I propose that the Congressional leadership re-introduce the Trouble Asset Relief Program accompanied by a major new policy: a small tax on securities market transactions. This will accomplish the political goal of aiming a silver bullet into the heart of the (understandable) popular outrage that blocked passage of the TARP bill on Monday. It will simultaneously accomplish the fiscal goal of raising revenue in the future. This is revenue that the federal government would have sorely needed even before the bailout arose and will need even more if the taxpayer is to be protected against the risk of heavily subsidizing the financial sector.
A tax on securities market transactions might sound like a wild populist policy that would damage the functioning of the economy. But in fact it is probably more sensible than such populist measures as banning short sales, which has already been tried (to no avail).
Proposals for financial transactions taxes have a distinguished pedigree, going back at least as far as Keynes. Best-known is the Tobin tax proposal, by Nobel Prize winner James Tobin, which was specifically aimed at volatility in foreign exchange markets. More relevant to what I am proposing are two articles by the pre-Treasury Larry Summers: “A Few Good Taxes” and “When Financial Markets Work Too Well: A Cautious Case for a Securities Transactions Tax” (1989). Add to the list of proponents another Nobel Prize Winner, Joe Stiglitz.
There is extensive experience with securities transaction taxes (STTs), especially in other countries. There have also been quite a few studies of their effects. Often the motivation for such proposals is to reduce short-term speculative turnover: a tax of 0.1% means nothing to a long-term investor, but is a strong disincentive to those traders who hold their positions for only minutes or hours. The idea is that reducing short-term speculation will reduce volatility. On the other hand, defenders of unfettered financial markets often argue that such a tax will reduce liquidity and thus hurt the customers who depend on the market.
The general historical experience seems to be that there is no discernible effect on volatility (though a couple of studies find effects on volatility, either upward or downward). In other words, the tax might not help the functioning of the financial markets — the original motivation — but neither does it hurt, according to a majority of the studies. In some cases the volume of trading within the country is affected. But what the tax does does usually do is raise money for the Treasury.
The UK long had a securities transactions tax, known as a stamp duty, which was set at 0.5% in 1986. Sweden introduced a 0.5 per cent tax on the purchase and sale of equities in 1984 — prompting some financial trading to move offshore — and kept it until 1991. (Froot and Campbell, studied these two examples in a 1994 book that I edited. The Swedish case is particularly relevant to the current US context because the popular motivation there was to “take down a peg” high-earning speculators.) Japan, Korea, Taiwan and Hong Kong have a long history with securities transactions taxes, and India introduced one in 2004; in these cases there were not significant reductions in either price volatility or market turnover. Other countries that have had financial turnover taxes of at least 0.1% include Australia, Austria, Denmark, Finland, France, Germany, Malaysia, and Singapore. (Germany abolished its turnover tax in 1991, and Japan in 1999.) In addition there are other countries that impose smaller fees.
Even the United States had a STT until 1965, and to this day imposes an SEC fee of .0033%. Thus we have already lost our virginity !
An important potential drawback, if the US were to impose a more substantial transactions tax alone, is that it might drive financial business offshore. There is an answer to this point. As noted, many countries already have taxes on financial transactions. Furthermore, lots would love to cooperate with the United States in an international program to harmonize such taxes internationally. This is precisely the sort of project in which many abroad have long asked Americans to participate, but which we have not hitherto wanted to do.
The level and longevity of the tax might be adjusted to achieve the goal of Section 134 of the TARP bill: that the taxpayer recoup the costs of the bailout. A 2004 study by the Congressional Research Service reported that an 0.5% tax on stock transfers could raise $65 billion a year. Others have produced higher revenue estimates. A tax extended to bonds and derivatives (especially derivatives!) would of course raise more. Remember that one does not compare this annual revenue to the $700 billion headline cost of the bailout. Rather, one compares the present discounted value of the annual flow of revenue to however much of TARP’s $700 billion is left over after the government (we hope) collects something on the troubled loans and also recoups something on warrants obtained from the participating banks.
The tax might on the margin contribute to a shrinking of the size of the financial sector; but this shrinking needs to happen anyway, as Ken Rogoff has pointed out. And most important politically, it would give expression in a non-damaging way to the blood lust that the public feels toward Wall Street, a venting that needs to take place if the bailout bill is going to be approved.
[To any readers who wish to post comments: I suggest you go to the RGE version of this post.]