By Mark Thoma
All opinions expressed are his own.
The financial transactions tax is back in the news today. According to reports, French President Nicolas Sarkozy and German Chancellor Angela Merkel will propose a financial transactions tax in September.
Is this a good idea? It would certainly provide needed revenue to cash strapped governments, but at what cost? Governments must raise revenue somehow, but is this the best way to get the cash they need? Some taxes have a large distortionary effect on economic activity -- with a financial transactions tax, the worry is that investment activity will be curtailed-- and others have a much smaller effect. Some taxes can even make markets work better, e.g. taxes that force firms to internalize pollution costs and other externalities improves the decisions firms make. From society's point of view, they are more, not less efficient. Thus, in designing a tax system, we should look for taxes that provide the most revenue at the least cost.
So is a financial transactions tax a highly distortionary, costly tax? The answer is no. The tax would discourage short-term speculative activity, but much of this activity provides little social value. It pushes money around among winners and losers, and traders like it for that reason, but if this activity is discouraged through taxation it would have little effect on long-term investment decisions by firms. For example, one thing this would discourage is high frequency computer trading to exploit minute differences in prices. Does it really matter for long-term investment if these differences persist for a few seconds or minutes more?
In fact, there's even an argument that this tax will improve the efficiency of financial markets. The late economist James Tobin, the originator of the tax, argued that speculative activity causes harmful fluctuations in financial markets. For example, pursuit of speculative gains can cause firms to increase leverage, and if a financial crisis hits it can be very disruptive to the economy when firm are forced to unwind that leverage quickly. That wouldn't be so much of a problem if the costs fell only on those making the decision to take on so much leverage. But, unfortunately, as we have seen in this crisis, the costs can be very large and spread beyond the firms and individuals making the decision to take on so much risk. Thus, just as with pollution there are externalities -- costs that fall on the innocent -- and to the extent that a transactions tax forces firms to internalize the costs of their decisions, it improves rather than hinders the efficiency of financial markets.
There is one potential problem however: the ability to avoid the tax by moving activity elsewhere. But I don't see this as a huge worry. Trading is mostly carried out on centralized exchanges, so keeping track of the transactions and taxing them isn't that hard (the UK has had a tax on stocks for some time, and that hasn't driven all activity elsewhere). Nevertheless, if the U.S. were to follow suit, as I think it should -- it could raise hundreds of billions a year in revenue with minimal distortions -- that would help to prevent evasive activity.