Which states came out ahead after Covid?
Not the Northeast.
I’ve come across a variety of news and pseudo-academic articles that have focused on the winners and losers of the pandemic. Very broadly speaking, I tend to see the winners as: Texas, Florida, Tennessee, the South generally, Arizona, South Carolina while the losers as California, Oregon, and Illinois. These tend to be due to things like population change, work from home relocation, and ad hoc stories of businesses moving headquarters or operations across states. It seems like Americans responded to: warmer weather, cheaper housing, laxer covid policies, and schools that opened earlier. But my sense emerged from quite the un-systematic reading of stories, so I thought I’d take a look at a variety of state-level measures to see who came out ahead after the Pandemic.
Today I’m going to look at GDP per capita, probably the foundation measure for changing post-pandemic fortunes. We can broadly think of this measure as a state’s level of development. I grabbed state level data from the Bureau of Economic Analysis Region Economic Accounts data from 1980 to 2023.
Where did economic growth occur?
The first question is: where did the largest economic growth rates occur? I took percentage changes in GDP per capita between 2023 and 2019 values (both adjusted for inflation to 2022 dollars). This should roughly show who experienced the biggest changes in economic development compared to just before the pandemic. Below is a map of how states fared. The darker red states had the highest percent increase in GDP per capita. The darkest blue states fared the worst.
Things aren’t looking great for the frozen north and northeast, aside from Maine. From Minnesota to Vermont, we see the lowest rates of economic growth.
The places that had the highest growth rates were Arizona, Utah, Nebraska, Tennessee, and Florida.
California and Oregon - not great. Washington and Idaho - surprisingly good. New Mexico, not great. Texas, Good. Louisiana and Georgia? Not great. South Carolina surprisingly not great, very much against my expectations.
My initial conclusion: the upper and the eastern tended to do worse, the Southern and the middle seemed to do the best.
But percentage growth relies on initial levels, right? California could grow by $10,000, Mississippi by $8,000, and we might see Mississippi as growing more because it’s starting from a lower GDP per capita level. Let’s look at the co-distribution of (1) 2019 GDP per capita levels and (2) 2019 - 2023 growth rates.
The best place to be in the map above is dark blue: these were places with high starting GDP and big growth: Washington State, Colorado, and, surprisingly, Nebraska hit this sweet spot.
Tragically, Minnesota is one of the states with a high relative GDP per capita level in 2019 that had a low growth rate, alongside many Northeastern states such as New York and New Jersey.
The states that really did the worst - low starting GDP per capita and low growth rates - were New Mexico, Michigan, and Vermont. I don’t hear much about these states or their pandemic experiences. But they’re the clear losers, falling even farther behind.
Texas and Florida are interesting - they had high growth rates but started in middle (Texas, 20th in GDP per capita in 2019) and low (Florida, 38th in 2019) starting points.
Of course, while the map is snazzy and visually appealing, its cut points kind of make artificial selections on GDP levels and percent changes. Let’s look at a simple scatterplot and highlight the mapped winners and losers:
Nebraska and Colorado look pretty similar to Texas and Utah to me. Washington remains unusual in its high growth relative to its high starting GDP level. New Mexico, Michigan, and Vermont didn’t do great, but look similar to Oklahoma, Louisiana, and South Carolina.
Hawaii, Delaware, Connecticut, Maryland, Pennsylvania, and Wisconsin look like they came out terribly after the pandemic.
Some of the pandemic success stories were relatively poor states (Florida, Arizona, Arkansas, Tennessee) that grew a whole lot.
The main pattern we see here is convergence: poorer states grew faster than richer states. Did we see this pattern of change - lower GDP levels with higher growth - in earlier periods? Let’s look at the preceding four year period: from 2015-2019 (I’m going to keep the “winners” and “losers” of the pandemic highlighted for an easy comparison).
Not really! Starting level and percent change are kind of unrelated here. What’s notable - just before the pandemic, tech hubs like Washington, California, New York, and Massachusetts were going wild with massive growth. Delaware and North Dakota were crashing. An everyone else seems to be kind of randomly scattered around some modest (Iowa) to large (Utah and Oregon) growth.
So it’s kind of interesting to see less developed states grow much faster than richer ones through the pandemic.
Growth beyond expectation
I’ve also been thinking about how states did during and after the pandemic, relative to the trends that they were on prior to the pandemic.
Here’s a simple example
Let’s say that we’re looking at three states: red state, blue state, and purple state. Each have the same GDP levels in 2019 and 2023, the white dots. The increase between 2019 and 2023 mean different things for the three states. For red state, this increase is simply on the trend it was on between 2015 and 2019. 2023 is a predictable outcome. For blue, and especially purple, states, the change between 2019 and 2023 are off the established trend - blue would expect stagnation and purple you would expect declining GDP over time. It’s probably good to bake these pre-established trends from before the pandemic into our analyses to get a sense of which states over and under performed.
So let’s do the following:
Predict trends of GDP per capita based on state-specific trajectories between 2012 and 2019.
Project these trends through 2023.
Compare projected trends versus what happened.
Let’s look at two states to get a sense of how our interpretations change:
Sadly, Minnesota represents a state whose development got dislodged by the pandemic. We see after the significant decline of GDP during 2020 a much more modest growth in 2021-2023. Minnesota ended up several thousands of dollars per capita lower than if the state had remained on its 2012-2019 trend.
Compare that to Texas. Texas GDP per capita went absolutely bonkers following the pandemic. Whereas Minnesota’s GDP per capita was a few thousand dollars lower than the trend, Texas’ GDP per capita was about $10,000 higher than its 2012-2019 trend! A massive change. Notice too that Minnesota and Texas had similar GDP per capita levels in 2012, Minnesota grew more rapidly through 2019, but Texas more than caught up, and surpassed, Minnesota in the post-pandemic period.
We are thus thinking of difference measures, how a state did relative to what we’d expect given how it was doing just prior to the pandemic:
Texas had more booms and busts around its trend, and had a massive boom after the pandemic. Minnesota was more stable around its 2012-2019 trend, got dislodged during the pandemic, and its growth has been lower than what we’d expect based on 2012-2019.
What happens when we look at these changes off the preceding trends in years 2020-2023?
A few clear winners stick out: North Dakota, Washington D.C., Wyoming, Texas, and Nebraska. These states experienced GDP per capita growth that was $8,500 to $15,000 above what their 2012-2019 trend would expect.
The next chunk of winners: Louisiana, Arizona, Oklahoma, Florida, Arkansas. A little above and below $5,000 above previous trends.
We see some states consistently showing up as winners. But some of the states receiving lots of focus, like Florida and Arizona, don’t jump off the page.
What about the losers? New York sticks out as the clear laggard, with GDP per capita about $7,000 below its 2012-2019 trend. Maryland, Connecticut, Wisconsin, Delaware, California, Massachusetts, and Pennsylvania are the next chunk of below-trend states, around $5,000 below their 2012-2019 trends. The northeast really struggled, no matter how you slice it.
Are there any geographical patterns to these off-trend changes?
Generally, the more Northeast you go - Maine excluded - the worse that states fared following the pandemic. New England, the Mid-Atlantic, and East North Central (the eastern part of the Midwest) all underperformed their previous GDP trend. The western Midwest (except Minnesota - boo!), the non-coastal South, and the Mountain states tended to all punch above their previous trend. The Pacific looks a lot like the Northeast, with all states underperforming.
It’s straightforward enough to see these regional patterns on the map.
The more Southern and middle you go, the more that states tended to overperform compared to the 2012-2019 period.
Surprises: I had assumed the southern coastal states - Carolinas, Virginia, and Georgia - would all be going gangbusters, but apparently not. At least not compared to what they were doing prior to the pandemic. Same with Idaho. Notice too that Alaska overperformed and Hawaii underperformed. I excluded them from the graphs above because both were VERY extreme compared to the contiguous United States.
Are we seeing an energy thing going on? What unites Texas, Nebraska, Wyoming, Alaska, and North Dakota?
Finally, how do these off trend changes compare to GDP per capita levels in 2019?
Here we’re looking at a dual-comparison: states that were highly developed in 2019 and had high off-trend growth rates are dark blue. States that had low development levels and got knocked below growth trends are pale beige. So the states that tended to most overperform were North Dakota, Nebraska, Colorado, Wyoming, and Alaska. The poor states that got the most off-trend growth were Arizona, Florida, Montana, Oklahoma, Arkansas, Tennessee, and West Virginia. The northeast and far-west coast were rich places that have underperformed after the pandemic.
It’s a tad easier to see who sticks out with a scatterplot rather than a map:
The strangest places most overperforming: Texas, Nebraska, Wyoming, Alaska, and North Dakota stick out like sore thumbs. Again, is this an energy thing? Wisconsin, Maryland, New York, South Carolina, Michigan, and Hawaii are distinct in how far below the trend they are.
Conclusions
The losers of the pandemic are a bit easier to identify, since they consistently show up across different analyses:
The Northeast, New York especially
The Pacific west (less so Washington State)
Wisconsin and Michigan
South Carolina
Some of these I’m familiar with, others not. Why did Wisconsin and Michigan struggle so much? And South Carolina doesn’t seem to have come out of the pandemic particularly well, but I often see it held up as a pandemic winner.
The winners:
Tier I - obvious winners
Alaska
North Dakota
Nebraska
Wyoming
Washington DC
Tier 2 - mostly winners
Texas
West South Central
Mountain Region
Tier 3 - a bit or little win, depending on your mood
Florida
My main lesson - I didn’t have a good sense of which states came out ahead and behind before looking at the actual trends. I think we’re mostly seeing an energy story defining the big winners, and a density/expensive housing story of the big losers. But this is super loose and suppositional. I often see pandemic state results framed as California versus Texas. But we would probably do better if we talked about Hawaii versus Alaska, Wisconsin versus North Dakota, Wyoming versus South Carolina. It’s a bit harder to make sense of these comparisons, but I think they’d probably be better to clarify the post-pandemic economy.