Scandalous overrreaction?

The various scandals over spending on government conferences, and other problems within the IRS, have put a spotlight on an important question: Does the attention given such happenings affect the quality of government management?

For the inspectors general whose reports typically follow these scandals and, I dare say, for the public, the answer is clear: Exposing scandal raises the price for wrongdoing and incompetence, and if you raise the price of something, you get less of it. Exposure of scandals is thus good for government management. QED.

Yet I am guessing that many government managers themselves – and, perhaps more importantly, outside experts on government management – are not so sure.

The downside of scandal was illustrated in concrete terms in a recent op-ed by Defense Logistics Agency manager Joe Bednar, entitled "Enough Bureaucracy Already," appearing in Federal Times. Bednar notes that in his agency, routine actions to complete a financial obligation went through 12 reviews as of 2010. To an outsider, that sounds like a lot. Obviously, there need to be checks and balances on dispersing money – but are 12 approvals really needed?

Wait… that was 2010. Now the number is more than 40!

In 2011, agency financial transactions procedures were revised – with the regulations increasing from 300 pages to 6,000.

A "simple letter to assign specific oversight responsibilities" used to require only one signature – now it requires three signatures, several certifications and reporting responsibilities.

We know where all this comes from. Scandals occur, and either Congress demands or agencies proactively impose new controls – that is, more approvals, documentation and so forth. After the GSA conference scandal last year, bills were introduced in Congress that would actually require the head of an agency to approve personally all conference spending. Though this never passed, the restrictions agencies have implemented on their own are onerous enough.

Maybe one way to answer whether this is too much is if I posed the following hypothetical: if an agency currently requires only two signatures on a financial obligation document, would it be a good idea to double it?

If the document currently requires four signatures, would it be a good idea to double it to eight?

Or if it currently requires 12 signatures – to take the real story from the Bednar column – would it be a good idea to increase it to 40?

The point of all this is that it is always easy to ask for more controls, but at some point this starts adding only marginal additional protection, while costing the organization more and more staff time, slowed decision-making, and staff demoralization. Indeed, when you get to these huge numbers of signatures, the actual control level is likely to be lower than with few signatures – due to the famous "diffusion of responsibility" effect, studied in social psychology. Simply put, people reduce their vigilance when a larger number of people are responsible for some outcome, each one assuming that the others will take a closer look.

Time to say it: Requiring anything even close to 40 reviews for approving a payment is absurd. Will anyone dare call a stop to this?