Insight on Incentives

Plus: Paying for snitches, the power of putting data on a map, and more
May 7, 2009 AT 3:00 AM
Barrett and Greene
By Katherine Barrett & Richard Greene  |  Columnists
Government management experts. Their website is

We've written a lot about incentives in the B&G Report (including the following item), and would very much like to hear from managers about their experiences with incentives for staff that have either worked well, failed dismally or fallen someplace in between. Please e-mail us with your examples!

A snitch in time. According to a study by University of Arkansas psychology researchers Jessica K. Swanner and Denise R. Beike, providing incentives for people to snitch on their fellow men -- a common and useful technique in many police departments -- may sometimes be counterproductive.

The researchers anticipated that incentives would create an increase in valid snitching -- secondary confessions, as they're called. But according to their article in the Journal of Law and Human Behavior, that's not necessarily the case. Turned out that participants in the study were no more likely to offer such a secondary confession when it was truthful. It was only when they had to stretch the truth to offer this information that the incentives had impact. Of course, when it comes to street-corner snitches, it may be that a little cash lubrication is important. But the study provides a potent cautionary note about the validity of what's said.

The report's conclusion: "It is essential for jurors, prosecutors and judges to be informed about the potentially biasing nature of incentives to confess. Snitches may indeed lie or come to believe a falsehood about another to be the truth. Jurors must be able to consider this possibility as they make their verdicts."

Back on the soapbox. Regular readers of the B&G Report will recall that we've repeatedly suggested that one of the best ways to use performance measures -- and to make sure they're used -- is by disaggregating the data along geographic lines. Folks don't really care how the schools are doing statewide. But they're fascinated -- and may be moved to action -- when they see how their district compares to others. What's more, this kind of data permits the weak to learn from the strong.

A recent report from the Rockefeller Institute provides fodder for just such a useful examination. Apparently, there's wide variation in the rate at which counties in New York State turn down Medicaid applications for nursing-home care, based on improper asset transfers (folks who purposefully impoverish themselves exclusively to get Medicaid benefits). The rate of Medicaid denials as a result of asset transfers made up nearly half the total in Franklin County, and only 0.5 percent in Westchester County. The report calls for additional research to explain the disparities. Is Westchester overlooking a lot of flim-flammery? Or is Franklin County overreaching? Or are there other extraneous, but interesting, factors to explain the difference? In any case, solid analysis of this kind of information is one of the keys to making real use of performance data.

Quote of the month: "Progress always involves risks. You can't steal second base and keep your foot on first." Attributed to Frederick Wilcox and quoted in a recent Management Insights column by Bill Eggers and Shalbh Singh, on

Manager's Reading List: Our ongoing feature about books to read, recommended by B&G readers

This entry is from Raelene Freitag, director of the Children's Research Center in Madison, Wisconsin: "I've been keeping my eyes open for good management books relating to child welfare, but I haven't seen any for a long time. It's a pretty thin bookshelf. But there's one book that I use a lot: The Quality Toolbox by Nancy Tague. It is very practical and gives great concrete ways for managers who have a problem and want to solve it. It shows how to start walking through the data that's available to make decisions about the solutions you want to try. It's not a book to read, but more of a reference book to use. I'm a huge fan of the approaches to quality improvement in that book."

Read the full archive of Manager's Reading List suggestions.

Vacancy rates at state health departments range from 2 percent to 17 percent. And 22 states have a vacancy rate of 10 percent or higher. That's according to a November-December survey by the Association of State and Territorial Health Officials (ASTHO). We know that some of these vacancies are the result of too few qualified candidates for rarified jobs. But it's pretty clear that many of these vacancies -- and the number is rising -- are the result of budget reductions. The ramifications could be pretty unpleasant. Take a look at "Shortchanging America's Health," a March report from the Trust for America's Health.

Our most recent Smart Management column talks about some reasons for frightening budget overruns on infrastructure projects. Subsequent to finishing that column, we came across an Associated Press article by Doug Simpson that points to some reasons for these problems, as illustrated in a new Mississippi River bridge project upriver from Baton Rouge, an improved Earhardt Boulevard in New Orleans and wider highways in rural Louisiana.

According to Simpson, "The original projections were so far off that William Ankner, state transportation secretary since last year, has openly scoffed at gross underestimation of inflation and the rising prices of steel, labor and construction materials. The most glaring example: widening the Huey P. Long Bridge in Jefferson Parish was initially estimated at $50 million -- but is now projected to cost $1.1 billion, more than 20 times the original estimate.

" 'I think they were apocryphal estimates,' Ankner told a panel of lawmakers recently. 'Well, they were basically reached out of thin air in some cases.'"

One of the biggest flaws in the estimating methodology? The forecasts became surreal after hurricanes Katrina and Rita hit. Then paying for the projects became even more problematic when some risky financing arrangements didn't work out. That's one good reason not to plan to pay real bills for materials and labor with financing arrangements that are somewhat less concrete.

Some months ago, we did a hearty number of interviews with people who were experts in economic development for the states. Pretty much all of them agreed that tax givebacks weren't really a great way to attract new business. There were many other, more effective things much higher on the list. In a completely different arena, we interviewed a few site-selection company representatives in the past week. All of them emphasized that states or cities had better offer tax givebacks if they were going to have any hope of drawing new businesses. We wonder how to reconcile the two.

Trust but verify. How many cities, counties and states devote the time and money to properly oversee the work done by service contractors? We strongly suspect that it's far too few. Consider this story from Milwaukee County, uncovered by a recent county audit. A security firm had been hired to monitor the transit system, according to the Milwaukee Journal Sentinel. The original contract called for 85 percent of the contracted-for hours to be spent actually riding the buses. That may have been somewhat higher than was necessary. But when the auditor took a look, it turned out security firm employees were only spending 3 percent of their time on buses -- that's less time than we spend on Amtrak most weeks, and we're not a security firm.

Transit officials pointed out that things had already improved somewhat, and the percent of time on buses was now up to 30 percent. But Sheriff David A. Clarke, Jr., argues that the number should still be far higher. "It's a travesty," he told the Journal Sentinel. "The taxpayers aren't getting what they are paying for."

Credit goes to the audit for verifying the problems raised by others. Other entities can learn a lesson here.

With nearly 40 years experience in working in or with the public sector, Jim Moore, director of government programs for The Rensselaerville Institute (TRI), has thought a lot about ways in which performance targets can be more effectively used in contracting. One of the big mistakes that governments make, he says, is to select outcomes, targets and delivery methods for a contract without providing flexibility or choice to providers. When goals are set out of the blue, it leaves contractors little chance of success, he says, citing his own experience as the executive director of a group of workforce agencies in Missouri in the 1990s.

"Missouri was transitioning from welfare to work and if you wanted to bid, you had to commit that 80 percent of the people they sent to you would be placed in jobs and you had to keep them there for 12 months," he says. "There was no rationale for that number. The best the state had been able to do was 46 percent placement success. It was totally irrational."

In his work with TRI, Moore advises governments to aggressively train providers on the outcome terms that might be used in the contract and then negotiate the outcomes together, including both targets and milestones that lead up to them. Once a contract is launched, he believes the government should view the money it puts into a contract as an investment and that the different parties to the contract should work in partnership to make sure that investment gets its desired returns. For more information about the TRI approach, take a look at its Web site.

Research Assistant: Heather Kleba