Taylor says alleged advantage to bail bonds is not backed by any evidence

Unaccountable, armed bounty hunters breaking down doors in Milwaukee, profiting on the criminal justice system.

State Sen. Lena Taylor (D-Milwaukee) raised that specter after a last-minute Republican budget action on June 5, 2013 that would authorize private bail agents to work in five Wisconsin counties at a judge’s discretion -- Milwaukee, Waukesha, Racine, Kenosha and Dane.

"Bail Bonds are back," Taylor wrote in a June 7, 2013 opinion piece in the Milwaukee Courier, a weekly newspaper. "Since the early 1970’s private bail bondsmen and bounty hunters have been illegal in Wisconsin. In the 2011 budget, Republicans tried to introduce them back to Wisconsin, but Governor Walker wisely vetoed them out of the budget."

Taylor then took on one of the key talking points used by bonding companies in favor of the GOP plan.

"There is no statistical evidence that bail bonds increase the likelihood of those awaiting trial returning to court for their scheduled hearings, etc.," Taylor wrote. "In fact, this provision is widely opposed by almost everyone in the criminal justice system in Wisconsin, especially in Milwaukee."

Wisconsin hasn't had a commercial bail bonds system since 1979, and many leaders within the state's justice system want to keep it that way.

A key issue in the debate is whether use of a bail bondsman gets more defendants to show up for court, so we wondered about Taylor’s claim that "no statistical evidence" supports the notion that bail bonds bring that about.

How bail bonds work

National research suggests that about 60 percent of felony defendants in large counties are released while their cases are pending; the figure was 75 percent in Milwaukee County during a period studied in 2012. Studies focus on felony cases.

Under the proposed bail bonds system, defendants would pay a bail agent 10 percent of the full cash bail amount. The company must pay 3 percent of the bail amount to the court. If the defendant fails to appear in court, the bonding company is liable for the full amount. But a bail agent usually has an opportunity to recover a no-show defendant using bounty hunters.

These "private bail bonds" or "surety" bonds are in use in 46 states and long have been the most common form of pretrial release.

Types of financial release include full cash bond, and in some states but not Wisconsin, "deposit bonds" with a partial bail paid to the court. The most common types of non-financial release are personal recognizance and conditional release, the U.S. Bureau of Justice Statistics said in a 2010 report.

In Milwaukee County, for example, 60% of felons in the 2012 study were released on personal recognizance bonds, meaning they aren’t liable to pay unless they fail to appear. The other 40% had to pay cash -- or stay in jail before trial.

Bail companies say their work needs no taxpayer funding and enhances public safety.

Critics say it’s an unfair monetary burden on lower-income people, and Milwaukee County judges say they are in a better position, using sophisticated risk analyses of defendants, to judge whether the accused should be held or walk free while their cases are in process. Milwaukee County’s no-show rate in that 2012 study period was 16 percent in felony cases.

Taylor’s claim

Asked for backup, Taylor’s office pointed us to a widely cited Bureau of Justice Statistics report issued in 2007. The arm of the Justice Department examined the pretrial releases of 71,000 state-court defendants in eight years from 1990 to 2004 in the 75 largest U.S. counties, including some Milwaukee County cases.

But that report cites an advantage for bail bonds, undercutting Taylor’s claim rather than bolstering it.

That study said the no-show rate for those on bail bonds was 18 percent, vs. 30 percent for those released on unsecured bond. The no-show rate was lowest for defendants who pledged property equal to the value of their bail.

In 2008, one of the authors of that study, Thomas H. Cohen, published a narrower follow-up study using the same county-based federal data.

He found that no-show rates were twice as high in five U.S. counties that make little use of private bail bonds, compared to five counties that use the approach a lot. No Wisconsin counties were included in that study.

Surveying the research as a whole on the topic, Cohen wrote: "Various studies conducted at both the academic and non-academic levels have also found that surety agents outperform other kinds of pretrial release at preventing court skips."

Case closed?

Not quite.

Researchers on both sides say a key question is whether this correlation between bail bonds and a lower no-show rate means that the type of bond causes the disparity. To gauge that, local differences in bail-setting practices must be accounted for.

Featured Fact-check

David A. Clarke Jr.

stated on September 30, 2011 in a radio interview

Says he’s "never asked for more" than inflationary budget increases in his 10 budgets as Milwaukee County sheriff

The Bureau of Justice Statistics itself issued an advisory to researchers and the media in 2010 warning that its data "are insufficient to explain causal associations between the patterns reported, such as the efficacy of one form of pretrial release over another."

The advisory said factors associated with a defendant’s pre-trial misconduct also would have to be considered -- things like a defendant’s community ties, employment status, income, educational background, drug abuse history, and mental health status.

Various studies have tried to determine what causes the bail-bond advantage by comparing similarly situated defendants and controlling for other factors.

Perhaps the most widely cited study of that kind was published in 2004 in the peer-reviewed Journal of Law and Economics. It was authored by Alexander Tabarrok and Eric Helland, economists at George Mason University and Claremont McKenna College, respectively.

The study suggested that defendants released on private bail bonds are 28 percent less likely to fail to appear than similar defendants released on their own recognizance (no money bail). It also found that if defendants released on bail bonds do no-show, they are 53 percent less likely to remain a fugitive for extended periods.

Tabarrok is the Bartley J. Madden Chair in Economics at the Mercatus Center, a free-market think tank at George Mason that has received major funding from lightning-rod conservative industrialist Charles Koch, a Mercatus director. Tabarrok is also research director at the Independent Institute.

Helland is an economist at Claremont McKenna, a former economic adviser to the George W. Bush administration and is currently a researcher at the  RAND Institute for Civil Justice, a non-profit research organization that seeks a more efficient and more equitable justice system.

Critics of private bail bonds, including the Pretrial Justice Institute, point to the Bureau of Justice Statistics own advisory and say that only local studies can draw hard conclusions on the reasons for varying no-show rates.

The institute, funded mainly by the Justice Department and the liberal Public Welfare Foundation, advocates to reduce pre-trial detention when assessments predict that defendants pose no threat and will return for court appearances. It cites other factors necessary to judge the value of different release methods.

"For example, do we want potentially dangerous defendants to buy their way out of jail? Do we want low risk indigent defendants to take up expensive jail space because they cannot afford the services of a bail bondsman?"  the group asked in a 2010 paper.

Our rating

Taylor said "there is no statistical evidence that bail bonds increase the likelihood of those awaiting trial returning to court for their scheduled hearings."

There’s no definitive evidence, to be sure, but various statistical studies have shown that bail bonds are associated with, and may be a cause of, lower no-show rates.

The studies are not without their limitations, and don’t settle the broad policy argument over bail bonds, but Taylor’s blanket assertion is off base.

We rate her claim False.