The Numbers that Tell the Real Story

We rely too much on aggregate data about our cities and states. It’s the differences among the numbers that are truly important.
by | March 19, 2012 AT 11:00 AM

All of us tend to do it: We use a kind of shorthand to talk about things in a way that relies on the over-aggregation of data. As in "the American people want..." or "cities are in for some tough times in 2012...." But in fact we know that there are vast differences in what the American people want or the circumstances cities will face in the coming year. Understanding those differences and the factors driving them is the key to development of effective policies.

Being careful about data in this way is a lesson I first learned from Harry Hatry, director of the Public Management Program at the Urban Institute and a co-author of the classic text "How Effective Are Your Community Services?" One of his principle lessons is to beware of aggregate data. He frequently uses the metaphor of "the man who drowned in a lake with an average depth of one inch."

A recent email exchange I had with Mike Eglinski, city auditor of Lawrence, Kan., brought Harry's warning to mind again. Mike did a simple analysis that suggested the rather startling scope of differences in the financial conditions of city governments across the nation.

Mike looked at Lawrence and 14 other similar cities (all college towns, all of them similar in population size, per-capita income and age of housing), calculating their net obligations for Other Post-Employment Benefits (retiree liabilities in addition to pension benefits) as a percentage of total expenses for the latest year. The median net OPEB obligation he found was 1.4 percent. But "in a group of just 15 similar cities," he wrote, "the range was from 0.08 percent to 8.82 percent (in other words, the largest net OPEB obligation was more than 100 times the smallest)." He also found a lot of variety within a state. "My list has 2 cities in California (Davis and Chico). One of those cities is at the median, the other is 6 times larger."

A broader, deeper analysis than the one Mike did is likely to find similarly profound differences among cities and, I suspect, among states as well. For example, a recent report by Bill Holland, the state auditor of Illinois, showed that his state was the "brokest" in the country with a statement of net assets that showed a deficit of $37.9 billion. Texas, on the other hand, had a statement of net assets with a positive balance of more than $90 billion.

I suspect that, just as the gaps among American households have grown in terms of income and wealth, so too have the differences in the financial conditions of our state and local governments.

Two sets of issues arise from this: First, the fact that the "new normal" in government is not a brief recession and quick recovery but a long period of fiscal scarcity makes it ever harder for governments to use the usual devices to skate by in a financial crunch. You can juggle the books a bit for a couple of years and defer maintenance on equipment for even longer, but you cannot do so for decades without some sort of collapse in the quality of life in your community.

Second, well-educated upper-income residents are more mobile than ever and have greater freedom to live where they choose rather than having to live where they can get a job. States and localities are not immune from market forces. They compete for residents. Financially constrained jurisdictions that suffer declines in quality of life will have less capacity to attract middle- and upper-middle-class residents and the tax base those residents represent, leading to a downward spiral in quality of life.

When you see data about states and cities — from homicide rates to unemployment rates to pension-funding status — remember Harry Hatry's warning: One size doesn't fit all. What differences lie within the numbers? What is driving those differences? What should we do about it?