A securities transaction tax is a tax imposed on securities transfers, including, generally, purchases and sales. The tax could apply to the value of trades in stocks, bonds, derivative instruments, mutual funds, exchange-traded funds (ETFs), and other securities.
While an STT can be structured in a variety of ways, ICI believes that any such tax could harm individual fund investors who are investing to meet retirement, education, and other financial goals. Click here for more detail on ICI’s position on STTs, particularly the following points:
Proponents of STTs cite the fact that an STT would raise revenue for the government. Supporters of STTs also believe that these taxes will improve the functioning of the markets and help long-term investors.
Supporters of STTs believe an STT would improve market functioning and help long-term investors in several ways: by reducing stock price volatility; discouraging what they perceive to be a short-term focus on the part of investors; and reducing “financial engineering,” or the creation of complex financial instruments. These arguments are advanced by such advocates as the Aspen Institute, the Economic Policy Institute (EPI), and the Center for Economic and Policy Research (CEPR). (See Claimed Benefits of a Securities Transaction Tax, Questions 13–15 below.)
The short answer is no, as is detailed in this FAQ and other materials.
There is no evidence that imposing a transaction tax reduces stock price volatility (see Question 13).
An STT could discourage trading, particularly short-term trades. But there is little evidence to support the view that long-term trades are better for markets or for investors than short-term trades (see Question 14).
It is not clear that an STT would discourage financial engineering, and it might even encourage the practice, as financial firms design products that reduce or avoid the STT (see Question 15 below).
An STT that applies broadly to stocks and bonds would raise funds’ costs and reduce investment returns for shareholders in mutual funds, ETFs, and closed-end funds.
Funds trade their portfolio securities routinely as they invest shareholder cash, meet shareholder redemptions, and adjust fund portfolios. An STT would raise transaction costs on those trades, and that cost would reduce shareholders’ returns.
If a securities transaction tax does not exempt trades in mutual fund shares, it would subject mutual fund shareholders to double taxation. A fund buyer would pay tax on both the purchase of fund shares and on the fund’s portfolio trades. Shareholders in money market funds could be hit especially hard by the tax because many of these shareholders buy and sell shares frequently, as they use these accounts to manage their cash balances.
In addition to its direct costs, such a tax is likely to reduce market liquidity and widen bid-ask spreads, further increasing investors’ transaction costs.
No matter how such a tax is structured, it would create market distortions that would reduce the efficiency of markets for all participants, including fund investors. The tax would reduce market volume and could have a negative impact on liquidity and price discovery—the process of determining market prices through the interaction of buyers and sellers.
If the U.S. imposed an STT, market volume on U.S. exchanges would decline as some trades would become too costly to execute. Other trades would move off of U.S. exchanges, as investors either move trading activity to other, lower-tax venues or trade more in alternate securities.
Securities transactions affected by the tax could move offshore. This happened, for example, after Sweden imposed such a tax and a large portion of trading in Swedish stocks migrated to London.
Volume on U.S. exchanges would also be reduced if investors can avoid the tax by trading alternate securities. If, for example, a tax were imposed on equity trading but not on bond transactions, Wall Street’s financial engineers would have an incentive to develop more debt instruments that imitate the returns of the equities subject to the transaction tax.
The tax could cause intermediaries, such as dealers and market makers, to be less willing to provide liquidity to the markets.
One way these intermediaries profit is by buying securities at a lower price (the “bid”) and selling them at a higher price (the “ask”). An STT would reduce intermediaries’ profits and discourage them from trading unless spreads between bid and ask prices widened by at least the amount of the tax.
Price discovery could be impeded because investors would also be discouraged from trading. Investors trade when they obtain information that affects their estimate of the fundamental value of a firm. To overcome the increased cost of trading, investors would require a higher expected rate of return before they trade. That is, to act on new information, investors would require either a greater discrepancy between the stock’s current market price and their new estimate of the fundamental value of the firm, or more certainty surrounding their estimate. The tax’s barrier to trading would result in less interaction between buyers and sellers. Generally, greater interaction in markets results in more accurate price discovery.
Short of a worldwide agreement on a uniform securities transaction tax, imposition of an STT in the United States will create incentives for investors to trade elsewhere. Indeed, imposition of an STT in the United States—the world’s largest securities market—has the potential to prompt the development of a dominant global tax-free exchange.
To operate efficiently, a mutual fund must make routine purchases and sales of securities to invest shareholder cash flows, obtain cash to meet investor redemptions, and adjust fund portfolios to implement the fund’s investment strategies as market conditions and the value of portfolio securities change. Introducing a tax incentive to avoid trading could distort funds’ decisions and make fund operations less efficient and more costly.
A stock fund’s adviser already has strong incentives to ensure that it doesn’t engage in portfolio trading—whether “too much” or “too little”—that harms the fund’s performance. If the cost of a fund’s trading activity exceeds the benefits in terms of improving the fund’s performance, investors may leave the fund.
Any tax has the potential to reduce the budget deficit. The question is, which taxes can raise revenue most efficiently with the fewest harmful side effects? A securities transaction tax would significantly impair the functioning of financial markets, imposing a very high cost on the economy relative to its revenues.
Backers of a securities transaction tax claim several additional benefits, including:
These hypotheses are premised on several beliefs regarding investor behavior held by proponents of the tax.
There is no evidence that imposing a transaction tax reduces stock price volatility.
In studies to date, there is no evidence of decreased stock price volatility in countries that implemented or increased STTs. For example, using data from 23 countries from 1987 to 1989, Roll in 1989 found that stock return volatility was not related to transaction taxes.1 Looking at price volatility in Sweden before and after imposition of transaction tax in 1984, Umlauf in 1993 found that price volatility did not decline.2 Saporta and Kan in 1997 found that the United Kingdom’s stamp duty did not affect volatility of securities’ prices.3 Examining the effect of allowing negotiated commission in the U.S. in 1975, Jones and Seguin in 1997 found no evidence that the lowering of commissions increased volatility.4 Looking at the effect of changes in transaction taxes in Hong Kong, Japan, Korea, and Taiwan from 1975 to 1994, Hu in 1998 found no significant effects on price volatility.5 Looking into smaller market segments, Habermeier and Kirelenko in 2001 found that transaction taxes have negative effects on price discovery, volatility, and liquidity and lead to a reduction in market efficiency.6
The expectation that stock price volatility will decline is founded on the assumption that short-term investors trade based on speculation (i.e., they are “noise traders”), whereas long-term investors trade based on the fundamental value of stock. There is no reason to believe this is a valid distinction. Short-term trades can be based on new information that alters market participants’ estimates of the long-term fundamental value of a stock. Conversely, long-term trades can be speculative in nature, betting that the fundamental value of a firm will grow over time.
No. That expectation is based on a misunderstanding of investor and market behavior.
The expectation that shareholders will become more patient is based on the belief that the market punishes corporations that undertake long-term investments. In practice, while the majority of stock trades are short-term, the stock market tends to reward corporations that undertake long-term investments. On average, announcements of long-term investments lead to an increase in a firm’s stock prices, 7 particularly at firms that have valuable investment opportunities.8
An STT may reduce some types of financial engineering, but it could also create new incentives to create complex financial instruments.
Engineered products that focus on short-term gains or require multiple trades in securities subject to the tax would be negatively affected by an STT. But much of what is referred to as financial engineering is not motivated by short-term gains, nor is it executed with multiple trades. A fair amount of financial engineering is undertaken in response to government policies—for example, creating securities that are treated as equity for accounting or regulatory purposes, but are treated as debt for tax purposes. As detailed in Question 21, imposition of an STT could actually increase financial engineering, as Wall Street devises methods to avoid the tax.
An STT would probably discourage high-frequency trading, but it is a very blunt instrument to use in reaching that goal. Any broad-based tax on securities transactions would have negative consequences for all investors.
Furthermore, the question of whether high-frequency trading harms or helps markets is not at all settled. High-frequency trading may, at times, benefit the markets by contributing liquidity and tightening bid-ask spreads. At the same time, high-frequency trading raises a number of regulatory issues that should be examined closely. That is why ICI supports the Securities and Exchange Commission’s (SEC’s) current broad examination of market structure issues, including high-frequency trading.
The most recent legislative proposal for an STT in the U.S. is H.R. 4191, titled “Let Wall Street Pay for the Restoration of Main Street Act of 2009.” It was introduced in the House of Representatives on December 3, 2009, by Rep. Peter DeFazio (D-OR). Senator Tom Harkin (D-IA) has introduced similar legislation (S. 2927) in the Senate.
H.R. 4191 would apply a 25 basis point (0.25 percent) tax on the value of transactions in stocks or stock options. It would also apply a 2 basis point (0.02 percent) tax on the value of transactions in futures, swaps, credit default swaps, or options on these securities. The tax would apply to all purchases and sales that occur in the United States on a trading facility or to any transaction where either the purchaser or seller is a U.S. person.
Like any other STT, the proposed tax in H.R. 4191 would impair the functioning of the financial markets. The proposed tax would reduce stock market volume and liquidity, resulting in wider bid-ask spreads, to the detriment of all investors. Any such tax would raise taxes on middle-class investors, depress stock prices, and increase the cost of hedging risk.
Specifically, H.R. 4191 could hurt up to 79 million investors in long-term mutual funds, including index funds, exchange-traded funds, and closed-end funds, by reducing their investment returns (see Question 5 above). The costs of the STT on portfolio transactions would be borne by shareholders in any fund that holds equities or uses financial instruments such as futures contracts to hedge risk (including interest rate risk).
Due to its design, H.R. 4191’s STT is likely to create greater market distortions than other STTs. Since the bill’s tax does not apply to bond transactions, Wall Street financial engineers are likely to develop securities that qualify as bonds to avoid the tax, but replicate the returns of stocks or stock indexes. Another possibility is using bonds and options to create portfolios that replicate the returns of holding a stock. Because bonds are not taxed, total transaction taxes on these portfolios would be lower than the tax on an equivalent portfolio of stocks. H.R. 4191 would also give corporations an additional incentive to finance capital investments with debt rather than equity.
H.R. 4191 exempts the purchase or sale of mutual fund shares from its tax. It exempts trading in tax-deferred pension accounts. It exempts an individual investor’s first $100,000 in transactions annually.
Unfortunately, no.
Under H.R. 4191, mutual fund investors would not pay tax on the purchase or sale of fund shares. However, H.R. 4191 would impose a tax on all portfolio-level equity transactions executed by a mutual fund. Fund investors would indirectly pay the tax on these transactions: the tax would increase funds’ cost of investing in portfolio securities and decrease funds’ investment returns.
The bill exempts pension funds from tax on trades in equity securities that they hold directly in their portfolio. However, a mutual fund held by a pension fund, such as a mutual fund held in a 401(k) account, would still be subject to the tax on portfolio trades.
Similarly, the bill exempts an individual’s first $100,000 of trades each year involving stocks that they hold directly. However, portfolio transactions undertaken by a mutual fund held by an individual investor would still be subject to tax.
Thus, the bulk of equity investors who own mutual funds but do not directly own stocks would not be protected under H.R. 4191 from the burden of the tax.
It’s not that simple. In almost all cases, it’s impossible to exempt funds that are held in retirement accounts from the tax. And creating a system to “pass through” taxes to shareholders to provide retirement savers (or any other subset of shareholders) with a credit for the transaction tax paid would be difficult and costly (see Questions 25–27). Such a system could also raise compliance issues for plans regulated by ERISA.
Most mutual funds are held by many types of investors. The same fund may be owned by individuals (in taxable accounts or in IRAs) and institutions (including 401(k) plans, pension funds, and other institutional investors). Under the Investment Company Act of 1940, a mutual fund’s adviser must conduct portfolio trading for the benefit of all its investors. The adviser cannot designate individual trades to a portion of the fund’s investors. Thus, it is impossible for some portfolio trades to be exempt from tax (those allocated to fund shares held by retirement accounts), while other trades are taxed.
Trying to implement such a tax credit would be very costly and impractical. Fund transfer agents and intermediaries (e.g., brokers, financial advisers, retirement plan recordkeepers, banks) would have to create a brand-new infrastructure and systems to capture, calculate, retain, and report this tax information to investors—all at great expense, which would all ultimately be borne by shareholders. So even if Congress created a tax credit for shareholders to compensate for the transaction taxes paid by their mutual funds, the costs of implementing systems to flow through and report the credit would reduce, and might even eliminate, the benefit of the exemption.
Allocating the transaction tax to individual investors would force mutual funds to create a costly new infrastructure that would serve no other purpose for investors.
Funds often do not know who the ultimate beneficial owners of their shares are because the shares are often sold through intermediaries. For funds to pass through transaction taxes paid on portfolio transactions to the beneficial owners of the shares, mutual funds would have to create systems to calculate transaction taxes paid on portfolio securities, capture this information, and distribute it to fund transfer agents and intermediaries who track the ownership of shares on a daily basis.
In turn, these entities would need to precisely allocate transaction taxes to each beneficial owner of mutual fund shares, whose share balances may change daily. To be precise, these calculations would have to be performed very frequently—most likely every trading day. Fund transfer agents and intermediaries would need to build systems and repositories to capture, calculate, and track such taxes paid and to report information to the shareholder or holder of record at year-end. And these systems must be robust enough to handle any restatements or corrections. All of the costs of creating and maintaining this infrastructure would lower investment returns for all mutual fund shareholders, whether they qualify for the tax credits or not.
Mutual funds typically distribute capital gains to shareholders once a year, but precise handling of an STT credit would require allocating the credit much more frequently (most likely daily).
Passing through an STT paid at the portfolio level to individual shareholders would require fund transfer agents and intermediaries to build systems that track this information for each beneficial shareholder on a frequent basis. If the value of the credit was not calculated and allocated to shareholders each day on which there was a portfolio trade, the accumulation of unallocated tax credits would create incentives for investors to buy and sell fund shares solely to capture tax benefits.
By contrast, realized capital gains are included in a fund’s net asset value (NAV) until they are distributed, typically once or twice a year. Funds distribute gains to shareholders who are the owners of record on the date of the distribution. Thus, it is unnecessary to allocate gain realizations to individual shareholders on a daily basis.
Currently, the SEC imposes a “Section 31 fee” on stock transactions, and the proceeds of this SST are used to fund the agency. As of January 15, 2010, that tax is 0.127 basis points (0.00127 percent).
The proposed 25 basis point transaction tax rate is almost 200 times the current Section 31 rate.
From 1914 to 1965, the federal government imposed a tax on stock transfers. This STT was initially 2 basis points (0.02 percent) in 1914, increased to a range of between 4 and 6 basis points from 1932 through 1958, and reverted to 4 basis points from 1959 until the tax was repealed in 1965.
H.R. 4191 proposes a tax that is significantly larger and broader—and thus potentially more harmful—than the earlier STTs.
First, the tax rate being proposed through H.R. 4191 is much higher than any in U.S. history. The proposed rate of 25 basis points is four to 12 times higher than the rate of the earlier tax.
Second, H.R. 4191’s tax base is different. From 1914 through 1958, the historical STT was assessed on the par value of stock. H.R. 4191 would instead apply the tax to the stock’s market value of a stock, instead of the par value of the stock. Par value is a legal concept that bears no relation to market value. A stock’s par value is typically less than its market value.
In 1959, the historical SST was changed to apply to the market value of a transferred stock, increasing the tax base. Just seven years later, Congress viewed the tax as complicating securities transactions and repealed it.
When the tax was implemented in 1914, the stated tax rate was 2 cents per $100. This is a rate of 0.02 percent, or 2 basis points. In 1932 the rate was doubled to 4 cents per $100, or 0.04 percent (4 basis points). The rate reached as high as 0.06 percent (6 basis points). Thus, the 25 basis point tax rate proposed by H.R. 4191 is four to 12 times greater than the rate of the historical STT.
The previous law stipulated the tax rate in terms of cents per $100 of transaction value, whereas H.R. 4191 stipulates the tax rate in terms of percentage points. There could have been some difficulty in translating the two measures.
No. Even at the lower 4 basis points rate that applied from 1932 through 1965, an STT would have a far greater impact on investor returns today than historically. Overall transaction costs have declined markedly since 1965, so a tax would represent a proportionately much greater increase in costs.
Investors should not be required to suffer decreased investment returns from a transaction tax, regardless of its magnitude, in addition to the current Section 31 tax. In fact, in 2002, legislation was enacted to reduce the Section 31 tax by 60 percent over several years because lawmakers concluded that a fee level in excess of what was necessary to fund the SEC was a tax on investors. At the time Congress voted for this reform, the savings to investors was projected to be $15 billion over 10 years.
Yes. The UK has a stamp duty. The tax rate is 50 basis points, or 0.5 percent.
There are two main differences between the UK tax and the one proposed in H.R. 4191, both of which make the proposed U.S. tax more sweeping.
First, the U.S. proposal taxes all securities purchases, including those made by intermediaries such as dealers. By contrast, UK tax law exempts intermediaries from the tax on securities purchases. This is a significant difference because intermediaries facilitate many stock exchange trades.
Typically, one long-term investor does not sell a block of shares to another long-term investor looking to buy the exact same number of shares. Instead, investors usually trade through market makers, dealers, or other intermediaries, who provide liquidity by constantly buying and selling shares on the market and from their own inventory. Under the UK stamp duty, intermediaries are exempt from tax, so such trades are taxed only once when an investor buys shares from an intermediary and pays the tax. Under H.R. 4191, such trades that provide market liquidity would be taxed twice—once when an intermediary purchases the shares from an investor, and again when another investor later buys the shares from the intermediary.
Second, the UK tax applies only to shares in a UK company or in foreign companies with shares registered in the UK. Thus, a UK citizen can trade unregistered foreign shares without paying tax, even if the trade is in London. The U.S. proposal would tax all trades on U.S. exchanges and would even attempt to tax trades by a U.S. investor offshore.
It is noteworthy that the UK stock market is much less liquid than the U.S. stock market. According to figures presented by Gus Sauter, chief investment officer of the Vanguard Group,9 in 2009, average daily turnover in the UK stock market was 0.37 percent—approximately one-fourth the 1.38 percent average daily turnover in the U.S. stock market. Moreover, turnover in the UK market would likely be even lower if liquidity-providing intermediaries were subject to the tax, as they would be under H.R. 4191.
The stamp tax has caused much of the trading in the UK to migrate to securities that are exempt from the tax, with several results:
Yes. The tax was introduced in January 1984. Both purchases and sales of domestic equities were taxed at 0.5 percent (50 basis points) for a combined 1 percent tax on each transaction. Trading in stock options was also taxed. In July 1986, the tax rate was doubled. In 1989, the tax was extended to trading in fixed-income securities and derivatives, but at lower rates than those applied to stock trades.
After Sweden doubled the transaction tax rate in 1986, Umlauf in 1993 found that 60 percent of the volume of the 11 most actively traded Swedish stocks migrated to London.13 That shift represented 30 percent of all trading volume in Swedish equities. By 1990, 50 percent of all trading in Swedish equities had migrated.
In a separate study, Campbell and Froot in 1995 found that only 27 percent of the trading volume in Ericsson, a major company based in Sweden, took place on the Stockholm exchange in 1988.14
Umlauf also looked at stock price volatility in Sweden before and after imposition of transaction tax in 1984, and he found that price volatility did not decline.
No. Sweden began phasing out the tax in 1990 and abolished it in December 1991.
1 Roll, Richard. 1989. “Price Volatility, International Market Links, and Their Implications for Regulatory Policies.” Journal of Financial Services Research 3, 211–246.
2 Umlauf, Steven R. 1993. “Transaction Taxes and the Behavior of the Swedish Stock Market.” Journal of Financial Economics 33, 227–240.
3 Saporta, Victoria, and Kahmhon Kan. 1997. “The Effects of Stamp Duty on the Level and Volatility of UK Equity Prices.” Working Paper, Bank of England.
4 Jones, Charles M., and Paul J. Sequin. 1997. “Transaction Costs and Price Volatility: Evidence from Commission Deregulation.” American Economic Review 87, 728–737.
5 Hu, Shing-yang. 1998. “The Effects of the Stock Transaction Tax on the Stock Market – Experiences from Asian Markets.” Pacific-Basin Finance Journal 6, 347–364.
6 Habermeier, Karl, and Andrei Kirilenko. 2001. “Securities Transaction Taxes and Financial Markets.” IMF Working Paper.
7 Kiefer, Donald W. 1990. “The Securities Transactions Tax: an Overview of the Issues.” CRS Report for Congress, July 25.
8 Chung, Kee H., Peter Wright, and Charlie Charoenwong. 1998. “Investment Opportunities and Market Reaction To Capital Expenditure Decisions.” Journal of Banking and Finance 22.
9 “Financial Transactions Tax: Taxing U.S. Investment, Savings, and Growth.” December 16, 2009, www.uschamber.com/webcasts/2009/091216_ccmc_ftt.htm (video) at 16:45 minutes.
10 Government of Canada, Depository Services Program. 1996. Financial Transactions Taxes:
The International Experience and the Lessons for Canada. Available at http://dsp-psd.pwgsc.gc.ca/Collection-R/LoPBdP/BP/bp419-e.htm.
11 Ibid.
12 EU Clearing and Settlement Fiscal compliance Experts' Group. 2006. Fact-Finding Study On Fiscal Compliance Procedures Related To Clearing And Settlement Within The EU. Available at http://ec.europa.eu/internal_market/financial-markets/docs/compliance/ff_study_en.pdf.
13 Umlauf, “Transaction Taxes and the Behavior of the Swedish Stock Market.” Journal of Financial Economics 33, 227–240.
14 Campbell, John Y. and Kenneth A. Froot. 1995. “Securities Transaction Taxes: What About International Experiences and Migrating Markets.” Securities Transaction Taxes: False Hopes and Unintended Consequences, Catalyst Institute.
March 2010