"Our vision is that our audit and assurance work improves the performance of, and the public's trust in, the public sector." This is a statement from the website of the Auditor-General of New Zealand. Similar statements can be found for auditing organizations in other countries, including Canada's Office of the Auditor General and the United Kingdom's National Audit Office.

This idea, that government auditing should aim to improve the public's trust in government, is something that state, local and federal auditors in the United States have not embraced, but they should.

Trust in government is, or course, at a perilously low point in the United States, and I think that the dominant way that we do performance auditing is a contributing factor. Unlike many of the audit organizations overseas, in the U.S auditors, inspectors general and other watchdogs do not see improving trust in government as an essential objective of their work.

Consider this recent headline from the Fort Worth Star-Telegram: "In Texas, food stamps for the dead, scofflaws, cost taxpayers $500,000 a month." By American standards, this audit of Texas' administration of the food-stamp program from the U.S. Department of Agriculture's inspector general's office is pretty well done. It is concise, well documented and clearly written. But the error rate uncovered by the audit was, at worst, 12 one-hundredths of 1 percent, and at least some of the error was because of delays between the time death notices are published on the Social Security Administration's death master file and the matches run by Texas program administrators. In my opinion, and in the opinion of the agency being audited, what this audit actually found was that the program works pretty well. What about the $500,000 a month? Yes, that's a lot of money, but it's a tiny fraction of the billions the program costs.

The audit did not render a conclusion as to whether or not the program was achieving its objective of minimizing the number of ineligible participants. That's not unusual for performance auditing in America, but it's in contrast to the approach of auditors in other countries. Auditors in the United Kingdom, for example, render an opinion on whether a program is "fit for service" and whether the program delivers "value for money."

This is what we need in the United States. Auditors should, in consultation with the management of the agency being audited, decide before doing the fieldwork what criteria will be used to draw conclusions about the agency's performance. In the case of the Texas food-stamp audit, for example, what level of compliance would have led the auditors to conclude that the eligibility process was working?

What I'm asking auditors to do would take courage on their part. For a program like food stamps, for example, going to a legislative hearing in the current political environment and saying that you've concluded that the program you audited is working well would expose the auditors to considerable heat. But the entire point of auditor independence is to be able to speak truth to power. Good auditors have to have the courage to call 'em as they see 'em and be willing to stand up to legislators as well as to top management at the agency under audit.

The National Intergovernmental Audit Forum is the most important organization of auditors in the United States. Chaired by the U.S. Government Accountability Office, it is the one organization that brings together state, local and federal auditors to work on issues important to the profession. At the biennial meeting of the forum later this month I will moderate a panel with the title "Communicating and Reporting for Impact in a Digital Age." I hope we can make some progress toward changing what auditors see as their role.

Impact is the number-one thing auditors want from their work. In America, auditors need to see improving trust in government as a major impact to be achieved and shape their work accordingly.