It’s become a commonplace to see yet another tech company pulling up stakes and leaving California … and the state government seems determined to get rid of many of their biggest economic contributors with new legislation to make life harder for companies and wealthy individuals (see the second item, for instance). But it’s not just California losing population to other states:

Universities in the NCAA Southeastern Conference.
Map from Wikimedia Commons

Americans have long expressed their political preferences at the ballot box and their economic preferences with moving vans. Increasingly, they are expressing a third preference — where their children attend college — and the destination looks remarkably like the other two.

In 2025, North Carolina gained more than 84,000 residents from other states, more than any state in the union. Texas added over 67,000. California lost more than 229,000, and New York lost nearly 138,000, continuing a migration that has by now acquired the character of a rout rather than a trend. Yet few have noticed that the same states drawing families and factories are drawing something else: students, and with them, the institutions built to educate them.

For most of the 20th century, the hierarchy of American public higher education was as fixed as the Electoral College map of a one-party era. Berkeley, Michigan, Wisconsin — these were the names invoked when a public university wished to be mentioned in the same breath as Harvard. They remain formidable.

But formidable is no longer synonymous with singular.

Consider what the Southeastern Conference, or SEC — an organization whose reputation was built on forward passes rather than research grants — has quietly assembled. Texas A&M enrolls more than 79,000 students, a figure that would have seemed a typographical error a generation ago (see here and here). The University of Florida enrolls roughly 56,000 (see here, here, and here). Collectively, SEC institutions now educate more than half a million students, with total enrollment around 620,000 (see here and here), a statistic that ought to interest university presidents in blue states rather more than it apparently does.

This is not the result of some coordinated Southern strategy hatched in a statehouse. Universities are lagging indicators of demography, not leading ones. States that attract families acquire, in due course, college students. States that generate jobs retain their graduates rather than exporting them, as Rust Belt states have exported so much else.

States that welcome capital acquire, eventually, the tax base with which to build ambitious public institutions. Ambition, it turns out, is portable. The family that leaves Illinois for Tennessee brings its expectations for its children along with the furniture, and both eventually find a permanent address in Knoxville or some other city in the Volunteer State.

Affordability, too, plays its part, and Americans have noticed that a nationally respectable degree from a Southern flagship costs a fraction of what state flagships elsewhere demand for a comparable credential, at a moment when the public’s confidence in the value of a degree generally has become, to put it charitably, provisional.

One of California’s most impactful proposed pieces of legislation has been their “Billionaire tax”:

California’s proposed “billionaire’s tax” on the assets of wealthy state residents enjoys a slim lead in the polls leading up to the midterm elections, though so do two measures that, if they draw more votes, could render the scheme unenforceable. That makes for a high-stakes battle as many prosperous Californians are already fleeing to escape a tax that could force them to surrender ownership stakes in companies they’ve founded. Even if it doesn’t pass, a new report warns that the proposal is based on faulty research and could do vast economic damage.

[…]

The Wealth Tax Scheme Is Based on Unusual Accounting

Now, a new report finds the proposed wealth tax is not only dangerous for California’s economic prospects; it’s also based on bad research. According to the report’s author, Independent Institute research fellow Kristian Fors, “this proposal has been heavily influenced by the work of UC Berkeley professors Emmanuel Saez and Gabriel Zucman to justify the concept of wealth taxation”. Saez and Zucman have generated news headlines with their claims that the wealthy are undertaxed in comparison to lower-income Americans. But as Fors points out, the estimate that drew so much news coverage claimed that “the overall tax rate for the top 400 richest households in the nation was 23 percent”, while just a year earlier, Saez and Zucman “estimated that the top 1 percent paid an overall tax rate of approximately 36 percent; the authors found a share of 41 percent for the top 0.001 percent of earners in the most recent year of their survey, a category encompassing billionaires and high multimillionaires”.

The discrepancy, according to Fors, comes from how the economists treat corporate income taxes. Conventional analysis assumes that corporate taxes burden shareholders, workers, and consumers as the tax gets passed on. That results in the earlier estimate of a 41 percent tax rate for the wealthiest. To reach their much lower 23 percent estimated rate, Saez and Zucman used a non-standard analysis — and while the results of that approach won news coverage, they weren’t subject to peer review.

“Saez and Zucman’s empirical work on the California billionaire tax proposal retains these same unconventional accounting practices from 2019 without addressing their conflict with the mainstream corporate tax incidence literature,” cautions Fors.

The Independent Institute report observes that privacy laws prevented Saez and Zucman from drawing on personal income tax and other financial records. They relied on the Forbes 400 list, and “a 2010 study by a group of IRS statisticians found that the Forbes 400 dramatically overestimates the net worth of individuals in their lifetimes when compared to probate records of their estates after death”. Adjusting for that discrepancy produces estimates of “an average effective tax rate of 38 percent between 2018 and 2020, as compared to the 24 percent claimed by Saez and Zucman for this period”.