Is It Time for a Trading Tax?
Knowledge @ Wharton
October 26, 2011
To its advocates, the idea is a no-brainer: Charge a tiny tax on each stock, bond or derivative trade to raise badly needed revenue, discourage dangerous short-term speculation and make Wall Street help clean up its own mess.
"It seems like an idea whose time has come," says Jack Bogle, founder and retired CEO of The Vanguard Group, the mutual fund firm, who argues that a transaction tax would help curb speculative trading. "Speculation has triumphed over investment, and the implications of that are very bad."
The concept has been around for decades. In 1972, Princeton economist James Tobin, a Nobel laureate, proposed a transaction tax to calm the currency markets. The idea has since been suggested for stock, bond and derivatives markets as well. "It's an old question in finance -- whether you want to throw some sand in the wheels of the financial markets," notes Wharton finance professor Itay Goldstein. "I can certainly see the benefit in having [a] tax, because in some cases I do think that speculation might be getting out of control."
Recently, the idea has gained momentum. In September, the European Union's executive body recommended a 0.1% tax on stock and bond trades, and a 0.01% tax on derivatives trades. In the United States, where members of Congress have introduced transaction tax bills several times, only to see them stall in committee, the issue is getting renewed attention.
What has changed? Following the financial crisis, many want to make Wall Street pay. And, of course, governments are eager to find new revenue. Also, the soaring growth of computerized "high-frequency trading" has triggered concerns that too much speculation is roiling the markets and hurting ordinary investors. The "flash crash" of May 2010 -- when stocks inexplicably plunged 700 points, then quickly rebounded -- raised concerns about the potentially damaging effects of high-frequency trading. A very small tax could curb the practice by wiping out the tiny profits produced on individual trades, millions of which are conducted daily.
"I think we're in an environment where people are looking for someone to blame," says GusSauter, managing director and chief investment officer at Vanguard, who opposes transaction taxes. "If it's not Goldman Sachs, it's high-frequency traders."
Transaction tax proposals have a number of critics, ranging from those who are against tax increases in general, to those who say that such a tax would actually make the financial markets less efficient, hurting ordinary investors by raising costs.
"This is one of the worst ideas of a truly great economist [Tobin] who seldom had a bad idea," notes Wharton finance professor Richard J. Herring, who feels the tax could create more problems than it would solve. "It is an idea that will not die because it has enormous appeal to politicians, who are generally suspicious about markets and who are often desperate to raise money without appearing to tax the middle class."
In the U.S. and Europe, surveys show the public has been highly critical of large financial institutions, which are widely seen as having caused the economic crisis. In the U.S., the recent spate of Occupy Wall Street demonstrations suggests that these views remain strong. And in Europe, there is growing concern that big banks will need more propping up because they are heavily invested in shaky bonds from Greece and other troubled countries.
To transaction tax advocates, this seems like fertile ground. In a September 28 guest column on The New York Times editorial page, Philippe Douste-Blazy, a former French foreign minister who now serves as a special adviser on innovative financing to the United Nations secretary general, argued for a 0.05% transaction tax "aimed at financial institutions' casino-style trading, which helped precipitate the economic crisis." Globally, such taxes could raise hundreds of billions of dollars a year, which Douste-Blazy said could fund development aid around the world.
"Governments, both rich and poor, urgently need a way to calm speculation in the financial markets and to raise revenue," he wrote, adding: "Fears about the feasibility of a FTT [financial transaction tax] are overblown. Indeed, more than 40 such taxes have already been put in place around the globe. Britain, for example, has a very successful FTT on shares, known as the stamp duty, which raises more than $6 billion a year and has not had a significant impact on the competitiveness of London's finance sector."
The same week Douste-Blazy's column was published, the European Union's executive body proposed a tax that would take effect in January 2014 and apply to transactions among financial institutions whenever at least one party was located in the EU. In announcing the proposal, Algirdas Semeta, European commissioner for taxation, said the financial sector had "somehow abused" its central role in the economy, and should therefore help pay to fix the damage. The EU's executive body estimates the tax would raise about $77.5 billion a year.
The idea has support from some powerful EU members, including France and Germany, but is opposed by others, such as the United Kingdom and Sweden. Enacting the proposal across the EU would take approval from all 27 member countries, though countries could impose it individually. Officials in the United States and Canada have also opposed a transaction tax, as have the governments of China and Australia.
In the U.S., a number of consumer groups and union organizations, including the AFL-CIO, strongly support a transaction tax. In a 2010 statement, a coalition of these groups estimated such a tax could raise $100 billion a year in the U.S. and "curb excessive speculation by big banks, but with minimal impact on long-term investors."
But business groups in the U.S., including the U.S. Chamber of Commerce, the Business Roundtable and the Investment Company Institute, which represents the mutual fund industry, have long opposed a transaction tax.
"A transaction tax will cycle through the entire U.S. economy, harming both investors and businesses," the groups wrote in a September 22 letter to Treasury Secretary Timothy Geithner, adding that a tax would "impede the efficiency of markets, impair depth and liquidity, raise costs to issuers, investors and pensioners, and distort asset flows by discriminating against asset classes."
The 'Rent-a-stock Society'
According to Bogle, high-frequency traders incur costs that they pass on to their customers, including mutual fund companies. They also have no long-term interest in the companies whose stocks they own, making them indifferent shareholders and contributing to poor corporate governance, he adds. "In the rent-a-stock society, traders who are only in the market for seconds, as some of these traders are, couldn't care less about executive compensation, political contributions [by corporations] and things like that."
Because the market's reactions are hard to predict, forecasts of transaction tax revenue are little more than guesses. If the tax succeeded in dramatically curtailing high-frequency trading, the revenue could well fall below projections -- and governments would also lose the revenue they now get by taxing the handsome profits earned by these trading firms.
An investor, large or small, who bought a stock or bond and held it for months, years or decades, would have little to worry about from a tax of one-tenth of one percent, charged once on the security's purchase and only once again after it was sold. But today, most ordinary investors don't buy individual stocks and bonds; they invest through mutual funds and exchange-traded funds. Though the investor may hold a fund's shares for many years, stocks or bonds in the fund may be bought and sold very frequently to put new investors' money to work, raise cash for redemptions, cash out its winners, dump losers or maintain the mix of holdings promised to investors. An additional expense, though small, can add up, with even a fraction of a percent significantly undermining the benefits of compounding over the years, notes Vanguard's Sauter.
"It's a huge misconception that Wall Street is going to pay this tax," he says. "It's actually going to be individual investors who pay the tax. It's going to be passed on to them."
Pension funds, college endowments and insurance company assets also would be affected by increased trading costs. If a fund turned over, or reinvested, its assets three times a year, as some aggressively managed mutual funds do, the 0.1% trading tax could reduce assets by 0.3% each year. Were this to cause the average annual return to fall to 7.7% from 8%, a $100 investment would grow to $441 over 20 years instead of $466, a cut of more than 5%.
For Vanguard founder Bogle, this simply underscores the folly of investing in funds with high turnover. He and Sauter agree, though, that a transaction tax would have less affect on investors in index-style funds, Vanguard's main market. Index funds buy and hold stocks or bonds in an underlying index such as the Standard & Poor's 500, and thus have minimal turnover.
Much of the debate over a transaction tax focuses on the ultra short-term traders, for whom the tax could be devastating. The issue: Are practices like high-frequency trading good or bad?
Generally, high-frequency traders strive to profit on very brief opportunities, such as when a stock trades at a slightly different price on one exchange versus another, or when an option or other derivative trades at a price slightly higher or lower than the actual stock, bond or currency it represents. Because modern electronic markets are so fast and efficient, these price differences can be extremely small and brief, and high-frequency traders make profits by trading in very large volumes, over and over, with a block of securities often sold at virtually the same instant it is purchased. Imposing even a very small tax on each of these trades, which may earn less than a penny a share, would take away the potential profits, probably wiping out the industry.
High-frequency trading was made possible by powerful computers, lightening-fast communications and regulatory loosening that permitted electronic trading at the traditional exchanges as well as on newer, outside computerized systems. It has grown dramatically in the last six or seven years, and studies find it now accounts for 60% to 70% of daily trading volume in the U.S. One study estimated high-frequency firms earned about $13 billion in 2009 and 2010.
Many experts believe this has been a positive development. By quickly detecting pricing disparities, high-frequency traders reduce the spread, or difference between bid and ask prices, defenders say. Wide spreads represent pricing inefficiencies, when the bid or ask prices do not accurately represent a security's true value according to fundamentals like corporate earnings. Narrow spreads mean the buyer and seller trade at the "correct" price.
In addition, defenders say that high-frequency trading enhances the availability of buyers and sellers, allowing others to trade when they want to. This desirable "liquidity" also helps to reduce spreads, and it supports market confidence, as investors know that money put into the market can easily be taken out. "Their actions do benefit us," Sauter says of high-frequency traders. "They do create liquidity, and they do create tighter spreads."
The mutual fund industry and other institutional investors like using the high-frequency system because it allows them to conduct large trades in small pieces, so big buy and sell orders do not drive prices up or down. High-frequency traders are looking for tiny price differences, Sauter adds. "They are not pushing the market around."
Liquidity is universally seen as a good thing. But how much liquidity is really necessary? If the system can already match a buyer and seller within five minutes, is it really better to do it in five milliseconds? Bogle says no -- that ordinary investors have nothing to gain from the hyper-liquidity provided by high-frequency trading. "Who says we need all that liquidity? I don't." He notes that investors did fine in past decades when high-frequency trading did not exist and the average stock changed hands much less frequently than it does today. "If you look at the mutual fund industry, [turnover of the average fund's holdings] was about 15% [a year] when I came into this business" in the 1970s. "Now it's 100%.... The market works fine at much lower levels of liquidity. I think liquidity is the last refuge of the scoundrel, a buzzword you throw out to defend your position."
False Data and Unplanned Effects
Some also question whether high-frequency trading does in fact provide all the liquidity its defenders claim. Many of these traders move into and out of the market, simultaneously buying and selling the same securities so quickly that they have little interaction with anyone except other high-frequency traders, according to Wharton finance professor Franklin Allen.
The industry's detractors also say the textbook arguments for high-frequency trading fail to give a full picture of the industry's actual day-to-day practices, some of them abusive. Regulators in the U.S. and Europe have fined high-frequency traders for using the system to manipulate prices. One such practice, called layering, involves issuing and then cancelling orders that the trader never means to complete, pumping false pricing data into the market.
Also, many traders' algorithms are extraordinarily complex, triggering trades when an array of factors line up just so, sometimes with unplanned effects. An investigation by the Securities and Exchange Commission and Commodities Futures Trading Commission found that the 2010 flash crash started with a large sell order from a mutual fund. That triggered a spiral of sell orders from flash traders. Then, when new conditions were established, the process reversed. To critics, the fact that it took months for regulators to uncover the sequence of events -- requiring a voluminous report to explain them -- indicates how poorly understood the process is, and how hard it would be to monitor potential problems in real time.
High-frequency traders also may be using the system to probe the market, discovering when other participants are buying and selling large blocks of specific stocks or bonds, Allen adds. That would allow them to "front run," or buy ahead of the big trader, profiting improperly when the other buyer's order pushes the price up. "My guess is a big portion of the [industry's] profits are related to this activity," Allen notes.
Some critics, including Bogle, blame high-frequency trading for the stock market's recent high volatility, but the topic is much debated. According to Herring, research has shown that increases in trading expenses actually lead to greater volatility, not less. A transaction tax would thus make volatility worse. "If you're really interested in dampening volatility, leverage is the more likely villain to pursue," he adds, referring to trading with borrowed money, which can amplify both gains and losses. By allowing a trader to control more shares, leverage lets the trader have a larger effect on supply and demand, producing bigger price swings as big blocks are bought and sold.
Among the long-running objections to transaction taxes is the likelihood that trading would simply move to countries that don't have them. "Taxing a footloose industry like the financial markets is certain to be an exercise in frustration," says Herring, noting that with electronic trading, it is easy to shift business to a system in another country.
"Financial markets are very fluid from one country to another," adds Wharton emeritus finance professor Marshall E. Blume. "If we make it unattractive to trade U.S. equities in the U.S., equities would be traded in another venue outside the country." Blume says the trading tax is a "dumb idea."
Indeed, Sweden experienced just such an exodus after imposing a tax in 1984. Tax supporters note, however, that this would not be a problem if the tax were implemented at the same time by all the major economic players in Europe and North America. The EU proposal is a major first step. Widespread adoption of a transaction tax, however, would take many more.
(c) Knowledge @ Wharton