Impose a Tax on Financial Transactions

The United States is home to large financial markets, with hundreds of billions of dollars in stocks and bonds—collectively referred to as securities—traded on a typical business day. The total dollar value, or market capitalization, of U.S. stocks was roughly $20 trillion in April 2013, and about $250 billion in shares is traded on a typical day. The value of outstanding bond market debt was about $38 trillion at the end of 2012, and average trading volume in debt, mostly concentrated in Treasury securities, amounts to over $800 billion on a typical day. In addition, large volumes of derivatives—contracts that derive their value from another security or commodity and include options, forwards, futures, and swaps—are traded on U.S. financial markets every business day. None of those transactions are taxed in the United States, although most taxpayers who sell securities for more than they paid for them owe tax on their gains.

This option would impose a tax on the purchase of most securities and on transactions involving derivatives. For purchases of stocks, bonds, and other debt obligations, the tax generally would be 0.01 percent of the value of the security. For purchases of derivative contracts, the tax would be 0.01 percent of all payments to be made under the terms of the contract, including the price paid when the contract was written, any periodic payments, and any amount to be paid when the contract expires. Trading costs for institutional investors tend to be very low—in many cases less than 0.10 percent of the value of the securities traded—so this option would generate a notable increase in trading costs for those investors.

The tax would not apply to the initial issuance of stock or debt securities, transactions in debt obligations with fixed maturities of no more than 100 days, or currency transactions (although transactions involving currency derivatives would be taxed). The tax would be imposed on transactions that occurred within the United States and on transactions that took place outside of the country as long as any party to an offshore transaction was a U.S. taxpayer (whether a corporation, partnership, citizen, or resident). The tax would apply to transactions occurring after December 31, 2014. This option would be effective a year later than the other revenue options analyzed in this report to provide the government and firms with sufficient time to develop and implement the new reporting systems that would be necessary to accurately collect the tax.

The tax would increase revenues by $180 billion from 2015 through 2023, according to estimates by the staff of the Joint Committee on Taxation. Those revenues would be lower if implementation of the option was phased in because of delays in developing the new reporting systems. (Because a financial transaction tax would reduce the tax base of income and payroll taxes, it would lead to reductions in revenues from those sources. The estimates shown here reflect those reductions.) The additional revenues from the option would depend importantly on the extent to which trading of securities fell in response to the tax.

One argument in favor of a tax on financial transactions is that it might reduce the amount of short-term speculation and computer-assisted high-frequency trading, and direct the resources now dedicated to those activities to more productive uses. Excessive speculation can destabilize markets and lead to disruptive events, such as the October 1987 stock market crash and the more recent “flash crash” that occurred when the stock market temporarily plunged on May 6, 2010.

However, the tax would discourage all short-term trading, not just speculation—including some transactions by well-informed traders and transactions that stabilize markets. Empirical evidence suggests that, on balance, a transaction tax could make asset prices less stable: In particular, a number of studies have concluded that higher transaction costs lead to more, rather than less, volatility in prices. (However, much of that evidence is from studies conducted before the rise of high-frequency trading programs, which now account for a significant share of trading in the stock market.)

The tax could have a number of negative effects on the economy stemming from its effects on trading and asset prices. However, because the tax would be only 0.01 percent of the value of the securities traded, most of those effects would probably be small. First, the tax could reduce private investment (leaving aside the effects of higher tax revenue on federal borrowing and thus on the funds available for investment). Specifically, the tax would raise the costs of financing investments to the extent that it made transactions more costly, financial markets less liquid, and financial risk management more expensive. Second, the transactions tax would reduce the value of existing financial assets because investors would not be willing to pay as much for assets that became more expensive to trade, lowering household wealth. And third, the cost to the Treasury of issuing federal debt would probably increase (again, leaving aside the effects of deficit reduction) because investors would pay less for Treasury securities that were less liquid.

In addition, traders would have an incentive to reduce the tax they must pay by moving their trading out of the country (although offshore trades by U.S. taxpayers would be taxed). Such effects would be mitigated if other countries enacted financial transaction taxes, as 11 members of the European Union are considering.