Suppose you wanted to judge which political party was a better steward of the national economy over the past 75 years.

One way would be to select a half-dozen common economic indicators, such as inflation, unemployment rate, GDP growth or per capita incomes, and compare the performance under Democrats and Republicans. If you did so, you’d quickly discover there’s no pattern.

If you did the same thing at the state level, up until about 1980, you’d find the same thing. Party affiliation played very little role in economic differences. In fact, if you picked just one variable that best measured prosperity—per capita income—there was no state-level correlation with a political party in 1980. There were rich states led by Republican and Democratic governors and poor states led by both as well.

That has changed. Of the 20 richest states today, 19 are solidly Democratic. Of the poorest 20 states, 19 are solidly Republican. The GOP dominates in poor, slowly growing states, while the Democrats dominate politics of prosperous, faster-growing states. The reason isn’t clear, and it certainly isn’t solely due to policy differences.

One big culprit is that political parties changed, erasing regional party differences. Up until the late 1990s, there were many conservative Democrats and liberal Republicans. Think Georgia Democrats and Maine Republicans. Without room for state-level differences in what it meant to be a Republican or Democrat, states began to align with national politics.

This trend is more extreme today. Even races for local municipal government tend to be nationalized. State and local issues are often ignored, or barely discussed in primary or general elections. The homogeneity of national politics will naturally cause parties to represent more similar places.

A second trend is household sorting by politics. Though most of sorting happens at the sub-state level, the nationalization of politics means that state borders now affect household location choice.

The nationalization of politics means that each party has been staking out positions that appeal to majorities in key states. As households sort, voters are choosing political landscapes they prefer. That will almost surely accelerate in the coming years.

The nationalization of state and local politics also means the nationalization of state and local policies. A quarter-century ago, when I first started serious research in state and local policy, state legislators rarely took their legislative marching orders from national think tanks or national parties. Today, elected leaders from both parties are expected to advance similar legislation, typically written by think tanks, everywhere at once.

The result is nearly identical legislation to raise the minimum wage, eliminate income or property taxes and create universal school choice appear in both Oklahoma and Massachusetts—the two most economically and electorally different states in the union.

The nationalization of American politics could have very uneven effects at the state and local levels. In affluent states and cities, policy mistakes usually don’t matter much. In poor places, they matter enormously.

San Francisco, Chicago and Seattle are arguably among the worst-governed cities in North America. Yet, their economies buzz, people visit, and even if a few people leave, home prices reflect the unending desire to live there.

A poor state or a declining rural town in the Midwest has no such luck. A bad decision or two could extend decline for decades. Deep tax cuts at a time when mobile residents are screaming for better services can end decades of hope for a stable population.

The most likely cause of divergence between rich and poor places is the fact that human capital—education, innovation and invention—replaced manufacturing and movement of goods as the primary source of prosperity. In other words, places that grow will collect more human capital. However, the educational policies pursued by both parties are vastly different.

The GOP has largely tried to adopt broad school choice, while cutting funding to both K-12 schools and higher education. The Democrats have largely eschewed school choice, but amply fund both K-12 and higher education. Today, 17 of the 20 states with high educational spending are Democratically controlled and 17 out of the 20 lowest funded states are GOP strongholds.

There’s more to education than the spending. Still, higher educational spending, even if it means higher tax rates, is leading to enrollment and population growth. Educational attainment differences alone explain about three quarters of the difference in per capita income between states.

At the same time, school choice effects are smaller than almost anyone hoped or expected. Today, it’s clear that the average student in private school underperforms their public school counterparts (charter schools tend to out-perform both). So, if poor states spend less on education and rely more on school choice, they will become poorer than states spending more on public education.

Economists have been saying this for three decades, with little effect. The prognosis is simply that poor states like Indiana are going to get poorer for decades to come while rich states will grow richer.

That’s terribly disappointing, but it is the most likely outcome.