Wednesday, January 8th, 2014

Convergence and Divergence In Per Capita Income

I was doing some research on the economic history of Rhode Island recently and decided to take a historical look at its per capita income levels vs. Massachusetts. Here’s what I found:

This is per capita income as a percentage of the US average. Both states started out far higher than the US average, peaking at 150% of it in the 1930s. But over the course of a mere decade or so, the both crashed to about the national average. (The stark reality of this collapse is evident in the wartime newsreel “Report From Rhode Island” which shows the Newport mansions in ruins). You can see also that up until the end of the war, the two states were virtually identical, then diverged. Both states had to spend some time wandering the wilderness, though Rhode Island had the worst of it. In fact, Rhode Island has never recovered. Its per capita income has simply oscillated between slightly above and slightly below the US average for close to 70 years. Massachusetts by contrast limped along until around 1977, then embarked on an uptrend and ever wider divergence from Rhode Island. A very telling graph I think.

I got curious and wondered about another state pair I’ve long looked at, Illinois and Indiana. So I ran the same graph and this is what I came up with:

Here we see an opposite pattern and one that surprised me. These two states were very divergent at the end of the Roaring 20s, with Illinois far above the US average and Indiana far below it. Over a similar time frame in which Mass and Rhode Island collapsed, these states converged, with Indiana gaining strongly and Illinois losing. This convergence left a gap to be sure but what I found most surprising is that it has basically remained constant over the decades. Both Indiana and Illinois have remained in parallel downtrends. Given the dominance of Illinois’ economy by metro Chicago, and the incredible divergence in policies between the two states, I wasn’t expecting them to track each other. Though maybe I shouldn’t have since I’ve noticed similar things in recent times before.

We hear a lot about income inequality in the US today. But if you look at those charts, it seems pretty clear that in the early parts of the century, the US was an extremely divergent economy in which some states where massively most prosperous than others. However well Massachusetts may be doing, for example, it hasn’t yet reached its pre-war peak. The Great Depression and World War II seem to have had an enormous leveling effect (looks like a lot of the New Deal was extremely effective over the long term, the TVA, REA, etc).

So I decided to plot the whole country on a map. In the map below, states in blue grew their per capita personal income at a higher rate than the nation as a whole since 1929. The ones in red grew more slowly. The color scaling is proportionate to the percentage change value. The 61.7 is the percentage value of the US average in decimal format, that is, 6,170%. (Keep in mind, this is a nominal value, not inflation adjusted).

This shows that although the traditional centers of prosperity in America may still be doing better than the rest, over the long term there has been a convergence in incomes as areas like the Deep South have modernized. Oklahoma is no longer Grapes of Wrath country.

But even among the traditional wealth centers, we see divergent paths. I plotted California, Illinois, Massachusetts, and New York to see what I would find:

We see big differences. California and Illinois have more or less headed straight downhill since the 40s. That surprised me because I’d always thought California enjoyed a sort of Golden Age in the post-war era. It did have a moment in the sun in the 1970s, but it’s interesting that it hasn’t done better vs. a state that got slammed hard by de-industrialization.

New York and Mass had very different paths, reaching 1970s nadirs then rebounding strongly. New York fell below both California and Illinois in the era when NYC nearly went bankrupt. Today it is far ahead of both. The turnaround in Massachusetts has been even more dramatic. Some of the superior performance of Mass may be due to demographic composition, however. It’s no secret that it’s good to be privileged and white in America. Of the traditional tier one big cities, Greater Boston is overwhelmingly the whitest, while LA and San Francisco are the most diverse. LA in particular has absorbed huge numbers of lower skilled Latino immigrants who, unlike the descendants of the early 20th century Irish and Italians immigrants to Boston, are still in the early stages of their development lifecycle. Down the road this should converge. New York managed to pull off its turnaround while retaining significant diversity and a large immigrant pipeline, which makes it even more impressive.

Note: All data from the Bureau of Economic Analysis, accessed and charted via Telestrian.

Topics: Demographic Analysis, Economic Development

13 Responses to “Convergence and Divergence In Per Capita Income”

  1. This is some really important analysis, in the same vein as the Big Theory on American Urban Development by Corner Side Yard –

    Get some grad students on this immediately!

  2. pete-rock says:

    @Peter Christensen, thanks for the plug. If this is sarcasm I refuse to see it. I wish I did have grad students at my disposal.

  3. Totally serious – we’re barely beginning to understand the life-cycle of modern cities, economic competitive advantages, etc.

    You could use these kinds of relative economic performance over time to test theories about different economic advantages (ports, education, integrated manufacturing, weather, stable political climate, housing stock age and quality, metro area municipal structures, etc) and make predictions about performance of different cities over the next decade. There’s a Richard Florida level of exposure (or at least a compelling thesis) available to the person that gets that right, in a convincing way.

  4. Jonathan says:

    Very interesting but the graph fails to take population into account. Graphing percentage of average per capita income means that the larger the state, the more closely that state will track the national average.

  5. pete-rock says:

    @Peter, I’m still pushing forward on the notion of the life cycles of cities. Earlier this week I wrote this

    to show that not only do metros develop during certain eras, they go through a series of stages, and that makes comparisons between metros a little dubious. I wouldn’t compare the mile times of a 23-year-old young adult and a 54-year-old middle-aged adult, and maybe we shouldn’t compare Charlotte with Cleveland.

  6. Chris Barnett says:

    It seems counterintuitive, but Jonathan’s right. The law of large numbers is at work here. At least one state did very well on this measure as its population growth seriously lagged the US average.

    North Dakota’s per capita income relative to the US grew initially because the state’s population declined over the 80-year stretch from 1930-2010, while agricultural enterprises became more productive and extensive (making individual farmers’ incomes higher). Then the oil and gas boom hit.

    In 2010, only 672,000 people lived in the whole state, less than in any of the main Lower Midwest metropolitan core counties (Cuyahoga, Franklin, Hamilton, OH; Wayne, MI; Marion, IN; Cook, IL; Milwaukee, WI; Hennepin, MN). Spreading the oil and gas-based increment in personal income over that small a base will yield an outsize per-capita income gain that is merely a statistical curiosity.

    And, of course, no amount of eds and meds, state-level tax-tinkering or economic development incentive will put oil under the ground.

  7. More productive = more money. Oil & gas = more money. More money / less people -> what’s not legitimate about that?

    North Dakota’s median household income is also higher than the US average, and that’s less susceptible to influence by outliers than per capita income. I do think North Dakota benefits from the fact that the end date (i.e., today) coincides with its boom period. Indisputably there’s a massive difference between the North Dakota of today and that of 1929. Keep in mind the conditions that existed in rural America and the south in the prewar era (i.e., no electricity, no running water, school until 6th or 8th grade, etc).

    What’s more, every urbanist touting the virtues of the NYC/SF model (e.g., Richard Florida, Ed Glaeser, Enrico Moretti) have focused on per capita income as their key metric.

  8. Jonathan says:

    Aaron, use of per capita income is uncontroversial, my issue is with using percentage of US per capita income as your measure. Use of the latter invites problems with relative population sizes.

  9. Jonathan, I gather your complaint is that since say California is so large, it disproportionately impacts the national average. Fair enough. However, comparisons of difference sized entities vs. national averages is done all the time and is the root of all these divergence narratives we see around big cities and such. (I actually have some state level analysis this weekend that recalculates state averages ex-metro).

    However, the same trends are clearly visible if I take the US average out and plot index or raw values. In the case of PCI, that becomes harder to see on the chart because of the dominance of inflation effects over time. And if I’d plotted indexes and showed say Indiana growing faster and higher than Illinois, someone would no doubt have complained about that too.

  10. Jonathan says:

    First, I do appreciate your insights here.

    Second, I have no problem with comparing different sized entities vs. national averages. Obviously RI is not as wealthy as Massachusetts on any given day.

    The very small problem I have is comparing different sized entities vs. national averages over time, where the sizes change over time as well. The relative population sizes are also a factor in the graphs you draw, but you are ignoring them.

  11. Sean S. says:

    More dangerous is presuming of course that this is connected to urban politics, as opposed to the housing preference of the wealthy. The fact that a large number of wealthy individuals, for instance, make the Upper East Side their home does not mean that New York City urban policies created that. I think median income would be a better measure, and controlling for population size.

  12. Matthew Hall says:

    I’ve long wondered why some of the booming sunbelt town have had worse per capita income trends than the staid old metros that people have presumably left to move to the booming sunbelt towns. It can’t be that migrants perceive they are moving to places with worse income trajectories, it must be that they are moving because they think their individual incomes will be higher in the place they are moving to, all other things being equal. How ironic.

  13. urbanleftbehind says:


    But consider that much of the booming sunbelt population consist of retirees, ranging from well-heeled, but often times on fixed pensions or annuities. Also, a steady stream of immigrants also keeps fixed wages down. If one goes to a “sunbelt” location (I think Dallas, Houston, and Austin have graduated from that designation for various reasons) assured of a high 5- or 6-figure salary, they will have a higher than average per-capita income. The weather also means everybody else wants to be there as well.

    The key to gaming this metric is to maintain or grow your high-income cohort while minimizing the growth of the attendant service sector and not becoming perceived as a destination for the lower-income and indigent classes. All of the reddish states on the US map in the graphic above were net receivers of the Great Migration from the South in the mid-1900s and/or expedited Latino immigration later in the century (California also had the Okies in the 1930s). Atlanta suffers in the comparison (see the other more recent post) in recent years because it has become a net receiver of lower income black in-migration from other failed metros.

    The far northern tier of states are have maintained or improved their comparability to the nations PerCapita income growth rate because of not only “unattractiveness” to lower and middle income in-migrants and immigrants, but perhaps also to cultural factors that frown upon conspicuous consumption (think Minnesota and flinty north New Englanders) and hence growth of a service sector.

The Urban State of Mind: Meditations on the City is the first Urbanophile e-book, featuring provocative essays on the key issues facing our cities, including innovation, talent attraction and brain drain, global soft power, sustainability, economic development, and localism. Included are 28 carefully curated essays out of nearly 1,200 posts in the first seven years of the Urbanophile, plus 9 original pieces. It's great for anyone who cares about our cities.

Telestrian Data Terminal


A production of the Urbanophile, Telestrian is the fastest, easiest, and best way to access public data about cities and regions, with totally unique features like the ability to create thematic maps with no technical knowledge and easy to use place to place migration data. It's a great way to support the Urbanophile, but more importantly it can save you tons of time and deliver huge value and capabilities to you and your organization.

Try It For 30 Days Free!

About the Urbanophile


Aaron M. Renn is an opinion-leading urban analyst, consultant, speaker, and writer on a mission to help America’s cities thrive and find sustainable success in the 21st century.

Full Bio


Please email before connecting with me on LinkedIn if we don't already know each other.


Copyright © 2006-2014 Urbanophile, LLC, All Rights Reserved - Copyright Information