Forty Years of Economic Freedom Winning
Forty Years of Economic Freedom Winning
Leftist policies have been failing for a very long time. We should take that seriously.
Whether Democrats or Republicans are better at governance is often treated as a matter of religious opinion. You’ll see conservatives debate one another over whether Trump or DeSantis is better, and liberals discuss whether they should push Medicare for all, but I can’t remember the last time I saw someone try to make an argument across party lines and tell the other side that they should come over. This is odd! Maybe on social issues you can’t convince people, but we have data on economics and population figures, and in theory we should be able to figure out which side does a better job building the kinds of communities people want to live in. There is a general consensus about wanting individuals and families to have higher living standards. Where people choose to live is moreover a very important kind of revealed preference.
And Americans move a lot. Between 2021 and 2022 alone, 8.2 million people moved states, which is 2.4% of the entire country. The American Community Survey in 2007 found that an American moves 11.7 times in their life. Of course, the nation also attracts a large number of immigrants, of which there are currently over 45 million, and those people also have to choose where to live. Where natives and newcomers go should in theory provide us some of the best evidence we can find regarding the effectiveness of different government policies.
Leftists believe in higher taxes, more regulation, and more government services, conservatives in leaving more money and decision-making in the hands of the private sector. Perhaps with only fifty states, we run into the issue of too small a sample size to say anything meaningful. Yet small sample sizes aren’t always an insurmountable problem if the variation in policy or outcomes we end up observing are extreme. If all Red States had better economies than all Blue States, or vice versa, that would be quite compelling evidence. Luckily for us, states have widely different policies. California has a 13.3% highest marginal tax rate, while nine states impose no income tax at all. If it turned out that this made no discernible difference in outcomes, it’s probably not worth debating economic policy at all. However, if these things matter, we should see some indication of that fact.
This article begins by briefly discussing some of the ways in which states differ across policy areas. I then go on to compare state outcomes when it comes to GDP growth and population change, covering the period from 1980 to 2022. The evidence is clear and overwhelming that states that have had more economic freedom have been doing much better. Moreover, although weather is a potential confounder here, it likely doesn’t provide an adequate explanation for the variation we see.
States Differ in Policy
The most basic economic policy difference revolves around the degree to which a state takes money that citizens earn. Unlike the federal government, states can’t print dollars, so there’s a direct connection between taxation and spending. Here is tax burden by state as of 2022.
The highest burdens are in New York, Connecticut, Hawaii, Vermont, and California. The lowest are Alaska, Wyoming, Tennessee, South Dakota, and Michigan. While this seems to roughly map on to Red/Blue orientation, note that there is a far from perfect correlation here with the ideological leanings of states. Pennsylvania is about equal to Montana and Idaho, even though the latter two states vote much more Republican.
Taxes are of course only one dimension of comparison. Here’s how much occupational licensing varies across states for selected professions. As you can see, some places require you to go to school or train for years to get the same job others let anyone try their luck at.
States differ in how much they regulate the labor market in other ways too. There’s a federal minimum wage at $7.25 per hour. Sixteen states set no minimum wage above that, while 34 do, with the highest being California at $16. Similarly, some jurisdictions have anti-discrimination laws that go beyond federal requirements while others do not.
Luckily, there are organizations that track economic freedom overall and give each state an aggregate score. Here’s the Cato Institute map for 2022, which takes into account all the factors I mentioned and some others to rank the states.
There’s another economic freedom index from the Fraser Institute, and it looks similar, with states like Tennessee at the top and California and New York near the bottom. ALEC, which works on conservative legislation, ranks states on 15 variables for economic freedom, giving top marks to Utah, North Carolina, and Arizona.
You might notice that all of these indexes and measures tend to be released by conservative organizations. While this may be a reason to mistrust the data, it begs the question as to why there are no leftist organizations ranking states and then bragging about how great all the ones run by Democrats are doing? If one side is doing all the measuring, it may be because the other knows that comparisons don’t make them look good. Republicans on the House Budget committee have published at least one document touting the successes that their party has had in state government, something Democrats appear never to do. Ezra Klein has to be honest and admit that the Blue State model has failed when it comes to housing. There are defenders of extensive economic regulation in the abstract, but few who want to defend it as actually practiced.
One 2008 paper argues that, when you control for state level factors, which party wins a gubernatorial election has no effect on most social welfare and policy outcomes. To take this to mean that policy doesn’t matter, however, would be a mistake. First there is a restriction of range, in that a Democratic governor who wins in a Republican state or vice versa is likely to be more moderate than the typical member of the national party. Louisiana, for example, had a Democratic governor from 2016 to 2024 who signed a six-week abortion ban. The Democratic governor of Kansas vetoed bills that would have cut state income taxes, but in 2022 signed a phaseout of the sales tax on food. More fundamentally, any one election is unlikely to have a major effect because policy is cumulative. Most gubernatorial administrations don’t rework the tax code from the ground up when they come into office. One would probably expect the overall impact of any one governor to be mostly limited.
Freer States Gain More People, Have Better Economies
Of the two fundamental measures of how states are doing, I think population growth is superior, because there’s a lot of subjectivity and potential measurement error that goes into GDP, while where people choose to live is among the most direct measures of their preferences. Nonetheless, we’ll see here that both tell a similar story.
Here’s US Census Bureau data on population growth from 2010 to 2020.
The best performing states are concentrated in the mountain West and the South. The middle of the country and the Northeast are generally growing much slower, and in a few cases losing people.
One thing to note is that while one sees a lot of variation when you consider all 50 states, it’s probably useful to weigh the data by population size. Many small states are subject to larger economic forces that determine their fates. North Dakota has been experiencing booms and busts based on oil production since the mid-2000s, and the price of that commodity is likely to be more important with regards to its fortunes than decisions made by policy makers in any given year.
Large states have more internal variation, which creates more room for government decision-making to play a role in outcomes. They tend not to be dominated by just one or two industries. Also, larger states are just simply more important because they have more people. About a third of the US lives in either California, Texas, Florida, or New York. Here’s population change from 2010 to 2020 for the 10 largest states, with political orientation indicated through color.
Instead of change in population size, one can look at GDP growth over time. I do that in the graph below, for the period between 2010 and 2022. Note that population change from 2010-2020 and GDP growth from 2010-2022 correlate at about 0.76, so you’re measuring much the same thing.
In the graph below, a red state is one where Republicans control the governorship, senate, or house 80% of the time. A blue state is the same for Democrats, while a purple state falls into neither category. For example, if in one year, a state has a Republican governor, a Democratic house, and a Democratic senate, Democrats have 66.67% control of the state. I do that every year from 2011 to 2020 for each state, and then arrive at a total to classify them.
There’s a clear pattern here, with red states growing faster. But we can certainly do better than just eyeballing which states are red and which are blue. Below, I show the relationship between economic freedom score as measured by the Fraser Institute and population growth for each state for every decade from 1980 to 2020. We see correlations that range from 0.27 in 1980-1990 to 0.39 from 2000-2010.
I use the Fraser Institute here because they have data going back to 1981. The Cato Institute rankings begin in 2000, and in the appendix I show that if you use them for the time period 2000-2022 the results are almost exactly the same. See also the appendix if you would like the four graphs above in one matrix, which is less readable in terms of the individual observations, though some might prefer it anyway.
The consistent nature of the results is what I find convincing. Within any particular decade, there may be variation due to factors like regional booms and busts, the discovery of natural resources, or wider changes in the economy that decimate a certain industry. But that stuff should generally wash out in the long run. The connection between population growth and economic freedom appears to be quite consistent, and clearest in the states with the largest and most diverse populations.
Here we see the value of looking at economic freedom scores rather than simply eyeballing which are red and blue states. Over the last few decades, Mississippi, Louisiana, and West Virginia have had quite poor performances, and one might be tempted to blame Red State governance. But they rank either middling or low when it comes to economic freedom, so this would be a mistake.
One complication of looking at population growth is that it also factors in birth and death rates, which are influenced by average age and may or may not tell us something about economic policy. For the time scales we are looking at, 10-year periods, this may not matter too much. Here’s a map of natural population growth in 2017, and it doesn’t seem like factoring that variable in would change the results much. California actually has a pretty decent natural growth rate because it’s a young state, while Florida is full of retirees who are always dying off, meaning that the overall changes in population for the latter state are even more impressive than they look at first glance. I haven’t done a formal analysis, but I would be shocked if taking into account natural population growth would make the picture look any different. This is most likely to make a difference in Utah, which has had the highest birth rate in the United States over the last several decades, but it doesn’t do that well in terms of economic freedom anyway.
Moreover, shouldn’t states with freer economies get credit for higher birthrates, if there actually is a connection? I guess you would then have to count death rates too. The population figures give you all of it: migration into the state (good), births (good), migration out of the state (bad), and deaths (bad). The only potentially unfair part of this is if states start out with different age pyramids. But I doubt that’s driving the results we see, as there doesn’t seem to be a strong correlation between age and economic freedom scores, and most states are pretty similar in how old they are anyway, within a few years of one another. Looking at long-run trends also solves this problem. States will often end up with unhealthy age pyramids precisely because they haven’t had growing economies. In other words, since we’re looking at a four-decade time period here, how old states are is an endogenous variable.
Below is the same analysis as above but for economic growth.
We find the relationship between economic freedom and GDP growth to be positive each year, with the correlation ranging from 0.12 in 2000-2010 to 0.41 in the 1990s. Note that the GDP figures from before 2000 are pegged to 1997 dollars, while those for 2000 and after are pegged to 2017 dollars. I carried out the same analysis from 2000 to 2022 using the Cato Institute data, and got almost identical results over those two decades. Again, I leave the charts to the appendix in order to not make the text here more cluttered than it already is.
The Last Few Years Look Even Worse for Blue States
There are already indications that if I carried out this analysis in another 5-10 years, economic freedom would still be winning. Note that above I only analyzed population trends up to 2020. Here’s what’s been happening over the last few years, post pandemic.
This looks like more of the same, but much worse for Blue States. California and Hawaii haven’t historically been particularly high performing, but they were always gaining people across every decade from 1980 to 2020. The fact that population is actually shrinking in both states indicates that their problems are getting worse. Moreover, Oregon and Washington have traditionally been high-performing states, bucking the trend of liberal governance being associated with poorer outcomes. Now Oregon is in the red, and Washington is barely growing at all. Meanwhile, Republican governed Idaho next door continues to boom.
The West coast states and Hawaii were particularly bad on covid, and trends of the last few years probably reflect that. Hawaii did not remove its indoor mask mandate until March 2022, a full year and a half after DeSantis banned even localities from imposing such restrictions. The fact that Blue States bungled covid that badly, after already bleeding people for other reasons in the years before, indicates that something is deeply broken. We can’t brush covid off as a one-off given the larger pattern of failure. Extended school closures and too much deference to the public health bureaucracy can be seen as symptoms of a major underlying problem, which is government captured by special interests.
The future should bring us more of the same, if not an even greater split between economically free and pro-regulation states. In California, the Department of Finance is projecting that the population of the state will remain basically unchanged until 2060. After decades of simply losing poorer residents, the wealthy are now leaving too. Things aren’t looking good for Illinois either, which has the worst pension crisis in the nation, meaning it basically mortgaged its future to get more expensive government employees. If this brought better services or somehow paid for itself in another way, I’m sure we would’ve heard about this by now. States that have allowed public sector unions to determine policy will be feeling the pain for generations.
Is It All Just Weather?
People want to live in dry places where the sun is shining. Maps of population movement and GDP growth look much like climate maps.
Wikipedia has a map of world locations with a Mediterranean climate, and you can see how blessed the West coast is, including Oregon and Washington. Yet I think there’s enough variation among states to say that we can still see the influence of policy. Consider:
Arizona and New Mexico. These are both extremely hot desert states that are right next to each other. Yet Arizona is economically freer, and has been growing much faster over the last several decades.
New Hampshire versus Vermont. Like Arizona and New Mexico, they’re neighbors that don’t look too different from the outside. Yet between 2010 and 2022, New Hampshire saw its GDP increase by 23.7%, compared to 10.5% for Vermont. The freer state in this pairing has also gained more people every decade from 1980 to 2020.
California and Hawaii. If you’re going to say it’s all about the weather, it’s a huge problem when Hawaii and California aren’t growing that fast. These states are probably too nice to ever suffer the fate of the northeast, but they’re definitely underperforming given their natural advantages. One can’t say that there are no Blue States with good weather. California and Hawaii are probably the top two on that measure!
Here’s GDP per capita for Vermont versus New Hampshire from 1958 to 2023. Remember, this underestimates how much better New Hampshire has been doing, because it has been gaining more people too. While both states started out just below the US average in 1958, today the state with more economic freedom is significantly wealthier while the one that favors more stringent regulations is slightly poorer.
We can make the same comparison between Arizona and New Mexico.
Arizona is doing much better, but it’s really not fair to look at per capita, because again, like with the New Hampshire/Vermont comparison, one state has been attracting a lot more people. So below, I present real total personal income across the time period, and here we can finally understand the true magic of capitalism.
Over 66 years, Arizona saw a gain of 2,317%, compared to 733% for New Mexico. Economic freedom draws more people, and makes them wealthier, lifting all boats. I’ll further note that New Mexico is the home to two national laboratories, Los Alamos and Sadina, and the subsidization of scientific research and the concentration of high human capital is often thought to be a way to spur economic growth. Yet Arizona by just leaving people alone has blown past its neighbor.
Maybe this is mostly motivated reasoning. As a thought experiment, I tried to imagine I’m a leftist here and see if I couldn’t explain this all away. The first thing I would cling to is Oregon and Washington. But as mentioned before, these are states with uniquely blessed climates, and there’s probably no reason they should be gaining fewer people than Georgia in most years. The South is muggy and hot, so I’m not even sure its weather is all that great. In 2017, 6.8 million Floridians had to evacuate their homes during Hurricane Irma. This was the biggest hurricane-caused exodus in American history, beating out the Houston evacuation of 2005. Major hurricanes aren’t that common, but they’re a huge disruption to people’s lives, and this is something that they may consider when deciding where to settle.
Historically the climate has been blamed for the South’s backwardness and lack of development. Sure, things changed with the invention of air conditioning, but still. I buy that it’s better than living in Minnesota, but I don’t know by how much.
One idea for future research is to look at counties that border one another across states with very different policies and compare how they’re doing. For example, here’s a county map I found for population change between 2020 and 2023.
Through this method, one can factor out weather and wider shifts in the economy like the natural tendency for rural areas to depopulate. Here, note the borders of Illinois. The whole state is basically emptying, but things look much better in neighboring counties in Indiana and Missouri, though Iowa appears similarly bad. You can likewise look at Idaho versus Oregon and Washington, or Nevada versus California. Finally, check out Texas and Louisiana. Both are Red States, but the latter is rated as much less economically free. There’s no reason a researcher can’t do this for population and GDP over longer time periods and see whether my theory about the impact of policy holds. I invite others to undertake this analysis, and if one is conducted well I’d possibly be willing to publish it here as a followup to this essay.
The World is Not that Mysterious
I think this data on variation between American states has to be considered alongside what else we know about the world. Consider how much wealthier the US is than Europe, and how that gap appears to be growing over time. Note also the massive turnaround for ex-communist countries after the collapse of the Berlin Wall, and also the rise of East Asian nations as they’ve moved away from top-down control. The evidence is consistent with a model of the world where the more pro-market your policies are, the faster you will grow, the more people and jobs you will draw, and the better off you will be.
I often hear people say that the most successful societies have adopted capitalism but with aspects of the regulatory and welfare state woven into it. The implication here is that the kind of mixed system most of the developed world has is better than a purer capitalism or purer socialism. Yet I see no evidence that anyone has too much capitalism. America does better than Europe, and within America, Red States do better than Blue States. Consider the experiment of South Korea versus North Korea. The relatively capitalist state is much better, but it would be nice to have an even more capitalist Korea that we could compare the South to. Since we don’t, there’s no reason to just assume that South Korea has achieved some kind of policy ideal, rather than it still being able to benefit from an even higher level of economic freedom. The entire US experienced the New Deal and the Great Society, and it would likewise be nice to observe the outcome of a different universe where the federal government did not expand. Things worked out relatively well for the United States, but I’d argue that this is in part because we were still economically freer than the other major economies. Variation between American states suggests the country would have been better off if the federal government had been even less expansive.
I think no country has gone beyond the optimal amount of capitalism because the exact same arguments against central planning that we apply towards communism work just as well against a mixed economy. If a purely communist state zaps the motivation to work hard, shouldn’t a mixed economy state do the same, just to a lesser degree? If you don’t believe that government should run the steel factories, why then should it run education? What about health care? Sure, you can come up with reasons why the latter two industries are different than most areas of the economy, but I don’t find the arguments very convincing. And if government-enforced cartels are bad in all other areas, what makes labor unions acceptable? I’m convinced that unions are so clearly offensive to the laws of economics that no one would defend them if they weren’t already established institutions.
I generally don’t like cross-national comparisons, and the same problems exist when analyzing a limited number of states. But 40 years of data all pointing in the same direction cannot be easily dismissed. I believe economic freedom matters, and it is important enough that one can find signal amongst the noise even in relatively small datasets. If I’m still writing in 2030, I commit to doing a followup analysis that I predict will find a continuing relationship between economic freedom and population and GDP growth.
Appendix
The main investigation here into the relationships between economic freedom and both population change and GDP growth uses the Fraser Institute scores for the former, but if you replace the numbers with those of the Cato Institute, the results are much the same for the periods 2000-2010 and 2010-2022, as can be seen in the graphs below.
The following is a matrix of the relationship between economic freedom and population growth for the four decades from 1980 to 2020, based on the Fraser Institute scores.
While I agree that economic freedom is a good thing, focusing on GDP growth instead of current GDP can lead to some pretty misleading conclusions. Internationally, Benin (in Africa) has a recent growth rate of 6% - much better than the US. Would you want to live there? (Nothing against Benin in particular, just I'd rather be in the US). Countries that are starting from really low GDP per capita points can show pretty spectacular growth as we see in the US in some of the red states. If you look at GDP per capita by state the top performers are New York, Massachusetts, Washington, California. Hardly a who's who of high growth rates, but that's where the money is.
Google the Kuznet's Curve. Basically, it predicts exactly what we see. The low GDP states tend to grow fast until they reach a certain level then growth slows. There are a lot of reasons for this including some of the ones that Hanania gives here, but it's pretty consistently observed and I'd be shocked if it didn't prove to be pretty durable in the case of the current low gdp, high growth states.
It'd be nice if you did this topic as an adversarial collaboration with someone smart who holds a more leftist stance. You can wave numbers around all day, but statistics can be manipulated and it takes an expert to really fact check them. I believe in economic libertarianism too, but I find the big picture stuff like communist states failing more convincing, because the small details can be endlessly manipulated by someone who knows what they're doing. Two opposed experts fact checking each other and producing a conclusion together would be great.