May 15, 2005 to May 22, 2005
The Estate Tax
Posner on Estate Taxation
I agree with Becker that inheritance taxes are preferable to estate taxes and that consumption taxes are preferable to income taxes. However, I do not share his strong opposition to the federal estate tax.
The government needs revenues, and taxation is the principal means of obtaining those revenues. An ideal tax from an economic standpoint is one that generates substantial revenue without distorting the allocation of resources. These turn out to be linked concerns. A narrow-based tax, such as a tax on sports cars, is undesirable from this standpoint because, unless the tax rate is very low, in which event little revenue will be generated, the tax will deflect consumers to close substitutes that are not subject to the tax, and thus little revenue will be generated, because of substitution away from the taxed product.
A recent article in the New York Times calls the estate tax "perfect" on the ground that it does not distort the allocation of resources because people can't escape death. Were this true, the estate tax would be analogous to a "head tax''say a tax of $1,000 a year on every U.S. citizen. Such a tax would be difficult to escape because substituting away from the taxed activity (or rather status) would require expatriation, which is very costly to an individual. The tax would generate almost $300 billion a year in revenue. The estate tax is not nearly so "perfect" as a head tax, because contrary to the Times article it is easily avoidable, as a head tax is not. I emphasize "easily"; for that is the reason the problem with the estate tax is not, as it might seem to be, that it is a tax on savings and can thus be avoided only by consuming all one's wealth before death and therefore is likely to distort people's consumption decisions, which would make it an inefficient tax. People who have no bequest motive'that is, who are not interested in having a positive balance when they die'are not greatly affected by estate taxation because they already have every incentive to dissipate their wealth before they die. People who do have a bequest motive are not much affected by the estate tax either, because they can transfer their assets to their children or to other intended heirs before death, reserving the income from the assets for their lifetime (the equivalent of an annuity, which expires on the annuitant's death, leaving nothing for distribution to heirs).
As a result of these incentives and opportunities, the estate tax does not generate much revenue for the government. It collects some, however'and I do not consider 1 percent of total federal tax revenues a trivial amount'and its distortionary effects are probably modest. Becker is correct that it is costly in effort expended by lawyers to minimize the impact of the tax on their clients (though he may exaggerate the cost by attributing the entire income of estate-planning lawyers to tax counseling, when even in the absence of estate taxation legal counseling would be required for the preparation of wills and other documents required for large estates). But in that respect, it is no different from the income tax. Were the estate tax abolished, the revenue it generates would have to be made up by some other tax, which is as likely to be as distortionary as the estate tax and would invite avoidance efforts by tax lawyers.
The more interesting question to me, though I have no good answer to it, is whether the estate tax should be stiffened, or other measures used to limit bequests. An article in last Friday's Wall Street Journal, echoing remarks that President Summers of Harvard made at a conference at the Kennedy School that same day, notes a possible, and possibly troublesome, decline in social mobility in the United States. Wealthy people seem increasingly able to guarantee that their children and even grandchildren will remain in the upper income tier, leaving fewer places for the children and grandchildren of the poor to occupy. Through "legacy" admissions (as at Harvard!), expensive private schooling and tutoring, including tutoring in taking college admission tests, as well as by means of direct transfers of wealth, wealthy people are able to "purchase" a secure place for their children and grandchildren in the upper class. Even if, as Becker argues, social mobility has not actually declined in recent decades, it is lower than it used to be and the conditions for a decline seem in place.
But the normative implications are unclear. For example, one concern with declining social mobility is a fear that rich kids won't work as hard as poor ones, so that economic growth will lag. But if they don't work as hard, they will lose jobs to the poor. They may continue to live quite well by clipping coupons, but the poor (or rather the former poor) will occupy the high-paying jobs. And because rich kids can take financial risks that the poor cannot, and risk-taking is important to innovation and hence to economic growth, bequests may lead to an increase rather than a reduction in economic growth, though this depends on the balance between the drag on growth from rich kids" working less hard and the spur to growth from their taking more financial risks.
Furthermore, the positive correlation between parents" and children's wealth may conceal the actual causality. It may be that the parents are wealthy in part because of a genetic endowment that they pass on to their children, who because of that favorable endowment would become wealthy even if they didn't inherit any money. But probably wealth does enhance the advantage in having such an endowment.
If lack of social mobility is a problem, nevertheless it is unlikely to be solved by trying to limit bequests, since wealthy people can transfer much of their wealth during their lifetime. To have an impact on the transmission of wealth across generations, therefore, a stiff tax on bequests would have to be complemented by a stiff tax on gifts, and "gift" would have to be broadly defined to include such things as paying $50,000 a year for tuition and expected donations at a fancy private school in New York City. And really stiff estate and gift taxation, even if feasible, would be undesirable because of the disincentive effect on the work effort of those people'and they are numerous'who, to a significant degree, are motivated to become rich by a desire to make their children and grandchildren better off than they would otherwise be.
There is a traditional concern with dynastic fortunes'that is, with accumulations of wealth that are so great that they confer disproportionate political power on a family. The founder will usually be too busy making money to participate heavily in public affairs, though there are exceptions, such as Joseph P. Kennedy, President Kennedy's father; Michael Bloomberg, New York's mayor; and George Soros, the billionaire backer of the Democratic Party. The next generation, the generation of the inheritors rather than the creators of wealth, may decide to devote full time to public matters, for good or for will. Concern with accumulating political power over generations lies behind the esoteric "rule against perpetuities," which forbids making a bequest that will not take effect until more than 21 years after the death of currently living persons'this to prevent transmitting wealth to one's remote unborn descendants. As a curious tandem to the movement of abolish the federal estate tax, many states are allowing the rule against perpetuities (a rule of state rather than federal law) to be undone by the device of the well-named "dynasty trust," whereby a wealthy person places money in trust with instructions that the trustee invest the money for the benefit of specified beneficiaries, such as the descendants'however remote'of the creator of the trust for as long as there are such descendants. Depending on the amount doled out by the trustee in each generation, the trust might over time accumulate enormous assets simply by the operation of compound interest. The device is quite recent yet already some $100 billion have been placed in dynastic trusts. See the study by law professors Robert Sitkoff and Max Schanzenbach, forthcoming in the Yale Law Journal.
Should we worry about the dynastic trust? Probably some degree of wealth inequality is potentially destabilizing politically. But on the other hand the creation of centers of private power acts as an offset to growing governmental power and so may actually serve to preserve liberty. Notice in this connection that abolishing the estate tax would reduce the incentive to make charitable bequests, which are tax exempt.
I should note finally two possibly illusory aspects of the proposed abolition of federal estate tax. One is that states may respond by increasing their own estate taxes, which are less efficient than federal estate taxes because it is easier to evade a tax by moving from one state to another than by expatriating oneself, and so such taxes affect locational decisions more than the federal tax. Second, the current estate tax gives heirs a "stepped up" basis in capital that they inherit. That is, should they later sell a capital asset that they inherited, the cost basis for computing how much capital-gains they owe will be the value of the asset at the death of their testator rather than the cost that the testator incurred to acquire it originally. So abolition of the federal estate tax would be offset to an unknown degree by increased capital-gains taxation of heirs, and also by increased administrative expense since it is often difficult to determine the original value of an asset that was acquired many years earlier.
Should the Estate Tax Go? BECKER
The tax cut law of 2001 included a slow phase out of the estate tax by 2010, but the tax is supposed to be reinstated in 2011 when the entire 2001 law expires. This strange political compromise on estate taxes presumably will not last, so it is a good time to consider what should be done about this tax. I believe taxes on estates should be permanently abolished since they do little to reduce income or wealth inequality, benefit a vast army of lawyers and accountants whose role is to find ways to cut taxes on the estates of the wealthy, create problems for some families with smaller businesses, and do not raise a lot of tax revenue. In April the House of Representatives rather strongly voted to repeal permanently the tax on estates, so the issue now goes to the Senate, where some Democrats are promising a filibuster.
In earlier times, bequests of assets, especially property, were the dominant way to pass economic wealth from parents to children. But after the knowledge revolution took off toward the end of the 19th century, bequests of financial and material wealth have become less important in the overall economy. Instead, the most important way for parents to "bequeath" economic position is through the transmission of knowledge in the form of education, training, and other human capital. Such capital embodied in people now comprises over 70 per cent of all "wealth" in economically advanced nations, far more important than material capital.
In modern economies children of better educated, higher earning, and more able parents on the average receive greater training and schooling, better health, and are more encouraged to develop their talents than are children in other families. Primarily for these reasons, children of parents with greater human capital form an economic elite that tends to have better jobs, less unemployment, and higher earnings. But this elite circulates over generations, and there is no convincing evidence that the degree of circulation, the degree of social mobility across generations, has been falling during the last couple of decades.
Some defenders of a sizeable estate tax rate claim not any major effect on inequality, but that it allegedly brings in lots of revenue with little disincentive to wealth accumulation and other behavior. However, estate and gift taxes in fiscal 2005 are expected to contribute only $24 billion in federal tax revenues, which is about 1 per cent of estimated total federal tax receipts, and just one third of federal revenue from excise taxes. The rise in exemptions may have reduced revenue from estate taxes, but this tax did not contribute much more even a decade ago. $24 billion is not small change, except to politicians, but if the estate tax were abolished, the lost revenue could be made up without difficulty with only modest increases in income or consumption taxes.
A main reason for the small yield of estate taxes is that very rich persons with large estates often pay little to the government since they employ skilled lawyers and accountants to help them find ways to sharply cut their estate taxes. Trust and estate planning is the specialty of about 20,000 lawyers who, along with accountants, spend their expensive time discovering ways to reduce the amount owed in estate taxes. These ways include gifts, trusts that may skip a generation, insurance trusts, and charitable trusts and foundations. These talented individuals should be spending their time in more economically productive ways. Since the average estate-planning attorney earns more than $150,000, spending on 20,000 of them would be in excess of $3 billion. If another $1 billion goes to estate accountants, total spending on tax avoidance would seem to be in excess of $4 billion. This is more than 1/6th of the revenue generated by this tax, a strikingly high percent.
The estate tax is a bad way to reduce the effect of inheritances on inequality in the distribution of incomes and wealth, even for families that do leave large estates. For this tax does not consider how many children, parents, other relatives, or friends share estate resources. A parent who leaves $10 million to an only child has a larger effect on the personal inequality of wealth in the next generation than does someone leaving the same amount to be divided among several children, nieces, nephews, friends, and employees. Similarly, a large bequest to successful children with high incomes raises inequality more than does the same size bequest to children with low or just average incomes.
This is why taxing inheritances rather than bequests would be a better way to reduce inequality in succeeding generations. That is, $10 million in bequests divided five ways should generally be taxed at much lower rates than the same amount given to one person, while $10 million divided among several well off children should also be taxed at higher rates than the same amount divided among children with modest incomes. Although an inheritance tax would be better than the estate tax, I am not advocating a direct tax on inheritances, for there is a better approach that indirectly does tax inheritances (see the discussion later of consumption taxes).
The estate tax also makes it harder for families to pass successful businesses on to their heirs. Despite the 2001 tax law that increased exemptions, families are still sometimes forced to sell more successful and profitable businesses upon death of the principal owner in order to pay estate duties. This is why farmers and other owners of small businesses continue to be active politically in advocating much lower estate taxes, if not their complete abolition.
High tax rates on estates may be thought to be universal, but in fact many countries have low taxes on estates. Moreover, some countries, including Canada and Switzerland, essentially have not taxed bequests to close family members, although they may tax capital gains on assets transferred to children.
I cannot go deeply in this discussion into the reasons why I believe a tax on consumption, perhaps a progressive one, instead of income and corporate taxes, should form the heart of the federal tax system. Suffice it to say that consumption taxes, unlike income taxes, do not distort savings decisions, a particularly important issue for the United States.
A general reliance on consumption taxes would, among other things, replace an estate tax by indirect taxes on inheritances. The tax on inheritances would be indirect because they would not be taxed when they are received, but only as they are spent. So if a family receives a large inheritance that raises their consumption several fold, the amount they would pay in consumption taxes would also increase several fold for as long as the family continues to consume at a much higher level.
So my conclusion is that the estate tax should go, or at least have greater reduced rates, since this tax has little effect on inequality in a knowledge economy, encourages costly avoidance behavior to take advantage of various tax loopholes, raises only a modest amount of government revenue, and reduces incentives to form family businesses and other entrepreneurial activity. Estate taxes do not even tax the right base if the aim is to reduce the effect of inheritances on inequality in the personal distribution of income and wealth. The energy and political capital spent on supporting high estate taxes is better spent on trying to raise opportunities to children from poor families by improving their education, training, and health.
Reply on Estate Tax Elimination-BECKER
I enjoyed reading all the comments. As the discussion makes clear, the issue of estate taxation raises many interesting questions that would take more than one posting to cover at all adequately. Let me try to respond to some of the points raised.
As some of you indicated, and others denied, the estate tax is not necessary to get a circulation of the very rich since wealthy families did regress over time even before the estate tax became important. Although many of the descendants of John D. Rockefeller are still quite rich, as a family they have fallen greatly in wealth compared to the Gates" and other much newer wealthy families.
There is nothing intrinsically regressive about consumption taxes. There can be various rebates, exemptions, and credits that would make the overall system progressive in the usual meaning of that term: that the average tax rate rises with level of consumption, or that marginal rates rise at some consumption levels and do not fall at any others. My preferred way to make a consumption tax progressive is via exemptions. The most efficient way to implement a progressive consumption tax along these lines is to allow all savings to be deducted from income, and then tax the residual (which is consumption), with an exemption at the lower end that should be quite generous.
The NYTimes article that referred to my 1987 Presidential address to the American Economic Association badly misstated what I said. I did not claim that children's income was not much related to the income of their parents. In fact I assumed for the sake of discussion that about 40% of the parents" income advantage was passed on to children. Note that grandchildren would then only have about 16% of the advantages of their grandparents. It is true that some recent work claims the fraction passed on might be as high as 50% rather than 40%, but that is controversial, as Dean Lillard of Cornell and others have argued. I also stand by my claim that there is no credible evidence that the degree of intergenerational mobility has fallen during the past few decades.
If consumption were taxed, the basis would automatically be "stepped up" since that true basis would determine consumption spending. Even if one does not like a consumption tax, surely an inheritance tax is much better from any equity standard than the estate tax. I agree that the value of the inheritance for tax purposes should be based on the market value of inheritances, not the purchase price.
I do not believe that if the Federal estate tax were eliminated, it would simply be replaced by much higher estate taxes by the states. For it is easy for many rich persons to change their state of residence by moving to states with lower estate taxes. That is certainly a lot easier than changing country of residence, and a considerable number of the very rich even do that.
Estate Taxation--Posner's Response to Comments
My posting precipitated an interesting debate in the comments about the ethical, political, and economic issues presented by inequality of wealth, which has become very great in twenty-first century America. These issues are important, but estate taxation is only peripherally related to them. Without unforeseeable increases in estate tax rates, coupled with extremely stiff gift taxes, estate taxation will continue to have negligible effects on wealth inequality. I agree with Becker, moreover, that even without explicit gifts wealthy people can transfer substantial wealth to their children by investing in the children's human capital (earning capacity)--not to mention the genetic endowment that high-IQ parents transfer to their children. I believe that these transfers are increasingly important, for two reasons. First, with the decline in the importance of strength and stamina as factors of production, the economic return to intelligence has risen. And second, with the breakdown of traditional barriers, such as religion and ethnicity, to assortative mating (likes with likes), there is more matching of IQs in marriage and so a greater production of highly intelligent people.
I do not think there would be an ethical objection to efforts to reduce the inequality of wealth. Even if one does not regard one's genetic endowment as a form of unearned luck (as I do not--"luck" to me refers to purely adventitious factors in one's success or failure in life), luck plays an enormous role in wealth; stated differently, the variance in wealth is much greater than the variance in intelligence, character, effort, or all these things combined. And if wealth could be equalized costlessly, there would be a net gain in economic welfare because of the phenomenon of declining marginal utility of income--that last dollar is worth more to a poor person than to a rich, so transferring a dollar from the rich to the poor will increase aggregate utility, and this effect could continue until incomes were equalized.
The objection to efforts to equalize wealth, including by drastic changes in estate and gift taxation, is that they are very costly. The have adverse incentive effects on both rich and poor, and a variety of other negative consequences as well. Paradoxically, equalizing wealth can increase envy, because one is more likely to envy someone who is slightly better off than one is than someone who is unimaginably better off. Few people envy Bill Gates, because they cannot imagine what they would do with so much money; but they know very well what they would do with the additional income of their slightly wealthier next-door neighbor.
It is important to recognize, moreover (a point that Oliver Wendell Holmes, Jr. stressed repeatedly), that personal wealth, no matter how great, is a part, and a constructive part, of aggregate social wealth. The rich do not burn their money, or put in boxes under their beds. If saved, the money is invested; if consumed, it provides incomes to the people who produce the goods that the rich buy; if given to charity or to politicians, it affects, not necessarily for the worse, the social, cultural, and political character of the society. The opportunity to amass wealth also channels the ambitions of aggressive people into relatively harmless channels, even if Samuel Johnson exaggerated when he said that people are rarely as innocently engaged as when they are making money.