains from trade (North 1991, Djankov et al. 2008). Djankov et al. (2003) identify legal origins and procedural formalism as main drivers of court inefficiency. For this reason, measures of the investment climate such as those found in the Doing Business report reward legal reforms aimed to simplify procedures and speed up adjudication. However, there is no evidence that legal reforms affect judges’ behaviour in a way that combats de facto formalism (La Porta et al. 2008).
This relates to the wider issue of bureaucratic performance. The drive to improve court efficiency and reduce procedural formalism relates to a more general problem in increasing government performance. While incentives may improve performance when civil servants face a single task, or in addressing the extensive margin of service provision (e.g. absenteeism), performance contracts may in fact be counterproductive in multi-tasking environments (Finan et al. 2017, Rasul and Rogger 2017).
The challenges of studying judges’ behaviour
The dearth of evidence on judges’ response to changes in incentives is not surprising. Judges are the epitome of high-level bureaucrats who benefit from far-reaching independence and authority, and perform a variety of complex tasks. Two main challenges are thus apparent:
- The complexity of their work means monitoring judges’ performance is particularly difficult.
- The need for an independent justice system means that it is hard to experiment with judges’ incentives.
The study: Can changing the rules of the game affect judges’ performance?
In a recent study, we empirically evaluate the impact of a legal reform on the efficiency and quality of adjudication in the civil and commercial court of Dakar, Senegal (Kondylis and Stein 2018). The reform gave judges the duty and powers to conclude pre-trial proceedings in four months (Decree no 2013-1071). At baseline, i.e. before the reform, pre-trial hearings lasted on average 157 days, or over two-thirds of the total length of court proceedings. The decree allowed judges to desk-reject cases based on insufficient evidence in the first hearing.
The application of the decree was staggered across the six chambers of the court. Combined with high-frequency caseload data, this allowed us to set up an event study with multiple cut-offs. In practice, this means that we expected to see a drop in pre-trial duration for cases that entered a chamber just after the decree was introduced. The size of the drop would capture the causal impact of the decree, provided that chambers do not undergo other changes at the time of decree application. Having multiple decree introduction cut-offs allows us to purge our estimates of any changes that may have affected the whole court over the study period, which increases the confidence we place in our estimations of the causal effect of the decree.
Digitising court data
Monitoring judges’ behaviour is not straightforward, and this research would not have been possible without a thorough digital capture of caseload data over the 2012 to 2015 period (here and here). Before we started this work, the digital court database contained scant case-level information, such as date of entry. We worked to add granularity to these data to retrace the full history of all civil and commercial cases over our study period, down to each hearing. This level of detail on legal proceedings allows us to track changes in judges’ behaviour across the reform introduction cut-off.
The first order question is whether a simple legal reform can change the behaviour of skilled, independent, and powerful bureaucrats such as judges. We begin by measuring the impact of the reform on the speed of pre-trial proceedings. Averaging across the six chamber cut-offs, we find that cases that entered the court after the introduction of the decree complete their pre-trial procedure about 43 days faster than those that entered before (Figure 1). In the same vein, these cases are 23 percentage points more likely to end pre-trial stages in under four months.1
Changes in procedural formalism
What do judges do differently to achieve these speed gains? They use their new powers to desk-reject cases that present extremely weak evidence at the first hearing (Figure 2). They are also more likely to fast track well-documented cases straight to deliberations. This is straight cut in procedural formalism: desk rejections and fast-tracking account for 44% of the reduction in delays, and cases experienced on average two fewer pre-trial hearings from a baseline i.e. before the reform of 8.3. However, desk rejections and fast-tracking are not the entire story. Judges were more likely to apply pressure on parties by issuing strict deadlines for adjournments (Figure 3).
These effects are not driven by certain types of cases (specialisation) or certain types of judges, suggesting the decree changed judges’ performance not by curbing individual inefficiencies but by solving a coordination problem among judges.
Speed and quality trade-off
A natural question is whether increased speed comes at the cost of quality of the pre-trial or of the judgement itself. Here, it is important to note that judges do not have the incentive to focus solely on speed in the pre-trial stage, bring blatantly weak cases to deliberations, and walk away. This is because after the pre-trial, all cases assigned to a chamber are reviewed and deliberated on jointly by all judges in the chamber, including the president, and they issue the judgment jointly. Poor pre-trial quality would thus immediately be noticed by a judge’s direct superior.
To gauge effects on the quality of the evidence, we exploit the fact that judges in a chamber have two options in dealing with a case that reached deliberations insufficiently prepared. They may either send it back to pre-trial, or delay their decision, thus prolonging the deliberations.2 Hence, these events can serve as pre-trial quality markers. We do not measure any significant change to these markers, which allows us to conclude that the decree did not negatively affect the pre-trial quality.
Similarly, we do not find that judges’ zeal in meeting the new pre-trial deadline hurts the speed of deliberations or the quality of the final judgement. The length of the decision justification, the number of articles cited, as well as parties’ decision to appeal serve as judgment quality markers. This is consistent with the finding that judges’ overall workload is not affected by the reform.
The economic value of faster justice
We interviewed firms that were involved in disputes over our study period to try and determine the economic value of faster dispute resolution. We elicited their willingness to pay for a lawyer that could bring down procedural delays on the same order of magnitude as the decree did, citing a dispute over a median size claim. We find that firms are on average willing to pay an extra 294,000 West African CFA francs ($555) for the fictional lawyer who works faster relative to the lawyer who works at pre-decree speed (for whom they would be willing to pay F559,000 West African CFA francs or $1056 (Figure 4).3 Taken together, these results suggest that firms see a positive value in faster proceedings.
While we find evidence that changing the rules of the game can affect the behaviour of powerful, independent judges without any measurable impact on quality, more work is needed to assess the overall welfare and distributional implications of such reforms. There may be winners and losers, and we cannot sign the net effect. However, our study shows that giving judges higher discretionary powers lead them to change their behaviours, and that firms seem to benefit from these changes.
Djankov, S, R La Porta, F Lopez-de-Silanes and A Shleifer (2003), “Courts”, The Quarterly Journal of Economics 118(2): 453-517.
Djankov, S, O Hart, C McLiesh and A Shleifer (2008), “Debt enforcement around the world”, Journal of Political Economy 116(6): 1105-1150.
Finan, F, B Olken, Band R Pande (2017), “The personnel economics of the developing state”, Handbook of field experiments, Volume II. North Holland: A Banerjee and EDuflo (eds).
Kondylis, F, and M Stein (2018), “The speed of justice”, World Bank Policy Research Working Paper no. WPS 8372
La Porta, R, F Lopez-de-Silanes and A Shleifer (2008), “The economic consequences of legal origins”, Journal of Economic Literature 46(2): 285-332.
North, D, (1991), “Institutions”, Journal of Economic Perspectives 5(1): 97-112.
Rasul, I and D Rogger (2018), “Management of bureaucrats and public service delivery: Evidence from the Nigerian civil service”, Economic Journal 128: 413–446.
 When we presented these results to the president of the court in May 2017, he was surprised that the decree had had such an effect; he does not use statistics to monitor effort, and there are no formal sanctions accompanying this decree, only potential informal complaints from parties.
 Pre-reform, cases being sent back to pre-trial and decisions delays occurred for 11.8 and 5.5 percent of cases, respectively.
 We also elicited stated preferences for court fees that can speed up the procedure, and find qualitatively similar results. In addition we asked the firms how much they would have to pay in lawyer and court fees and found no significant effect of the reform.