I want to thank Chairman Sessions for requesting this written testimony on the U.S. Department of Justice's use of performance-based management, with a particular focus on their first required Performance Report, from the 1993 Government Performance and Results Act, to tell the taxpayers what their programs are accomplishing.

My name is Virginia Thomas. I am a Senior Fellow in Government Studies at The Heritage Foundation where my mission is to help the Congress and taxpayers focus on government accountability through the use of performance-based management techniques and effective congressional oversight. I must stress, however, that the views I express are entirely my own, and should not be construed as representing any official position of The Heritage Foundation. Although I am not a specialist in Justice programs, your invitation has given me the opportunity to review Justice Department reports on their activities concerning FY 1999, as well as related reports from the Congress, GAO and the Inspector General's office.

Your effort to focus your oversight authority on what is working and what is not working at Justice is a commendable one that has the potential to elevate policy debates about Justice's performance.

The presidential election elicits candidates who claim credit or tout new government spending for society's problems, when in truth, little is known about the effectiveness of existing federal programs. Americans should use this opportunity to demand credible information on such claims and promises. No federal program should continue or be spawned based on good intentions, political power, popularity or benign neglect. Washington has typically established programs and provided money for laudatory purposes, but rarely is there adequate follow through to judge whether the effort is achieving the desired results.

This paper is an effort to assess the Justice Department's effectiveness in achieving results for the American people based on their own reports that come from the 1993 Government Performance and Results Act.

PROMISE OF PERFORMANCE-BASED MANAGEMENT

Seven years ago, when Congress passed and President Clinton signed the Government Performance and Results Act (the Results Act), few realized the practical consequences that this law could have on our government. The Results Act requires that agencies prepare Annual Performance Reports that include:

A description of the agency's actual performance compared with the performance goals set in the latest annual Performance Plan for the applicable fiscal year (FY 1999 in this case);

An explanation of which goals were not met and why;

A description of what the agency will do in the future to achieve unmet goals; and

A description of actions to address performance goals that proved to be impractical or infeasible.¹

The first set of these reports were due this last spring.

Performance measurement need not be a mere paperwork exercise, or an exercise in gimmickry. Instead it should focus policy-makers on how the government can deliver higher quality, more effective assistance to more people at less cost.

Does Congress know which Justice Department programs are most and least effective? Improved performance-oriented congressional oversight may reveal that programs are duplicative and do not aid those for whom the programs were intended. And, most importantly, such oversight may well deliver real results for those for whom programs were originally created.

Consequences of Failing to Identify and Reform Ineffective Justice Programs

Americans don't care if the government is training more people, arresting more criminals, getting more "hits" on their web sites or even putting more cops on the street. The real measure of success is whether federal activities and taxpayer dollars are preventing crime, maintaining our prisons, securing our borders, protecting our civil rights, reducing illegal drug supply and use and maintaining civil order while honoring the rule of law and balancing our civil liberties.

Few agencies, including the Justice Department, have developed the necessary performance measures and reliable data sources to demonstrate the results they are achieving to benefit the public. Yet neither the Justice Department, nor America can afford to not ferret out wasteful, ineffective programs and better replicate programs that work based on their important function in our society.

The federal government spent approximately $26.2 billion in FY 1999 out of a total discretionary budget of $575 billion on administering justice, a function that involves at least 27 other departments and agencies², as well as the judicial branch.³ Over the past 10 years, the share of federal outlays for this governmental function has grown from less than 1% to almost 1.5% - which in a period of growing entitlement spending represents a large proportionate increase in taxpayer funding.

JUSTICE'S FIRST PERFORMANCE REPORT

Summarizing Justice's efforts to measure its performance, we see:

• The Department created 7 functions, 32 performance goals, and 175 performance measures to judge their own results.⁶

• There are 69 self-proclaimed successes out of 175; 39 failures and 59 instances where no one could verify the results due to data problems or changed measurements.

• Out of the 175 measures, only 17 are results-oriented. Another 143 are measures of activities and processes that are easy, yet inappropriate, for purposes of understanding the bottom line impact of federal efforts.

In reviewing the Department's goals, one might expect to see different measures than the ones used. For example:

• To reduce violent crime, Justice should measure each crime rate, particularly federal ones or violent crimes by number of incidents, and then also by arrests, prosecutions, convictions and % of sentencing served for crimes. Based on some reports on the outstanding warrants to be served that have been backlogged, Justice may want to consider measuring warrants served in this category as well. Another peculiar measure Justice uses in this goal is the measurement of reduction in La Cosa Nostra membership - a measure that does not pass common sense questions of how Justice would be able to know the population of LCN members.

• To reduce availability and use of drugs, Justice is comfortable using a variety of process-oriented measures. Instead, they should consider measures placed in statute for the Office of National Drug Control Program such as: reduced drug use from 6.1% to 3%, reduced adolescent drug use by a specific percentage, reduced drug-related emergency visits by a specific percentage and reduced juvenile or property-related drug crime.

• To reduce terrorism and espionage, Justice needs to measure reduced incidents of espionage. Terrorism acts that do occur would be a preferred measure over those that do not occur. One would wonder whether the data that supports these incidents are valid.

• To improve white collar crime or other civil law enforcement areas, measurements should be based on success of government litigation and enforcement efforts in the state and federal courts.

• Management goals should be consolidated, as they appear sporadically throughout Justice's report.

• Justice needs to set measures carefully to balance the use of intrusive technology with proper respect of privacy. Wiretapping and similar invasive investigative techniques should be judged based on whether or not this evidence was used in a successful prosecution within a reasonable timeframe, such as one year.

• Grants are proliferating out of Justice and they should be judged with strict accountability for the impact or benefit the public or the government is able to receive from each grant. Currently, there is little or no obvious reporting of results on grants in the Department's Performance Report.

• Customer satisfaction measures or surveys should be used in appropriate functions, not, as appears now to be the case, with prisoners who are assessed on whether they found their transportation acceptable, but rather with the victims of crime who were assisted or should have been assisted with timely, quality service from public officials.

• Programs focusing on missing children or violent crime against women should use common sense measures such as reduced incidences of missing children or reduced incidences of reported cases of violence against women. Certainly, the Justice Department should not be, as it appears to be, using performance management to force a potentially counter-productive law enforcement technique on local officials, such as # of grantees to implement a mandatory arrest policy in cases of violence against women.

• In measuring the anti-trust impact of Justice's actions, a better set of measures would include whether consumers now have more choices, with higher quality and lower prices than they did prior to Justice's intervention/actions?

• In Justice's report, there is no evidence of Justice cooperating with other federal offices to ensure successful governmental performance. Is Justice effectively sharing responsibility for various goals with other departments, such as IRS (for fair enforcement of tax laws) or Labor (for curtailing union corruption) or other Departments?

• Contrary to the concern raised by Attorney General Reno in early sections of the performance reports about the perverse incentives that measures could raise so as to establish bounty hunting in the law enforcement arena, Justice's plan includes several such measures. These include measurements that count collected fines, # of crimes investigated, amount of assets seized or # of prisoners moved.

• Prisoner maintenance is measured with poor and benign measurements. These could be replaced with measures such as: % of prisoners serving their complete sentence; # of escapes (which is currently used); adequacy of facilities and adequate maintenance of existing facilities.

• Inmate service programs such as drug abuse, education, job training should be measured with techniques that verify that there has been a lasting impact from any federal assistance. These impacts should then be transparent to see which programs are working the best.

• Immigration should measure the speed and accuracy by which citizenship is determined and assistance is given. Criminal aliens need to be removed expeditiously and measured in such a manner. INS could also better measure the prevalence or reduction of illegal aliens inside our borders by hospital records, school records and employment-related measures.

Some of the best and worst measures used at Justice are highlighted below:

If Congress has given the Justice Department more than it can humanly do, the public needs to recognize this with credible, reliable, objective and transparent data on what is working and what is not working at Justice.

In a report to the Senate Governmental Affairs committee, the General Accounting Office (GAO)said:

"Overall, DOJ's progress in achieving desirable program outcomes cannot be readily determined since the agency has yet to develop performance goals and measures that can objectively capture and describe performance results. DOJ's performance measures are (1) more output than outcome oriented, (2) do not capture all aspects of performance, and/or (3) have no stated performance targets."⁷

The Mercatus Center at George Mason University ranked the Justice Department Performance Report 21^st of the 24 largest federal departments and agencies with a grade of 23 out of 60 points. Specifically, they said of Justice's report:

"Tedious reading with lots of insufficiently explained acronyms and tables…The report is inwardly focused - not on Americans safety, freedom and access to dispute resolution, but on the `diverse activities and major accomplishments of' the DOJ"⁸

Justice should be able to demonstrate progress, but doesn't, in resolving the Department's most serious internal management problems, as documented by its own Inspector General.

Longstanding Problems at Justice

The first Performance Report should have demonstrated progress in solving many of the longstanding problems identified by Justice's Inspector General or the General Accounting Office, yet, there was little if any mention of these:

• INS's organizational structure impedes its ability to both enforce immigration laws and provide services for immigration and citizenship.

• Justice's INS cannot provide up-to-date information on immigration law.

• INS can't pass an audit.

• Vulnerabilities in INS' naturalization procedures and practices, coupled with a higher volume of naturalization applications, have increased the risk that U.S. citizenship will be granted to ineligible applicants. INS has allowed criminal aliens to be released into communities, and have increased the risk of granting citizenship to ineligible applicants.

• Justice has significant financial management weaknesses. They are unable to track a new flood of federal funding into Justice, such as the $8.8 billion grant program for state and local law enforcement agencies to hire or re-deploy 100,000 additional officers to perform community policing.

• Justice's DEA has such internal weaknesses as to allow embezzlements.

• Justice's Asset Forfeiture program, as well as a similar Treasury program, operate asset forfeiture (such as illegal drugs, property and cash) programs which have inventories valued at several billion dollars. The programs are vulnerable to theft and misappropriation due to internal control weaknesses.

• Justice's computer systems have been poorly planned, wasted time and money in delayed implementation and were unable to provide timely, useful and reliable data, as well as put vast amounts of sensitive data at risk of unauthorized disclosure.

• Justice now houses 118,000 inmates in Bureau of Prisons facilities which have a capacity of only 89,675.

• Detention space and infrastructure for criminals and illegal migrants are increasing problems. The INS and USMS are experiencing rapid growth in the use of detention space, from an average of 31,966 beds in 1996 to a projected 55,000 to 67,000 beds in 2001.

• The INS program to deport illegal aliens is largely ineffective. Only a small percentage of non-detained illegal aliens with final deportation orders are actually deported. The INS fails to identify many deportable criminal aliens, including aggravated felons nor do they initiate Institutional Hearing Program (IHP) proceedings for them before they are released from prison. Ineligible aliens, including convicted felons, are inappropriately granted voluntary departure.

• Justice faces significant difficulty attracting and retaining qualified ADP auditors, Information Technology professionals, and Financial Statement audit professionals.

Financial mismanagement is bad enough, but government ineptness has resulted in unnecessary deaths. On June 2, 1999, New Mexico Border Patrol agents had a man, known as the railway murderer, in custody who was wanted for several murders. After being released back into Mexico, he is suspected to have murdered at least four more people, including Dr. Claudia Benton in Houston, Texas. Despite being wanted by the FBI⁹, Texas Rangers and the Houston police, INS agents let him go simply because of their lack of knowledge on how to use a database that includes fingerprints, photographs and criminal background information for those caught entering the country illegally. Attorney General Janet Reno told Congress that 37% of the 35,000 criminal aliens released over the last five years from INS custody went on to commit additional crimes.¹⁰

Clinton Performance Promises

President Clinton, in his Government Wide Performance Plans made a variety of related promises to improve government management and performance in the following ways:

Government wide Performance plan: 
President Clinton's promises 1999 and 2000

1999 Government Wide Performance Plan included the following priority performance goals:

• Justice will maintain commitment to reduce the incidence of violent crime below the 1996 level of 634 offenses per 100,000.

• Justice will provide funding for communities to hire and deploy 16,000 more officers in 1999.

• Justice will reduce specific organized crime and its influence on unions and industries from the 1997 level, while intensifying efforts to prevent emerging organized crime enterprises.

• Justice will ensure no judge or witness is the victim of an assault.

• The Marshals Service will apprehend 80% of violent offenders within 1 year of a warrant's issuance, and will reduce fugitive backlog from 1998 by 5%.

• Reduce the availability and abuse of illegal drugs.

• Reduce the average time between application and naturalization of qualified candidates from 24 months to 6-10 months.

• Increase the removals of illegal aliens from 111,794 in 1997 to 134,900 in 1999.

• Identify over 38,500 unauthorized workers.

• Ensure convicted violent offenders serve at least 85% of their sentences.

2000 Government Wide Performance Plan included the following priority performance goals:

• Fund an additional 50,000 officers by 2005.

• Provide law enforcement with latest technology for fighting and solving crime.

• Fund new prosecutors.

• Largest gun enforcement initiative in history to improve speed and accuracy of Brady background checks, add an additional 500 Bureau of Alcohol, Tobacco and Firearms agents, expand crime gun tracing and ballistic imaging systems, hire 1,000 gun prosecutors, fund smart gun technology research, fund over 100 prosecutors and 20 enforcement coordination teams, fund media campaigns on violence and gun safety, and get local law enforcement to end the reselling of used and seized firearms.

• Establish the first national ballistics network to trace bullets and shell casings.

• Lower recidivism rates

• Promote responsible fatherhood for the nearly 500,000 inmates who leave prison or jail this year.

• Combat gender-based crime by strengthening the justice system's response, support and enhance services for victims, fund a hotline and battered women shelters and expand outreach to under-served populations.

• Reduce the rising Indian violent crime rate by increasing the number of law enforcement in Indian country, improve the quality of the criminal justice system, enhance substance abuse programs and combat youth crime.

• Reimburse state and local governments for the cost of incarcerating criminal illegal aliens.

• Combat money laundering and financial crime through interagency efforts and use of technology.

• Combat terrorism with $11 billion.

• Reimburse telecommunications manufacturers and carriers for retrofitting in conversion to digital to ensure wiretapping is effective in future.

• Reduce overcrowding in the prisons.

• Increase prisoner literacy.

• Complete 1.3 million applications for citizenship and reduce processing time to 6 months by the end of 2000. (Backlog was over 1.8 million cases in 1999.)

• Address application backlog in other immigration service programs.

• Fund new Border Patrol agents so that there will be 9,800 agents on north and south borders, compared to 3,965 in 1993.

• Add new state of the art technology - night scopes, ground sensors, fencing, lighting, new or improved road construction - to enhance border patrol enforcement.

• Add 1,038 detention beds to 20,000 capacity. [Removed 178,168 aliens, including 62,838 criminal aliens]

• Help fight drug problem with drug treatment, prevention, domestic law enforcement, supply reduction efforts, including Columbia assistance.

• Implement or expand tough drug testing, treatment and graduated sanctions for prisoners, parolees, probationers, and ex-offenders reentering society.

OFFICE OF JUSTICE PROGRAMS

Office of Justice Programs has 849 employees (up from 334 in FY 1993) along with 110 contractors (up from 36 in FY 1993) working in 5 Bureaus, 7 Program offices and 7 Support offices - all of which were created since 1984. For fiscal year 1999, the Office of Justice Programs awarded about 4044 grants, providing $4.1 billion in FY 1999, up from 564 grants in FY 1993 totaling $800 million.¹¹

OJP receives the fifth largest segment of funding for Justice activities (14.1%), following INS, Office, Boards and Divisions, FBI, and the Bureau of Prisons.¹² Its mission is to prevent and control crime, improve the criminal and juvenile justice systems, increase knowledge about crime and related issues and assist crime victims.¹³ Funding for OJP has grown from $1.1 billion in FY 1995 to approximately $4 billion in FY 1999.¹⁴

In FY 1999, OJP received:

• $3.4 billion in direct funding
• $324 million in prior year collections of Crime Victims Fund
• over $405 million from reimbursable agreements.¹⁵

COPS

The Congress authorized $8.8 billion over 6 years for grants to local law enforcement agencies to add 100,000 new police officers to the streets in an effort to reduce crime. In fact, despite claims to the contrary, the number of officers funded by Washington is more like 40,000, a figure similar to the growth in numbers of police officers had there never been a new program. In addition, the police newly hired were largely in low-crime areas, giving them little ability to impact national crime rates. In fact, although national crime rates have decreased, there is no evidence of a correlation between grants awarded in high crime areas to reduced crime or even to more police officers on the street.¹⁶

Drug Court program

This program aims to improve public safety, reduce prison and jail overcrowding and reduce criminal recidivism through the use of intensively supervised drug treatment for drug addicted, non-violent offenders. Since 1989, an estimated 140,000 drug offenders have entered drug court programs.¹⁷ There are over 250 Drug Court Program Office federally funded drug courts, 60% of the total 415 operational drug courts nationwide. In FY 99, they received $40 million for these purposes. ¹⁸

Success in the program appears to be assessed with varying standards. Many drug court programs vary in their approach, population served, treatment techniques¹⁹, time at which participation is made available, program costs and completion and retention rates²⁰. Although local flexibility is admirable, focus on results and outcomes should be made consistent so that programs can be critiqued based on their ability to achieve the clear objectives established in the programs' origination - reduced criminal behavior and reduced drug use.

• Success is touted by the Drug Court Clearinghouse at American University in its claims that 80% of drug court program participants do not commit crimes while enrolled in the program. Yet, the Drug Courts had 85% of their participants in FY 1998 not commit crime while participating in the program. In a peculiar public policy exercise, the Office decided to reduce the goal for FY 1999 to 80%, rather than increase it beyond that achieved the previous year.

A GAO report issued in July, 1997 on Drug Courts to the Senate Judiciary Committee indicated that as of that date, 40% of participants were enrolled in the program, 31% of those surveyed had completed the program and 24% had failed to complete the program because they withdrew, died or were terminated from the program. In addition, an average of 48% completed the program. ²¹ Thus, with the Drug Court performance measure, we don't know if within the "80% of participants who do not commit crime while in the program" these are individuals who are terminated, who may die or who may be otherwise unsuccessful in the program.

• Surveyed graduates of one drug court program who were evaluated stated that the most important aspects of the program to them were: (1) close judicial monitoring, (2) staff support, (3) urine tests, (4) sanctions and (5) the opportunity to have their charges dismissed. Yet not all programs emphasize these same components.

Whereas federal efforts should allow local communities flexibility:

(1) The programs' goals and measures should relate to breaking the cycle of substance abuse and crime for those who participate in Drug Courts. Thus, this program is ripe for longitudinal evaluative work - monitoring the long term progress of participants' drug and crime-related behavior who have passed through the program against those who were never in the program in the first place. Measures should focus on:

• reducing drug dependency and use
• reducing recidivism in criminal behavior, and
• graduation and retention rates

(2) Program officials should be setting the threshold for performance higher than the previous year, not engaging in low-balling measures for the potential good publicity of meeting goals or other possible reasons.

(3) Serious performance-management would never rely on self-reported information for evaluation; but rather would ensure that credible, objective data is collected about the performance of a tax-funded program.

CONCLUSION

Could we, with better planning, better goals and measurements, and more honest data, get less crime, better law enforcement, and better Justice-related results than we are getting today? Crime should be going down; costs to administer justice should be reducing; technology should be applied fairly, prosecutions should be supported by convictions, victims of crime should believe they have an ally in government, legal immigrants should be processed swiftly, and borders should be secure.

According to the Department of Justice's last Strategic Plan, one of its eight core values is:

Accountability to the Taxpayer. We are committed to serving as effective and responsible stewards of the taxpayers dollars that are entrusted to us. We are efficient and results oriented. We measure and report on our progress in achieving goals.²²

Justice's first report on its performance with taxpayer dollars, although a work in progress, is a poor exhibit of this stated core value. We can only hope that future Performance Reports will live up to the expectations that Congress, in its wisdom of passing the Government Performance and Results Act, had.

Government should improve on the delivery of public services for purposes clear in the core value stated above. Yet, equally obvious is that Justice's funding is subject to increasing fiscal pressures. The discretionary budget - an increasingly pressurized portion of total federal spending - is being squeezed. In examining federal budget projections, entitlements will continue to take a growing percentage of total federal spending (66.6% projected in 2005), with defense spending decreasing (15.6% in 2005), interest payments on our national debt fluctuating (7.7% in 2005) and the discretionary budget is being squeezed from 21.3% of the federal budget in 1980 to only 10.1% in 2005. Congress will have no choice in the near future but to scrutinize Justice's budget based on these common sense performance concepts.

There is no doubt that political agendas, personnel and funding greatly influence the creation of a mission, measures and results of any administration, yet this analysis discusses how the Justice Department could better measure their performance and results.

To improve upon this initial effort, Congress and the Administration should:

1. Reduce the number of measures at Justice, consolidating them and making them more outcome-oriented and using greater common sense. Having too many measures is the equivalent of having no measures or priorities at all. Excessive measuring, especially in grants or delegated to state and local officials could end up causing a reign of paperwork terror - a reign that would have counterproductive impact on the ability of people to get results.

2. Improve the credibility of Justice's data for policy-makers. Truth matters.

3. Coordinate with other federal offices who are also focused on the same goals and outcomes and use similar or the same measures wherever possible. Policy-makers should know which tools of governance work best: tax incentives, grants, loans, educational campaigns, training or subsidies.

4. Ask Justice managers questions such as:

• Have you developed the right measures to assess a program's success?

• Which of these are achieving or failing to achieve their objectives?

• Can you specify why each of your programs were initially established - what they were designed to address or solve?

• For each program, can you specify the magnitude of the problem, the seriousness of the problem and the numbers of the population affected who are to be impacted by your program?

• For each program, do you know what other public or private entities are attempting to solve the same or similar problem, if they are effective and what you are doing to coordinate with these other entities if appropriate?

• For each program, can you demonstrate whether or not it has been proven to be effective in achieving its results? If not, can you give us a schedule by which such methodologically sound evaluations will be conducted to ensure that we are not wasting taxpayer dollars on non-productive or even counter productive programs?

• For each program proven to be effective by an independent, credible authority, can you demonstrate that it is using the most cost efficient techniques to accomplish its goals?

• Can you describe the success, in human terms, of your programs?

Public office holders or candidates should refrain from asking the federal government to do more than it can possibly, realistically achieve. With tenacious, persistent, results-oriented congressional oversight, government can be reformed, downsized, streamlined and improved based on credible, objective information about program results.

Such oversight may expose the fact that Congress and past Administrations have layered program upon program, earmarking initiatives that don't add value as much as they help a political constituency of one sort or another. But such a focus would bring a new level of accountability and understanding to the American public that may stop candidates from over-promising and further waste.

Thank you for requesting this preliminary research and I look forward to working to assist you and your oversight of the Justice Department in additional ways.

Virginia L. Thomas is Senior Fellow, Government Studies at The Heritage Foundation.