States work on reducing errors - food stamp program

Anyone who's ever run a business--or managed a family budget, for that matter--knows mistakes can be costly. When you're talking about a $12-billion operation like the Food Stamp Program, mistakes can cost plenty, and not just in terms of money. They can also jeopardize the public support the program needs to survive.

At the national level, USDA has taken a number of steps to help eliminate the source of errors. For example, to reduce the incidence of overpayments caused by inaccurate reporting of income, federal rules now require states to match their food stamp rolls with state employment records. Some states, like Massachusetts, are using similar techniques to comb bank records of food stamp applicants in search of unreported assets.

But while the common belief is that most errors are caused by recipients committing fraud, such as hiding assets and income and lying about the number of people in their households, this is not the case.

The majority of errors are due to a variety of other causes, such as math errors by eligibility workers, failure of eligibility workers to apply food stamp rules correctly, computer errors, and unintentional mistakes by food stamp recipients in reporting income. These are problems that need to be tackled--and are being tackled--by state and local food stamp managers.

While error reduction tactics vary from state to state, there are similarities in the way managers are approaching their problems. The following examples from five Western states illustrate several approaches that are getting results: Nevada

Nevada has been a leader in eliminating errors caused by eligibility workers' mistakes. The state has maintained one of the nation's lowest error rates--ranging from 1.5 to 3 percent in the past 2 years.

Nevada has maintained this low error rate, which has earned them several awards from USDA, by directly linking performance standards for workers to the number of errors they make in a case.

Supervisors of eligibility workers inspect a sample of 25 cases per month to see if there are any mistakes. Acceptable error rates are clearly defined--0 to 0.5 percent is outstanding; .6 to 1.5 percent is above standard; 1.6 to 4 percent is acceptable; 4.1 to 6 percent is a "must improve"; and above 6 percent is unacceptable.

Special training and reviews are provided to workers who show up with ratings below acceptable. If, after special training and counseling, the worker continues to have an unacceptable error rate, then the worker can be fired.

"This system works well because workers will never receive a surprise on how they are doing," says June Young, a Nevada official who has worked with the performance standard system. "We are very clear about what is expected of them."

As punitive as this system may sound at first, Nevada has had to fire few workers and actually has a low staff turnover rate. The key here is strong support to insure that the worker is not left without guidance.

Food stamp managers in Nevada devote a great deal of time and energy to making sure that the policy manuals given to workers are written as simply as possible. The manuals help workers understand program policy, and they are valuable reference tools for working on difficult cases.

The state has also developed a computer system that performs many routine calculations that were previously done by workers. This removes tedious chores and lessens the chance of errors. Arizona

Arizona has taken a similar tact to reducing error rates, but with some key differences. Faced with federal sanctions for an excessive error rate, Arizona implemented a quality assurance system in May 1982.

As in Nevada, eligibility worker supervisors review a sample of completed cases. However, in Arizona the reviews are completed before case information is put into the computer and food stamps are issued.

"The supervisory reviews help us determine what problems a worker is having," says Diane Ross, who heads Arizona's Food Stamp Program. "Completing the reviews before benefits are issued in a case prevents loss to the program and means we don't have to go through the claims process."

Of course, the quality of supervisory reviews depends on how much supervisors know. To make sure supervisors were catching all the errors, Arizona set up a small independent quality assurance unit to check supervisors' reviews. State directors found that supervisors' emphasis on management often left them weak in knowledge of the subtleties of program rules.

"Many supervisors didn't know all the details of eligibility rules," says Ross, "so not all errors were being found during the first-line inspection. The second-line review told us where we should provide special training for supervisors, and it was a good way to make sure policy was enforced the same way in different offices." Hawaii

Hawaii has discovered a way to ensure the uniform application of policy, using an adaption of the Japanese "quality circle."

Every time an error is found, a worker must prepare a written report analyzing the causes of the error. But the process does not end there. The worker must discuss his report during part of the monthly staff meeting. Workers then see what mistakes others are making to see if they have the same misunderstandings of policy or procedure.

Larry Higa, a section administrator for the island of Oahua, explains the philosophy of this approach: "What is the sense of just one person correcting an error since other workers are often making the same error or misapplying the same policy?

"Almost immediately after we began these meetings, workers became more aware of the types of errors that were occurring in the unit and the causes. Workers got a better handle on how to remedy problems."

While there was some initial resistance to the practice of discussing errors in open staff meetings--who likes to discuss their mistakes in front of their colleagues?--workers soon began to see that the practice was not designed to be punitive.

"Initially this system was unpopular with the workers. They felt they were being blamed for all the errors," says Higa, "but we worked with supervisors to instill a sense that we weren't singling anyone out and simply wanted to solve the problem of recurring errors." Alaska

Several states are developing special techniques to focus on client errors. Alaska is one state that has had problems with extremely high error rates. At one time, one quarter of food stamp dollars in Alaska were issued incorrectly, and state managers attributed much of it to client error.

To solve this problem, they began an intensive program of home visits to make sure client circumstances were exactly as stated on food stamp applications.

During home visits, eligibility workers verify: that the household members listed on the application actually reside at the address given; that there are no unlisted household members; that income is as declared; and that the household situation generally conforms to what is listed on the application.

"Initially home visits were made in cases where the eligibility worker or a supervisor suspected a problem," explains Bob Clem, one of the Alaska officials responsible for the Food Stamp Program.

"Later, we concentrated our visits on types of cases that frequently had errors, such as households that declared no income or that had employable adults who were not working. Now we review many cases that are simply randomly selected."

The results of the state's efforts are hopeful--from an error rate high of nearly 25 percent, Alaska has lowered its error rate to 13.9 percent, a 44-percent reduction. While the current rate is still well above the national average, Alaska seems to be well on the way to solving many of its error rate problems.

The cost of systematic home visits may not be for every state, however. Says Alaska official Clem, "Home visits are far and away the best way to go in reducing client error, but it's very expensive. The commitment of large amounts of staff time makes the system extremely costly." Washington

Finally, the state of Washington is working to reduce client error by improving the interview techniques of workers. State managers have produced a 30-minute videotape called "Focused Interviewing."

The training in the tape shows workers that by simply asking clients direct and clear questions about household circumstances, they can often elicit information that was not immediately available.

Washington officials tried this approach for several reasons. With the complicated definitions and deductions in welfare programs, and with the irregular sources of income that low-income people often have, it is very easy for clients to overlook or misunderstand the different types of questions on applications. Clear questioning by workers can often clear up such misunderstandings.

Direct questioning is also a way to deal with problems caused by the minority of clients who are trying to hide income or resources. Washington officials have found that much of this kind of cheating is simply omission, and clients will respond truthfully to direct questions.

All of these techniques, from home visits to focused interviewing and supervisor reviews, have contributed to the nationwide reduction in the food stamp error rate.

For more information, write: Director, Food Stamp Division Western Regional Office Food and Nutrition Service, USDA 550 Kearny Street San Francisco, California 94108

COPYRIGHT 1984 U.S. Government Printing Office
COPYRIGHT 2004 Gale Group