As a native of the buckeye state I like to keep an eye on what is going on there. As part of a paper I am currently working on I collected GDP/capita data and population data for US Metropolitan Statistical Areas (MSA). As I was collecting the data I thought it would be interesting to examine the differences between Ohio MSAs. But first a brief background on MSAs.
A MSA is considered the economic city. A MSA usually consists of one large primary city and the surrounding urban counties, but some MSAs have more than one main city (e.g. Dallas-Fort Worth, San Francisco-Oakland, Minneapolis-St. Paul). Each MSA will contain many distinct political cities, where political cities are delineated based on government boundaries. The land area of an MSA is determined by economic forces rather than political ones. In particular, the commuting patterns of workers determine the geographic size of a MSA. If a certain proportion of workers in a county commute to one of the primary cities of an MSA then that county will be considered a part of that MSA/economic city. Thus distinct political cities (i.e. each city has its own government) that are economically intertwined will be part of the same MSA. Since economic activity often crosses political borders seamlessly it is more useful for economists to use the economic city as the unit of analysis rather than the political city. In fact, economists often use the term city and MSA interchangeably, which can be confusing to non-economists.
So now that we have a better understanding of what an MSA is, lets look at the data for Ohio MSAs. The first graph (below) shows the populations of eight major Ohio MSAs from 1990 - 2010 (click to enlarge)
Only the Columbus MSA and Cincinnati MSA experienced significant population growth over the last 20 years. Cincinnati was the largest city in Ohio in 2010, just surpassing Cleveland. The Cleveland and Youngstown MSAs experienced a decline in population while the remaining MSAs held fairly steady.
But even a steady population is good in a state like Ohio since overall the state shrank from 11.5 million people to 10.8 million people from 1990 - 2010. This means that Columbus and Cincinnati both contain a larger proportion of Ohio's population in 2010 than they did in 1990; i.e. the population of Ohio has become more concentrated in those two cities. In 1990 28% of Ohioans lived in Columbus or Cincinnati. In 2010 37% of Ohioans lived in Columbus or Cincinnati.
In the graph below I show the Real Gross Domestic Product (GDP)/capita in 2001 and 2011 for the same MSAs.
The three largest cities - Cleveland, Columbus, and Cincinnati - were also the three wealthiest cities. Out of the three though, only Cleveland experienced an increase in GDP per capita from 2001 to 2011. Both Cincinnati and Columbus, the two fastest growing cities, experienced a decline.
One explanation for this could be that the people already in and moving to Ohio view Cincinnati and Columbus as places of economic opportunity. This means that poorer people may be inclined to move to one of those cities in order to improve their lives. Growing cities can appear poorer when looking at measures like average income or GDP because they are better at attracting relatively poor migrants who are searching for economic opportunities. An influx of relatively poor migrants will decrease the average GDP/capita. Columbus and Cincinnati may appear poorer when looking at averages, but we know by the population numbers above that whatever may be going on in those two cities it is not discouraging people from moving there.
Meanwhile, places like Toledo, Akron, Lima, and Cleveland appear to be getting richer on average , but they are also stagnant or shrinking. If lower income people are leaving those cities in order to find better opportunities elsewhere it can have the effect of increasing average GDP/capita in those cities.
Cities that are thriving economically will always attract relatively poor people who are looking for better lives. That is why it is important to look beyond average incomes and similar statistics in order to see what is really going on. I would have to do more research before I could conclude that the story I just told is really occurring in Columbus and Cincinnati but it is a plausible story.