Research Findings Cast Doubt on Argument That Estate Taxes Harm State Economies
January 10, 2007
In June 2001, President Bush signed federal legislation to phase out the federal estate tax. This legislation repeals the federal estate tax by 2010 and also effectively repealed by 2005 the state "pickup" taxes through which states share in federal estate tax collections. States can prevent this loss of revenue by "decoupling" from the federal change and, as of December 2006, 17 states plus the District of Columbia were decoupled from the federal changes.
In a number of states the estate tax has generated considerable debate. Opponents of state taxes on inherited wealth -- estate or inheritance taxes -- often argue that these taxes will raise little revenue and will harm state economies. The logic underlying their argument is that these taxes will make the state unattractive to wealthy families and cause them to leave the state or discourage them from moving in. A number of recent studies on elderly migration challenge these assumptions. At most, recent studies find that estate taxes have a small effect on the residence decisions of the very wealthy elderly -- few people are likely to be affected and the size of the effect is small. This suggests that the impact, if any, on a state's economy and revenue collections is also likely to be small. The loss of revenue from the estate tax, on the other hand, is likely to harm state budgets and state economies by endangering funds for public services important to the state's economy, including health care, education and infrastructure.
In a paper published in 2006, Professor Karen Conway of the University of New Hampshire and Professor Jonathan Rork of Vassar College used a multi-year framework to study migration patterns of elderly residents of all income levels. They concluded, "Our research casts doubt on the view that the elderly react to state EIG [estate, inheritance and gift] tax policies in making their migration decisions."
A 2004 study by Professor Jon Bakija of Williams College and Professor Joel Slemrod of the University of Michigan Business School focused on the impact of taxes on just the most wealthy elderly -- those that file federal estate tax returns. Their study is often cited by estate tax opponents because they found that high state estate and inheritance taxes reduce the number of federal estate tax returns filed in the state. But, the effect they found is very small. They concluded that any resulting revenue losses "would not be large relative to the revenue raised by the tax."
One reason that state estate taxes are unlikely to have any measurable effect on state economies or people's decisions about where to live is that very few estates are subject to the tax and the tax rate is low for large estates that do owe estate tax. Bakija and Slemrod also found that the average effective rate of state estate taxes in states that have decoupled from the federal tax is 4.5 percent. This is in contrast to the rate of 16 percent often cited by estate tax critics who generally emphasize only the top marginal rate.
Census and Internal Revenue Service data on migration also suggest that state tax policy has little impact on the size of a state's elderly population. A relatively small percent of seniors -- one half of one percent -- move across state lines each year for any reason and most cite family, job or housing related reasons for the move. In addition, separate studies by Wisconsin and Iowa of the patterns of in- and out- migration of seniors conclude that tax levels appear to have no effect on these patterns.
In summary, recent research does not support the argument that states should reduce or eliminate their estate or inheritance taxes as an economic development measure designed to attract or retain elderly residents. The number of elderly residents in a state is more likely to be affected by other factors such as climate, the closeness of relatives, the availability of jobs and access to services such as health care than by tax policy. These new studies reinforce the conclusion reached in a paper published in 2000 by economists at Syracuse University that "states should focus on marketing their amenities, rather than using fiscal policy to recruit retirees."
Two other points about state estate taxes are worth noting from an economic point of view. First, if a state repeals its estate tax, other states may reap a substantial share of the benefit. This is because the estate's heirs may actually live in other states and are likely to spend or invest their inherited dollars in those other states.
Second, the revenue loss from repealing the estate tax is typically quite substantial and may harm, not help, a state's economy. This is because states must balance their budgets. A state must compensate for the loss of estate tax revenue by spending less on public services important to its economy such as education or health care or by collecting more in other taxes on state residents than it otherwise would have.
Retaining estate tax revenue could make a state a more attractive place to live by helping pay for public services and allowing the state to avoid raising other taxes.
Conway and Rork, 2006
In a paper published in 2006, Professor Karen Conway of the University of New Hampshire and Professor Jonathan Rork of Vassar College studied the effects of state estate and inheritance taxes on elderly migration rates using census data from 1970 to 2000. They conclude that estate and inheritance tax rates have little impact on the migration decisions of people over age 65. To the extent that there is any connection between estate or inheritance taxes and migration, their findings suggest a quite different relationship. Rather than low estate or inheritance taxes causing the elderly to move to a state, it appears that an influx of elderly residents can result in the state subsequently lowering these taxes. This result could be explained by elected officials responding to the increased political clout of the elderly as their numbers grow. It can also explain why other researchers that rely on data from one point in time rather than from a series of years, as these authors do, find a relationship between tax levels and the number of elderly residents in a state.
In the conclusion of their paper Conway and Rork state, "Our research casts doubt on the view that the elderly react to state EIG [estate, inheritance and gift] tax policies in making their migration decisions. In fact, using two different analyses we find evidence that the causality may instead run in the other direction -- states that experience high elderly net in-migration are more likely to subsequently eliminate or reduce their EIG taxes."
Bakija and Slemrod, 2004
A 2004 study by Professor Jon Bakija of Williams College and Professor Joel Slemrod of the University of Michigan Business School focused on the impact of taxes on the most wealthy elderly. They did find that high state estate and inheritance taxes have a statistically significant, but small, negative effect on the number of elderly in a state, as measured by the number of federal estate tax returns filed in the state.
Because this paper is often cited by opponents of state estate taxes, it is important to put the results in perspective. Bakija and Slemrod examined the implications of their findings on the concern that decoupling from the federal estate tax would end up reducing rather than increasing state revenues. This could occur if the estate tax caused a significant number of the elderly to leave the state and thus reduced collections from other state and local taxes such as income and sales taxes enough to offset the estate tax revenues that are gained by retaining the estate tax. According to the study, that is unlikely to be the case: "... our results imply that in the case of a decoupled estate tax, revenue losses and deadweight losses from these particular forms of behavioral response [tax-induced migration and related avoidance activities] are unlikely to be large relative to revenues collected." See box for more details on their estimates. Their finding that any such revenue losses would be small relative to the revenue raised by an estate tax is a reflection of the modest impact of state estate taxes on elderly migration.
Bakija and Slemrod note that the federal estate tax filing data they use does not allow them to distinguish between people who completely move out of a state and those who simply change residency for part of the year. The impact on a state's economy of the decision of an elderly person to "move" out of state could be very small if these wealthy elderly residents are merely changing their legal residences by adjusting where they live for a few more months of the year rather than moving out of the state entirely, because they would continue to pay property, sales, and perhaps even income tax to the state. This is a very real possibility for wealthy families, as some two-thirds of people over 65 who are in the top two percent of net worth own second homes according to the Federal Reserve's 2001 Survey of Consumer Finances.
Any State Revenue Losses That Might Result From Decoupling From The Federal Estate Tax Cut Would Be Small
In their 2004 paper on the impact of state taxes on the wealthy elderly, professors Bakija and Slemrod did find a statistically significant but fairly small effect of state taxes on the location decisions of wealthy elderly. In order to put these results in context, they examine what the results of their analysis imply about the revenue effects of decoupling a state's estate tax from federal law.
First, they estimate that behavioral responses (i.e. either moving or other tax avoidance measures such as changing reported state of residence) of the level they found could reduce the amount of revenue collected from a state estate tax by 6.2 percent to 13.5 percent
In addition, they make rough estimates of the additional amount of revenue that could be lost if you include the effect on taxes other than the estate tax. To do this they assume first that people move five years before death and that all revenue from income, sales and property taxes that would have been generated by these residents is lost. (As the authors note this is a worst-case scenario as in reality it is unlikely that all of these individuals would completely pull up roots, sell their in-state property and spend no more time in the state.) Using the assumptions that people leave five years before death because of concern over a state estate tax and incur no taxes in the state in those years, they find that the total amount of tax revenue lost would range from 10.7 percent to 23.2 percent of the revenue generated by decoupling. When they assume that elderly residents leave 10 years before death, these estimates increase to 15.2 to 33.1 percent. Some critiques of decoupling -- including a Wall Street Journal editorial -- have cited these results as showing that as much as one-third of new revenue would be lost as a result of out-migration if a state decouples. While this is technically accurate it does focus on the absolute worst case scenario and ignores the lower and likely more realistic estimates.
Another conclusion of the Bakija/Slemrod paper is that the average effective rate of state estate taxes in states that have decoupled from the federal tax is 4.5 percent. This average rate is fairly consistent across estates of various sizes above $675,000, because of the deductibility of the state estate tax and the increasing number of other deductions -- such as for charitable contributions -- that larger estates take. This is considerably less than the 16 percent rate cited by estate tax critics who generally emphasize only the top marginal rate.
Other Analyses and Data
The Census Bureau prepares a report annually on the number of people who move. This survey includes information on characteristics of these movers as well as a question asking about the reason for the move. The most recent survey covers moves between 2004 and 2005. Data on migration patterns from the Census Bureau's survey also lead to the conclusion that the impact of taxes on the number of seniors in a state is likely to be small.
First, the vast majority of people over 65 do not move. And when they do move, very few seniors -- only 0.5 percent -- move across state lines each year for any reason. The most common reasons for moving cited by those over 65 are family-related reasons. The next most common reasons given are housing-related such as to move into new or better housing or less expensive housing. Seniors also report moving for work-related reasons, as well as because of climate or health. The Census Bureau survey does not specifically ask whether taxes are the reason for a move.
Officials in Wisconsin analyzed the interstate migration patterns in the Census data in order to test the theory that seniors move in or out of their state as a result of tax levels. In a brief analysis included in a memo accompanying testimony on three bills submitted to the Wisconsin legislature in 2005 that would have reduced or eliminated that state's estate tax, the Revenue Department of the state of Wisconsin addressed the question of whether tax levels have caused the elderly to leave the state. Because one motivation for these bills was a concern that too many elderly were leaving the state as a result of taxes, analysts in the Wisconsin Department of Revenue compared elderly migration patterns in their state to nearby states with different tax structures. They concluded that there is "little evidence that tax treatment has a significant effect on residence decisions by the elderly."
The Fiscal Services division of the Iowa Legislative Services Agency used data from the Internal Revenue Service on taxpayer migration for a similar analysis. Their focus was on the income tax but the results apply to the debate over the estate tax as well. Specifically they examined migration into Texas and Arizona from Iowa and other surrounding states to see if tax differentials appeared to be playing a role in these moves. They concluded that, "While Iowa loses taxpayers and taxable income to both Texas and Arizona, Internal Revenue Service state-to-state migration data indicate that differences in state income tax policy do not explain Iowa's out-migration to those states, as both higher and lower income tax states around Iowa have similar or even greater relative losses than Iowa."
In conclusion, recent research does not support the argument that states should reduce or eliminate their estate or inheritance taxes as an economic development measure designed to attract or retain elderly residents. The number of elderly residents in a state is more likely to be affected by other factors such as the closeness of relatives, the location of their places of work and access to services such as health care than by tax policy.
There are two other points from the general economic literature on taxes and economic growth that are relevant to this discussion. First, some portion of the benefits from estate tax repeal likely will accrue to heirs who live in other states. Those heirs likely will spend those dollars in those other states, with the result that only a portion of the dollars lost from estate tax repeal will remain within state borders. By contrast, public expenditures and even many other types of tax cuts are much more likely to lead to spending within a state's borders, thereby providing some economic stimulus within the state.
Second, states must balance their budgets, so the loss of revenue from estate tax repeal must be offset by lower public expenditures, increases in other taxes, or a combination of the two. Since a large share of state expenditures are for services that are important to economic growth such as education, health care, and infrastructure, repeal of an estate tax runs the risk of hurting rather than helping a state's economy.
In short, the negative effects of estate tax repeal on a state's future economic growth could easily outweigh any small positive impact of preventing migration that eliminating the estate tax might produce.
 The estate tax in Virginia will expire effective July 2007, in Wisconsin in 2008 and in Kansas in 2010.
 Aging Studies Program Paper No. 22, "Chasing the Elderly: Can State and Local Governments Attract Recent Retirees?" William Duncombe, Mark Robbins, and Douglas Wolf, Center for Policy Research, Maxwell School of Citizenship and Public Affairs, Syracuse University, September 2000.
 Conway, Karen Smith and Jonathan C. Rork, "State 'Death' Taxes and Elderly Migration -- The Chicken or the Egg?" National Tax Journal, Volume LIX, No. 1, March 2006 available at http://pubpages.unh.edu/~ksconway/NTJ_Conway_Rork_Finalv.pdf.
 Bakija, Jon and Joel Slemrod, "Do the Rich Flee from High State Taxes? Evidence from Federal Estate Tax Returns". NBER Working Paper No. 10645, July 2004.
 Bakija, Jon and Joel Slemrod, "Do the Rich Flee from High State Taxes? Evidence from Federal Estate Tax Returns", Abstract, NBER Working Paper No. 10645, July 2004
 Geographical Mobility, U.S. Census Bureau, Population Division, Tables revised September 24, 2006.
 State of Wisconsin Department of Revenue memo dated April 13, 2005 on "Senate Job Creation, Economic Development and Consumer Affairs Committee Hearing, April 19, 2005"
 State of Iowa Fiscal Services, Legislative Services Agency, Taxpayer Migration - Iowa to Texas and Arizona, February 10, 2005.