The Costs of International Justice Author(s): David Wippman Source: The American Journal of International Law, Vol. 100, No. 4 (Oct., 2006), pp. 861- 881 Published by: Cambridge University Press Stable URL: Accessed: 19-07-2018 02:21 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide

range of content in a trusted digital archive. We use information technology and tools to increase productivity and

facilitate new forms of scholarship. For more information about JSTOR, please contact

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

Terms and Conditions of Use

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to The American Journal of International Law

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to



By David Wippman *

Critics ofinternational criminal justice commonly point out that the International Criminal

Tribunal for the Former Yugoslavia (ICTY) and the International Criminal Tribunal for Rwanda (ICTR) have held relatively few trials despite substantial annual budgets. In short, they claim that international justice is costly and slow.

Although formed in 1993 with the modest annual budget of $276,000, the ICTY soon mushroomed into an institution with twenty-eight judges (sixteen permanent and twelve ad

litem) and over eleven hundred staff. Its biennial budget for 2006 and 2007 is $276,474,100.1

The ICTR has also grown dramatically. It now has sixteen permanent judges, nine ad litem judges, and an authorized staff of over one thousand. Its biennial budget for 2006 and 2007 was $269,758,400.2 The combined biennial budget of the two tribunals thus exceeds $545 million, an amount equal to approximately one-seventh of the regular UN budgets for a com-

parable period.3 Over the course of its first thirteen years, the ICTY alone (including the 2006-

2007 budget) has been allocated over $1.2 billion.4

The budgets of the two institutions have grown to keep pace with the escalating number and

cost of legal proceedings. At the same time, the number of cases actually tried by the two tri-

bunals remains small. The ICTY tried only 6 offenders in its first six years ofoperation. As polit-

ical changes within the former Yugoslavia and increased international support enabled the Tri-

bunal gradually to expand its operations, the numbers have increased significantly, but remain

low overall. As of October 10, 2006, the ICTY had indicted 161 people, and tried 51; 46 were convicted and 5 acquitted.5

The ICTR has conducted very few trials. As ofJune 2006, the ICTR had issued twenty-two

judgments involving 28 accused. Of these cases, however, five were resolved by guilty pleas.

* Vice Provost for International Relations and Professor of Law, Cornell University. The author wishes to acknowledge receipt of U.S. government funding for an earlier version of this Note; the views expressed in it, how- ever, are solely those of the author. The author also wishes to thank Larry Johnson for helpful comments on an early draft of the paper.

1 ICTY at a Glance (Aug. 10 & Sept. 13, 2006), at <>. ICTY doc- uments cited in this Note are available online at the Tribunal’s Web site, <>.

2 ICTR, About the Tribunal (n.d.), at <>. 3 The regular budget of the United Nations for 2006 was approximately $1.755 billion; on average, the regular

budget runs to about $1.9 billion, although total UN spending is closer to $15 billion, taking into account spending on peacekeeping operations, the programs and funds, and specialized agencies. Questions and Answers About the United Nations, ch. 5 (Mar. 2006), at <>.

4 ICTY at a Glance, supra note 1. Actual spending is somewhat less, as not all moneys allocated are expended. 5 Id.


This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


These numbers have generated substantial resentment on the part of many of the UN mem-

ber states asked to pay for the tribunals, and increasing criticism from commentators inside and

outside the UN system. For example, Ralph Zacklin, UN assistant secretary-general for legal

affairs, argues that the tribunals “have grown into enormous and extremely costly bureaucratic

machines that outstrip or rival in size many of the specialised agencies of the United Nations,

and dwarf core offices and departments of the UN Secretariat.”‘ He adds that “the expenditure

of such resources is not wholly justified in terms of discharging their principal task of bringing

to justice those responsible for the most serious crimes in a timely and expeditious manner.”7

While many disagree with Zacklin’s assessment, concerns over cost have “precipitated a financial crisis at the ICTY,” and (along with the failure of some states to pay their assessed

contributions) led to a June 2004 recruitment freeze.8 Moreover, cost concerns played a major

role in the adoption of the ICTY’s “completion strategy,” designed to bring the work of the

Tribunal to a close by 2010.9 The underlying question debated at the United Nations and elsewhere is “whether the ICTY

(and ICTR) [a]re providing value for money.”‘0 One way to approach the issue is to compare the cost of criminal trials in an effective national legal system with the cost of international

criminal trials. Making such a comparison is, at least in part, an apples-and-oranges exercise,

since national legal systems for the most part do not handle the kinds of cases prosecuted by

the ad hoc tribunals, and operate in a very different legal and political environment. None-

theless, such a comparison can help put the costs of international tribunals in perspective.

In general, while the average criminal trial in the United States costs far less than an average

criminal trial in the ICTY or the ICTR, a major U.S. criminal trial involving numerous victims

may cost far more. Thus, while the millions spent per trial in the ICTY may appear exorbi-

tant,11 those numbers pale in comparison to the sums spent on the cases of Timothy McVeigh

and Terry Nichols, which cost the United States government over $82.5 million in prosecution

costs alone, not counting the costs on appeal.12

6 Ralph Zacklin, The Failings ofAd Hoc International Tribunals, 2 J. INT’L CRIM. JUST. 541, 543 (2004). 7 Id.

8 Dominic Raab, Evaluating the ICTY and Its Completion Strategy-Efforts to Achieve Accountability for War Crimes and Their Tribunals, 3 J. INT’L CRIM. JUST. 82, 96 (2005). The hiring freeze was lifted in January 2005, but pressures to control costs remained strong.


SLAVIA 3 (AI Index EUR 05/001/2005, June 2005), available at < engeur050012005>, which stated:

The Tribunal’s “completion strategy” appears to be mostly dictated by financial constraints influenced by a changing geopolitical setting, where countries of the former Yugoslavia have become less of a priority in the international scene, and based on the assumption that local courts in former Yugoslav countries would be able to perform the Tribunal’s tasks at a lower cost.

10 Raab, supra note 8, at 88.

” Dividing the ICTY budget by the number of trials concluded produces a figure of some $18 million per trial, but that figure omits the costs of proceedings related to individuals indicted but not tried and the costs related to cases now in progress.

12 See Kevin Flynn, Bombing Case Cost $82.5 Million, ROCKY MTN. NEWS (Denver), Nov. 3, 1998, at 4A, avail- able in LEXIS, News Library, Major Newspapers File; Howard Pankratz, Bomb Trials Cost $82.5 Million, DENV. PosT ONLINE, Nov. 3, 1998, at < 103.htm>.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


Part I of this Note offers an analysis of the relative costs of criminal trials in the United States

and the ICTY,13 with the caveats indicated above in mind. This analysis finds that per-trial-day

costs do not differ dramatically between the two legal systems; as discussed in part II, the high

cost of ICTY proceedings derives principally from the length, complexity, and international nature of its proceedings. Part II examines some of the critical differences between ICTY and

U.S. criminal trials and what is counted as a judicial cost in each, as well as some of the structural

and procedural features of the ICTY that help explain the length and cost of its trials. Part III

touches briefly on the quality of ICTY proceedings. Ultimately, justification for the costs of international justice may fall outside a strict utili-

tarian analysis. As Judge Richard Matsch of the Oklahoma City bombing case observed in response to a juror who questioned the cost of the Nichols trial, “We don’t use a cost-benefit

analysis in deciding what’s necessary for a fair trial.”14


Attempting to compare U.S. and ICTY criminal trial costs is in many respects a frustrating

exercise. The United States compiles and publishes an enormous amount of data concerning the U.S. law enforcement and criminal justice system. Unfortunately, the data are generally aggregated in ways that do not yield per-case, per-trial, or per-trial-day numbers. Moreover,

the data are collected in categories that for the most part do not correlate readily with the fairly

extensive information contained in annual reports and other documents detailing ICTY costs and expenditures. Further, although the Administrative Office of the United States Courts

(AO) distributes funds directly to each court in accordance with a spending plan approved by

the Judicial Conference, I have not been able to obtain budget or spending information for individual federal district courts.1

A handful of studies provide data on the costs of criminal trials in individual U.S. states.

These studies focus on the difference in cost between capital and noncapital cases, and use dif-

ferent assumptions, cover different time periods, and aggregate data in different ways. A few

provide data on costs associated with cases as a whole, including pretrial, trial, and postconviction

proceedings. The ICTY data, however, are not aggregated in ways that yield per-case costs.

13 For simplicity, the analysis here focuses on the ICTY, the longest-running and most productive of contem- porary international criminal tribunals, rather than on the ICTR or other tribunals.

14 Karen Abbott & Lynn Bartels, Teacher Rips Bomb Trial as Too Costly, ROCKY MTN. NEWS, Oct. 17, 1997, at 22A.

15 The Administrative Office describes the budget process for the federal courts as follows:

The Administrative Office, in consultation with the courts and with various Judicial Conference committees, prepares a proposed budget for the judiciary for each fiscal year. The proposal is reviewed and approved by the Judicial Conference and is submitted to the Congress with detailed justifications …

After Congress enacts a budget for the judiciary, the Judicial Conference approves a plan to spend the money, and the Administrative Office distributes funds directly to each court, operating unit, and program in the judiciary. Individual courts have considerable authority and flexibility to conduct their work, establish budget priorities, make sound business decisions, hire staff, and make purchases, consistent with Judicial Con- ference policies.


This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


In general, the most useful comparison would probably be between the average cost of com-

plex criminal cases in the United States and the average cost of ICTY cases, but the available studies on state courts in the United States are limited to just a few states and yield cost estimates

that vary significantly from one state to another;’16 moreover, the available federal and ICTY

cost information does not readily lend itself to such a comparison. Accordingly, the analysis

below focuses on the average cost of ICTY and federal trials on a per-day basis, occasionally

turning to state court data for additional points of comparison, and then offers some general observations about nontrial costs.

To complicate matters, the ICTY’s budget is a moving target. Early budgets include one- time start-up costs, and more recent budgets reflect costs associated with the ICTY’s comple-

tion strategy (e.g., those pertaining to the transfer of cases to national courts). In addition, the

ICTY has gradually implemented a series of reforms designed to control costs and expedite trials.

Nonetheless, it is possible to make some helpful comparisons between the two systems. The

next section compares the per-trial-day costs in the United States with those in the ICTY.

Daily Administrative Costs of Criminal Trials in the United States and the ICTY

The Administrative Office of the U.S. Courts, after considerable prodding, provided a one-

page summary of the administrative costs of a criminal jury trial in U.S. federal courts for fiscal

year 2002. Factoring in the trial judge’s salary and benefits, those of the judge’s staff and the

courtroom deputy, miscellaneous expenses, space and facilities costs, and court security, the

Administrative Office reaches an hourly cost of $409.76; at an average of five hours spent per

day on trial, it finds a daily trial cost (excluding prosecution, defense, U.S. marshal, jury, and

jury clerk costs) of $2049. Jury costs amount to $1000 per day except for the first day of trial

when costs of prospective jurors must be included. Accordingly, total administrative trial costs

(excluding first-day juror costs) come to $3049 per five-hour day, or $4278 per eight-hour day.17

16 The most comprehensive state study examines the cost of murder cases in North Carolina. PHILIP J. COOK & DONNA B. SLAWSON, THE COSTS OF PROCESSING MURDER CASES IN NORTH CAROLINA (1993) (with Lori A. Gries), available at <>. This 1993 study finds that the average non-death-penalty murder trial costs $17,000, the average capital trial ending with the guilt phase costs $57,000, and the average bifurcated capital trial (that is, the average capital trial including both the guilt and sentencing phases) costs $84,000. Id. at 47. A 2002 Indiana study finds that the average cost of the death penalty phase (from filing of charges to sentencing) is $63,095 for a life-without-parole murder case, and $272,975 for a death penalty case, excluding most prosecuting attorney and court personnel costs. M. Goodpaster, Cost Comparison Between a Death Penalty Case and a Case Where the Charge and Conviction Is Life Without Parole, in THE APPLI- CATION OF INDIANA’S CAPITAL SENTENCING LAW: FINDINGS OF THE INDIANA CRIMINAL LAW STUDY

COMMISSION 122A, 122F (2002), available at <> [here- inafter INDIANA REPORT]. A 2003 Kansas study finds that the median cost of a noncapital murder trial (excluding overhead costs such as space) is $32,000, and that the median cost of a capital trial ending in the death sentence is $508,000. COSTS INCURRED FOR DEATH PENALTY CASES, A K-GOAL AUDIT OF THE DEPARTMENT OF COR- RECTIONS 13 (Dec. 2003), at <> [herein- after K-GOAL AUDIT]. A 2004 Tennessee study yields an average life-with-parole murder trial cost of $31,622, and an average bifurcated capital trial cost of $46,791. EMILY WILSON, BRIAN DOSS, & SONYA PHILLIPS, TEN- NESSEE’S DEATH PENALTY: COSTS AND CONSEQUENCES 16 (July 2004), at <http://www.comptroller.state.>. Most of the cases considered take much less time to try than ICTY cases. For example, in Kansas, the average non-death-penalty trial lasted nine days; the average death penalty trial lasted twenty-eight days. K-GOAL AUDIT, supra, at 15.

17 AO, Daily Cost of a Criminal Jury Trial-FY 2002 (Apr. 16, 2002) [hereinafter AO Estimate].

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


Within limits, a similar calculation can be performed for the ICTY, using data for 2001 (2002 introduces complications including the move to a biennial budget and the introduction

of ad litem judges).

Judicial salary and expenses. The AO calculation lists the annual salary and benefits of a dis- trict judge for fiscal year 2002 at $161,324.'” The annual salary of an ICTY judge for 2001 was

$160,000.19 Common costs for the judges, including travel on home leave, education allow- ance, pension benefits, and costs associated with installation and relocation amount to an addi-

tional $67,121 per judge, for a total annual judicial salary and expenses cost of $227,121.20 Since ICTY trials require three-judge panels, the annual recurring cost of judges’ salaries and

expenses for comparison purposes would be $681,364, as opposed to $161,324 for a U.S. dis- trict court judge. On a per-trial-day basis (assuming an eight-hour day and 260 workdays per

year),21 the comparison would be $620.48 for a U.S. district court judge as opposed to $2620.63 for a three-judge ICTY panel. Adding in the cost of the jury makes the comparison $1620 per day for a U.S. judge and jury against $2620.63 for a three-judge ICTY panel. As noted, the ICTY common cost figures used above may overstate the annual recurring cost

of items such as relocation expenses.22 The ICTY has provided a more recent estimate of $4,162,100 for twenty-three judges’ salaries and costs; that works out to $180,960 per judge.23

Using these figures, the annual cost for three judges would be $ 542,882; on a per-trial-day basis

(assuming an eight-hour day and 260 workdays), that would amount to $2088 per day for a three-judge panel as compared to $1620 per day for a U.S. judge and jury. Judges ‘staffsalary and benefits. The Administrative Office calculates an annual recurring cost

for judges’ staff salary and benefits of $311,331. This number apparently includes a judge’s law

clerks and secretary; it is not clear whether it includes any other staff (e.g., the clerk’s office staff

or the court reporter). The number of judges’ staff associated with a particular trial for the

ICTY is difficult to calculate with any precision. Each trial chamber (consisting of three judges)

is typically supported by approximately five professional staff; the staff costs are included in the

budget for the Registry rather than the budget for chambers. According to a former member

of the Office of the Prosecutor, the support staff of a trial chamber typically would include

(using UN standard salary-scale designations) one staff member at the P5 level (a senior judicial officer), two to three at the P3 level, and one or two at the P2 level. Salaries within each of those

18 Id. This number seems to understate benefits, since the annual salary of a federal district court judge in 2002 was $150,000.


20 This figure may overstate the recurring annual cost, since installation and relocation expenses (over half of the common cost amount) presumably are not incurred annually. U.S. district court judges, who live in the general vicinity of their courts, generally do not incur home leave travel or relocation expenses, which together account for over two-thirds of the common costs for ICTY judges.

21 The use of an assumed eight-hour day is for comparison purposes only. In fact, trials rarely occupy a full eight hours per day in either system. Critics of the ICTY often complain that the judges work fewer hours than in the United States or elsewhere, but neither U.S. nor ICTY judges record their hours. ICTY judges do take two three- week recesses each year, and are entitled to another two weeks when they may take vacation.

22 See note 20 supra. 23 David Pimentel, chief, court management, ICTY, e-mail to author (Mar. 12, 2005).

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


categories vary depending on years of experience. In addition, each chamber will have a sec-

retary, interns, and others. Judges receive additional legal support from the Tribunal’s Registry.

Unfortunately, the ICTY budget information does not break out these support staff costs from

other Registry costs in a way that permits a ready comparison to the AO figures. However, the

proposed 2001 budget does contain an estimate of the costs required to add six adlitem judges.

This proposed budget specifies fourteen new posts for chambers’ legal support, consisting of

two P3 legal officers, six P2 legal officers, and six GS staff (secretaries).24 Using those figures,

and the median salaries on the 2001 UN salary scale (with an average 2001 post adjustment multiplier of 3.35), the staff costs for the six ad litem judges would be $159,756 (salaries for

two P3 legal officers), plus $381,252 (salaries for six P2 legal officers), plus $275,046 (for six

GS secretaries), for a total of $816,054. Divided among the six ad litem judges, these costs would amount to $136,008 per judge, or $408,024 per three-judge trial chamber. On a per- day basis (assuming an eight-hour day and 260 workdays per year), the comparison on staff costs would be a daily trial cost of $1197.43 for a U.S. court, and $1569 for the ICTY.

Courtroom deputy. The Administrative Office estimates an annual daily cost for a courtroom

deputy’s salary and benefits at $218.90. The ICTY does not use a courtroom deputy as such, but it does have a courtroom usher, who acts more or less like a bailiff. In 2005 the courtroom

usher was paid $64,600 annually, which (assuming an eight-hour day and 260 workdays per year) amounts to $248.46 per trial day. In addition, the ICTY has courtroom officers who rep-

resent the registrar in the court; also in 2005, a courtroom officer was paid $89,100 annually,25 which on a per-trial-day basis amounts to $343.

Miscellaneous expenses. The Administrative Office estimates an annual recurring cost for “travel, transcripts, supplies, telephones, etc.,” as $171.97 per trial day.26 I have been unable

to identify with any confidence the ICTY costs that correspond to the AO “miscellaneous” cat-


Space andfacilities. The Administrative Office calculates the annual recurring cost of space

and facilities associated with a criminal jury trial at $859.91 per day, and explains the calcu- lation as follows:

The annual rent cost for the District Court courtrooms and chambers of Article III Dis-

trict Judges in Space and Facilities Division’s Unit Cost model is based on rent charges to an identifier for this specific type of space from the most current rent bill. This identifier includes rental charges for courtrooms, chambers, and ancillary space (e.g. jury deliber- ation suites, attorney conference rooms, judges restricted corridors, shared conference rooms, restricted elevators, and lobbies). Note that District Court courtrooms and cham-

bers rent data does not include Magistrate Judges courtrooms and chambers rent data as that information is identified as a separate rent cost item.

24 2001 REPORT, supra note 19, Annex IX, para. 11, at 104. UN professional staff salaries “are set by reference to the highest-paying national civil service.” OFFICE OF HUMAN RESOURCES MANAGEMENT, UNITED NATIONS, SALARIES & POST ADJUSTMENT (2005), at<http:l// salary.htm>. As a practical matter, that means salaries are pegged to U.S. civil service salaries, since the “federal civil service of the United States ofAmerica has to date been taken as the highest paid national civil service.” Id. In addi- tion, salaries are subject to a post adjustment multiplier, to compensate for the difference in living costs between New York City (the United Nations base) and the relevant post. Id.

25 Pimentel, supra note 23. 26 AO Estimate, supra note 17, n.2.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


Using this rent bill data we obtain an average annual rent cost per rentable square foot. Then we apply this average rent cost to the rentable square feet for an Article III District Court Judge’s courtroom and chambers based on square footage allowances in the U.S. Courts Design Guide. This calculation results in the average annual rent cost for an Article III District Court Judge’s courtroom and chambers only, not including any annual rent cost for ancillary space.

Since the unit cost data represents annual rent cost, a daily rent cost could be determined by dividing the annual cost by 365. Then the rent cost for a trial could be calculated using the number of days including weekends that the average trial covers or the comparable period of time of the trials being used in the comparison.27

The ICTY has three courtrooms, which vary significantly in size. The 2001 space rental costs

for courtrooms and chambers amounted to $100.68 per square meter for a total of 2963 square

meters at an annual cost of $298,314, or a daily average per trial cost of $272. Using more recent rates of 120 euros per square meter, the average daily space rental cost for a trial would be $324.28

Court security. The Administrative Office estimates an eight-hour per-day cost for court

security at $209.40. I have not been able to obtain comparable data for the ICTY. As noted earlier, the above comparisons use figures from the Administrative Office and the

ICTY that may mask differences in what is included and how the figures are calculated. None-

theless, it seems possible to make some “back-of-the-envelope” observations.

The largest components of a U.S. federal criminal trial as calculated by the Administrative

Office are the salary and benefits of the district judge, those of the judge’s staff, space and facil-

ities costs, and courtroom deputy and juror costs. Together, those costs constitute approxi-

mately 90 percent of the total administrative cost as determined by the Administrative Office.

Looking at those components of the daily trial cost (and excluding components such as “mis-

cellaneous” and court security, for which relevant comparisons cannot be made on the cur-

rently available data), one finds that ICTY costs are higher, but principally because ICTY trials

require three judges rather than the single judge of a U.S. criminal trial.

ICTY judges and U.S. judges are paid essentially the same amount, although ICTY judges receive significant additional sums as part of their “common costs,” to account for much higher relocation and home-leave travel costs. This discrepancy is to be expected given the location

of the Tribunal and the use of judges from countries around the world. In combination, these

two differences mean that a three-judge ICTY trial costs significantly more per trial day in

terms of judges’ time than a U.S. criminal trial, though the excess is partly offset by the costs

of a jury in a U.S. trial. Costs for judges’ staff at the ICTY are also higher, for similar reasons,

although the difference is not dramatic. The greater parity in staff costs may reflect the ability

ofl ICTY judges to share staff, which U.S. judges, who are not working on the same cases, can- not do.

Space and facilities cost considerably less in the ICTY than in U.S. courts. This observation

seems counterintuitive but may reflect donations that are not counted into the ICTY cost structure.

27 Staff of Administrative Office of the Federal Courts, e-mail to author (Feb. 2, 2005).

28 Pimentel, supra note 23.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


Overall, on basic administrative costs-judges, staff, and space (and jury costs in the United

States)-the two systems seem roughly comparable, allowing for the difference between single-

judge and three-judge trials. Adding up the figures above, and using the lower of the two figures

provided for judges’ salaries and costs, we find that an ICTY trial on an eight-hour-day basis

costs approximately $4572 per day, while a U.S. criminal trial on an eight-hour-day basis costs

about $3894 per day. On this analysis, an ICTY per-trial-day cost is only about 15 percent higher, a calculation that has to be taken with a grain of salt for the reasons noted above.

Only one of the state court studies mentioned earlier estimates the cost ofa trial day. Accord-

ing to that study, per-trial-day costs in the Superior Court of North Carolina in 1993 for judges, staff, and space amounted to $1416;29 adjusted for inflation, that would amount to $1763 in 2002 (assuming that judicial costs rose at the rate of inflation reflected in the con-

sumer price index). This figure reflects somewhat different assumptions about the number of

workdays in a year, modestly lower judicial and staff salaries, and significantly lower space costs

($216 per day in 2002 dollars compared to $860 per day in U.S. federal court). Even if jury costs are added to the North Carolina figure, the total would still be less than the costs of a trial

in federal court and significantly less than the per-trial-day cost in the ICTY. But given the vari-

ability of costs in different U.S. states and the 1993 date of the North Carolina study, the fed- eral court data seem to offer a better comparison.

Of course, even if the above numbers are fully reliable, these comparisons have only limited

value because the costs just described constitute a small percentage of the overall costs asso- ciated with criminal cases. In complex cases, much of the cost derives from investigation and other pretrial costs and from postconviction proceedings. Judges’ salaries and expenses con- stitute roughly 10 percent of the U.S. federal judiciary budget.30 At the ICTY, the amount of

the assessed budget for 2001 allocated to the trial chambers was only 3.3 percent of the total

assessed budget; the Office of the Prosecutor was allocated 29.3 percent, and the Registry 67.4 percent.31 The following sections assess some of the costs associated with the activities of those offices.

Prosecution and Defense Salary Costs

The AO cost estimate does not include prosecution or defense costs, though these may far outstrip costs associated with judges and their staff. Comparing prosecution and defense costs

on a per-trial or per-trial-day basis is not feasible; cases differ in complexity and staffing, and I have not been able to obtain from the Administrative Office or anywhere else current data

indicating average prosecution or defense costs for federal courts.32 Moreover, most of the cost

29 COOK & SLAWSON, supra note 16, at 37.

30 Where the Money Goes: A Look at How theJudiciary’s FY2005 Budget Is Divided, THIRD BRANCH AT 1 (AO,

Washington, D.C., Feb. 2005), availableat<http:ll//>. The judiciary budget cannot be compared directly to the ICTY budget, since the two budgets include different categories of costs.

31 As discussed below, these offices perform functions beyond those performed by U.S. Attorneys’ offices and clerks’ offices in the United States.

32 In 1995 the Department ofJustice “reported an average total cost per prosecution [in death penalty cases] of $365,296, but this figure does not include the cost ofinvestigation or the cost ofscientific testing and expert evaluations performed by law enforcementpersonnel.” Subcommittee on Federal Death Penalty Cases, Committee on Defender Services, Judicial Conference of the United States, Federal Death Penalty Cases: Recommendations Concerning the Cost and Quality of Defense Representation, pt. I ?B(5) (May 1998), available at < dpenalty/1COVER.htm> [hereinafter Federal Death Penalty Cases]. Some of the state court studies also provide

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


in complex cases is often incurred in pretrial investigation and proceedings. But some poten-

tially helpful general comparisons can be made. U.S. prosecution costs. Assistant United States attorneys’ pay is administratively determined

on the basis, in part, of the number of years of professional experience. Basic pay for 2005 ranged from $40,105 to $118,327, plus locality pay. (The pay ceiling in U.S. Attorneys’ offices in 2005 was $142,500.) A single assistant U.S. attorney will handle the typical simple criminal trial. More complex

criminal trials may require two to four assistant U.S. attorneys, as well as a case agent (e.g., an

FBI or Internal Revenue Service agent); an extremely complex case may require a substantially

larger prosecution task force, with agents, paralegals, and other staff. The McVeigh case, for

example, involved “dozens” of federal prosecutors and staff.33 Although the trial lasted less than

six weeks from opening statements to the jury’s verdict,34 the prosecution is reported to have

spent $10.1 million trying the case.35 Most of that expense related to work done in preparing for trial, which took over two years.

Even so, the bulk of the costs associated with prosecuting McVeigh stemmed from law enforcement investigation efforts. According to the Department of Justice, “[T]he FBI had 2,592 agents investigating the case in 1995; 760 in 1996; 552 in 1997 and 228 in 1998,” at a cost of $60.6 million.36

Of course, McVeigh’s was an exceptional case. Among other things, prosecutors sought the

death penalty, a punishment not permitted in international tribunals. As the U.S. Supreme Court has acknowledged, “death is different.” Death penalty cases in the United States typ- ically cost substantially more than other cases, for a variety of reasons.37 Some of these reasons

are unique to the death penalty and the procedures by which it is sought. Death penalty cases

usually involve bifurcated trials, with separate guilt and penalty phases, forcing in effect two

separate, though related trials. Jurors must be “death qualified” through an extensive voir dire

process; constitutional and other legal issues unique to the death penalty must be litigated; and

attorneys on both sides often feel obligated to be unusually thorough given the stakes.38

per-trial or per-case breakdowns of some prosecution and defense costs, but these costs are based on estimates pre- pared by surveying prosecution and defense attorneys, vary significantly by case type (capital or noncapital), and are associated with trials of relatively brief duration. For example, a Tennessee study showed average capital case prosecution trial costs of $11,427 and average life-with-parole case prosecution trial costs of $8,923. See WILSON, DOSS, & PHILLIPS, supra note 16, at 16. These costs cannot be readily correlated with ICTY cost data, which are not broken out by case. 33 STEPHEN JONES & PETER ISRAEL, OTHERS UNKNOWN: THE OKLAHOMA CITY BOMBING CASE AND

CONSPIRACY 112 (1998).

34 See Oklahoma Bombing Chronology, WASHINGTONPOST.COM, 1998, at < wp-srv/national/longterm/oklahoma/stories/chron.htm>. 35 Pankratz, supra note 12. 36 Id.

37 Estimates on the extent to which death penalty cases are more expensive vary considerably. According to the Kansas study, “Cases in which the death penalty was sought and imposed cost about 70% more than cases in which the death penalty wasn’t sought.” K-GOAL AUDIT, supra note 16, at ii. The Tennessee study suggests that death penalty murder trials are roughly 50 percent more expensive than non-death-penalty murder trials. WILSON, DOSS, & PHILLIPS, supra note 16, at ii. The Indiana study indicates that death penalty cases are about four times as expen- sive as non-death-penalty cases during the trial phase. K. Janeway (Reporter), INDIANA REPORT, supra note 16, at 119. In the federal courts, one 1998 study suggests that the costs of representation in death penalty cases are about four times greater than in non-death-penalty cases. See Federal Death Penalty Cases, supra note 32, Executive Summary.

38 See, e.g., Federal Death Penalty Cases, supra note 32; K-GOALAUDIT, supra note 16, at 14-15; Janeway, supra note 37, at 11-15.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


But perhaps the principal reason for the high cost of federal death penalty trials is the com-

plexity of the guilt phase. In this respect, as a study of federal death penalty representation costs

observes, “federal death penalty cases have more in common with complex drug conspiracy cases than with non-capital federal homicide cases, many ofwhich are comparatively simple.””39

Federal death penalty cases “typically involve investigations stretching over years, and encom-

passing numerous acts of violence. They often include several homicide charges, many wit- nesses, and evidence in the guilt phase derived from wiretaps, video surveillance, informants,

and experts.”40 In some respects, then, these cases resemble in complexity the kinds of cases

brought before the ICTY. Yet large, complex cases that do not involve the death penalty can also run up extraordinary

prosecution costs. In the Enron case, for example, the Justice Department assembled a large

prosecution task force to try Kenneth Lay and others. Unfortunately, prosecution costs on such

cases are for the most part not publicly available.

ICTYprosecution costs. ICTY trial teams, and therefore ICTY prosecution costs, also vary from case to case. In 2000 the ICTY had eight trial teams. Each team consisted of a senior trial

attorney at the P5 level, two co-counsel at the P4 level, and one legal officer at the P2 level, plus

a support staff that included a case manager, a trial support assistant, and three secretaries, all

at the GS level.41 Salary ranges for these positions resemble those for comparable positions in

the U.S. Attorneys’ offices. For example, the 2002 salary range of a senior trial attorney (P5)

amounted to $101,084 to $127,602, depending on experience, while that of a junior attorney (P1) in the same year was $42,944 to $56,432. These salaries are subject to a post adjustment multiplier.

Because salaries do not differ sharply between the U.S. and ICTY systems, the principal dif-

ferences in prosecution costs during trial relate to the length and complexity of trials, a subject

discussed more fully below. Similarly, prosecution costs both before and after trial vary sub-

stantially depending on the length and complexity of the pre- and post-trial proceedings.

Moreover, prosecution costs in the ICTY cover investigation. While U.S. Attorneys’ offices

may rely on federal and state law enforcement agencies to carry out most of the investigative work necessary to charge and prosecute someone for a crime, the ICTY must obtain much of

that work from in-house investigators. Thus, for the 2004-2005 biennium, the ICTY budget allocated $100 million for prosecution expenses;42 and for 2006-2007, $85 million.

U.S. criminaldefense costs. Indigent defendants in the United States are entitled to appointed

counsel under the Criminal Justice Act. In some cases, counsel will be provided by salaried members of the federal public defender service; in other cases, assigned counsel are provided.

Since the ICTY uses only assigned counsel for indigent defendants, assigned counsel are used here for comparative purposes. Assigned counsel in U.S. federal courts are entitled to be paid

up to $60 per hour for time spent in court on noncapital cases, though that amount may increase to $75 per hour in some districts.43 Case maximums on noncapital felony cases are set

39 Federal Death Penalty Cases, supra note 32, pt. I ?B(2) (footnote omitted). 40 Id.

41 2001 REPORT, supra note 19, Annex II, para. 13. 42 ICTY Budget for the Biennium 2004-2005, UN Doc. A/58/226, tbl. 2, at 7 (2003). 43 7 AO, OFFICE OF DEFENDER SERVICES TRAINING BRANCH, GUIDELINES FOR THE ADMINISTRATION

OF THE CRIMINAL JUSTICE ACT (18 U.S.C. 3006A), ch. 2, pt. C ?2.22 (2005), available at <

Publications/GenRef/Vol7/Volume_7/SectionA/Chapter_2.html#ptc> [hereinafter GUIDELINES].

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


at $7000 at the trial level and $5000 on appeal, though these caps may be exceeded in lengthy

and complex cases.44 In capital cases, assigned counsel may be paid “an amount not to exceed $125 per hour for in-court and out-of-court time.”45 They are also entitled to reimbursement for staff, investigators, expert witnesses, travel, and other expenses, subject to approval by the court in advance. In death penalty cases, the defendant is automatically entitled to two assigned counsel, but in a complex case, the trial team may turn out to be much larger. In the McVeigh trial, the court approved a team of thirty-five people, including numerous lawyers, investigators,

assorted experts (on explosives, architecture, pathology, psychiatry, trace analysis, terrorism, and

even anthropology), researchers, and others.46

The principal determinant of the cost of defense representation is the length and complexity of the case. In the period from 1988 to 1997, the average cost for defense counsel and related services in cases in which federal prosecutors were authorized to seek the death penalty was $218,112, as compared to $55,772 for non-death-penalty cases.47 The average cost of repre- sentation in drug conspiracy death penalty cases was $244,185.48 In complex, high-profile cases, defense costs can skyrocket. The total cost of McVeigh’s

defense through sentencing came to $13,780,835, of which approximately $6.7 million went to attorneys’ fees. Another $1.5 million covered staff, housing, and security. Defense inves- tigators received $2 million, and expert witnesses $3 million.49 In the Enron case, some esti- mates place the cost of the defense of the company’s former chief executive, Jeffrey K. Skilling, in connection with his four-month, non-death-penalty criminal trial at over $65 million,50 an

amount that greatly exceeds the cost of even the most expensive ICTY cases. ICTYdefense costs. In an effort to ensure “equality of arms,” the ICTY “provides, as a matter

of course, for teams for each defendant consisting of two counsel and up to three legal assistants

and investigators.””51 In its early trials, the ICTY capped payment to the defense at $200 per day.52 The Tribunal then moved to a system in which defense counsel were paid an hourly fee

at a rate “assessed by the Registrar on the basis of the seniority and experience of counsel.”53 Until recently, lead counsel received $110 per hour, subject to a determination of reasonable- ness and necessity.54 In addition, defense counsel were entitled to a daily subsistence allowance and travel expenses, as well as reimbursement for expenses incurred for expert witnesses, inves-

tigators, secretarial support, and other items. In short, assigned counsel were treated similarly to assigned counsel in federal courts in the United States.

44 Id.

45 GUIDELINES, supra note 43, ch. 6 $6.02, available at <

Volume_7/Section_A/Chapter_6.html#6.02>. 46 JONES & ISRAEL, supra note 33, at 42-43. 47 Federal Death Penalty Cases, supra note 32, Executive Summary. 48 Id., pt. I ?B(2). 49 McVeigh ‘s Defense Tab Tops $13-Million, ST. PETERSBURG TIMES (Fla.), June 30, 2001, at 4A, available in

LEXIS, News Library, Major Newspapers File. 50 Carrie Johnson, After the Enron Trial, Defense Firm Is Stuck with the Tab, WASH. POST, June 16, 2006, at D 1.

Skilling’s private counsel billed as much as $850 per hour. 51 See Sylvia de Bertodano, What Price Defence? Resourcing the Defence at the ICTIY 2 J. INT’L CRIM. JUST. 503,

504 (2004). 52 Id. at 503.

53 ICTY, Directive on the Assignment of Defense Counsel, UN Doc. IT/73/Rev.10, Art. 25, as amended July 28, 2004, available at <http:l//>.

54 Id., Annex I. Lead counsel with fewer than twenty years’ experience received less than $110 per hour, down to a minimum of $80 per hour for those with nine or fewer years’ experience. Co-counsel received $80 per hour.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


Unfortunately, the ICTY system was subject to abuse by defense attorneys, as it gave them

a monetary incentive to string out proceedings to run up their fees. In some cases, attorneys

engaged in fee-splitting arrangements with ICTY defendants. To prevent abuse and exercise some control over the length of proceedings, the ICTY in 2001 placed a cap on total hours. In 2003 the ICTY moved to a lump sum payment system. Under the lump sum system, the registrar, in consultation with trial chambers and the parties, estimates the duration and com-

plexity of the various phases of the case and allots a sum to cover defense costs for each phase.

The system varies depending on the assessed complexity of the case. Calculation of the monthly

allotments is based on the monthly salaries for a step-VII (midsalary scale) P5 lead counsel and a

step-VII P4 co-counsel, plus 40 percent of their joint salary for office costs, and funding for one,

three, or five staff members, depending on the projected complexity level of the case. As a result, the

defense team will receive the amounts noted in table 1, depending on the difficulty of the case.


Monthly Allotment for

Level of Difficulty Monthly Allotment Interpretation and Translation

1 (Difficult) C25,738 + ?3,000.00 ?1000 maximum 2 (Very Difficult) ?25,738 + ?9,000.00 ?1000 maximum 3 (Extremely Difficult/Leadership) ?25,738 + ?15,000.00 ?1000 maximum

Source: ICTY, Directive on Defence Counsel-Trial Legal Aid Policy, para. 25 (May 1, 2006).

Thus, defense teams are paid approximately $35,000 to $50,000 per month during trial, according to the complexity of the case. A different lump sum arrangement is used for pretrial

work.55 The lump sum is intended to cover “[a]ll aspects of representation except for necessary

travel and [daily subsistence allowance].””56 This system, though designed to prevent abuse, is also intended to permit counsel reasonable reimbursement for their time and expenses; the amounts themselves do not appear dramatically different from what could be expected in a comparably staffed U.S. criminal trial.

As in the United States, the principal determinants of total defense costs are the duration and

complexity of proceedings. For the 2004-2005 biennium, for example, the ICTY budget allo- cated $29.5 million to the defense, somewhat less than one-third the amount allotted to the



The main explanation for the perceived high cost of ICTY cases consists of an assortment of factors independent ofjudicial or staff salaries and expenses. These factors include the length

and complexity of international criminal trials; the inclusion of investigation, detention unit,

and other nonjudicial costs in the ICTY budget; translation and travel expenses necessitated by the international character of the Tribunal and its location; unusual witness relocation costs;

relatively modest use of plea bargaining; and the like.

55 See ICTY, Directive on Defence Counsel-Pre-trial Legal Aid Policy (May 1, 2006). 56 Id. BA(3).

57 De Bertodano, supra note 51, at 507.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


Length and Complexity of Trials

Length, complexity, and cost are related, but not always directly correlated. The McVeigh

trial was extraordinarily expensive in large part because of the high profile and complexity of

the case. The destruction of the Alfred P. Murrah federal building and the loss of 168 lives, coupled with concerns over possible foreign terrorist connections, prompted the U.S. govern-

ment to devote extraordinary resources to the investigation, apprehension, and prosecution of the perpetrators. Nonetheless, despite the extraordinary amounts ofwork that went into inves-

tigation and case preparation, the trial itself took less than six weeks; the defense case lasted only

four days. The relative brevity of the trial reflected the active management of the case by the

trial judge and tactical decisions by the prosecution and defense, both ofwhich ended up calling far fewer witnesses and introducing far fewer exhibits than either had originally anticipated. Thus, the trial of an exceptionally complex and resource-intensive case need not last months

or years to generate high costs. Conversely, even relatively simple cases may take surprisingly long to try. As the ICTY has


The complexity of a case is not necessarily correlated with the estimated duration of a case. It is possible to have a case in which the Trial Chamber permits many witnesses to testify because there are multiple defendants, which is nonetheless neither factually nor legally complex, since the accused may be lowly ranked and there may be few counts in each respective indictment.58

In general, however, complex cases take longer to try and generate higher overall costs. Most U.S. criminal trials are relatively simple cases, requiring only a single prosecuting attorney and a day or two to prosecute. By contrast, almost all ICTY trials, because of the nature of the pros-

ecutions undertaken, entail lengthy and expensive court proceedings and pretrial investiga- tions. This is the single biggest difference between the average ICTY trial and U.S. criminal trials.

U.S. criminal trials-length and complexity. Of the 13,558 cases tried in U.S. federal district

courts in the twelve-month period ending on September 30, 2001, 7045 were criminal trials.59 Of these, over half (3657 trials) lasted only one day. Most of the rest (3121 trials) lasted two to nine days. Only 201 lasted ten to nineteen days, and only 66 lasted more than twenty days.60

In other words, in 2001 less than 1 percent of federal criminal trials lasted more than twenty

days. The numbers are essentially the same for subsequent years; in 2003 only sixty-seven of

7118 criminal trials lasted more than twenty days.61 Of course, only a small percentage of criminal cases-less than 10 percent-actually goes to trial.62 Many are resolved through plea bargaining or are dismissed.

58 ICTY, Directive on Defence Counsel-Trial Legal Aid Policy, para. 20 (May 1, 2006).

59 AO, 2001 JUDICIAL BUSINESS, U.S. District Courts-Length of Civil and Criminal Trials Completed, by District, During the 12-Month Period Ending September 30,2001, tbl. C-8, available at <http://www.uscourts. gov/judbus2001/appendices/c08sep01.pdf>.

60 Id.

61 U.S. DEP’T OF JUSTICE, BUREAU OF LABOR STATISTICS, SOURCEBOOK OF CRIMINAL JUSTICE STATIS- TICS ONLINE, Length of Civil and Criminal Trials Completed in U.S. District Courts, tbl. 5.42, at 445 (31st ed. 2003), available at <> [hereinafter SOURCEBOOK].

62 Dan Ackman, Quattrone Fought Long Odds, and Lost, FORBES.COM, Sept. 9, 2004, at <http://www.> (citing AO statistics).

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


The types of criminal cases tried vary widely, from migratory bird offenses to homicide. The

Judicial Conference of the United States employs a weighting system to reflect the broad dif-

ference in complexity associated with various case types. Average criminal cases receive a weight

of approximately 1.0; more time-consuming cases receive higher weights. Thus, a death penalty

habeas corpus case is assigned a weight of 12.89 (which implies it will take almost thirteen times

as long to handle as an average case), while a defaulted student loan case is assigned a weight of

0.031 (indicating it will take less than one-third the time required to handle an average case).

Most U.S. criminal cases are not difficult under these criteria or as compared to the average

ICTY case. Of the new cases filed in 2001, there were far fewer homicides (0.53 percent of the

total), kidnappings (0.167 percent), and assaults (1 percent) than traffic violations (7.9 per- cent) and larceny and theft (5.2 percent).63 As a general matter, simpler cases require less pre-

trial investigation and skirmishing, fewer attorneys and staff, fewer witnesses and exhibits, and

correspondingly lower overall expenditures.

ICTYtrials-length and complexity. A March 2000 report by then-ICTY president Claude Jorda on behalf of all of the ICTY’s judges notes that the average length of ICTY trials up to

that point was a little over twelve months (with average pretrial preparation lasting slightly over

ten months).64 The shortest trial to that point had lasted over six months, and the longest (later

outstripped by the Milo3ievic trial) lasted two years and seven months.65 By May 2003, the aver-

age length of trials had reached approximately seventeen months.66 Responding to consider-

able criticism, the ICTY adopted a series of measures designed to expedite trials; by May 2004,

they averaged twelve months in length.67 Nevertheless, as the ICTY increasingly focuses on leadership cases as part of its completion strategy, the duration of trials is unlikely to decline

and may even go up. The length of ICTY trials may be attributed in part to flaws in the design and operation of

the Tribunal. Recognizing this shortcoming, the ICTY has adopted a series of mitigative mea-

sures, some recommended rather recently by working groups on speeding up trials and appeals.

These measures include reducing the number of offenses charged, limiting live testimony, join-

ing defendants, and controlling the proceedings to reduce delaying tactics.68

63 SOURCEBOOK, supra note 61, tbl. 5.10, available at < pdf>.

64 Claude Jorda, Report on the Operation of the International Tribunal for the Former Yugoslavia, UN Doc. A/55/382-S/2000/865, Annex I, at 9. Figures on trial lengths, however, have to be viewed with some caution. The ICTY prosecutor argues that

the true measure of the duration of a trial was not the time period over which the trial extended, but rather the actual number of days of trial, excluding periods when courtrooms or judges were unavailable for various reasons and excluding trial suspensions owing to interlocutory appeals, the need to deal with motions, illness or unavailability of attorneys or other necessary parties, etc.

Report of the Expert Group to Conduct a Review of the Effective Operation and Functioning of the International Tribunal for the Former Yugoslavia and the International Criminal Tribunal for Rwanda, UN Doc. A/54/634, para. 171 (1999) [hereinafter Expert Group Report]. The prosecutor goes on to argue that “by comparison with comparable complex civil or criminal litigation in national courts, there were not likely to be dramatic differences between the actual length of trials before the Tribunal and those in national courts.” Id.

65 Jorda, supra note 64, at 9. 66 Raab, supra note 8, at 89. 67 Id.

68 For a discussion of some of the specific measures adopted, see Assessment and Report of Judge Fausto Pocar, UN Doc. S/2006/353, Annex I (2006).

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


The principal explanation for lengthy trials, however, lies in the type of cases the ICTY is

mandated to pursue. As a group of experts formed to consider means open to the International

Criminal Court (ICC) to manage the length of its proceedings observed:

Due to the fact that international crimes typically involve atrocities committed on a mas- sive scale, international criminal justice has to cope with cases which are more extensive and complex than most national cases. In particular, hundreds of witnesses will have to be interpreted and heard and volumes of documentary evidence will have to be translated and evaluated. The complexity will be multiplied whenever more than one conflict fall to be addressed concurrently.69

As noted above, the ICTY itself rates its cases “as being either Level 1 (difficult), Level 2 (very

difficult), or Level 3 (extremely difficult/leadership).'”70 To make that assessment, the ICTY uses the following factors:

* the position of the accused within the political/military hierarchy;

* the number and nature of counts in the indictment;

* whether the case raises any novel issues;

* whether the case involves multiple municipalities (geographical scope of case);

* the complexity of legal and factual arguments involved; and

* the number and type of witnesses and documents involved.71

As of early 2006, the proposed ICTY budget for 2006-2007, reversing the order of complexity

of the three levels of cases, projected “level 1” cases as featuring 500 witnesses and some five thousand exhibits; “level 2” cases about 300 witnesses and three thousand exhibits; and “level 3” cases about 150 witnesses and one thousand exhibits.72

Relative to most national criminal court trials, ICTY trials are inherently complex. The nature of the crimes at issue- genocide, crimes against humanity, and war crimes-may force

prosecutors to introduce evidence on questions such as the nature of the conflict in the locality

where the crime took place (e.g., was it international or noninternational, or both at varying

times?), “the link between the conflict and the crime,” “proof that the crime was committed

against a member of a protected class of persons,” the place of the accused in the relevant political and military hierarchy, the number and type of atrocities in a given area, and sim- ilar matters.73

The nature of the ICTY’s mandate as a tribunal established to prosecute serious violations of international law tends to produce large-scale proceedings in yet another way, for its goals

include fostering accountability, deterring future atrocities, providing acknowledgment to vic-

tims, and building a historical record of the conflict. With those goals in mind, the Office of

69 ICC Office of the Prosecutor, Informal Expert Paper, Measures Available to the International Criminal Court

to Reduce the Length of Proceedings, para. 8 (2003), available at < length_ofproceedings.pdf>. 70 Directive on Defence Counsel, supra note 58, para. 23. 71 Id., para. 19. 72 Report of the Secretary-General, Budget for the International Tribunal for the Former Yugoslavia for the Bien-

nium 2006-2007, UN Doc. A/60/264, para. 49 (2005).

73 Expert Group Report, supra note 64, para. 166.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


the Prosecutor has generally felt obliged to issue broad indictments, covering the full range of

acts for which a given accused is responsible, rather than selective indictments that might focus

on a few of the gravest crimes and lead to more expeditious trials.74 By contrast, in a national

legal system, prosecutors might choose to try a serial killer for only one of the murders he alleg-

edly committed. The ICTY prosecutor’s more inclusive approach has occasioned some con- troversy,75 particularly in the Milos’evii trial. Arraying a broad range of charges in multiple

counts of often-amended indictments complicates both the investigation and the trial of cases

and enlarges trial lengths and overall costs.

Moreover, ICTY counsel and judges, particularly in the Tribunal’s early years, lacked a well-

established body of either substantive or procedural law to apply; as a result, numerous issues

had to be litigated and resolved, ranging from the authority of the Tribunal itself, to the ele-

ments of the crimes it was prosecuting, to novel procedural questions stemming in part from

the Tribunal’s status as a hybrid civil-common law body. In developed legal systems, many such issues need not be litigated because the applicable law is clear.76

In addition to the inherent complexity of international criminal trials, a host of other factors

help explain both the length and the cost of ICTY proceedings. Some of the most important of these factors are outlined below.

Reliance on State Cooperation

In well functioning national legal systems, prosecutors and courts may generally rely on law

enforcement agencies to make timely arrests and assist in the identification and production of evidence needed for trial. The ICTY, like other international criminal tribunals, has no coer-

cive powers to arrest suspects or compel testimony or the production of evidence. It must depend on the cooperation of states to carry out many of its tasks. Unfortunately, only on occa-

sion can the ICTY count on state cooperation in making arrests and gathering evidence. Even

then, the ICTY must do much of its own investigative legwork. As a result, the Office of the

Prosecutor, which receives almost 30 percent of the ICTY budget, has to maintain a large staff

of “investigators[,] . . . military and criminal intelligence analysts, researchers, documents analysts and forensic specialists.”77 Indeed, the Investigation Division constitutes the largest

part of the Office of the Prosecutor.78

The complexity of the necessary investigations far exceeds what is normally required in national prosecutions. Investigators must visit distant locales and interview numerous witnesses in a given case, often in a difficult postconflict security environment. Major inves-

tigations “have involved over 40 municipalities and hundreds ofwitnesses.”79 Often mass grave

74 Id., para. 65; see also Pocar, supra note 68, at 9, para. 28 (“The philosophy behind the Prosecution’s pleading practices is its obligation to victims.”).

75 See Raab, supra note 8, at 90. Indeed, the ICTY judges themselves have “decided to address the problem of lengthy and complex indictments” from a concern for the right of all accused to be tried without undue delay. Pocar, supra note 68, at 10.

76 Cf Federal Death Penalty Cases, supra note 30, pt. I OB(4)(b) (noting with reference to recently enacted death penalty laws that “newer statutes tend to produce more constitutional and interpretive issues than statutes that have already been the subject of extensive appellate and Supreme Court consideration”).

77 2001 REPORT, supra note 19, para. 30. 78 Id., Annex II, para. 29.

7 Expert Group Report, supra note 64, para. 133.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


exhumations are required, years after the crimes at issue took place. Investigation costs are included in the ICTY’s budget, although most such costs would normally form part of law enforcement agency budgets rather than court budgets in the United States.80 Moreover, long delays in making arrests make it difficult for the ICTY “to consolidate sev-

eral indictees into a single case, thereby saving time and resources, if they are transferred to The

Hague months or even years apart.””81 Consequently, the same witnesses may perforce be recalled on multiple occasions and considerable proportions of trials repeated, creating “enor-

mous operational inefficiencies.”82 Further, investigating crime scenes and conducting foren- sic exhumations in foreign countries are hard enough; when the governments involved hesitate

to cooperate and a substantial segment of the population is opposed to the work, investigations become even more difficult and costs increase as evidence decays or disappears and witnesses disperse or refuse to cooperate.83

Translation and Travel Costs

Because the ICTY is an international tribunal, it necessarily incurs expenses for translation and travel far in excess of those typically associated with national courts. Unlike most criminal

trials in the United States, in which translation costs and associated delays are typically small or nonexistent, all ICTY trials require the presentation of testimony and evidence in at least three

languages: French and English, the official languages of the court, and Bosnian-Croatian-Serbian, the language of the defendants. Even material that is not introduced at trial must often be translated

for review by the prosecutor and the accused. The Registry maintains a large staff of its own trans-

lators, as well as others on private contracts. In 2005 the annual cost for verbatim reporting during

trials alone was $3,029,000, not far off the $4,162,100 spent on judges.84 In 1999 the Language Section accounted for 10 percent of the Registry’s budget, which in turn amounted to almost 69

percent of the total ICTY budget.85 The costs of translation are often magnified when delays in get-

ting material translated hold up pretrial or trial proceedings. According to the Expert Group Report,

“translation is a bottleneck causing delay on a wide scale, including delay in the issuance ofTribunal

decisions after they have been agreed upon by a Trial or Appeals Chamber.”‘8

80 As an example, the United States deployed five hundred FBI agents to East Africa following the embassy bomb- ings there. According to then-FBI director Louis Freeh:

Over the course of the eight-weeks immediately following the bombings, the FBI:

* logged 61 flight missions, of which 16 were transoceanic;

* transported over 1,000 agents, technicians, examiners, and other employees deployed to and from East Africa; and

* moved 295 tons of equipment, supplies, and related items to support our investigative teams.

Statement for the Record of Louis J. Freeh, Director, FBI, Before the Senate Comm. on Appropriations, Subcomm. for the Departments of Commerce, Justice, and State, the Judiciary, and Related Agencies (Feb. 4, 1999), available at <http:ll// 999hr/990204-freehct2.htm>.

81 Raab, supra note 8, at 92.

82 Mark B. Harmon & Fergal Gaynor, ProsecutingMassive Crimes with Primitive Tools: Three Difficulties Encoun- tered by Prosecutors in International Criminal Proceedings, 2 J. INT’L CRIM. JUST. 403, 411 (2004).

83 Id

84 Pimentel, supra note 23.

85 Expert Group Report, supra note 64, para. 179. 86 Id., para. 37.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


Travel costs are also high as a result of the need for the prosecutor’s staff to carry out its work

in the former Yugoslavia. The ICTY’s 2001 budget, for example, allocated $2,432,500 for

travel,87 approximately 7 percent of the total budget for the Office of the Prosecutor.

Witness Costs

Witnesses whose testimony may be important often do not wish to cooperate with prose-

cutors or are afraid to do so because of ongoing ethnic tensions in the areas where they live. The

Tribunal must therefore spend large amounts each year on travel and support costs for wit-

nesses, their security, and often their relocation (including that of their families). In addition,

Registry officials must spend significant amounts of time negotiating with governments for

entry and exit permits and other travel documents, and work with host governments on pro- tection and relocation issues. For 2001, the Tribunal estimated witness relocation costs at

$790,000.88 The Tribunal estimated the costs for an estimated 670 witnesses to travel to The

Hague at $1,960,000.89


In considering whether the Tribunal has “provided value for money,” some thought must

be given to the quality of ICTY legal proceedings. The challenge facing the ICTY was enor- mous: to create a functioning, free-standing international tribunal from scratch, one that

would render high-caliber legal judgments while fully respecting the rights of the accused. At the outset, the Tribunal had “no seat, no courtroom, no prison, no budget, no computers,

no law clerks, no secretaries and no set rules governing the criminal procedure.”90 It had few

precedents to serve as models for its opinions. It had to blend a largely common law, adversarial

justice approach with elements of the civil law system, and to develop internal practices that

would enable judges from different backgrounds and legal traditions, speaking different lan-

guages, to work together.

For the most part, the ICTY seems to have succeeded in conducting credible trials while

simultaneously developing a large body of case law of generally high quality. Most observers

of the Tribunal credit it with producing a steady stream of precedents that will guide other

international tribunals for many years. In particular, the ICTY has generated a substantial body

of “[s]ubstantive international humanitarian law, international criminal law, and interna-

tional criminal procedure.””‘ As one commentator has observed:

In addition to the Judgments delivered by the Trial and Appeals Chambers, thousands of pre-trial, trial, and post-trial decisions have been rendered, ranging from contesting appointed counsel, to challenges to the form and substance of the indictment, to requests

87 2001 REPORT, supra note 19, para. 51, at 32. 88 Id., para. 54, at 39. 89 Id., para. 51, at 32.

90 Antonio Cassese, The ICT: A Living and Vital Reality, 2 J. INT’L CRIM. JUST. 585, 585 (2004). 91 Kelly D. Askin, Reflections on Some ofthe Most SignificantAchievements ofthe ICTY, 37 NEW ENG. L. REV.

903, 907 (2003).

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


by detainees for pre-trial release, to requests for protective measures for witnesses, to plea agreements.92

Although individual judgments have attracted criticism, and sometimes substantial criticism,

most observers would agree that the work of the Tribunal has generally been of high quality.

Moreover, the Tribunal is generally credited with striving to respect due process and the

rights of the accused. Judge Patricia Wald may overstate the case when she argues that “it would

be almost impossible to find an impartial commentator in 2003 in Europe, Africa, Asia or the United States who would contend that ICTY trials are not fair and in accordance with inter-

nationally accepted criminal-law standards,”93 but for the most part the Tribunal’s trials have

been accepted as fair and impartial. Some defendants have been acquitted, some convictions

reversed on appeal.

In part, the Tribunal’s reputation stems from the high caliber of its judges. Many have been

well respected national court judges or prominent legal academics. In part, the Tribunal’s rep-

utation stems from the high caliber of the Office of the Prosecutor, which has counted among

its members some prominent jurists and a fair number of attorneys seconded by the U.S. Department of Justice.

Of course, the prosecutor’s office necessarily learned by experience as it struggled to meld

prosecutors and staff from different legal traditions into a cohesive team. Inevitably, different

lawyers and investigators “brought with them dramatically different notions about how a wit-

ness statement should be written, what constitutes exculpatory material, whether and how a

witness should be prepared for cross-examination, and whether intent had to be proven by an

objective or a subjective standard.”‘9 The result has been “inconsistency and significant mis- understandings, both internally and externally.””9 At times, errors by the Office of the Pros-

ecutor have jeopardized the rights of the accused. In the Barayagwiza case, for example, the prosecutor was at least partly responsible for a prolonged period of pretrial detention in which

the accused was not notified of the charges against him or provided with a timely opportunity

to challenge his detention.96 In general, however, the Office of the Prosecutor appears to have

operated in accordance with a high standard of professionalism. If the Tribunal suffers from a substantial weakness when it comes to quality and due process,

it has been the variable, and sometimes very low, caliber of some of the attorneys serving as

defense counsel, particularly in the early years. At the outset, the prerequisites for selection as

a defense counsel were minimal. Many defense lawyers from the region had little or no expe-

rience with cross-examination and other features of the adversarial system, little knowledge of

92 Id.

3 Patricia M. Wald, ICTY]udicial Proceedings–An Appraisal firom Within, 2 J. INT’L CRIM. JUST. 466, 466 (2004).

94 Minna Schrag, Lessons LearnedJifrom ICTYExperience, 2 J. INT’L CRIM. JUST. 427,432 (2004) (footnote omit- ted).

95 Id.

96 In Barayagwiza v. Prosecutor, Decision, No. ICTR-97-19-AR72, paras. 101, 106 (Nov. 3, 1999), available at <>, the Tribunal terminated proceedings against Barayagwiza for abuse of process by the Office of the Prosecutor. Following a public outcry, however, that decision was reversed. Id. (Prosecutor’s Request for Review or Reconsideration) (Mar. 31, 2000).

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to


international humanitarian or criminal law, and little familiarity with the Tribunal’s proce-

dures.97 The result was “shockingly poor performances by some defense counsel,” the “nadir

. . [having been] reached in the Erdemovic case when the Appeals Chamber essentially found

that the attorney did not understand the concept of a guilty plea, much less explain it to his client.”98

The Tribunal has gradually improved its handling of assigned counsel, as it has moved to

streamline and strengthen its procedures generally. Moreover, problems in this area should not

be allowed to obscure the broader picture of a tribunal that is generally deemed fair and impar-

tial, at least by those outside the former Yugoslavia.

At the same time, as the Tribunal has struggled to expedite its proceedings and to implement

its completion strategy, it has had to balance its commitment to fair trials with competing con-

siderations, raising new questions. Judge Wald, for example, wonders whether the Tribunal’s

increasing reliance on witness statements rather than live testimony may jeopardize “the ICTY’s most precious asset-its reputation for fairness and truth seeking.””99 On balance, how-

ever, and viewed over the life of the Tribunal, the quality of its proceedings must be seen as high,

particularly in light of the challenges it has faced.


The perception that international criminal trials are costly and slow is accurate but mislead-

ing. On average, ICTY trials do cost much more than an average criminal trial in the United

States. But the reasons relate principally to the inherent complexity of the cases being tried, the

dependence of the Tribunal on international cooperation, and the costs implicit in its inter-

national nature (including translation and travel). Undoubtedly, structural issues and mismanagement have played a role in escalating costs

(this is especially true of the Rwanda Tribunal). In the past, for example, lack of courtroom

space led to lengthy trial delays and increased costs. Similarly, inadequate oversight of defense

counsel and broad-spectrum indictments have pushed up trial times and costs.

The ICTY has responded to its critics by adopting a series of measures to try to control costs

and reduce the length of trials. Greater use of plea bargaining in particular has enhanced the

Tribunal’s capacity to handle a greater volume of cases on its existing budget. But there is an

irreducible minimum that must be spent-well in excess of the amounts required for an aver-

age domestic criminal trial-if the quality of the proceedings is to be maintained. Few trials of the sort that come before the ICTY or that will come before the ICC can be handled, even

by the most expeditious court proceedings, in weeks rather than months. Accordingly, the

price of international justice will remain high. Whether we are getting value for money depends

on the extent to which one believes that the trials serve their purposes, which include not just

justice for victims, but larger societal goals, such as deterrence and national reconciliation.

97 See David Tolbert, The ICTYandDefense Counsel: A Troubled Relationship, 37 NEW ENG. L. REV. 975, 975 (2003).

98 Id. at 977, 979.

99 Wald, supra note 93, at 473.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to




In their masterful centennial essay, The Past and Future of the Claim of Preemptive Self- Defense (100 AJIL 525 (2006)), Michael Reisman and Andrea Armstrong observe that the claim of preemptive self-defense, particularly if applied unilaterally, poses a challenge to the international legal order. It may be instructive to note what Nuremberg judges had to say on that subject sixty years ago. The judgment of the International Military Tribunal of September 30, 1946, declared:

It was further argued that Germany alone could decide … whether preventive action was a necessity, and that in making her decision her judgment was conclusive. But, whether action taken under the claim of self-defense was in fact aggressive or defensive must ultimately be subject to investigation and adjudication if international law is ever to be enforced.

It was contended for the defendants that the attack upon the U.S.S.R. was justified because the Soviet Union was contemplating an attack upon Germany and making prep- arations to that end. It is impossible to believe that this view was ever honestly entertained.

… It was plain aggression.

The IMT judgment was reinforced in the dozen subsequent trials at Nuremberg headed by General Telford Taylor, who later became a professor at Columbia Law School. In the Ein- satzgruppen trial, after careful consideration of the contention that it was necessary to defend Germany against a presumed attack by the Soviet Union, the three U.S. judges rejected the arguments ofputative self-defense againstJews advanced by the defendants on the grounds that Jews were “known” to be sympathetic to Bolsheviks and had to be totally exterminated.

Justice Robert M. Jackson, chief architect of the Nuremberg trials and chief prosecutor for the United States before the International Military Tribunal, considered the condemnation of aggressive war as the greatest achievement of his life. Telford Taylor felt the same way. “We must never forget,” said Jackson in his opening statement on November 21, 1945, “that the record on which we judge these defendants today is the record on which history will judge us tomorrow.

TheJournalessay quotes the National Security Strategy ofthe United States announced by the White House in September 2002, which states: “[W]e will not hesitate to act alone, if nec- essary, to exercise our right of self-defense by acting preemptively against such terrorists … ” (100 AJIL at 530). A claim of legal entitlement by one nation to exercise preemptive self- defense, as your authors note, may encourage others to assert the same right. The world legal order is teetering on the brink of an abyss. International lawyers should try harder to define not merely terrorism, but also aggression and the proper functioning of the Security Council as envisaged by the UN Charter. It might reassure others if the United States would end its cam- paign against the new permanent International Criminal Court in The Hague, which is based on the Nuremberg principles and the search for a more humane and peaceful world under law.


* The writer was chief prosecutor in the Einsatzgruppen trial at Nuremberg.

This content downloaded from on Thu, 19 Jul 2018 02:21:26 UTC All use subject to

Comments are closed.