Steven Davidoff really doesn't like financial transactions taxes (FTT) but is not honest enough to acknowledge this fact. Instead he tells readers that proponents of a tax haven't thought about its consequences and uncritically repeats every piece of nonsense produced by the financial industry to attack the idea.
In the course of a 1300 word essay, we get assessments of the tax from Credit Suisse, Blackrock, and the Partnership for the City of New York, which is effectively the New York City Chamber of Commerce. All of these accounts are presented uncritically, as though the purveyors of this information had no interest other than conveying the truth. We are also told that the New York Stock Exchange "threatened to jump across the Hudson River to New Jersey" in reaction to a plan to increase the city's stock tax in the 1966 (interesting image). Davidoff apparently has never heard of businesses making threats to extract concessions from governments.
The NYT running a column like Davidoff's is like the Iowa City Press Citizen running a column on a plan to cut back farm subsidies where the views of the state's leading wheat and corn farmers are presented as unquestioned truth, along with a study from the corn growers trade organization. I suspect that the Press Citizen has higher standards.
Meanwhile when it comes to the proponents of the tax, Davidoff lectures:
"advocates of this neat idea conveniently ignore the century of less-than-successful experience with this tax, including New York State’s own failed attempt."
This comment is more than a little bizarre. Davidoff writes as though proponents of the tax are completely ignorant of economics and have not done research into the history of financial transaction taxes.
Contrary to this assessment, the proponents of the tax include some of the world's most prominent economists. Furthermore, there is extensive research on the history of financial transactions taxes. Much of it can be found right here on the European Commission's (EC) website.
Contrary to Davidoff's bizarre comment, implying the New York tax is a rare example of a government implementing such taxes, nearly all financial markets operated with financial transactions taxes for long periods of time (more than 300 years in the case of London's market). Most of the world's major financial centers, including London, Switzerland, Hong Kong and Singapore, still have financial transactions taxes on their stock exchanges. Perhaps Davidoff should be lecturing these governments on how their taxes really don't work.
As far as the substance, Davidoff tells us that research shows that the tax will increase rather than decrease volatility. There are two different notions of volatility at play here. One is the volatility associated with normal price fluctuations over the course of a day or week. This is likely to be increased by a tax since it will increase the costs for arbitragers to enter a market. That means that we may see somewhat larger divergences between prices than would otherwise be the case. This could mean that the gap in the price of oil between two markets may rise to 0.4 percent rather than 0.3 percent before arbitragers whittle it down again.
Proponents of FTTs are probably not much concerned about this sort of volatility. The economic consequences are likely close to zero. Furthermore, since the levels of taxation being debated would just raise transactions costs back to where they were 10-15 years ago, it is difficult to believe that the effects could be too severe. (We did have very liquid capital markets in the 1990s.)
The type of volatility that more likely concerns proponents of FTTs are the sharp movements that are not driven by fundamentals, such as the 1987 crash and the flash crash in the spring of 2010. While it is difficult to prove that a FTT will reduce the likelihood of such sharp movements, it is worth noting that such events did not occur in the 50s, 60s, and 70s, when trading costs were much higher than in the last three decades.
As far as the incidence of the tax, Davidoff gives us the assessment of Blackrock:
"if the financial transaction tax were set at 0.1 percent per trade, an investor putting $10,000 in its global equity fund would lose more than $2,300 in expected returns over a 10-year period. This amount would rise to $15,000 if the money were invested in a more actively managed European fund."
Incredibly, Blackrock assumes that its trading does not in any way respond to the tax. If this were true then Blackrock's funds would quickly go out of business since their cost would be far higher than others in the industry. There have been a range of trading elasticities estimated by various studies (see the EC research), with most estimates close to -1.0. (None are near zero.) If the elasticity is near -1 then trading volume would decline by roughly the same amount that the tax increases trading costs.
This means that if the tax doubled trading costs, then trading volume would be roughly cut in half. That means that if Blackrock's fund managers responded as the research suggests, then they would cut back the number of trades by enough so that the non-tax trading costs for their $10,000 account would fall by roughly $2,300 over the course of a decade or $15,000 in the case of its more actively managed European fund. This would be revenue lost to Blackrock, not to its clients.
In this respect, it is worth noting that the sharp decline in trading costs over the last four decades has not been associated with higher returns to investors, but rather to a more than proportionate increase in trading volume. This has caused the total amount spent on trading financial assets to rise sharply relative to the size of the economy. These trading costs are money out of investors' pockets and a drain on the economy.
Davidoff's effort to claim that the tax could not raise any revenue approaches the bizarre. He tells readers:
"In Britain, for example, where the financial transaction tax has fluctuated from half a percent to 2 percent, the tax has raised significantly less revenue than one might expect, about £3 billion a year. The reason is that investors who trade regularly in Britain use options to avoid the tax, which applies only to trading in stock. The result may be that the tax pushes investors into more risky securities in their efforts to avoid it."
First, it is worth noting that £3 billion comes to 0.2 percent of UK's GDP. (The UK had raised almost 0.3 percent of GDP from this tax before the 2008 crash [Table 2].) This would be the equivalent of almost $400 billion over the 10-year budget horizon in the United States. That is almost 3 times as much as President Obama has proposed to save by cutting the Social Security cost of living adjustment. In other words, in the current budget debates it would be regarded as real money.
Second, the decision to not tax derivatives like options is a political one made by governments that have been closely allied with the financial industry. The tax being put in place by 11 countries in the European Union would tax options and other derivatives. In the 1980s, Japan had a broadly based tax that was imposed on a wide range of financial assets including options. This tax raised an amount of revenue that was close to 1.0 percent of its GDP. This would amount to $2 trillion over the 10-year budget horizon in the United States.
At one point Davidoff tells readers:
"As for seeking revenue gains to solve budget problems, if the tax is too small, it will have no effect."
Huh? The Securities and Exchange Commission imposes a tax of 0.002 percent on stock trades in the United States. This tax raises roughly $1.2 billion to finance its budget. Is Davidoff suggesting that the SEC should get rid of this tax because it is not really raising money?
As I said, Davidoff doesn't like FTTs, that's pretty clear from reading this piece even though he tells us:
"This is not to say that a financial transaction tax by itself is such a terrible idea."
He has a case built with non-sequiturs (one example of the horror of FTTs is that traders fled a tax imposed by Sweden in the 1980s and instead did their trades in London, which also had a tax). And he ignores all sorts of evidence that FTTs can and do raise large amounts of revenue without disrupting capital markets. This piece lets us know where Davidoff stands on FTTs, it doesn't provide much information on the merits of the policy.
Gary Burtless adds the following on Blackrock's calculations for the cost of the tax to investors.
"I cannot see how a $10,000 investment held over a 10-year period could incur $2,300 in transactions-tax charges if the transactions tax were set at 0.1% per trade, as stipulated in Davidoff’s piece. Let’s think about this a minute. According to Vanguard’s website, its Total Stock Market Index fund has turnover equal to 5% of its portfolio per year. That means in the initial year the transactions tax will raise approximately $10,000 x 5% x 0.1% = $0.50. Assuming nominal returns are 6.5% a year, a straightforward calculation shows the transactions tax will raise about $6.75 from our hypothetical retirement saver over the stipulated 10-year holding period. In the absence of the transactions tax, our hypothetical saver would realize $8,771 in gross investment returns. With the transactions tax, his net (after-transactions-tax) returns will be $8,761, a difference of about $10. (Recall that our investor gives up returns he would have earned on the tax collected early in his holding period. Hence, his investment loss is larger than the amount of transactions tax paid.)
"The only conceivable way our retirement saver would give up $2,300 in expected returns on his equity holdings is if his fund manager had an extraordinarily large number of trades. No doubt BlackRock engages in more trades than Vanguard, which unlike BlackRock probably tries to minimize its trading expenses. But even if we go to the Fidelity Spartan Total Market Index fund, turnover is still only 17%. (Fidelity is not famous for holding down trading costs.) The total transaction tax over a 10-year period is only about $23, and our retirement saver gives up about $35 in 10-year returns."
(Only one link allowed per comment)