Inequality paper

LEGAL DISPARITIES IN THE CAPITAL OF CAPITAL PUNISHMENT Phillips, Scott . Journal of Criminal Law & Criminology ; Chicago  Vol. 99, Iss. 3,  (Summer 2009): 717-

755.

ProQuest document link ABSTRACT  

Death penalty opponents charge that wealthy defendants who can hire legal counsel are exempt from capital

punishment, but that indigent defendants who receive court-appointed counsel are frequently condemned to

death. The critique is based on sensational stories, but anecdotes cannot establish a causal relationship. To

explore the issue systematically, the current research examines the impact of legal counsel on the District

Attorney's decisions to seek the death penalty and juries' decisions to impose death sentences against adult

defendants indicted for capital murder in Harris County (Houston), Texas from 1992 to 1999 (n=504). Harris

County is the largest jurisdiction in the nation to use the appointment method rather than the public defender

method to deliver indigent capital defense, though by no means the only such jurisdiction. The empirical

comparison of hired counsel to appointed counsel in Harris County reveals three central findings: (1) Defendants

who hired counsel for the entire case were never sentenced to death; (2) Defendants who hired counsel for a

portion of the case were substantially less likely to be sentenced to death; (3) Hiring counsel is not the province of

the wealthy, as almost all of the capital murder defendants in this study were poor. Though not the focus of the

research and a finding that must be considered tentative, the data also reveal that defendants who hired counsel

for the entire case were much more likely to be acquitted. To be clear, the findings are not an indictment of

appointed attorneys, but rather an indictment of the structural deficiencies inherent in the appointment method.

The research concludes with a call for Harris County-the capital of capital punishment-to establish a Public

Defender Office with a specific Capital Defender Office. Though not a panacea, the public defender method comes

much closer to the adversarial ideal of evenly matched partisans doing battle to produce justice. [PUBLICATION

ABSTRACT] FULL TEXT  

Headnote

Death penalty opponents charge that wealthy defendants who can hire legal counsel are exempt from capital

punishment, but that indigent defendants who receive court-appointed counsel are frequently condemned to

death. The critique is based on sensational stories, but anecdotes cannot establish a causal relationship. To

explore the issue systematically, the current research examines the impact of legal counsel on the District

Attorney's decisions to seek the death penalty and juries' decisions to impose death sentences against adult

defendants indicted for capital murder in Harris County (Houston), Texas from 1992 to 1999 (n=504). Harris

County is the largest jurisdiction in the nation to use the appointment method rather than the public defender

method to deliver indigent capital defense, though by no means the only such jurisdiction. The empirical

comparison of hired counsel to appointed counsel in Harris County reveals three central findings: (1) Defendants

who hired counsel for the entire case were never sentenced to death; (2) Defendants who hired counsel for a

portion of the case were substantially less likely to be sentenced to death; (3) Hiring counsel is not the province of

the wealthy, as almost all of the capital murder defendants in this study were poor. Though not the focus of the

PDF GENERATED BY SEARCH.PROQUEST.COM Page 1 of 22

research and a finding that must be considered tentative, the data also reveal that defendants who hired counsel

for the entire case were much more likely to be acquitted. To be clear, the findings are not an indictment of

appointed attorneys, but rather an indictment of the structural deficiencies inherent in the appointment method.

The research concludes with a call for Harris County-the capital of capital punishment-to establish a Public

Defender Office with a specific Capital Defender Office. Though not a panacea, the public defender method comes

much closer to the adversarial ideal of evenly matched partisans doing battle to produce justice.

I. LEGAL DISPARITIES IN THE CAPITAL OF CAPITAL PUNISHMENT

Death penalty opponents charge that wealthy defendants who can hire legal counsel are exempt from capital

punishment. Former Supreme Court Justice William Douglas, for example, noted: "One searches our chronicles in

vain for the execution of any member of the affluent strata of this society."1 Justice Douglas also compared

American capital punishment to ancient Hindu law, suggesting that wealthy defendants who can hire legal counsel

are immune from execution in practice, just as Brahmans were immune from execution by law.2 Noted abolitionist

Sister Helen Prejean has argued succinctly, "[R]ich people never go to death row."3

Death penalty opponents also charge that indigent defendants who receive court-appointed counsel are frequently

condemned to death. Sister Prejean commented that "capital punishment means them without the capital get the

punishment."4 Indeed, anecdotes of inept appointed counsel abound: counsel have been caught drunk in court,

asleep in court, and at times offer no defense during the punishment phase of a capital trial.5 David Dow

summarized the issue: "[Rjace matters in the death penalty system, but socioeconomic status matters even more.

Wealth matters because in many cases trial outcomes depend less on what really happened than on an advocate's

skill."6

The heart of the critique is a question of arbitrariness. The term arbitrary has two meanings: death sentences are

imposed randomly or death sentences are influenced by legally irrelevant factors.7 The latter interpretation can be

illustrated through an anecdote that Stephen Bright, the President of the Southern Center for Human Rights,

recounts: "A member of the Georgia Board of Pardons and Paroles has said that if the files of 100 cases punished

by death and 100 punished by life were shuffled, it would be impossible to sort them out by sentence based upon

information in the files about the crime and the offender."8 Put differently, death penalty opponents argue that the

relevant legal facts of a capital case cannot fully explain whether a defendant lives or dies, as irrelevant facts such

as having the resources to hire counsel also matter.

Despite the seriousness of the critique and the potential implications for equal justice, social science research on

the relationship between legal counsel and capital punishment is limited. The few available studies, based on data

from the 1970s, provide mixed results. Drawing on a sample of defendants convicted of first degree murder in

Florida from 1973 to 1977, Bowers reported that indigent defendants represented by appointed counsel were more

likely to be sentenced to death than defendants represented by hired counsel.9 Yet Foley examined all defendants

indicted for first degree murder in twenty-one Florida counties from 1972 to 1978 and reached the opposite

conclusion: the defendant's form of legal counsel was unrelated to the imposition of the death penalty.10 Moving

from Florida to Georgia, Baldus and his colleagues concluded that defendants with court-appointed counsel were

more likely than those with hired counsel to receive a death sentence, though the authors did not consider the

issue in depth.11 In a subsequent re-analysis of the Baldus data, Beck and Shumsky examined case outcomes for

all 476 defendants convicted of capital murder and deatheligible in Georgia from 1973 to 1978, and reported that

the adjusted odds of a death sentence were 1.87 times higher for defendants with courtappointed counsel

compared to defendants with hired counsel.12

This Article draws on modern data to evaluate death penalty opponents' two main premises: (1) Defendants who

can hire counsel are less likely to be sentenced to death; and (2) Only the wealthy can afford to hire counsel. To

explore such issues empirically, the research focuses on the population of adult defendants indicted for capital

murder in Harris County, Texas from 1992-1999 (n=504).

Harris County - home to Houston and surrounding areas - is an interesting place to conduct the research. Harris

County is the largest jurisdiction in the nation to use the appointment method rather than the public defender

PDF GENERATED BY SEARCH.PROQUEST.COM Page 2 of 22

method to deliver indigent capital defense,13 though by no means the only such jurisdiction. Under the

appointment method, the judge appoints a member of the private bar to represent an indigent defendant. In

contrast, public defender offices are comprised of a salaried staff of government attorneys who handle indigent

cases - the bureaucratic parallel to the District Attorney's office.

Harris County is also arguably the capital of capital punishment. With 106 executions in the modern era, defined as

the Supreme Court's reinstatement of capital punishment in 1976 to the present, Harris County has often captured

the national and international spotlight in the death penalty debate.14 As Table 1 demonstrates, if Harris County

were a state, it would rank second in executions after Texas. In fact, Harris County has executed as many

offenders as all of the other major urban counties in Texas, combined. The period from 1992 to 1999 is also critical

because the number of death sentences in Harris County climbed to historic highs. From 1976 to 1991, Texas's

death row received an average of six offenders per year from Harris County. But from 1992 to 1999, the average

almost doubled to eleven offenders per year. The average then dropped to five offenders per year from 2000 to

2007. 15

Harris County officials reject death penalty opponents' arguments. District Judge Doug Shaver, in an interview with

the Texas Lawyer, noted that appointed counsel might be superior to hired counsel: "From where I sit, the

appointed attorneys may even be better than the paid attorneys ___ "16 Indeed, District Judge Michael McSpadden

suggested in an interview with the Houston Chronicle that appointed attorneys are definitely superior to hired

attorneys: "If you are charged with a criminal offense in Harris County, you would be much better off in our court,

and many of the other courts, with a court-appointed rather than a retained attorney."17 Offering a different

response, John Holmes, the Harris County District Attorney from 1980 to 2000, contended in an interview with the

Houston Chronicle that appointed and hired counsel are equally ineffective in capital cases: "I don't think it makes

a hill of beans what kind of lawyer you are on these cases. These crimes are so horrible Clarence Darrow's not

going to help these guys."18 In the absence of sustained research, the relationship between legal counsel and

capital punishment remains an open question.

The current research findings both support and refute death penalty opponents' arguments. Defendants who hired

counsel for the entire case were never sentenced to death. Even defendants who hired counsel for a mere portion

of the case were substantially less likely to be sentenced to death. Yet hiring counsel does not appear to be the

province of the wealthy. Instead, virtually all capital defendants seem to be poor. Poor defendants do occasionally

hire legal counsel, presumably because friends and relatives pool resources in the hour of need. Such patterns

provide strong evidence of arbitrariness: even after using multivariate statistical techniques to control for other

factors that might influence the outcome of a case, the defendant's ability to hire counsel transforms the legal

landscape. To be clear, the findings are not an indictment of appointed attorneys; the appointment system is

flawed, not the individuals who work within the system.

Though not the central focus of this Article, the data also reveal that defendants who hired counsel for the entire

case were much more likely to be acquitted. Because the number of acquittals is very small, the relationship must

be considered provisional. Nonetheless, the relationship is strong. Moreover, the potential implications for

erroneous acquittal and wrongful conviction suggest that the relationship demands attention.

Some might wonder whether the research has any relevance beyond Houston. To answer the question, consider

the following: 252 of the 254 counties in Texas use the appointment method of indigent defense,19 and 426 of the

1,141 executions in the modern era have occurred in Texas.20 Understanding capital punishment in America

requires a close examination of Texas, and understanding capital punishment in Texas requires a close

examination of the appointment method of indigent capital defense.

This Article is organized as follows: Part II describes the appointment method of indigent capital defense in Texas,

including an historical overview and a review of existing critiques; Part III describes the research protocol,

including the population of cases, measurement strategies, and statistical techniques; Part IV reports the bivariate

and multivariate findings; Part V situates the findings within a broader discussion of arbitrariness and argues for

the creation of a Public Defender Office and a specific Capital Defender Office in Harris County.

PDF GENERATED BY SEARCH.PROQUEST.COM Page 3 of 22

II. INDIGENT DEFENSE IN TEXAS

A. DESCRIPTION OF APPOINTMENT METHOD: THE CURRENT RESEARCH AS A CONSERVATIVE TEST

To understand the relationship between legal counsel and capital punishment, it is important to describe how

indigent defense operates in Texas generally, and in Harris County specifically. The State of Texas does not fund or

administer indigent defense. Instead, each county is responsible for developing a method to provide counsel to the

poor. In the arena of capital punishment, two methods have evolved: appointed counsel and public defender

offices.21 Of the 254 Texas counties, 252 use the appointment method.22

The standards for being appointed to a capital case have changed over the years.23 Prior to 1991, state law was

silent regarding standards - Texas judges could appoint any member of the bar to represent an indigent capital

defendant. Most judges appointed members of the bar who were criminal defense experts. But others were more

cavalier, appointing friends who had no experience in the area, such as real-estate specialists or local state

legislators.

In 1991, Judge Jay W. Burnett spearheaded a capital certification program in Harris County designed to strengthen

appointment standards. To be eligible for the program, a defense attorney had to be licensed in Texas for a

minimum of five years, devote at least 50% of his or her practice to criminal law, and have tried to verdict either five

or more first degree felony cases or one or more capital cases. Those who met the eligibility requirements and

were interested in being appointed to capital cases had to enroll in a class on capital litigation that culminated in a

100question, multiple-choice exam. Defense attorneys who failed the exam could re-enroll in the class and re-take

the exam up to two more times. In addition to passing the exam, defense attorneys were required to complete

twenty hours of Continuing Legal Education (CLE) on capital litigation each year. To enforce the capital

certification program, the Texas Supreme Court passed an order stating that the Harris County Auditor could not

pay a defense attorney in a capital case unless that attorney had completed the program. The implementation of

the capital certification program gave Harris County the most rigorous standards for appointment to a capital case

in the State of Texas.24

Judge Burnett's efforts to strengthen appointment standards in Harris County soon spread across the State of

Texas. The Presiding Judge in the Second Judicial Region of Texas asked Judge Burnett to expand the capital

certification program to the remaining counties in the region.25 Later, the Texas legislature passed the 2001 Fair

Defense Act (FDA) modeled after the reforms made in the Second Judicial Region.26 The FDA set minimum

standards for appointment to a capital case; all judicial regions and constituent counties were required to develop

a plan to meet the promulgated standards.27

A 2003 report by the Equal Justice Center and Texas Defender Services concluded that the Second Judicial Region

remains the leader in reforming indigent capital defense in Texas, noting: "Region 2's efforts to apply the FDA are

the most commendable among all of the regions. Region 2 has provided meaningful interpretation of most FDA

requirements, consistent with legislative intent."28 The report also provides an in-depth examination of the most

active death jurisdictions in Texas, defined as the thirty-three counties that sentenced at least five defendants to

death between 1995 and 2003.29 Of the counties reviewed, Harris County has the most stringent standards for

appointment to a capital case.30 The exact standards for appointment to a capital case in Harris County after the

passage of the FDA are enumerated in Table 2. The objective criteria are as follows:

* General Criteria:

* Receive approval from a peer-review committee;

* Pass the capital certification exam; and

* Participate in CLE courses on capital litigation;

* Specific Criteria for Lead Counsel:

* Eight years of experience in criminal law;

* Tried a minimum of fifteen felony jury trials to verdict as Lead Counsel;

* Tried a minimum of two capital cases to verdict as Lead Counsel or Second Chair (must be defense counsel in

one of the two cases unless five years of experience in criminal defense); and

PDF GENERATED BY SEARCH.PROQUEST.COM Page 4 of 22

* No judgment of ineffective counsel in a prior capital case;

* Specific Criteria for Second Chair:

* Five years of experience in criminal law; and

* Tried a minimum often felony jury trials to verdict as Lead Counsel.

The fact that Harris County had the most rigorous standards for appointment to a capital case in Texas during the

1990s suggests that the current research is a conservative test, meaning the disparities that existed in Houston

during the 1990s probably also existed across the rest of the state. Moreover, because the central elements of the

2001 Fair Defense Act were already in place in Harris County during the 1990s, the passage of the act probably did

not remedy the disparities observed in the current research.

B. CRITIQUES OF APPOINTMENT METHOD

Despite attempts at reform, some have argued that the appointment method of delivering indigent defense is

fundamentally flawed. Two major reports have examined the shortcomings of the appointment method: Muting

Gideon's Trumpet (MGT) and The Fair Defense Report (FDR). MGT, a report prepared for the State Bar of Texas in

September 2000 by Allan Butcher and Michael Moore, political scientists at the University of Texas at Arlington,

draws on data from mail questionnaires distributed to a random sample of Texas criminal defense attorneys, all

Texas prosecutors, and all Texas judges.31 FDR, a report prepared by the Texas Appleseed Fair Defense Project in

December 2000, draws on interviews with pivotal criminal justice actors (for example, defense attorneys,

prosecutors, judges) and archival records (such as expenditure reports and case files) for a stratified random

sample of twenty-three Texas counties (including Harris).32

The problems surrounding the appointment method described in MGT and FDR can be divided into the following

five categories: (1) flat-fee compensation, (2) the potential for insufficient support services, (3) a potential conflict

of interest for the defense attorney, (4) a potential conflict of interest for the judge, and (5) questionable

appointment practices. Each issue is considered in turn.

Under flat-fee compensation, the defense counsel receives a standard fee for a capital case disposed at trial

regardless of the number of hours worked (if a plea-bargain is reached, then the judge has the discretion to reduce

the fee).33 The American Bar Association discourages flat-fee compensation because of the potential for abuse: a

rational actor could go to trial, but limit the number of hours worked to maximize profit.34 Although most defense

teams would not engage in such a cold economic calculus, the flat-fee arrangement creates an inevitable conflict

between the legal interests of the client and the economic interests of the defense team because each hour of

work reduces both the rate of pay and the time available to work for paying clients.35

The appointment method also suffers from the potential for insufficient support services. Appointed counsel must

request approval from the judge to hire support services, such as an investigator or expert witness.36 Judges do

not rubber stamp defense requests. In fact, criminal defense attorneys report that 32% of requests for support

services are denied.37 Judges can also approve requests but limit funding. Judges in Harris County, for example,

only tend to provide enough money to hire experts from the Houston area.38 This practice forces the defense team

to use local experts who might not be the most qualified and simultaneously allows prosecutors to "build a book"

on experts who testify in numerous cases.39 Judges in Harris County have even refused to compensate experts, or

only paid a portion of the total bill, putting the defense team in the untenable position of covering the balance or

alienating an expert who might be needed in the future.40 Though charges of insufficient support services from

the mouths of appointed counsel might be dismissed as self-serving, MGT reports that 27% of Texas judges agree

that appointed counsel do not receive sufficient support services.41

The appointment method also has the potential to create a conflict of interest for the defense team. The defense

team must balance the adversarial mandate to provide the most rigorous defense possible with the need for

continued personal income. Defense teams who fight too hard risk losing future appointments. One defense

attorney summarized the conflict as follows: "An attorney who files a lot of motions and asks a lot of questions

creates a problem for the judge. You tick off the judge and don't get any more appointments."42 Another defense

attorney explained in an interview with the Houston Chronicle that an appointed attorney who works in the best

PDF GENERATED BY SEARCH.PROQUEST.COM Page 5 of 22

interest of his or her client might be committing career suicide: "As a hired attorney, I work in the best interest of

my client and that often puts me at odds with the judge. But if an appointed attorney gets at odds with the judge,

he doesn't get any more court appointments."43 The problem is simple and serious: the defense counsel's

personal income depends on remaining in the good graces of the judge, a proposition that might not be in the best

interest of the client.

The appointment method also has the potential to create a dual conflict of interest for the judge who must decide

whether to approve support services. The judge must balance the mandate to fund indigent defense with the need

to placate county commissioners and the personal desire to get reelected. Among judges who responded to the

MGT questionnaire, 50% reported that other judges in their jurisdictions had been asked by the county

commissioner to control expenses related to indigent defense.44 Moreover, judges running for reelection do not

want to be perceived as writing a blank check for indigent defense and thus "soft on crime."45

Finally, some judges appear to engage in questionable appointment practices. The MGT questionnaire asked

judges whether certain factors influence other judges' appointment decisions. The judges agreed that legal

considerations are pivotal: more than 95% of judges reported that peers consider the difficulty of the case and the

potential appointee's knowledge and experience.46 But a substantial number of judges also noted that irrelevant

factors play a role: 52% reported that peers consider whether the potential appointee needs income, 40% reported

that peers consider whether the potential appointee is a friend, 35% reported that peers consider whether the

potential appointee is a political supporter, and 30% reported that peers consider whether the potential appointee

contributed to the judge's election campaign.47 A criminal defense attorney from Harris County confirmed the

charge of political partisanship: "I have been refused appointments because I cannot afford to give money to the

judge's reelection campaign . . . those attorneys who contribute the most money receive the most work."48 In fact,

budget records indicate that funds spent on appointed counsel increase during election years, raising the

possibility (though clearly not proving) that judges become more generous with the expectation of a quid pro

quo.49

The current state of indigent capital defense in Texas is paradoxical: the state has made major efforts at reform,

but some argue that the reforms have done nothing to change the structural deficiencies inherent in the

appointment method. The appointment method creates perverse economic incentives for the defense team, limits

access to the support services that are needed to mount a rigorous defense, creates potential conflicts of interest

for the defense team and the judge, and sometimes results in the appointment of defense counsel to a case for all

the wrong reasons.50 But does the appointment method alter case outcomes? This Article illuminates whether the

appointment method is merely procedurally flawed or truly a matter of life and death.

III. Research Methods

A. DEPENDENT VARIABLES: DECISIONS TO SEEK AND IMPOSE THE DEATH PENALTY

The path from the commission of a murder to the pronouncement of a death sentence in Harris County includes

four major decisions: the intake prosecutor's decision to charge a defendant with capital murder, the grand jury's

decision to indict a defendant for capital murder, the District Attorney's (DA) decision to seek the death penalty,

and the jury's decision to impose a death sentence.51 Because the charging and indictment decisions do not

appear to exhibit enough variation to warrant an investigation, this Article focuses on whether the defendant's

form of legal counsel influences the DA's decision to seek death or the jury's decision to impose death - the

trajectory and disposition of a case.52

The data include the population of adult defendants indicted for capital murder in Harris County, Texas from 1992

to 1999 (n=504).53 The Harris County District Clerk (HCDC) used the Harris County Justice Information

Management System (JIMS) to identify the defendants. The HCDC also provided a JIMS file that contained public

information about each case, including whether the case resulted in a plea bargain or trial and the disposition. The

Harris County District Attorney's office provided archival documents that were used to verify the list of defendants

and determine if the DA sought death.

Figure 1 traces the trajectory, disposition, and current status of the 504 adult defendants indicted for murdering

PDF GENERATED BY SEARCH.PROQUEST.COM Page 6 of 22

614 victims in all (defendants aged seventeen or older at the time of the crime were eligible for the death penalty in

Texas during the time period under consideration54). The figure reveals that the DA sought death against 129 of

the 504 defendants. The 129 defendants in question were adjudicated as follows: 98 were sentenced to death, 29

were sentenced to life imprisonment, 1 was sentenced to confinement in the Texas Department of Corrections

(TDC) for some period of time less than life, and one was acquitted.55 Of the 98 condemned defendants, 36 have

been executed to date, 48 remain on death row, and 14 will not be executed (10 were commuted to life

imprisonment due to the Supreme Court's 2005 decision regarding juveniles in Roper v. Simmons56 and 4 died of

natural causes on death row). The figure also reveals that the DA sought a life sentence against 218 defendants

and reached a plea bargain with 157 defendants.

B. LEGAL COUNSEL

JIMS data indicate that the defendant's form of legal counsel can be divided into three categories: appointed

(n=369), hired (n=31), and mixed (n=104). Defendants with mixed counsel conform to one of the following

scenarios: the defendant is declared indigent and appointed counsel, but later secures the funds to hire counsel; or

the defendant hires counsel, but exhausts all funds and must be appointed counsel.57 Unfortunately, JIMS data do

not distinguish between these scenarios. Therefore, mixed counsel simply means the defendant had appointed

counsel and hired counsel during different stages of the case.

C. CONTROLS

To control for potential confounders, the multivariate models examine the legal dimensions of the case and the

social characteristics of the defendant and victim. Data for the controls were collected and merged from the

following archival sources: Harris County Justice Information Management System (JIMS), Texas Vital Statistics

Mortality File (VSMF), Grand Jury Indictments (GJI), the Harris County Medical Examiner (HCME), the Houston

Chronicle newspaper (HC), and the website www.publicdata.com. Table 3 reports measurement strategies, means,

and the data source for each variable.

1. Legal Dimensions of the Case

Data regarding the legal dimensions of the case were drawn from grand jury indictments, the Harris County

Medical Examiner, and the Houston Chronicle newspaper. The multivariate models control for the heinousness of

the crime (as discussed below), whether multiple defendants were indicted (l=yes), the form of capital murder (as

discussed below), and the method of murder (dichotomous indicators for shot, stabbed, beaten, and asphyxiated).

Controls for heinousness and the form of capital murder require elaboration.

To measure the heinousness of the crime, newspaper articles about each case were collected from the Houston

Chronicle online archive (an average of 6.75 articles per case, for a total of more than 3,400 articles). The

aggravating and mitigating circumstances in each case were coded based on a list drawn from Baldus and his

colleagues' landmark research on race and capital punishment.58 Table 4 lists the aggravating and mitigating

circumstances in question. The following formula was used to construct a scale of heinousness: number of

aggravating circumstances minus number of mitigating circumstances (the scale ranged from -3 to +7). The

original scale was transformed into three dichotomous indicators: Level 1 Heinousness (bottom quartile of scores

ranging from -3 to 0), Level 2 Heinousness (middle fifty percent of scores ranging from 1 to 2), and Level 3

Heinousness (top quartile of scores ranging from 3 to 7).

The heinousness measure included missing data because the Houston Chronicle did not report on twenty-eight of

the cases. To address the problem, missing cases are assumed to be Level 1. This assumption is based on

compelling patterns. To begin, the cliché "if it bleeds it leads" encapsulates the media's obsession with sensational

crimes. Considering the fact that the Houston Chronicle reported on 476 of the 504 cases, the twenty-eight capital

murders that did not attract media attention are almost sure to be the least heinous of all. The DA did not seek

death against any of the twenty-eight defendants in question, bolstering the assumption of minimal heinousness.

Because the substantive results are the same regardless of whether the missing cases are excluded or coded as

Level 1, the models presented in the results section use the revised indicator of heinousness to ensure complete

data for all cases. Thus, the original scale was transformed into three dichotomous indicators to facilitate a

PDF GENERATED BY SEARCH.PROQUEST.COM Page 7 of 22

solution to the missing data problem (and because several values on the original scale had no cases or just one

case).

Using newspaper articles to code heinousness is not ideal, but all other avenues were closed.59 Nonetheless,

focusing on newspaper articles is a reasonable approach. To begin, the measure has face validity: 12% of cases

coded as Level 1 heinousness resulted in a death sentence, compared to 17% of cases coded as Level 2

heinousness and 35% of cases coded as Level 3 heinousness. In addition, the aggravating and mitigating

circumstances listed in Table 4 are arguably the type of facts a newspaper would tend to report. Perhaps most

importantly, this Article improves upon most capital punishment studies which rely on Supplemental Homicide

Report data, and therefore do not include any measure of heinousness.60

Grand Jury indictments were used to determine the form of capital murder. Of the forms delineated in the Texas

capital murder statute, the following appear in the data: robbery, burglary, multiple victims, kidnapping, rape,

remuneration, child zero to five years old, police officer, arson, and obstruction/retaliation. The form of capital

murder is measured through dichotomous indicators coded l=yes, 0=no (other includes police officer, arson, and

obstruction/retaliation). Because a case can be a capital murder for multiple reasons, the indicators are not

mutually exclusive.

2. Defendant and Victim Social Characteristics

Data regarding the defendant's social characteristics were drawn from the JIMS file. The multivariate models

control for the defendant's race/ethnicity (dichotomous indicators for White, Black, Hispanic, and Asian),61 sex

(l=male), age (dichotomous indictors for teens seventeen to nineteen years old, young adults twenty to twenty-nine

years old, and adults thirty years old or more), prior violent conviction (l=yes), and prior nonviolent conviction

(l=yes).62

Data regarding the victim's social characteristics were drawn from the Texas Vital Statistics Mortality File (VSMF)

and the website www.publicdata.com.63 The multivariate models control for the victim's race/ethnicity

(dichotomous indicators for White, Black, Hispanic, and Asian), the victim's sex (l=female), whether the victim was

vulnerable due to age (l=six to sixteen or over sixty; children less than six are considered separately as a form of

capital murder), and whether the victim had a prior conviction (l=prior violent or non-violent conviction).64

Coding victim characteristics required a procedure that could accommodate cases with multiple victims. The data

reveal that death is more apt to be sought and imposed on behalf of victims who are white, female, and vulnerable

due to age. But death is less apt to be sought and imposed on behalf of disreputable victims with a prior criminal

record. Thus, a case with multiple victims is coded 1 if any of the victims meet the criterion in question (for

example, a case is coded as female victim if any of the victims are female).

D. MODELING

Logistic regression is used to estimate the impact of legal counsel on the odds of the DA seeking death (l=seek

death; 0=all other trajectories) and the jury imposing death (l=death sentence; 0=all other dispositions). In a

logistic model, odds ratios represent the effect of a unit change in the independent variable on the odds of the

outcome occurring. An odds ratio greater than 1 denotes a direct relationship, an odds ratio less than 1 denotes an

inverse relationship, and an odds ratio of 1 suggests that the independent variable is unrelated to the outcome.65

Recall that the data include a population of cases, not a random sample. The question of whether statistical

significance should be applied to population data remains contested.66 The answer depends, in part, on the

definition of a population. One interpretation is that an "apparent population" includes the census of relevant

events. Therefore, an apparent population should be treated as a true population, meaning tests of statistical

significance are irrelevant. But another interpretation is that an apparent population is just one "realization" of all

the populations that could have occurred if historical events were replicated numerous times.67 Therefore, an

apparent population should be treated as a random sample, meaning tests of statistical significance are

relevant.68

The current research follows Bollen' s interpretation that (1) ignoring statistical significance in population data is

legitimate and appropriate if a researcher is attempting to describe the population rather than draw inferences, and

PDF GENERATED BY SEARCH.PROQUEST.COM Page 8 of 22

(2) researchers should focus more on substantive significance and less on statistical significance.69 Thus, the

central purpose of the research presented in this Article is to describe the magnitude of the relationship between

legal counsel and capital punishment for the population of cases.70

Nonetheless, the use of statistical significance in arguably inappropriate contexts - including convenience samples

and population data - is so common that ignoring significance has become unacceptable to most readers.

Capitulating to convention, I report significance levels for legal counsel in the text for the multivariate models. Yet I

maintain that all the population parameters are meaningful, regardless of statistical significance.

IV. RESULTS

A. BIVARIATE FINDINGS

Table 5 presents a cross-tabulation of the DA's decision to seek death and the jury's decision to impose death by

the defendant's form of legal counsel. The percentage distribution suggests a strong relationship. The DA sought

death against just 3% of defendants with hired counsel, compared to 26% of defendants with mixed counsel and

27% of defendants with appointed counsel. Moreover, no defendant with hired counsel received a death sentence,

compared to 14% of defendants with mixed counsel and 23% of defendants with appointed counsel.

How did all thirty-one defendants with hired counsel escape the ultimate state sanction? The central mechanism

was to negotiate a plea bargain: 68% of defendants with hired counsel negotiated a plea bargain compared to 30%

of defendants with appointed counsel and 24% of defendants with mixed counsel.

Hiring counsel seems to eliminate the chance of being sentenced to death. But perhaps hired counsel is not the

driving force. Perhaps defendants who hired counsel were accused of committing murders that were less worthy

of the death penalty. To examine the alternative explanation, Table 6 considers the relationship between hired

counsel and the pivotal legal considerations in a case. The findings refute the alternative explanation. It is true that

defendants who hired counsel were less likely to have a prior violent conviction and slightly less likely to have been

accused of acting alone. Yet defendants who hired counsel were more likely to have been indicted for killing

multiple victims. Perhaps most importantly, defendants who hired counsel were just as likely to have been indicted

for the most heinous murders. Such counterbalancing forces suggest that the nature of the case cannot account

for the fact that defendants who hired counsel were never sentenced to death.

Some might argue that it is inappropriate to draw conclusions about the role of hired counsel in capital

punishment based on just thirty-one cases. This is a valid concern. But three responses should ease the concern.

First, the data represent a population of cases, not a random sample. Thus, the results are real for the time period

in question. Second, it is extremely unlikely that the pattern has changed over time. Consider the following

example. Given that the period from 1992 to 1999 included thirty-one defendants with hired counsel, it is

reasonable to assume, for the purposes of argument, that the period from 2000 to 2007 also included thirty-one

defendants with hired counsel. For the death sentence rate among defendants with hired counsel and mixed

counsel to reach parity, the DA would have had to secure a death sentence against nine of the next thirty- one

defendants with hired counsel (9/62=14%). For the death sentence rate among defendants with hired counsel and

appointed counsel to reach parity, the DA would have had to secure a death sentence against an extraordinary

fourteen of the next thirty-one defendants with hired counsel (14/62=23%).71 Could the DA go from securing a

death sentence against 0% of defendants with hired counsel to 29% (9/31) or even 45% (14/31) of defendants with

hired counsel? Parity is technically possible, but appears extremely improbable. Parity appears even more

improbable in light of the DA' s past record of allowing defendants with hired counsel to plea bargain. Third, the

data include 104 defendants with mixed counsel. If hiring counsel for a mere portion of a case reduces the chance

of a death sentence then the influence of hired counsel is confirmed based on a larger number of cases. The

findings, considered below, confirm that hiring counsel for a portion of a case is enough to change the outcome.

Before turning to the multivariate models, it is important to discuss the statistical problem of quasi-complete

separation. Quasi-complete separation occurs if almost all (or all) of the cases at one level of a dichotomous

independent variable have the same value on the dependent variable.72 Here, thirty of the thirty-one defendants

who hired counsel are coded 0 on seek death, and all thirty-one are coded 0 on death sentence. The absence of

PDF GENERATED BY SEARCH.PROQUEST.COM Page 9 of 22

variation precludes a multivariate comparison of defendants with hired counsel to defendants with appointed

counsel, though such a comparison is arguably not needed given the magnitude of the bivariate relationships. But

it is possible to conduct multivariate comparisons of (1) defendants with appointed counsel to defendants with

mixed counsel, dropping the thirty-one defendants who hired counsel (l=appointed, 0=mixed; n=473), and (2)

defendants with appointed counsel to defendants who hired counsel during some or all of the case (l=appointed,

0=mixed or hired; n=504). Both solutions to the problem of quasi-complete separation are considered below.

B. MULTIVARIATE FINDINGS

Does the defendant's form of legal counsel influence the DA' s decision to seek death? Among the multivariate

comparisons that can be made, the answer is no. The findings, reported in Table 7, reveal that the odds of seeking

death are about the same regardless of whether defendants with appointed counsel are compared to defendants

with mixed counsel (Model IA) or defendants with mixed/hired counsel (Model 2A). Neither odds ratio is

significant.

But the defendant's form of legal counsel does influence the ultimate outcome - being sentenced to death. The

odds of being sentenced to death are 1.549 times higher if the defendant has appointed counsel, as compared to

defendants with mixed counsel (Model IB; odds ratio not significant). The odds of being sentenced to death are

2.154 times higher if the defendant has appointed counsel, as compared to defendants with mixed/hired counsel

(Model 2B; odds ratio significant at p=.025).

The empirical patterns seem curious. Why does the defendant's form of legal counsel have no influence on the

DA's decision to seek the death penalty, but a substantial influence on the jury's decision to impose the death

penalty? The most obvious explanation is that the DA is not swayed by defense attorneys, but jurors are. Indeed,

during the time period under consideration the DA, John Holmes, is on record arguing that the type of defense

attorney in a capital case doesn't amount to "a hill of beans."73 But the obvious explanation is problematic: the DA

almost never sought the death penalty against defendants who hired counsel for the entire case, opting instead for

a plea bargain in most cases. The DA appears to have been more attuned to the presence of hired defense

attorneys than his comments to the media suggest.

The more convincing explanation revolves around the meaning of mixed counsel. Recall that defendants with

mixed counsel can either (1) be appointed counsel but subsequently hire representation once someone provides

the necessary funds, or (2) hire representation but subsequently exhaust all resources and be appointed counsel.

Though the JIMS data do not distinguish between such scenarios, the Chief Counsel to the Harris County DA

suggests that the former scenario - moving from appointed to hired - is more common.74 If true, then defendants

with mixed counsel would tend to have appointed counsel in the initial stages of a case when the DA is deciding

whether to seek death, but hired counsel in the final stage of a case when the jurors are deciding whether to

impose death. Seen from this perspective, the findings make sense. The DA treated defendants with appointed

counsel and mixed counsel the same because, in most cases, the form of counsel is the same. Jurors treated

defendants with appointed counsel and mixed counsel differently because, in most cases, the form of counsel is

different. The number of defendants who hired counsel for the entire case is too small to change the basic

patterns at either stage, but does amplify the legal disparities observed in the death sentence stage. Future

research should examine whether the temporal sequence of mixed counsel - hired to appointed versus appointed

to hired - explains why the impact of legal counsel depends on the stage of the case in question. Regardless, the

bottom line remains: legal counsel shapes death sentences.

C. USING PREDICTED PROBABILITIES TO EXAMINE THE MAGNITUDE OF POPULATION PARAMETERS

How substantial are the disparities in case outcomes between defendants with appointed counsel and

mixed/hired counsel? To provide a more interpretable metric, Table 8 presents predicted probabilities for seek

death and death sentence based on models 2A and 2B (confounders held constant at the mean). The predicted

probabilities are also used to calculate the conditional probability of jurors rendering a death sentence at trial.75

To illustrate the calculation of conditional probabilities, consider the following example. For defendants with

appointed counsel, the predicted probability of the DA seeking death is .20 and the predicted probability of a death

PDF GENERATED BY SEARCH.PROQUEST.COM Page 10 of 22

sentence is .17. Thus, the conditional probability of jurors rendering a death sentence at trial is: .2Ox=. 17;

x=.17/.20; x=.85.

The magnitude of the adjusted disparities can be illustrated through a set of hypothetical cases. Assume that 100

defendants with appointed counsel and 100 defendants with mixed/hired counsel were indicted for capital murder.

The predicted probabilities suggest the following: the DA would seek death against twenty defendants with

appointed counsel and jurors would return a death sentence against 85% of them - meaning seventeen defendants

with appointed counsel would be condemned to die; the DA would seek death against seventeen defendants with

mixed/hired counsel and jurors would return a death sentence against 47% of them meaning eight defendants with

mixed/hired counsel would be condemned to die. The probabilities translate abstract numbers into human lives. In

this scenario, nine defendants would be condemned as a result of not having the resources to hire counsel for

some or all of the case.

D. LEGAL COUNSEL AND SOCIOECONOMIC STATUS

Death penalty opponents imply a close connection between the defendant's form of legal counsel and the

defendant's socioeconomic status, suggesting that defendants who can hire counsel are wealthy, but indigent

defendants who must accept court-appointed counsel are, by definition, poor. But the issue turns out to be more

complicated.

To investigate the relationship between legal counsel and socioeconomic status, defendants were geo-coded to

census block groups based on addresses contained in the JIMS database. Census block groups represent the

closest possible approximation of neighborhoods. Linear interpolation was used to estimate median household

income in the defendant's neighborhood based on 1990 and 2000 census data (if, for example, median household

income for a block group increased from $30,000 in 1990 to $40,000 in 2000, and the defendant was indicted in

1995, then median household income for 1995 is estimated to be $35,000).

The results provide a surprising twist - there appears to be no relationship between legal counsel and

socioeconomic status. As detailed in Table 9, defendants with appointed counsel lived in neighborhoods with an

average household income of $25,493, compared to defendants with mixed counsel and hired counsel who lived in

neighborhoods with average household incomes of $27,310 and $29,707, respectively. Although defendants with

mixed and hired counsel lived in neighborhoods with slightly higher incomes, the difference of $2,000 to $4,000 is

small - and surely not enough to pay for a defense attorney in a capital case. Capital defendants were also below

average regardless of the form of legal counsel in question, as median household income for all Harris County

residents in 1995 was approximately $37,947.76 With few exceptions, capital murder defendants in this study

appear to be uniformly poor - meaning disparities based on legal counsel do not equate to disparities based on

socioeconomic status.77 The fact that some defendants from such poor neighborhoods can hire counsel

suggests that others, perhaps relatives and friends, have pooled resources in the hour of need.78

The census data are far from perfect: 155 of the 504 defendants (31%) could not be matched to neighborhoods

due to missing or incomplete addresses in JIMS, rural routes, homelessness, or the defendant's address being

listed as prison. The data also reflect median household income for the neighborhood, so the defendant's personal

income might be quite different from the neighborhood average (though neighborhoods defined as census block

groups tend to be fairly homogenous). The limitations of census data suggest that the findings regarding

socioeconomic status should be interpreted with caution. Indeed, a more nuanced measure of socioeconomic

status might uncover more variation. Yet prior research using a different methodological approach reached the

same conclusion. Drawing on data from the Uniform Crime Reports and a comprehensive search of articles in the

New York Times, Green and Wakefield reported that just 20 of the 13,000 defendants arrested for murder in New

York City from 1955 to 1975 were middle class or upper class.79 In fact, the involvement of elites in homicide has

virtually disappeared over the past several centuries.80

E. ONE FINAL CONSIDERATION: LEGAL COUNSEL AND ACQUITTALS

Figure 1 demonstrates that only 8 of the 504 defendants indicted for capital murder were acquitted - not enough to

estimate a multivariate model. But the relationship between legal counsel and acquittals can be examined using

PDF GENERATED BY SEARCH.PROQUEST.COM Page 11 of 22

percentages. The results suggest that hiring legal counsel for the entire case dramatically increases the probability

of an acquittal: 30% of defendants with hired counsel who went to trial were acquitted (3 of 10), compared to 2% of

defendants with appointed counsel (5 of 258) and 0% of defendants with mixed counsel (0 of 79). Thus,

defendants who hired counsel were about twenty times more likely to be acquitted at trial than all other

defendants (3 of 10 versus 5 of 337). Remarkably, if the acquittal rate for defendants who hired counsel and were

disposed at trial had prevailed for all the cases disposed at trial then the total number of acquittals would have

catapulted from 8 to 104 (30% of 347=104).

The relationship between legal counsel and acquittals is troubling. It does not seem plausible to conclude that

defendants who hired counsel were actually twenty times more likely to be innocent. Instead, the results suggest

that defendants with hired counsel are being erroneously acquitted, or defendants with appointed counsel and

mixed counsel are being wrongly convicted, or both. The specter of wrongful conviction is real: from 1973 to the

present, 130 defendants who were convicted and sentenced to death in the United States have been released due

to innocence.81 Given the number of mistakes caught just in time, the most reasonable conclusion is that

wrongful executions have occurred.82 Even Ernest Van Den Haag, one of the most ardent and articulate

supporters of capital punishment, admits that wrongful execution is inevitable.83 But such a horrendous fate is

out of the question for the defendants represented in the current data set who hired counsel - none was sentenced

to death.

V. CONCLUSION

Death penalty opponents charge that socioeconomic status shapes the administration of capital punishment -

wealthy defendants who can hire legal counsel are treated in a fundamentally different manner than indigent

defendants who receive court-appointed counsel. But this critique has been the subject of more speculation and

anecdote than research. The current research begins to address the issue systematically by examining the fate of

all adult defendants indicted for capital murder in Harris County, Texas from 1992 to 1999. To evaluate death

penalty opponents' arguments, consider the central research findings:

* Hiring counsel for the entire case eliminates the chance of a death sentence. Among the thirty-one defendants

who hired counsel, none was sentenced to death despite committing murders that were just as heinous as those

committed by defendants with mixed and appointed counsel. Interestingly, defendants who hired counsel were

much more likely to negotiate a plea bargain, suggesting that the DA viewed hired attorneys as more formidable

opponents.

* Hiring counsel for a mere portion of a capital case substantially reduces the chance of a death sentence. Among

the 104 defendants with mixed counsel, a combination of hired and appointed, 14% were sentenced to death. But

among the 369 defendants who had appointed counsel, 23% were sentenced to death. The differences in case

outcomes for defendants with mixed and appointed counsel are confirmed in the multivariate statistical models.

Even some hired counsel is better than none.

* Defendants who hired counsel for the entire case were twenty times more likely to be acquitted at trial than

defendants who did not. In fact, if the acquittal rate for defendants who hired counsel and were disposed at trial

had prevailed for all the cases disposed at trial then the total number of acquittals would have catapulted from 8 to

104 - a troubling figure given the number of innocent defendants released from death row in recent years. The

findings regarding acquittals must be interpreted with caution: the number of defendants with hired counsel who

were disposed at trial, and the total number of acquittals, are both quite small. Nonetheless, to find such a strong

pattern in a population of cases suggests that the relationship between hired counsel and acquittals is probably

real.

* Geo-coding (matching) defendants to census block groups revealed that almost all capital murder defendants

live in poor neighborhoods.

Such patterns support one of the premises advanced by death penalty opponents and refute the other. Those who

can hire counsel for the entire case, or even a portion of the case, appear to be treated in a fundamentally different

manner than those who cannot. But hiring counsel is not the sole province of the affluent. Presumably, poor

PDF GENERATED BY SEARCH.PROQUEST.COM Page 12 of 22

defendants can sometimes hire counsel because friends and relatives pool resources in the hour of need, an

assumption that needs to be tested in future research.

The fact that almost all capital murder defendants are poor does not answer the charge of arbitrariness. The

charge of arbitrariness suggests that relevant legal facts cannot fully explain case outcomes. The defendant's

ability to hire counsel is not a relevant legal fact, but nonetheless has a substantial influence on case outcomes.

Regardless of whether the money comes from a personal account or is cobbled together from friends and

relatives, the point is that money matters. Retribution is supposed to be proportionate to harm, not proportionate

to financial resources.

It might be tempting to dismiss the claim of arbitrariness. Some might argue that the research findings are only

applicable to Harris County. But 252 of the 254 counties in Texas use the appointment method, and Harris County

had the most rigorous standards for appointaient to a capital case in the state during the 1990s. Thus, the

disparities uncovered in Harris County are probably a conservative estimate of disparities across the rest of Texas

during the 1990s. Others might argue that the research findings are only applicable to the time period before the

passage of the 2001 Fair Defense Act. But Harris County was a leader in the indigent capital defense reform

movement, meaning the major elements of the Fair Defense Act were already in place for capital cases in Harris

County during the 1990s. Moreover, the Fair Defense Act did not remedy the structural deficiencies inherent in the

appointment method, including: flat fee compensation, the potential for insufficient support services, potential

conflicts of interest for the defense attorney and judge, and the potential for questionable appointment practices.

Logic suggests that arbitrariness based on the defendant's form of legal counsel is probably a widespread and

continuing problem in Texas. If so, then the findings of the current research have important implications for the

national death penalty debate, as Texas accounts for 426 of the 1,141 executions in the modern era.

Understanding capital punishment in Texas is a prerequisite for understanding capital punishment in America.

Does the finding that indigent capital defendants are disadvantaged in Harris County mean that indigent capital

defendants are disadvantaged in all times and places? Not necessarily. Harris County uses an appointment

method that has been roundly criticized. In 2005, Harris County spent just $5.27 per capita on indigent defense,

less than half the national average of S11.88.84 In fact, Harris County is the largest jurisdiction in the United

States that does not have a public defender office.85 Conversely, some jurisdictions have renowned capital

defender programs, such as those of Colorado, New York, and Philadelphia.86 In Colorado between 1980 and 1999,

prosecutors sought the death penalty 110 times but secured just 13 death sentences and one execution to date.87

Prosecutors in New York sought the death penalty in only 39 of 500 potential cases from 1995 to 2000 resulting in

5 death sentences and no executions.88 And the Defender Association of Philadelphia has handled almost 1,000

murder cases without a single death sentence.89 Most indigent capital defense systems probably fall between the

extremes noted in the comparison of Harris County to Colorado, New York, and Philadelphia. The fact that indigent

capital defense varies tremendously from one jurisdiction to the next means that the question of whether indigent

defendants are universally disadvantaged cannot currently be answered. Much more research is needed to paint a

complete picture of the complex connections between indigent defense and capital punishment.

Despite the need for more research, one conclusion seems warranted: public defender offices have a better

performance record than appointed counsel. Dow' s research comparing both methods of delivering indigent

capital defense indicates that the prosecution's rate of securing death sentences ranges from 0% to 50% in

jurisdictions with public defender offices, compared to 50% to 100% in jurisdictions with court-appointed

counsel.90

Although it is not exactly clear why public defender offices provide a more rigorous defense than appointed

counsel, the following hypotheses have been advanced: (1) Public defenders are more aggressive advocates due

to a stronger ideological commitment to indigent defense;91 (2) Public defender positions are highly competitive

so the office can be selective in hiring top legal talent;92 (3) Public defenders have a state budget to hire

investigators and expert witnesses (like the prosecution) rather than asking a judge to approve such support

services, particularly given that an elected judge might not want to seem "soft on crime" by providing large sums

PDF GENERATED BY SEARCH.PROQUEST.COM Page 13 of 22

for indigent defense or the judge might be under pressure from a county commissioner to cut indigent defense

costs;93 (4) Public defenders receive an annual salary rather than a flat fee per capital case, so each hour of work

for a public defender does not reduce the rate of pay or detract from paying clients;94 (5) A public defender's

personal income does not depend on remaining in the good graces of the judge, so public defenders can present

the most rigorous defense possible without risking personal financial repercussions;95 (6) Public defenders are

assigned to handle cases based on expertise rather than the dubious factors that appear to sometimes influence

the judge's choice of appointed counsel, including whether the potential appointee is a friend and whether the

potential appointee contributed to the judge's re-election campaign.96 Future research should examine which of

the hypothesized mechanisms operate to produce such important differences in the performance records of public

defenders and appointed counsel.

If Harris County officials are serious about providing rigorous indigent defense then the county should create a

public defender office with a specific capital defender office. Indeed, all jurisdictions in Texas that use the

appointment method should consider public defender offices.97 To be effective, public defender and capital

defender offices must be independent of the judiciary and have resources proportionate to the District Attorney's

office. Noble attempts have been made to improve the appointment method of delivering indigent capital defense,

but the appointment method is fraught with structural inadequacies. The current research demonstrates that the

inadequacies are not just a problem of procedure, but rather a matter of life and death. Harris County officials who

contend that appointed counsel is superior to hired counsel, or that the defendant's form of legal counsel doesn't

amount to a hill of beans, are incorrect. Hired counsel is far superior. To bolster the point, consider one last finding:

the current research does not focus on executions because the process remains ongoing, but it is instructive to

note that thirty-three of the thirty-six defendants in the data who have been executed to date had appointed

counsel (meaning 73% of all defendants had appointed counsel, but 92% of defendants executed to date had

appointed counsel). Though public defender and capital defender offices would not be a panacea - some amount

of arbitrariness in the administration of capital punishment based on the defendant's ability to hire counsel might

remain - such organizations would surely narrow the gap and come much closer to the adversarial ideal of evenly

matched partisans doing battle to produce justice.

Footnote

1 Furman v Georgia, 408 U.S. 238, 251-52 (1972).

2 Id. at 255-56.

3 HELEN PREJEAN, THE DEATH OF INNOCENTS: AN EYEWITNESS ACCOUNT OF WRONGFUL EXECUTIONS 190

(2005).

4 Id.

5 See, e.g., ROBERT M. BOHM, DEATHQUEST III: AN INTRODUCTION TO THE THEORY AND PRACTICE OF CAPITAL

PUNISHMENT IN THE UNITED STATES (3D ED. 2007); DAVID R. DOW, EXECUTED ON A TECHNICALITY: LETHAL

INJUSTICE ON AMERICA'S DEATH ROW (2005); RAYMOND PATERNOSTER ET AL., THE DEATH PENALTY:

AMERICA'S EXPERIENCE WITH CAPITAL PUNISHMENT (2008); Stephen B. Bright, Counsel for the Poor: The Death

Sentence Not for the Worst Crime but for the Worst Lawyer, 103 Yale L.J. 1835 (1994) [hereinafter Counsel for the

Poor]; Stephen B. Bright, The Politics of Capital Punishment: The Sacrifice of Fairness for Executions, in

AMERICA'S EXPERIMENT WITH CAPITAL PUNISHMENT 127 (James R. Acker et al. eds., 2003).

Footnote

6 Dow, supra note 5, at 7.

7 DAVID C. BALDUS ET AL., EQUAL JUSTICE AND THE DEATH PENALTY: A LEGAL AND EMPIRICAL ANALYSIS 14-

15 (1990).

8 Bright, Counsel for the Poor, supra note 5, at 1 840.

9 WILLIAM J. BOWERS, LEGAL HOMICIDE: DEATH AS PUNISHMENT IN AMERICA 1864-1982 (1984).

10 Linda Foley, Florida After the Furman Decision: The Effect of Extralegal Factors on the Processing of Capital

Offense Cases, 5 Behav. Sci. &L. 457 (1987).

PDF GENERATED BY SEARCH.PROQUEST.COM Page 14 of 22

11 BALDUS ET AL., supra note 7, at 157-58.

Footnote

12 James C. Beck &Robert Shumsky, A Comparison of Retained and Appointed Counsel in Cases of Capital Murder,

21 Law &Hum. Behav. 525 (1997).

13 CAROL J. DEFRANCES &MARIKA F.X. LITRAS, U.S. DEP'T OF JUSTICE, INDIGENT DEFENSE SERVICES IN

LARGE COUNTIES, 1999 (2000), available at http://www.ojp.usdoj.gov/ bjs/pub/pdf/idslc99.pdf.

14 See, e.g., AMNESTY INT'L, AI INDEX NO. AMR 51/125/2007, UNITED STATES OF AMERICA: ONE COUNTY, 100

EXECUTIONS (2007), available at http://www.amnesty.org/ en/library/info/AMR5 1/1 25/2007.

15 Tex. Dep't of Criminal Justice, Death Row Home Page, http://www.tdcj.state.tx.us/ stat/deathrow.htm (last

visited May 15, 2009). The annual number of death sentences from Harris County was calculated from the Texas

Department of Criminal Justice website, which lists the county of conviction for each offender and the date the

offender was received on death row. Id.

Footnote

16 Mark Ballard, Gideon's Broken Promise, Tex. Law., Aug. 28, 1995, at 17, 19.

17 Bob Sablatura, Study Confirms Money Counts in County's Courts: Those Using Appointed Lawyers Are Twice as

Likely to Serve Time, HOUS. CHRON., Oct. 17, 1999, at Al.

18 Mary Flood, What Price Justice? Gary Graham Case Fuels Debate Over Appointed Attorneys, HoUS. CHRON.,

July 1, 2000, at Al.

Footnote

19 TEX. APPLESEED FAIR DEF. PROJECT, THE FAIR DEFENSE REPORT: ANALYSIS OF INDIGENT DEFENSE

PRACTICES IN TEXAS 48 (2000), available at http://www.texasappleseed.net/pdf/projects_fairDefense_fairref.pdf.

20 Death Penalty Info. Qr., Number of Executions by State and Region Since 1976,

http://www.deathpenaltyinfo.org/number-executions-state-and-region-1976 (last visited May 15,2009).

Footnote

21 In non-capital cases, contract counsel is also used. Under the contract method, an attorney enters into a

contract with a judge to represent all indigent defendants who appear before the court for a specified period of

time.

22 TEX. APPLESEED FAIR DEF. PROJECT, supra note 19, at 48. The two counties that have a sizeable public

defender office are Dallas County and El Paso County. Id.

23 The following information comes from personal correspondence with Judge Burnett: the description of indigent

defense prior to 1991; the development of the capital certification program; and the relationship of the capital

certification program to the subsequent Fair Defense Act. E-mail from Jay W. Burnett, J., Harris County, to Scott

Phillips, Professor, Univ. of Denver (Aug. 20, 2007, 1 1:00 AM) (on file with author).

Footnote

2 Id.

25 Texas has nine Judicial Regions. Tex. Courts Online, Administrative Judicial Regions,

http://www.courts.state.tx.us/courts/ajr.asp (last visited May 15, 2009).

26 E-mail from Jay W. Burnett, supra note 23.

27 Id

28 EQUAL JUSTICE CTR. &TEX. DEFENDER SERV., TEXAS DEATH PENALTY PRACTICES: QUALITY OF REGIONAL

STANDARDS AND COUNTY PLANS GOVERNING INDIGENT DEFENSE IN CAPITAL CASES 25, available at

http://www.equaljusticecenter.org/Report%20on%20 Texas%20Capital%20Defense%20Procedures%20-

%202d%20ed.pdf.

29 Id. at 40.

30 Id. at 46.

Footnote

31 ALLAN K. BUTCHER &MICHAEL K. MOORE, COMM. ON LEGAL SERV. TO THE POOR IN CRIMINAL MATTERS,

PDF GENERATED BY SEARCH.PROQUEST.COM Page 15 of 22

MUTING GIDEON'S TRUMPET: THE CRISIS IN INDIGENT CRIMINAL DEFENSE IN TEXAS 5 (2000), available at

http://www.uta.edu/pols/moore/indigent/last.pdf.

Footnote

32 TEX. APPLESEED FAIR DEF. PROJECT, supra note 19, at 6, 1 1 .

33 See id. at 99-101.

34 AM. BAR ASS'N, ABA STANDARDS FOR CRIMINAL JUSTICE: PROVIDING DEFENSE SERVICES 5-2.4 cmt. (3d ed.

1992) ("Since a primary objective of the payment system should be to encourage vigorous defense representation,

flat payment rates should be discouraged. The possible effect of such rates is to discourage lawyers from doing

more than what is minimally necessary to qualify for the flat payment." (footnote omitted)).

35 See id.

36 BUTCHER &MOORE, supra note 31, at 1 7.

37 Id. at 18.

38 TEX. APPLESEED FAIR DEF. PROJECT, supra note 19, at 119.

Footnote

39 Id

40 Id. at 120-21.

41 BUTCHER &MOORE, supra note 3 1 , at 1 8.

42 TEX. APPLESEED FAIR DEF. PROJECT, supra note 19, at 1 8.

43 Sablatura, supra note 17, at Al.

44 BUTCHER &MOORE, supra note 3 1 , at 20-2 1 .

45 Id.

Footnote

46 Id. at 13.

47 Id.

48 Id.

49 Ballard, supra note 16, at 21 .

50 Although researchers have not examined the effect of legal counsel on capital punishment in Harris County,

researchers have examined the relationship between legal counsel and punishment for other crimes. Examining all

criminal cases in Harris County from June 1992 to December 1994, for example, the Texas Lawyer reports that 25%

of defendants with court-appointed counsel were sentenced to jail or prison for an average of 7.6 years, compared

to 8% of defendants with hired counsel for an average of 2 years. Ballard, supra note 16, at 19. Other studies have

considered specific crimes. Focusing on defendants charged with first-offense possession of less than one gram

of cocaine in 1996, the Houston Chronicle reports that 57% of defendants with court-appointed counsel were

sentenced to jail or prison, compared to 25% of defendants with hired counsel. Sablatura, supra note 17. Having

the resources to hire counsel also matters in DWI cases. An examination of DWI cases disposed between January

and August of 1999 reveals that 98% of defendants with appointed counsel were convicted, and, among those, 95%

were sentenced to jail; but just 86% of defendants with hired counsel were convicted, and, among those, 32% were

sentenced to jail. Tana McCoy, Indigent Defense: An Application of Conflict Theory in the Analysis of Driving While

Intoxicated Cases 158 tbl.24, 177 tbl.37 (Aug. 2002) (unpublished Ph.D. dissertation, Sam Houston State

University) (on file with ProQuest).

Footnote

51 E-mail from Scott Durfee, Chief Counsel to the District Attorney, Harris County, to Scott Phillips, Professor, Univ.

of Denver (Aug. 11, 2006, 5:40 PM) (on file with author).

52 The Harris County intake division prosecutor must determine whether a homicide can be charged under the

Texas capital murder statute. Id. Despite repeated attempts, collecting the data needed to examine the impact of

legal counsel - whether the defendant had counsel and, if so, the form of counsel - on the charging decision proved

impossible. But the charging decision does not appear to exhibit much variation. To begin, the Texas capital

PDF GENERATED BY SEARCH.PROQUEST.COM Page 16 of 22

murder statute delineates narrow categories of murder that are death-eligible. Tex. Penal Code Ann. §19.03

(Vernon 2003 &Supp. 2008). The precision of the statute simplifies the charging decision, as opposed to states

that define heinous murders as death-eligible. For a listing of such states and examples of the heinous language

used to define death-eligible crimes, see Death Penalty Info. Ctr., Aggravating Factors for Capital Punishment by

State, http://www.deathpenaltyinfo.org/aggravating-factors-capital-punishment-state (last visited May 15, 2009).

Moreover, the Houston Chronicle reports in a February 2001 special series that the intake prosecutor has "standing

orders" to file capital murder charges in all possible cases. Mike Toison, A Deadly Distinction: Harris County Is a

Pipeline to Death Row, Hous. Chron., Feb. 5, 2001, at Al. Nonetheless, the inability to examine the charging decision

remains a potential weakness of the current research. The grand jury must return a "Bill of Indictment" for capital

murder in order for the DA to seek death. E-mail from Scott Durfee, supra note 51. This step borders on a formality,

as data from the Harris County district clerk indicate that grand juries returned a "No Bill" in just seven capital

cases from 1992-1999.

Footnote

53 Defendants were excluded if the case was dismissed, the case was disposed but expunged, the defendant was

never arrested, the victim's remains could not be identified, or the case had not been disposed at the time the list

of cases was requested from the Harris County District Clerk in December 2001. The two Native American

defendants were also excluded.

54 See Roper v. Simmons, 543 U.S. 551, 579 app. A.I (2005) (listing Texas as one of the states that permitted the

death penalty for juveniles).

55 The inmates sentenced to life imprisonment are eligible for parole because Texas did not pass a life without

parole (LWOP) statute until 2005. Defendants in the data who were convicted in 1992 must serve thirty-five years

before becoming eligible for parole; defendants in the data who were convicted between 1993 and the passage of

LWOP must serve forty years before becoming eligible for parole.

56 543 U.S. 551 (holding that the Eighth and Fourteenth Amendments barred the imposition of the death penalty

upon juvenile defendants).

Footnote

57 E-mail from Scott Durfee, supra note 51.

Footnote

58 BALDUS ET AL., supra note 7, at 526-35.

Footnote

59 The DA denied access to the capital murder memorandum as a confidential work product; police reports

contained limited information or had substantial amounts of information redacted; and case files and transcripts

were not an option because of the number of cases disposed through plea bargains.

60 The Research Assistant also coded heinousness based on a visceral reaction to the facts of the crime, just as a

DA or juror would do. Each case was assigned to Level 1 (relative minimal), Level 2 (intermediate), or Level 3

(extreme). The Baldus measure of heinousness - based on aggravating and mitigating circumstances - and the

visceral measure of heinousness produce the same substantive results in the multivariate models. See Baldus ET

AL., supra note 7.

Footnote

61 The JIMS file included separate indicators for race (White, Black, Asian) and ethnic origin QHispanic), but

important clues suggested that JIMS did not distinguish between Hispanic defendants and non-Hispanic

defendants in a consistent manner. An informal examination of defendants' names suggested a problem of under-

inclusion: defendants coded as Hispanic tended to have Spanish surnames, but some defendants with Spanish

surnames were coded as non-Hispanic. Moreover, the same defendants who appeared to be miscoded often

murdered Hispanic victims, a pattern that supports the presumption of coding errors in JIMS considering the intra-

racial nature of most murder. The problem was addressed with a two-pronged approach: if a defendant was coded

as Hispanic in JIMS, then the original code remained the same; if a defendant was coded as non-Hispanic in JIMS,

PDF GENERATED BY SEARCH.PROQUEST.COM Page 17 of 22

then the defendant's name was compared to the U.S. Census Bureau's 1990 Spanish Surname List. David L. Word

&R. Colby Perkins, Building a Spanish Surname List for the 1990s - A New Approach to an Old Problem (U.S.

Bureau of the Census, Technical Working Paper No. 13, 1996), available at

http://www.census.gov/population/documentation/twpnol3.pdf. The list classifies 12,215 surnames as "Heavily

Hispanic," meaning more than 75% of Census respondents with the surname reported being Hispanic. Id. Using a

conservative standard, capital murder defendants were recoded as Hispanic if at least 80% of Census respondents

with the same surname reported being Hispanic. A total of fifty-seven defendants were recoded as Hispanic. The

substantive results remain the same regardless of whether the original or revised race codes are used.

62 JIMS criminal record data are limited to Harris County, but the DA has access to national criminal record data.

To address the problem, JIMS data were supplemented with information from the website PublicData.com,

Homepage, www.publicdata.com (last visited May 15, 2009). The public data website uses the Freedom of

Information Act and Sunshine Laws to purchase public records from states. Users are charged a fee to access

criminal record data compiled from forty-three states, including Texas. The accuracy of the website is a function of

the accuracy of state records. To test the reliability of the website, I drew a random sample of forty defendants

from my data. I then examined whether the defendants' criminal records within Harris County would be coded the

same based on the two different data sources-JIMS and the public data website. The codes matched for thirty-nine

of the forty defendants. Because the public data website appeared to be reliable, I used the website to conduct

searches on all 504 defendants in an attempt to capture convictions outside of Harris County. Among the

defendants who had a clean record in JIMS, thirteen had a prior violent conviction and thirty-two had a prior non-

violent conviction on the public data website. Data from JIMS and the public data website were merged to

measure prior violent and non-violent conviction.

Footnote

63 If data were missing in the VSMF, then HCME records were used to code the age, sex, and race of the victim.

64 Searches were conducted on the public data website for all victims. See supra note 62 for more information on

the website. The data do not include enough Asian defendants or Asian victims to produce robust parameters. To

preserve the population of cases, Asian defendants and Asian victims are included in the multivariate models. But

the parameters for Asian defendant and Asian victim are reported in table footnotes and should not be interpreted.

65 Clustering occurs because multiple defendants are often indicted for the same crime. To adjust for the

clustering of defendants within cases, robust standard errors were calculated using the sandwich estimator in

Stata. Adjusting for clustering produced no substantive changes in the parameter estimates or ? values for legal

counsel. William Rogers, Regression Standard Errors in Clustered Samples, 13 Stata Technical Bull. 19 (1993);

Jeroen Weesie, Seemingly Unrelated Estimation and the Cluster-Adjusted Sandwich Estimator, 52 Stata Technical

Bull. 34 (1999).

66 See, e.g., Richard A. Berk et al., Reply to Bollen, Firebaugh, and Rubin, 25 SOC METHODOLOGY 481 (1995)

[hereinafter Reply to Bollen]; Richard A. Berk et al., Statistical Inference for Apparent Populations, 25 Soc

Methodology 421 (1995) [hereinafter Statistical Inference]; Kenneth A. Bollen, Apparent and Nonapparent

Significance Tests, 25 Soc. Methodology 459 (1995); Charles D. Cowger, Reply to Allen Rubin's Significance

Testing with Population Data, 59 SOC Serv. Rev. 521 (1985); Charles D. Cowger, Statistical Significance Tests:

Scientific Ritualism or Scientific Method? , 58 Soc. Serv. Rev. 358 (1984); Glenn Firebaugh, Will Bayesian Inference

Help? A Skeptical View, 25 Soc. Methodology 469 (1995); Donald B. Rubin, Bayes, Neyman, and Calibration, 25 Soc.

Methodology 473 (1995).

Footnote

67 Berk, Statistical Inference, supra note 66, at 426.

68 Id. at 426-28.

69 Bollen, supra note 66, at 464, 468.

70 As indicated, the purpose of this Article is to describe population parameters, not draw inferences. Nonetheless,

I argue that the disparities which existed in Harris County in the 1990s probably also existed across the rest of the

PDF GENERATED BY SEARCH.PROQUEST.COM Page 18 of 22

state, and the disparities which existed in Harris County in the 1990s probably persist in Houston today. Such

claims are logical suppositions rather than data driven inferences.

Footnote

71 The example assumes, for the purposes of argument, that the death sentence rate for defendants with mixed

counsel and appointed counsel remains the same over time.

Footnote

72 PAUL D. ALLISON, LOGISTIC REGRESSION USING THE SAS SYSTEM: THEORY AND APPLICATION 41-48 (1999).

Footnote

73 Flood, supra note 18.

74 The Chief Counsel was careful to note that his perception is based on anecdotal evidence. E-mail from Scott

Durfee, supra note 51.

Footnote

75 The 129 cases in which the DA sought death cannot be modeled separately because the group is not large

enough to support the number of independent variables. But the conditional probabilities reported in Table 8

provide insights regarding the effect of legal counsel on the decision to impose death among cases in which death

was sought.

Footnote

76 Linear interpolation was used to estimate median household income in Harris County in 1995. The Census

Bureau reports that median household income increased from $30,970 in 1989 to $42,598 in 1999. U.S. Census

Bureau, Harris County: Texas Quicklinks, http://quickfacts.census.gov/qfd/states/48/482011k.html (last visited

May 15, 2009). The year 1995 was chosen because it is the mid-point of the range of years considered in the

current research.

77 The only truly wealthy defendant appears to be Robert Angleton. Robert Angleton was accused of hiring his

brother, Roger Angleton, to kill his wife Doris to prevent her from acquiring his considerable fortune in a pending

divorce. Doris was killed in her River Oaks home, unquestionably the most affluent neighborhood in Houston

(Robert's census block group had the highest median household income of any defendant by a large margin).

Robert was a business owner with real estate holdings, though he also worked as a bookmaker and had millions of

dollars stored in safety deposit boxes at different banks (Robert was also a confidential informant for the Houston

Police Department vice division). The evidence against Robert was strong. Roger was found with a tape that the

prosecution claimed contained a conversation between the brothers regarding how to kill Doris, though the

defense had experts who disputed whether the voice on the tape was truly Robert. The State's case became more

difficult after Roger committed suicide in jail and left a note claiming that Robert was innocent. Roger maintained

that he killed Doris without prompting in order to blackmail Robert into paying him money that was owed. Robert,

who had hired counsel, is the only defendant in the data who was acquitted in a case in which the DA sought the

death penalty. For a summary of the Angleton case, see Steve Brewer et al., Jury Hands an Acquittal to Angleton:

Wealthy Ex-Bookie Cleared of Role in Wife's '97 Slaying, Hous. Chron., Aug. 13, 1998, at Al.

Footnote

78 The substantive findings remain the same if the mean is used to calculate the average. The means of median

household income for defendants with appointed, mixed, and hired counsel are: $27,780, $31,147, and $36,800.

The mean of hired drops from $36,800 to $30,490 if the most extreme outlier, Robert Angleton, is removed from the

calculation. See supra note 77 for further discussion of Angleton.

Footnote

79 Edward Green &Russell P. Wakefield, Patterns of Middle and Upper Class Homicide, 70 J. Crim. L. &Criminology

172, 175 (1979). Some might argue that the real issue surrounding socioeconomic status and capital punishment

is the definition of capital murder - the forms of murder that the poor tend to commit in the context of predatory

crime are defined as capital murder, but the forms of murder that the rich tend to commit in the context of

PDF GENERATED BY SEARCH.PROQUEST.COM Page 19 of 22

corporate crime are not. See, e.g., Böhm, supra note 5, at 305-06. This argument raises a host of fascinating

questions regarding the social construction of crime, the role of discretion in punishing the powerful and the

powerless, and the meaning of intent. But such questions are beyond the scope of the current research.

80 Mark Cooney, The Decline of Elite Homicide, 35 Criminology 381, 388-92 (1997).

81 Death Penalty Info. Ctr., Innocence: List of Those Freed from Death Row,

http://www.deathpenaltyinfo.org/innocence-list-those-freed-death-row (last visited May 15, 2009). Defendants on

the list fall into one of the following categories: the conviction was overturned and the defendant was acquitted at

retrial; the conviction was overturned and all charges were dropped; or the defendant was given a pardon based on

evidence of innocence. Id.

Footnote

82 For more on wrongful conviction and wrongful execution, see PREJEAN, supra note 3; MICHAEL L. RADELET ET

AL., IN SPITE OF INNOCENCE: ERRONEOUS CONVICTIONS IN CAPITAL CASES (1992); BARRY SCHECK ET AL.,

ACTUAL INNOCENCE: WHEN JUSTICE GOES WRONG AND HOW TO MAKE IT RIGHT (2003); Samuel R. Gross et al.,

Exonerations in the United States 1989 Through 2003, 95 J. Crim. L. &Criminology 523 (2005); Michael L. Radelet

&Hugo Adam Bedau, Miscarriages of Justice in Potentially Capital Cases, 40 Stan. L. Rev. 21 (1987); Michael L.

Radelet &Hugo Adam Bedau, The Execution of the Innocent, in AMERICA'S EXPERIMENT WITH CAPITAL

PUNISHMENT, supra note 5, at 325.

83 Ernest Van Den Haag, Justice, Deterrence and the Death Penalty, in AMERICA'S EXPERIMENT WITH CAPITAL

PUNISHMENT, supra note 5, at 233.

Footnote

84 JENNIFER M. SAUBERMANN &ROBERT L. SPANGENBERG, THE SPANGENGBERG GROUP, STATE AND

COUNTY EXPENDITURES FOR INDIGENT DEFENSE SERVICES IN FISCAL YEAR 2005 (2006), available at

http://www.abanet.org/legalservices/sclaid/defender/downloads/

FINAL_REPORT_FY_2005_Expenditure_Report.pdf; Task Force on Indigent Def, Financial Report,

http://tfid.tamu.edu/public/default.asp (follow "Expense Report Results" hyperlink; then select "2005" reporting

period) (last visited May 15, 2009). Per capita spending on indigent defense (PCSID) is calculated as follows: total

expenditures divided by population (total expenditures are listed id. at 35-38; state populations are listed in the

brief discussion regarding indigent defense in each state, id.). Harris County, with a population of 3,673,089, spent

a total of $19,361,692 on indigent defense (the expenditure figure is the sum of appointed counsel fees,

investigation expenses, expert witnesses, and other litigation expenses). To be clear, neither data source

disaggregates indigent defense and indigent capital defense. But, it seems reasonable to assume a strong

correlation between per capita spending on indigent defense and per capita spending on indigent capital defense.

85 DEFRANCES &LITRAS, supra note 13.

86 BOHM, supra note 5; PATERNOSTER et al., supra note 5.

Footnote

87 Stephanie Hindson et al., Race, Gender, Region and Death Sentencing in Colorado, 1980-1999, 77 U. COLO. L.

REV. 549, 573 (2006).

88 BOHM, supra note 5, at 270.

89 PATERNOSTER ET AL., supra note 5.

90 David R. Dow, Teague and Death: The Impact of Current Retroactivity Doctrine on Capital Defendants, 19

Hastings Const. L.Q. 23, 72 (1991).

91 Id.

91 Id.

93 Id.; BUTCHER &MOORE, supra note 31; TEX. APPLESEED FAIR DEF. PROJECT, supra note 19, at 37.

Footnote

94 TEX. APPLESEED FAIR DEF. PROJECT, supra note 19, at 35.

95 Id. at 21-22.

PDF GENERATED BY SEARCH.PROQUEST.COM Page 20 of 22

96 Butcher &Moore, supra note 31.

97 For more on Public Defender Offices, see Rebecca Copeland, Getting It Right from the Beginning: A Critical

Examination of Current Criminal Defense in Texas and Proposal for a Statewide Public Defender System, 32 St.

Mary's L.J. 493 (2001).

AuthorAffiliation

SCOTT PHILLIPS*

AuthorAffiliation

* Department of Sociology and Criminology, University of Denver, 2000 E. Asbury Avenue, Denver, CO 80208-2948.

E-mail: [email protected]. I am indebted to Scott Durfee, Chief Counsel to the Harris County District Attorney,

who explained numerous office processes and provided archival data. I am also indebted to Kim Bohannon from

the Harris County District Clerk's office who made the research possible by providing data from the Harris County

Justice Information Management System (JIMS). I would like to thank Russ Curtis for discussions about capital

punishment in Texas during the earliest stages of the project. For providing comments on earlier versions of this

manuscript, I thank Mark Cooney, Bob Böhm, David Dow, and Mike Radelet. For technical assistance in Stata, I

thank Lisa Martinez. Most importantly, I could not have completed the project without help from the following

undergraduate Research Assistants: Breck Garrett, Mor Regev, Shaefali Pillai Rogers, and Ivan Zapata. DETAILS

Subject: Capital punishment; Public defenders; Attorneys; Trials; Political appointments

Publication title: Journal of Criminal Law &Criminology; Chicago

Volume: 99

Issue: 3

Pages: 717-755

Number of pages: 39

Publication year: 2009

Publication date: Summer 2009

Section: CRIMINOLOGY

Publisher: Northwestern University (on behalf of School of Law)

Place of publication: Chicago

Country of publication: United States, Chicago

Publication subject: Law, Law--Criminal Law, Criminology And Law Enforcement

ISSN: 00914169

CODEN: JCRLA

PDF GENERATED BY SEARCH.PROQUEST.COM Page 21 of 22

Database copyright  2019 ProQuest LLC. All rights reserved. Terms and Conditions Contact ProQuest

Source type: Scholarly Journals

Language of publication: English

Document type: Feature

Document feature: References

ProQuest document ID: 218404715

Document URL: http://0-

search.proquest.com.wizard.umd.umich.edu/docview/218404715?accountid=14578

Copyright: Copyright Northwestern University School of Law Summer 2009

Last updated: 2016-04-23

Database: Sociology Database,Sociological Abstracts,Research Library

PDF GENERATED BY SEARCH.PROQUEST.COM Page 22 of 22