Research shows that people leave and avoid high tax states

Yesterday I wrote about how Minnesota is losing residents to other states. Many factors go into an individual’s decision as to where to live, work, and play. But, as we explain in our new report, ‘Taxes and Migration Minnesotans on the Move to Lower Tax States,’ one of these is the comparative tax burdens that different jurisdictions will subject them to.

There are those, policymakers included, who deny this. Empirical research indicates that those people are wrong.

The effect of tax burdens

Since the 1970s, economist Richard Cebula has published studies attributing interstate migration flows to a range of factors, including state economic growth, housing costs, taxes, climate, crime rates, and public school spending. He has consistently found that income and property tax burdens are statistically significant in explaining interstate migration. Looking further at overall burdens, economist Yu Hsing calculated an ‘optimal’ value for the state and local tax burden and found that “Interstate in-migration rates are likely to increase (decrease) given a change in tax burdens, if the current level is below (above) the optimal tax level.” Economist Robert R. Preuhs examined “the effects of state policy on net interstate migration” and found that “states with low taxation levels, high investment-consumption ratios, and more liberal ideologies relative to other states, tend to experience more population growth via interstate migration.”

The effect of income taxes

One of the most studied taxes is the individual income tax.

A study by economist Mark Gius found that “income taxes have an effect on migration for most races and age groups. Individuals move from states with high income taxes to states with low income taxes.” Economists Karen Smith Conway and Andrew J. Houtenville looked at “whether elderly migration is affected by state fiscal policies” and found that, in addition to other factors, “Low personal income and death taxes also encourage migration.” A study by economists David Clark and William Hunter finds that “working males in their peak earning years are detracted by high income taxes.” A paper by economists Antony Davies and John Pulito concludes that “higher state income-tax rates cause a net out-migration not only of higher-income residents, but of residents in general.” The results of a study by economists Roger Cohen, Andrew Lai and Charles Steindel “indicate that variations in differential average marginal tax rates are associated with small but significant effects on net outmigration from a state.” In another paper, Cohen, Lai and Steindel found that New Jersey’s increase in its top individual income tax rate in 2004 induced net out-migration of 80 or more millionaires a year. This is a small number of people, but given the progressivity of income taxes, they have an outsized impact on state tax revenues. These findings were supported by research by economists Joshua Rauh and Ryan J. Shyu which documented “a substantial one-time out-migration response to increased state tax rates” with “strong behavioral responses to income taxation amongst high earners.”

A recent paper by economists Henrik Kleven, Camille Landais, Mathilde Muñoz and Stefanie Stantcheva that “review[ed] what we know about mobility responses to personal taxation” found that:

There is growing evidence that taxes can affect the geographic location of people both
within and across countries. This migration channel creates another efficiency cost of
taxation with which policymakers need to contend when setting tax policy.

Kleven, Landais, Muñoz and Stantcheva do, however, offer two arguments “against overusing these empirical findings to argue in favor of an ineluctable reduction in the level of taxation or progressivity.”

First, they point out that “while the mobility responses documented in some of the recent literature are striking and perhaps surprisingly large, they pertain to specific groups
of people and to specific countries.” While true, it is also true that high tax rates only affect relatively small numbers of people of which the groups studied are not unreasonable samples and who research shows tend to be more responsive to tax rates. It should also be pointed out that four of the twelve studies they cite relate to movement between states of the union.

Second, they note that “the strength of the mobility response to taxes…depends critically on the size of the tax jurisdiction, the extent of international or sub-national tax coordination, and the prevalence of other forces that foster or limit the movement of people, all of which can also be affected by policies.” While states remain free to set their own tax rates, the scope for “sub-national tax coordination” will be limited. Furthermore, movement between the states is much easier than movement between different countries, so we ought actually to see greater mobility responses resulting from state tax policy.

Neither of these points, then, should stop us from applying the findings of empirical research to Minnesota’s tax policies: people move in response to taxes.

The effect of wealth taxes

Kleven, Landais, Muñoz and Stantcheva look at papers estimating mobility responses to personal income taxes, but there is also evidence that taxes on wealth and capital income drive migration.

In addition to Conway and Houtenville’s find ings on the effects of the estate tax cited above, in another paper they find that “all elderly age groups avoid moving to states with high estate/inheritance/gift taxes” and that “the younger elderly appear to be ‘shopping around’ for destinations with a temperate climate and favorable government policies regarding income taxes and welfare spending, whereas the older elderly are more likely to be ‘driven out’ of their origin state by a high cost of living and income and property taxes.” Clark and Hunter, also cited above, find that “all migrants aged 55 to 69 avoid counties in states with high inheritance and estate taxes.” A paper by economists Jon Bakija and Joel Slemrod finds that high inheritance and estate taxes have statistically significant, if modest, negative impacts on the number of federal estate tax returns filed in a state.19 A study focusing on individuals from the Forbes 400 in the United States suggests that mobility responses to estate tax incentives might be larger at the very top of the wealth distribution. This paper, by economists Enrico Moretti and Daniel J. Wilson, weighs the revenue brought in by state estate taxes against the revenues lost when people leave the state and take future payments of income and sales taxes with them. They find that, while “the [revenue] benefit [of an estate tax] exceeds the cost for the vast majority of states,” this was not the case for Minnesota. Ours is one of four states where the costs in terms of lost revenues from other taxes outweigh the benefits in terms of estate tax revenues. Those states are the ones with the highest top rates of income tax: Hawaii, Minnesota, Oregon and Vermont. This echoes the findings of our own study in 2018 and strongly suggests that a state can impose either an estate tax or a high top rate of personal income tax, but it cannot impose both without people leaving and taking their revenues with them.

Taxes are (dis)incentives

These findings should not be surprising. After all, taxes are (dis)incentives and much public
policy, from cigarette taxes to carbon taxes, is based on the notion that people respond to the (dis)incentives provided by taxation. But, while policymakers generally embrace this logic when it comes to the fight against smoking or global warming, all too often many of those same policymakers abandon it when it comes to taxation of incomes or wealth. For some reason, people who think that smoking will decline if the tax on smoking is increased think that there will be no effect, that people will just go on working and investing exactly as before, if taxes on labor or investment income are increased. This discontinuity in logic is hard to account for, but it is an obstacle to sound, evidence based public policy.