Yesterday I attended a presentation by ACSI and ForeSee, offering federal agencies the results of the customer satisfaction surveys they use to measure "citizen-customer" relations. Most of the people in the room were investigating the results of the surveys as they relate to the agencies they represent, since customer satisfaction surveys are a metric by which their work is evaluated.

The studies confirmed two things that I suspected were true, but are more helpful when statistically verified. First, there is an expectations gap between what American citizens think about their government, and how satisfied they are with the government’s services when interacting with them. In other words, there is an unjustified pessimism toward government’s effectiveness, when judged by comparing self-reported expectations with self-reported satisfaction scores. This strikes me as very similar to congressional approval ratings, where Congress is historically consistently very low, and members are ranked very high for their own districts.

Second, and perhaps more relevant to our work, is that interactions online are apparently "the most satisfying piece of citizens’ interaction with the federal government." (quoted from the ForeSee E-Government Satisfaction Index, December 17, 2007.) As someone who is personally inclined to look for services online first, I’m not surprised to see others reporting that Web-based interactions with government are more satisfying. It should be helpful for convincing others about the importance of their Web site, or the importance of IT spending, to be able to point to real citations about citizens changing expectations of government.

The federal employees have a tough job ahead of them, stuck between the enormous piles of data and complex services, representing unwieldy legacy systems unchecked by market forces, staring into sets of statistical analysis about how government Web sites compare. Very different systems being compared with similar criteria.

Not that they shouldn’t be. Watching this particular form of accountability brought the main difference between the executive branch and the legislative branch into stark relief: the executive is supposed to be the efficient tyranny — the top down dictatorship of execution — controlled by the objective legislature — the staid bottom up deliberator. The big difference comes in accountability. The executive is built to be controlled and measured (whether this works in practice or not), and the legislature is built to be independent and accountable only electorally. When congressional oversight of the executive, and electoral oversight of the legislative both prove insufficient, popular pressure is all that is left as a check.

This raises the question: to what degree are the separation of powers devolving into built in accountability mechanisms like Inspectors General, or ethics committees?