Don’t Read Too Much into Census Numbers (Milwaukee)

Alan Berube and
Alan Berube Interim Vice President and Director - Brookings Metro

Bruce Katz
Bruce Katz Founding Director of the Nowak Metro Finance Lab - Drexel University

July 19, 2005

Is Milwaukee really hemorrhaging people?

New estimates from the U.S. Census Bureau indicate that on population, at least, recent declines in Milwaukee leave it outside the top 20 cities in the United States for the first time in decades. And that finding seems to have precipitated some soul-searching among the city’s elected officials and business leaders.

Our advice to Milwaukeeans: Take these numbers with a grain—if not a teaspoon—of salt.

In 1999, the Bureau’s Population Estimates Program—the same unit that produced the most recent estimates—figured the city’s population at 572,000. That signaled an alarming 9.8% decline from its 1990 population of 628,000.

With the advent of Census 2000, however, the city worked closely with the federal government to ensure a full and accurate count of its residents, and the results were quite a surprise. The city’s 2000 population came in at 597,000—representing a decline in the 1990s but one only about half as bad as the estimators predicted.

Now come the Census Bureau’s 2004 estimates, which show Milwaukee losing about 12,000 people in the first four years of this decade. Similar trends reported for other cities have generated newspaper headlines such as “Inner-city resurgence is over” and “Cities losing residents after ’90s influx” from coast to coast.

Are the numbers any better this time around?

Realize first that the Census Bureau produces population estimates for some 40,000 places across the U.S. every year and probably does the best it can with the data it collects. But local experts often have access to more detailed information with which to generate estimates. On that count, Wisconsin’s state demographer finds that Milwaukee’s population has basically held steady since 2000.

It turns out that the same methods that produce good estimates in suburban counties like Waukesha, which grow largely through the addition of new housing, may produce less-reliable estimates in places like Milwaukee that are adding new immigrants and their children and reclaiming vacant land and property.

Like Denver, Oakland, New York and other big cities, some of the Census Bureau’s methods effectively “cheated” Milwaukee out of population gains in the their mid-1990s estimates—and they may be doing the same thing this decade.

It’s also worth asking: Do these estimates really matter?

They do.

First, they help frame a national debate about the health of urban areas. Ten years is a long time between counts, and population change is one of the few indicators available for cities during that interval.

Second, the estimates drive program and planning decisions at all levels of government—including the allocation of federal and state funding to localities. That’s why 20 to 40 cities challenge the census estimates each year.

But when it comes to assessing the well-being of cities, size isn’t the only thing that matters. In fact, we’d argue that household change is a better indicator of a city’s trajectory than its population change. Housing demand, size of the tax base and local services needs are all better measured through household estimates than population estimates.

On this count, local estimates indicate that Milwaukee is holding onto its households. Moreover, increased population doesn’t necessarily mean increased well-being. As researcher Paul Gottlieb has noted, the Milwaukee region is “growing without growth,” achieving rising incomes without attracting significant numbers of new residents.

Of course, none of the quibbles over the recent estimates detract from the larger story: Milwaukee and most metropolitan areas continue to spread out across the landscape. Far more people are heading for the exurban hinterlands than are taking up residence in downtowns.

And while Milwaukee may still be holding steady, some big cities clearly lost population in the early 2000s—San Francisco and Boston, for instance, are still experiencing a hangover from the dot-com bubble of the late 1990s.

In the end, though, we’re confident that the current decade hasn’t altered the fundamentals that produced city population gains in the 1990s—reduced crime, increased immigration, vigorous city leadership and a dynamic real estate market.

So, Milwaukee, relax and take heart. Top 20 or not, you may be in for yet another surprise when the 2010 census rolls around.