Skip to main content
The Economic Burden of the Death Penalty in Maryland
Testimony Before the
Maryland Commission on Capital Punishment

Kenneth R. Stanton, Ph.D.
Research Associate, Jacob France Institute,
Merrick School of Business, University of Baltimore
Associate Professor (Finance and Economics)
Coppin State University

September 22, 2008

Introduction

My name is Kenneth Stanton. I am a Research Associate at the Jacob France Institute within the Merrick School of Business, at the University of Baltimore, where I was also an Assistant Professor of Finance from 2001 until August of 2008. I recently joined the faculty at Coppin State University, in Baltimore, as an Associate Professor of Finance and Economics.

I have had the good fortune to have studied under some of the world’s leading economists and finance professors and I have taken ten or more statistics and econometrics courses, mostly at the graduate level. My mentors have included presidential advisors, a Nobel laureate1, and arguably two of the world’s top 5 econometricians2 under whom I first studied the techniques employed by the authors of the Urban Institute’s death penalty cost study that was presented in previous testimony. I have been teaching finance, statistics, and economics courses at the college level, including advanced econometrics, since 1998.

I have extensive econometrics experience in efficiency and cost estimation. As an undergraduate, I was called upon to do the computational work associated with computable general equilibrium models, for faculty members’ research that was submitted to several leading economics journals. As an MBA student, I conducted significant portions of research and cost estimates for the revision of the book, Government Regulation and Performance of the American Economy3, in which we estimated the costs of various US regulatory policies. In other research, I estimated the efficiency of loan portfolio managers in one of Canada’s largest banks, the labor costs of a state highway authority, in addition to conducting a wide array of estimates of the values of businesses, and the economic losses associated with accidents and other factors. In estimating the factors that drive the cost efficiency of loan portfolio managers, the econometric models have much in common with the Urban Institute Study’s approach since the data structure is quite similar. The state highway project had even greater similarities to the current task since the costs associated with every task had to be benchmarked through structured interviews and online surveys of the state employees involved. The additional challenge was that many employees serve more than one function, making it necessary to estimate what fractions of their time were allocated to each task. In spite to what appear to be daunting challenges, experience has shown that this type of activity based cost accounting is in fact quite reliable when done with care. In my opinion, the Urban Institute study demonstrates the kind of care that is necessary to make the estimates reliable.

As one of three editors of the book, Obesity, Business, and Public Policy, and coauthor of three of its chapters, I estimated the cost projections to the US economy associated with the obesity epidemic and analyzed the economic issues underlying the obesity epidemic and potential solutions.

I have served in an advisement capacity to all three levels of government4 and aided the passage of legislation in Maryland, Illinois, North Carolina, Texas, Virginia, and others. Since 2003, in relation to my research or expertise, I have been cited or quoted in more then 300 newspapers and magazines, in many cases multiple times and, I have appeared many times on national television networks, and local radio and television stations across the country.

In short, I am particularly well trained not only to evaluate complex econometric cost estimates, but to explain these estimates, their strengths, weaknesses, and significance, to people who may not have the familiarity that I have gained over the past two or three decades working on these issues. Most of my testimony will be a general explanation of the Urban Institute’s study, The Cost of the Death Penalty in Maryland (2008), as well as explaining why the study is both, important, and of very high quality. As a practical matter, and as a very brief summary of what follows, in practice, Maryland does not have a death penalty since so very few death notice cases end in execution. What Maryland does have however, is an economically unjustifiable expenditure on prosecuting death notice cases during a period when the State of Maryland is looking to slash the budgets of programs that could reduce the need for death notice cases in the first place.

Economics

I have been studying and teaching economics since the mid-1970s. Nearly all human behavior can be explained or understood through the lens of this discipline and it permeates all aspects of everyday living. In particular, justice is almost entirely an economic issue. The basic economic tool of cost versus benefit is at the heart of our justice system. We do not, for instance, seek $5,000 fines for a minor parking infraction, nor for that matter do we ever administer $20,000 fines for jaywalking. The laws appropriately weigh the potential costs of the infraction to society and set a penalty that reflects those costs.

When we reach more serious levels of offence, it becomes more apparent that we seek penalties that reflect not only the costs of the harm done, but following the economic model we include an additional penalty above and beyond the cost of the harm done, to serve as a deterrent to those who may be considering similar conduct. Again, even the deterrent effect is an economic issue. To make economic sense, the cost of the deterrent measure, must equal the expected benefit of deterring future wrongdoing. If not, then it is a waste of resources.

In the State of Maryland, we quite clearly do no not have unlimited funds, and we are now facing the hard choices of reducing or eliminating expenditures on programs and projects that may adversely affect our future. Everything from spending on new schools and drug addiction treatment programs, to spending on the construction of new courtrooms and additional police training, and expenditures on crime laboratories should be held up to scrutiny to ensure that we get the greatest possible benefit from every dollar spent. When funding is tight, the economically inefficient use of funds, or waste, becomes more intolerable. Again, the obvious economic analysis is to consider the total costs versus the total benefits of every undertaking. And, every single undertaking generates some cost since it employs resources that could be used for some other purpose. For instance, the American Bar Association requires that the accused in a capital case be represented by two lawyers, and the bifurcated structure of these cases also requires other additional resources that have real costs. These costs rightfully include the costs of any real estate, such as courtroom space, that eventually must be paid for by the taxpayers of Maryland.

Very obviously, money not spent on the construction or maintenance of any additional courtroom space could be redirected to the construction of additional classroom space, infrastructure improvements, or even programs that more directly intervene to reduce violence and other crimes such as training and salaries for crime laboratories. A common error in understanding is to assume that since the building is already there, that it is costless. Remember that surplus building space can be reallocated or sold. Backlogs in other court activities that may not even involve felonies, such as family court issues could also undoubtedly be reduced if the resources freed up from capital cases are redeployed.

The arithmetic is quite simple: If we have a process that requires significant additional resources for the prosecution of death notice cases that essentially never succeed, then it makes far more sense to direct those resources to better uses—including investments that are likely to prevent violence in the first place—than to waste them in nonproductive efforts.

Economists use the jargon of opportunity costs, to explain this concept. The interpretation of opportunity costs is that every additional resource, every additional lawyer, every additional clerk, and every additional square foot of courtroom space, has an alternative use that must be given up when those resources are used to try a death notice case. Economists can easily convert these resources into dollar costs making the points even simpler to understand. In the situation at hand, the dollar costs over the period of the study likely exceed $186 million.

Activity Based Cost Accounting

In studies that require estimates of costs, one of the most widely used approaches is activity based cost accounting, which is the approach employed by the Urban Institute Study of the costs of capital cases. This approach requires a careful assessment of all resources used, from labor, to real estate and other capital investments. Once a reasonable estimate of the quantity of each resource, lawyers, and clerks and so on, is obtained, then each of those quantities is multiplied by its respective unit costs. The problem at hand is a typical textbook application of this approach and the authors from the Urban Institute have skillfully and carefully employed it.

The authors have used high quality data sources and they have taken extraordinary steps to verify that those data sources were in fact accurate–even to the extent of auditing the databases that they used to verify accuracy. For time estimates, they started with reasonable benchmarks and subsequently interviewed people who make their living doing this work to verify and adjust where necessary the time required for each process. All of their verification procedures indicate to me that the estimates are reasonable, and add to my confidence that they understand the techniques of activity based accounting.

For costs that are spread over long time periods, it is necessary to adjust for the effects of inflation and for the time value of money. The inflation adjustment was done at reasonable inflation rates, and the authors correctly adjusted health care costs at a different rate since the inflation rate in health care expenditures is known to be higher5. The latter adjustment simply reflects that if we need to pay some cost at a future date, we can put aside a smaller amount of money today, and provided we can earn interest on it, it will be sufficient to cover the costs when needed at the future date. Again, based on the risk associated with state bond issues and current interest rate, the authors’ estimates appear reasonable.

The most obvious bias in the study arises from ignoring the additional pretrial costs associated with filing and later revoking death notices. They also ignore the incremental costs associated with not guilty verdicts in death notice cases and the incremental costs of appeals to US Supreme Court. All of these factors introduce a downward bias to the estimates of the additional costs associated with adjudicating capital cases. For these reasons, the costs estimates are in all likelihood higher than the Urban Institute study indicates. Potentially even a greater factor in introducing bias is that the data ends at 1999. It is extremely plausible that the cost differential has increased since that time, in large part due to the lowered tendency of capital cases to end in an actual execution.

The authors conducted suitable sensitivity analyses of all key assumptions. One of the assumptions that is often of issue, is the choice of discount rate in accounting for the time value of costs. This is of particular importance in comparing the prison costs. But, given the outcomes, very little of the cost differential is driven by prison costs and nearly the entire differential is the result of adjudication cost differences. Since the difference in time frames of adjudication is smaller, the discount rate is of less importance and the results combined with the sensitivity analyses, bear this out.

Econometric Analysis

As with most econometric studies, the researchers in this study face the problem of missing or incomplete records in the available data. If for example, all of the least egregious cases were missing from the data, then the difference between the cases where the death penalty was obtained and the cases, in which a death notice was never filed, would be grossly underestimated. The authors of the Urban Institute Study correctly adjusted for this missing data problem by assigning weights to the available complete data to obtain a more appropriate representation of the universe of death penalty eligible cases. Their adjustments follow standard accepted practices and their diagnostic tests verify that they have not introduced any bias in the process.

Since econometric analyses of this type can be complex, some explanation of the approaches used is probably beneficial. One of the more complex aspects is the adjustment for propensity. This particular variable captures the effects that are likely to be related with more egregious cases That is, those cases which based on the characteristics of the accused and the crime, are more likely to be more costly to adjudicate whether they are death notice cases or not. The authors capture these effects in using a reasonable model, which generates an indicator of the propensity of the case to be more egregious. When this variable is included in the estimation of the cost factors, any additional cost associated purely with the propensity, will be attributed to this factor, rather than to the variables that indicate whether the case involved a death notice, or death sentence.

As with most econometric studies, the authors employ a sizeable number of models in their efforts to estimate the costs. Their final choice of model–model number 7 in their study–is driven by appropriate selection criteria6 but, it is perhaps more important to note that the difference in cost estimates across all of the models is surprisingly small.

Table 1 summarizes the cost differentials. The base case includes all of the trials which were death penalty eligible, but in which the death penalty was not sought. This set includes not only life without parole cases, but also those which ended in virtual life without parole, which are cases that drew sentences so lengthy, that release from prison has a near zero probability. This column provides the base case, with adjudication costs of $250,000 and prison costs of $870,000 against which the other two sets are compared. It is notable that the prison costs are not as dramatically different as the adjudication costs. In particular, the death notice cases have adjudication costs nearly three times higher than the base case if the death penalty is unsuccessfully sought, and nearly seven times as high if the death penalty is actually delivered.

After evaluating the quality of the study and comparing the estimates to the higher quality studies done previously, it is my opinion that the $179 million in extra adjudication and imprisonment costs associated with the cases in which a death notice was filed, including those in which a death sentence was delivered, is in all likelihood a conservative estimate. Combined with the $7 million costs associated with the operation of the Capital Defense Division, and the downward biases of the incremental cost estimates discussed above, my view of the Urban Institute’s Study remains, that the total cost to Maryland taxpayers in all likelihood exceeds $186 million.

Summary

First and foremost, it is important to understand that the State of Maryland faces a tight budget and there is no doubt that prosecuting capital cases is a costly process using money that could be redirected to highly valuable uses. The quality of the Urban Institute Study is very high, and if it makes any error, it is biased toward underestimating the cost of capital cases. The $186 million costs estimate is in my view an understatement of the true costs to the State of Maryland. It is also evident that the most significant portion of additional costs occurs at the adjudication stage, not through imprisonment costs. Therefore, given that so few cases actually deliver a death penalty, and such an extremely small number actually result in execution, it makes no economic sense to spend all this additional money to seek the death penalty. The extra money going to lawyers and other court related costs would be far more efficiently spent in trying to prevent these cases from arising in the first place.

Table 1: Capital Eligible Cases

 

Death Notice

Cases

Average costs generated through:

Base cases:

Death penalty not sought

Death penalty unsuccessfully sought

Death penalty successfully sought

Adjudication

Prison costs

 

$250,000

$870,000

$850,000

$950,000

$1,700,000

$1,300,000

Total:

Costs in excess of no death notice base case:

Adjudication:

Prison:

Total increment:

$1.12 million

 

$1.8 million

$600,000

$ 80,000

$0.68 million

$3.0 million

$1.45 million

$0.43 million

$1.88 million

Number of cases:

Number of cases × incremental costs over base cases:

Sample size is 1,136 murder cases between 1978 and 1999 where the guilty verdict was delivered.

974

(1,136 – 162)

$0

106

$72 million

56

$105 million

1 Sir John Hicks.

2 Dr. Russell Davidson and Dr. James MacKinnon.

3 MacAvoy, Paul, Government Regulation and Performance of the American Economy, 2nd edition, Norton Books, 1992.

4 Most recently as Co-chair of the Baltimore City Council Task Force on the Prevention of Childhood Obesity.

5 I have taught graduate level courses in health care economics and these facts can be found in any textbook on the subject.

6 The standard test is the Akaike Information Criterion which balances the tradeoff between the number of variables and the reduction of entropy in the model.