The European Union is starting a big debate about fiscal federalism. There are many questions involved. The one I want to focus on is whether a Europe-wide fiscal arrangement with common transfers would help smooth economic performance across participating countries while at the same time helping them to converge on similar levels of income per capita. The line you often hear is that this is how federal transfers work in the United States: rich states like New York and Massachusetts bankroll poorer states in the south and west of the country both when times are tough and in order to foster the whole of the U.S. economy.

In reality, the U.S. federal transfer system does not work that way. The richer states in the north-east of the country get more federal transfers per capita than the poorer states in the south and west. The reason is that the U.S. federal fiscal system was designed to support people as individuals (or households) and not as clusters or places on the map. Moreover, that design reflects important differences across state and local governments. State governments that believe in more redistribution tend to get more redistribution; state governments that do not believe in redistribution tend to leave people to fend for themselves. In this sense, state sovereignty and democratic legitimacy are powerful influences even when the ‘states’ in question are U.S. states rather than national states (or Member States).

This is a counter-intuitive argument and so I am going to build it up using four state-by-state comparisons. Along the way, I am going to pay specific attention to federal transfers to individuals – meaning all transfers to personal income for health, education, veterans benefits, pensions, income maintenance, and unemployment except for the unemployment benefits provided by state governments. My analysis ignores federal taxes and federal spending on goods and services, including public investment. In that sense, the picture I offer is only one-half of the automatic (non-discretionary) influence of the federal fiscal system. No doubt many people will complain that I left the best parts out. Nevertheless, what it reveals is strange enough to raise questions about the conventional wisdom that common transfer arrangements in large federal organizations like the United States play a role in binding the different parts of the country together. Europeans will need to think carefully about the answers to those questions if they are to avoid importing the very divisive debates over fiscal federalism that we have in the United States.

The first comparison is between Ohio and Florida and it largely confirms what most people believe are the advantages of common institutions. For those unfamiliar with U.S. economic geography, Ohio is a large mid-western state with a mature industrial base while Florida is a large southern state with a more mixed economy. Ohio is also cold while Florida is warm. Hence people used to go to Ohio to get a job and to Florida to retire – or at least that is the cliché. And this cliché shows up in the data for income per capita and per capita federal transfers. What I have done is take the ratio of Ohio income or transfers per capita to Florida income or transfers per capita and multiplied by 100. When the ratio is 100, then the per capita income or personal transfers in Ohio are the same as they are in Florida; when the ratio is above 100, Ohio has more than Florida; when the ratio is below 100, Florida has more than Ohio.

What you see in this comparison is what you would expect. In the late 1960s and early 1970s, Ohio and Florida have roughly the same income per capita but Florida has higher personal transfers because it also has more pensioners. Then during the crisis of the 1970s, the big steel and automotive industries in Ohio took hit and income fell while transfers rose. This experience marks the start of a see-saw pattern where transfers rise when incomes fall and fall when incomes rise with the result that Ohio and Florida both end up in much the same place. The two state economies have gone through dramatic transformations over the past four decades and they are probably as different from each other now in terms of what they produce as they ever were in the past. Nevertheless they appear to share a common economic fate.

A second comparison is between Maine and Mississippi. These are relatively small agricultural economies – Maine’s population is about one-third the size of Mississippi – with a heavy dependence on pulp and paper products, although Mississippi also has some gas and oil on the coast. Both are poor. But as the figure shows, Mississippi is poorer than Maine in per capita terms. It also starts off with significantly lower levels of personal transfers per capita. This situation is fairly constant until Mississippi gets hit by an economic shock in the late 1980s (as a result of a sharp decline in oil prices) such that income goes down and transfers rise relative to Maine. When things eventually stabilize Maine is still richer than Mississippi and Mississippi gets more personal transfers than it did in the past. These are all what you would expect according to the conventional wisdom. But the two parts of the country are not becoming more similar in terms of performance and Maine still has more of both income and transfers than Mississippi receives. Hence while we see some of the dynamics we would expect, there is still something not right with this picture.

A third comparison is between New York and California. These are large and complex states with dynamic economies that are hard to characterize. Suffice it to say that they are the twin engines of the U.S. economy. Nevertheless, they have very different relationships in terms of per capita income and personal transfers. The two states start out with roughly the same income per capita at the end of the 1960s; the difference is that New York got significantly more in personal transfers. Then New York went into crisis in the 1970s and the transfers rose to compensate. The transfers continued to rise as New York emerged from the crisis and its income per capita rose above that in California. And those transfers remained much higher for New York than for California even though income per capita in New York continued to increase. When California went into its own crisis in the 2000s, the relative balance of transfers changed a little but this was hardly enough to compensate for the difference in income per capita. Despite the fact that they exist within the same federal fiscal system, New York and California do not share the same economic fate. New York gets more out of the system than California does.

A fourth comparison is between Massachusetts and Texas. These states are very different both in terms of industrial structure and in terms of political ideology. Massachusetts has a mixed economy that combines mature industries with high technology; Texas has a lot more emphasis on resource extraction and refining. Massachusetts is more centre-left; Texas is on the opposite end of the spectrum. This shows up clearly in the data. In the late 1960s, Massachusetts was richer and it engaged in significantly more redistribution. The point to note is that this redistribution operated through federal transfers (meaning, once again, it does not include state funding for unemployment insurance). Moreover, when the Texas economy boomed and the Massachusetts economy suffered in the 1970s, those transfers increased. When the situation reversed in the 1980s, the transfers reversed as well. But the balance still favours Massachusetts and the relative rise in Massachusetts income was roughly twice what it lost in the 1970s even as relative transfers moved back to roughly the same high rate. During the past two decades, Massachusetts has witnessed a gradual reduction in personal transfers relative to Texas but it still benefits from roughly 30 percent higher per capita income and 30 percent higher per capita transfers by the end of the period. Like New York, Massachusetts gets more out of the federal system. Meanwhile, Texas shares the same federal arrangements but lives in a different economic universe.

The simple explanation for the difference between what these pictures reveal and the conventional wisdom is that state governments play an important role in determining how much redistribution takes place between rich and poor households in different parts of the country. This is true even if we take out specifically state-funded personal transfers, like I did with state funded unemployment benefits in compiling the data for these pictures.

Where state governments want to engage in redistribution, they put up more money to be matched by federal funding and they invest more resources to make sure that their people get access to federal entitlements; where state governments are less interested in redistribution (or less competent), they leave the advantages that the federal tax and transfer system has to offer lying on the table. As a result, it is simply not true that the federal fiscal system automatically makes economic conditions across the United States more equitable. Indeed, there are situations where federal transfers make differences across states bigger and not smaller: New York and California are one example; Massachusetts and Texas are another. The differences are even felt across relatively poor states like Maine and Mississippi.

So the question to consider is: Why didn’t the United States build a better federal transfer system? The answers lies in how the conventional wisdom is framed. You often hear that New York and Massachusetts pay for the rest of the country – places like Mississippi in particular. This assertion is actually a complaint made by those who benefit most from the arrangements that are in place. On the other side of the debate, you have politicians in places like Mississippi looking to repeal the affordable care act, cut back on social security, and defund the federal government. If the main beneficiaries complain about federal fiscal arrangements; those parts of the country that benefit less actively hate it. Proponents of fiscal federalism in Europe should pay close attention to these U.S. debates before they risk importing them into an already fraught political situation.

None of this is to say that I am opposed to either supranational resource pooling or cross-border transfers. My concern is that the conventional wisdom is pointing the European debate in the wrong direction based on a mistaken understanding of how things work in the United States. We should clear up that misunderstanding as soon as possible because the trouble with transfer unions is the very high political cost of making a mistake in designing how the system should operate.

Follow @Erik_Jones_SAIS