Epstein: Farewell to Capital Gains Tax

Richard A. Epstein

January 5, 2007

As a new Congress convenes, the divisive issue of tax reform looms large on the agenda. The Democratic push for higher taxation, especially on the rich, is spurred by the mistaken belief that any increase in tax burden works a painless redistribution of wealth, without undermining wealth creation. The Democrats' overarching position constitutes an economic blunder of Newtonian proportions: It assumes that profound government actions will produce no private reactions.

One of the big topics will likely be capital gains taxes, which many will argue are now too low. But there are dangers in that position. Notwithstanding President Bush's many economic sins, revenue from capital transactions has surged in light of the 2003 reforms that reduced the maximum rate on long-term capital gains to 15%. The explanation is simple enough. Capital gains taxes apply only when the owner of a capital asset chooses to sell. The lower the tax burden on the sale, the more the owner retains for either reinvestment or consumption. The increased velocity of transactions more than offsets the loss in taxes per transaction. The public treasury and the private investor both win.

However successful the 2003 reform, it is still unwise to view tax policy chiefly in terms of revenue generation. The goal of tax policy is not to maximize tax revenue as such, but to maximize the level of overall social welfare. This difference really matters. Further cuts in the level of capital gains taxation could easily reduce tax revenue. Indeed a capital gains tax of zero would generate zero capital gains revenue. But don't assume such radical action necessarily would lead to a reduction in overall social welfare. It would reduce tax revenue from a single source, but changes in underlying patterns of investment could easily improve the overall revenue situation.

As a matter of sound finance theory, we should first figure out the optimal structure of taxation for overall growth or welfare, and then set rates within that structure to achieve any public revenue target. As a matter of transitional politics, we should next ask what feasible changes in the current tax structure can move us closer to that social optimum.

On the first question, there is much reason to believe that a consumption tax, preferably flat, could outperform even the ideal income tax. The great vice of all income taxes is that they distort the choice between present and future consumption. By taxing savings, the income tax, flat or progressive, induces all taxed individuals excessively to substitute present for future consumption. This, in turn, drives down the capital stock, which, among other things, keeps wage levels high. Moving to a flat consumption tax would reduce the size of the tax base by excluding savings, and thus require somewhat higher rates. But at the same time, its flat structure would vastly reduce administrative costs, leaving society a systematic gain. Under such a system, of course there is no capital gains tax.

Getting to that consumption tax regime is not easily done in one leap. But it can be done in several small steps. The traditional Keogh and IRAs are one step in that direction, because they defer the capital gains tax on all assets that are segregated in retirement plans. There is, quite simply, no reason to tie the forgiveness of capital gains to the age of the stockholder. We could introduce a simple rollover provision, for example, that removes the capital gains tax - or capital losses - from all capital transactions kept in segregated accounts, until these funds are voluntarily distributed to some unrestricted account that is freely available for personal consumption. Not only would we simplify tax administration, but we also would increase the mobility of capital in a sensible way.

To see why, consider the position of any person who is thinking of making a new investment that requires withdrawing capital from other investments. If the new investment yields a return that is 10% higher, giving an 11% return instead of a 10% one, than the current investment, with the current 15% tax rate in place, the person would not make the switch, even though it would be socially justified. His current holding generates, say, a return of 10% on a $1,000 investment, for a net yield of $100. The new investment base shrinks to $850 after paying the 15% tax. Reinvested at 11%, the principal yields a return of $93.50, even before switching costs are taken into account.

The present investor may still make that new investment by borrowing $850, tax free, on his old holdings. But there are two real social shortcomings to this approach. First, by holding the first stock, the investor does not signal to the market his preference for the second investment. Capital markets therefore work less efficiently. Second, by borrowing, the investor is forced to leverage his basic position and perhaps take on an unacceptable level of risk. Neither of these two results should be dictated by the tax system, although investors may choose to leverage any investment by finding a broker who deals on margin. But socially speaking, zero capital gains taxation on reinvested assets is likely to outperform the current system, even if it drives the revenue from capital gains close to zero.

Members of Congress on both sides of the aisle need to master these fundamental truths.

Mr. Epstein is a professor of law at the University of Chicago and a senior fellow at the Hoover Institution.