27 Jan Guest Post: Ford–Complexity and Efficiency at International Criminal Courts
[Stuart Ford is an Assistant Professor at The John Marshall Law School.]
It is common to see people criticize international tribunals as too slow, too expensive, and inefficient. Professor Whiting even argues this is now the consensus position among “policymakers, practitioners, and commentators (both academic and popular).” But are these criticisms accurate? At least with respect to the International Criminal Tribunal for the former Yugoslavia (ICTY), I believe the answer is no.
Most of those who have criticized the ICTY are implicitly comparing the ICTY to trials in domestic courts. And indeed, ICTY trials take much longer than the average domestic criminal proceeding. For example, in 2011 nearly 70% of criminal trials in federal courts in the United States took one day or less to try and there were only 37 trials that lasted more than 20 days. See here at Table T-2. In comparison, the average ICTY trial has lasted 176 days. So, it is true that trials in the U.S. are much quicker than trials at the ICTY, but it is also true that ICTY trials are vastly more complex than the average domestic trial, and we generally expect more complex trials to be more expensive. As a result, it is misleading to compare the cost and length of the ICTY’s trials to those in other courts without first accounting for the complexity of those trials.
Consequently, I propose a method for measuring trial complexity based on the number of trial days, trial exhibits and trial witnesses needed to complete a trial. The figure below shows the relative complexity of trials at the ICTY and in the U.S. As you can see, the average domestic trial barely registers on the chart, and even the Lucchese trial, one of the most complex trials ever conducted in the U.S., is only about half as complex as the ICTY’s most complex trial. But measuring complexity is just the first step to understanding whether the ICTY is too slow and expensive.
Efficiency is often defined as the ratio of work performed to energy expended. The principal function of international criminal courts is to decide the guilt or innocence of those accused of having committed violations of international criminal law. They do this primarily by conducting trials. Thus, the work they perform can be measured by the complexity of the cases they try. Their cost is a measure of the energy they expend. Efficiency then becomes total trial complexity divided by cost. Thus with a complexity measure and cost data it is possible to calculate the overall efficiency of the ICTY and compare it to other courts. If the ICTY is more efficient than comparable courts, then it has not been too slow or expensive given the complexity of the cases it has tried.
Next, I looked for data on the efficiency of domestic courts in the United States. Some of my findings are shown in the table below. A number greater than 1 indicates better efficiency than the ICTY, a number lower than 1 indicates worse efficiency.
|
Relative Efficiency of Various Courts and Trials |
|
|
Relative Efficiency |
|
| Murder Trial in North Carolina |
14.9 |
| Bifurcated Death Penalty Trial in NC |
11.3 |
| Murder Trial in Kansas |
9 |
| Death Penalty Imposed in Kansas |
3.56 |
| Murder Trial in Maryland |
1.88 |
| Scott Peterson Prosecution |
1.24 |
| Death Penalty Imposed in Maryland |
1.18 |
| ICTY |
1 |
| O.J. Simpson Trial |
0.78 |
| Timothy McVeigh Defense |
0.05 |
| Timothy McVeigh Prosecution |
0.02 |
One thing that stands out from the data is that there is no standard efficiency for trials in the U.S. Efficiency varies over time and by location. More importantly, efficiency appears to be inversely related to trial complexity. In all the studies I found, cost increased faster than trial complexity causing efficiency to decrease as complexity increased. Thus, you cannot compare the efficiency of a court that hears simple cases to a court that hears complex cases. All other things being equal, the court that hears the simple cases will appear to be more efficient, but this is just an artifact of its lower average trial complexity. To make a meaningful comparison of inter-court efficiency, one must compare cases of similar complexity.
The question then becomes not whether the ICTY is more efficient than domestic criminal trials, but rather which trials to compare it to? The ICTY is much less efficient than the average murder trial in the United States, but the average murder trial is not a useful comparator. First, it is much less complex than the average ICTY trial – the average murder trial is only 3 or 4 days long. Second, it looks very different from the ICTY’s trials. It usually involves a single perpetrator and a single victim, a simple theory of liability, and the trial is concerned with a relatively small number of locations and a short period of time. In contrast, the ICTY’s trials usually involve multiple defendants, many more counts, crimes that take place at a large number of locations and over a longer period of time, hundreds to thousands of victims, indirect modes of liability, and the necessity of proving the criminal acts of many individuals other than the accused.
Death penalty trials are more complex than the average murder trial, but they are still dramatically less complex than the average ICTY trial making them a poor comparator. The charged conduct also generally looks more like the average murder trial than the ICTY’s trials. Nevertheless, the ICTY was almost as efficient as death penalty cases in Maryland, although it was less efficient than death penalty cases in North Carolina and Kansas. It was also roughly as efficient as a number of high profile murder cases, including the trials of Scott Peterson (see here at page 21 and following) and O.J. Simpson.
There is one type of trial that approaches the ICTY’s trials in complexity and looks like the ICTY’s trials in terms of the charged conduct – domestic trials of mass atrocity crimes. However, these are so rare in domestic courts that I had to look beyond the U.S. to find even a handful of cases that had data on cost and complexity. What I found was that the ICTY was very likely more efficient than the Spanish prosecution of those responsible for the 2004 train bombings in Madrid and more than three times as efficient as the Norwegian prosecution of Anders Breivik. Finally, the ICTY was about fifty times more efficient than the prosecution of Timothy McVeigh for the bombing of the Oklahoma City federal building.
Ultimately, the ICTY looks pretty good. It is more efficient than its closest domestic comparator – mass atrocity trials. It even turned out to be roughly as efficient as some death penalty cases, even though the ICTY’s trials have been significantly more complex than the average death penalty case. This suggests that the ICTY’s trials have not been too slow or expensive given their complexity. Indeed, it seems likely that domestic prosecution of the same cases would have been slower and/or more expensive.
Next, I intend to use the data I have collected on the ICTY’s indictments and my complexity measure to explore what causes the complexity of international criminal trials. Many efforts to modify the rules of procedure and evidence at international tribunals to produce faster or more efficient trials have been based on assumptions about what causes the complexity of their trials. Better understanding the causes of complexity should help us understand why many of these efforts have seemingly failed as well as point the way to better solutions.
very interesting. you might also want to compare trials of violations of international law in domestic courts (universal jurisdiction cases) to see how they compare to trials of the same kind in international tribunals.
I like the idea. I would guess the difficulty would be finding data on domestic trials. That has been the most surprising part of my research. I initially thought that collecting data on the ICTY would be the hard part. It wasn’t until I had collected the ICTY data that I realized how hard it was going to be to find domestic data.
This is an interesting and necessary defense in a framework where numbers matter more than the humanitarian reasons behind the judgments. Have you looked into the ICTR as well / are there plans for doing so?
An alternative to universal jurisdiction cases, which may cover only a small fraction of the crimes committed (ie, Spanish citizens victims of the Pinochet regime) instead of being comprehensive, what about looking into the mixed tribunals/commissions which are rather local, internationally-assisted bodies, such as Guatemala or Cambodia?
Indeed, I would say that comparing the ICTY to American courts is probably doing it a disfavour, in light of what you say about there being only ’37 trials that lasted more than 20 days’ in the whole of the US. Here in Australia, 20 trial days does not seem to be an unusual number for a single count of murder, at least where the case is circumstantial. And that is just on a relatively common garden domestic violence or low-level drug debt killing. If one looks at white collar criminal matters, involving complex transactions and corporate structures, then the trial time will often be measured not in days but weeks or months.
Pablo,
I would like to look at the ICTR. Unfortunately, I would guess that data collection would take about a year. I plan on doing it someday, but not right now. As to the ECCC, I can’t be certain until the trials are over, but I am pretty confident predicting that it will be less efficient than the SCSL (and thus also less efficient than the ICTY).
I haven’t looked at the Guatemala trial or any of the trials of the Pinochet victims. That does sound interesting though. Perhaps someone with a better command of Spanish would be willing to undertake that analysis?
Rob,
That is certainly interesting. Given that ordinary murder trials in the US usually take between 3 and 4 trial days, I would be kind of surprised if Australian trials were really 5 times longer. Perhaps because murders are so much rarer in Australia each one is treated as special whereas in the US they are treated as routine? Otherwise, procedurally trials in Australia and the US seem close enough to each other that it seems hard to explain such a large discrepancy.
I did look for statistical data on Australian criminal trials, but I know relatively little about Australian law (or Australian data sources) so it is entirely possible I missed something. Do you know of any data sources for average trial lengths in Australia?
There was a book written about the length of criminal trials here about 10 years ago, which is probably now out of date (I would guess that trials have gotten longer if anything). That was pretty much confirmed several years ago by a report on ‘trial efficiency’ in one State, being New South Wales.