Tax Rates

In most cases, STTs are proportional ad valorem taxes on the value of the purchase (or the sale) of a financial instrument. In some countries, however, the rate varies according to the value of the transaction. In France, for example, transactions below a certain threshold are exempt from the STT; above that threshold, a rate of 0.3 percent is applied to both purchases and sales, up to a certain value; transactions above that value are subject to a lower rate of 0.15 percent; finally, there is a ceiling on the total amount of tax for each transaction. Belgium also has a ceiling: for transactions above the ceiling, the STT is a fixed charge for each transaction, irrespective of its value.

Tax Base

STTs vary widely in respect of the base to which the tax rate is applied.

• Equity and debt. Most countries apply an STT only (or primarily) to transactions involving equity shares, but some also apply the tax to transactions involving debt instruments such as bonds and debentures. In countries such as Belgium that apply an STT to debt instruments, the rate is usually lower than the rate applied to shares.

• Primary and secondary markets. The STT may be applied only to transactions of existing financial instruments in the secondary market or to new issues in the primary market as well. In some countries, new issues are subject to a separate tax.

• Cash versus derivatives markets. STTs may be applied both to cash transfers of securities such as shares and bonds and to transfers of derivatives such as options and futures. Thus, some countries levy STTs on purchases or sales of an option, according to the option price (in addition to the tax on the transaction in the underlying security, in cases where the option is exercised).

• Domestic and foreign securities. STTs may be levied only on transfers of domestically registered securities or on transfers of both foreign and domestic securities. In the latter case, different rates may be applied. For instance, Switzerland’s stamp duty is levied at 0.15 percent on transactions in Swiss securities, and at 0.3 percent on transactions in foreign securities by those who are not members of the Swiss exchange.

• Domestic or foreign market participants. The STT is sometimes applied differentially, depending on whether purchasers or sellers are residents or nonresidents. In Belgium, for example, nonresidents are exempt from the transactions tax; in France, they are exempt when trading on the Paris Bourse.

• Exemptions. Market makers who act as intermediaries between buyers and sellers are commonly exempt from the STT or subject to reduced rates. In Italy, the exemption is broader: the STT is applied only to transactions that take place outside a formal stock exchange. In other countries, however, the STT is confined to transactions made through a broker. Exemption may also be extended to transactions by certain institutional investors (such as investment funds, social security funds, private pension funds, and life insurance companies).

Volume Reduction

An STT increases the cost of making the relevant transactions. The cost of sales of equity shares from one investor to another through a broker (including brokers’ commissions on both the sale and the purchase, but excluding STTs) may be as low as 30 basis points for typical transactions by institutional investors in the United States, where financial markets are most highly developed.³ In the European Union (EU) and Japan, corresponding figures lie between about 50 and 70 basis points. Even when levied at what might appear to be modest rates, STTs can substantially increase such costs. For example, in the United Kingdom, the 0.5 percent stamp duty that is levied on purchases of shares roughly doubles transactions costs—from about 50 to almost 100 basis points.

This increase in cost will affect the volume of transactions in three main ways. First, if the tax cannot be avoided, many transactions that would have taken place in the absence of the STT will simply not take place at all. Consider a single transfer of a particular share from investor A to investor B. A places a value of not more than V_A on the share; B places a value of at least V_B.⁴ Transfer of the share from A to B will take place where

V_B - V_A > C

where C is the total cost of the transfer. In any period of time, the volume of market transactions will therefore depend positively on factors that cause investors to differ in the valuations that they place on particular shares and negatively on the total cost of making those transactions.

Second, it may be possible to avoid STTs by moving the same transaction elsewhere—for instance, outside the regular exchange or to an exchange overseas. In general, additional costs are associated with such a move, but they may well be lower than the STT.

Third, it may be possible to avoid STTs by changing the transaction to something that is roughly equivalent in its effects. For instance, if an STT is imposed on transactions in some assets (such as company debentures) but not on those in other, similar assets (such as government bonds), the STT is likely to decrease the volume of transactions in the former market but increase it in the latter market. In some cases, financial markets may be able to create new assets that are very close substitutes but that are not themselves subject to the STT.

A number of econometric studies have investigated the responsiveness of trading volume in financial markets to changes in transactions costs (including STTs)—in particular, in Sweden, the United Kingdom, and the United States. In general, the results of these studies are consistent with a significant negative response. The estimated elasticity of response in the long run varies quite widely between different studies, however, from about -0.25 to -1.7.⁵

Price Reduction

Owners of financial instruments generally do not expect to hold them indefinitely. At some point, those instruments will be sold. If the purchasers are required to pay an STT, they will—other things being equal—demand a lower price for the asset. If future prices are reduced in this way, however, the present price of the asset would also be lower than it would otherwise have been. Thus, the market prices of the financial instruments to which STTs are applied can be expected to drop. From a different perspective, the cost of raising capital will go up when new financial instruments subject to STTs are issued.

There is substantial empirical support for the view that higher trading costs, in general, tend to reduce asset prices. Studies of the effects of STTs, in particular, also support this finding. Simple observations of the behavior of equity prices on the announcement of new STTs or rate increases are also consistent with this result. In Sweden, for instance, the All-Equity Index fell by 2.2 percent on October 24, 1983, when a 1 percent STT was first announced; and it fell again by 0.8 percent on March 11, 1986, when it was announced that the rate was to be raised to 2 percent. On the day in 1974 when the stamp duty rate in the United Kingdom increased from 1 percent to 2 percent, the stock market index declined by 3.3 percent.

Derivatives Markets

The effects of STTs on derivatives markets, such as the options market, are very complex, but a general conclusion is not likely to be controversial: even if the STT is levied only on transactions in the underlying assets and not on sales or purchases of the derivative, it is likely to severely affect the growth of such markets.

An options contract, by itself, is very risky for both the buyer and the seller. A financial institution seeking to create a market for options contracts will, therefore, hedge its risk, generally by requiring that the market maker engage in transactions in the underlying asset. If those transactions are subject to an STT, the cost of supplying an option contract—and, hence, the price of the contract—will increase.

Benefits

The most obvious benefit of an STT (as with any tax) is the revenue that it raises, which reduces the need for the government to raise revenues from other sources or to cut spending—neither of which is a costless option. In addition, advocates of STTs view the reduction in trading volume that the tax would bring as likely to be of benefit to the economy. In an early statement of this position, J. M. Keynes wrote:

It is usually agreed that casinos should, in the public interest, be inaccessible and expensive. And perhaps the same is true of Stock Exchanges…. The introduction of a substantial Government transfer tax on all transactions might prove the most serviceable reform available, with a view to mitigating the predominance of speculation over enterprise in the United States.”⁷

There are two main aspects to the possible benefit of a reduction in trading volume. The first is that it would reduce the resources that are devoted to some types of trading that STT advocates see as having no social value. For example, if a trade takes place only because the buyer and the seller have different expectations about the future performance of a security, then the expectations of at least one of them must be wrong. In this case, it could be argued that society as a whole should be indifferent as to which of the two holds the security and that any resources that are devoted to transferring it from one investor to the other are wasted resources. Against this view, a standard presumption in economic analysis is that, if two individuals freely choose to exchange an asset at a given market price, then it may be inferred that both gain “utility” from the trade—and the welfare of society is therefore increased. The different circumstances in which individuals may find themselves provide ample grounds for expecting that they will, at any time, attach different values to the same asset.

Second, it has been claimed that a reduced trading volume will reduce the volatility of share prices, since the STT on share transfers particularly penalizes speculative “noise traders,” whose purchase and sale decisions are based on price movements rather than fundamental information. But this claim, too, has proved controversial. A number of theoretical models have been developed in which the effect of an increase in transaction costs (such as an STT) has the effect of increasing, not reducing, the volatility of prices. Ultimately, however, the issue is an empirical one and there has been no clear empirical support for the view that STTs reduce volatility in financial markets.

Costs

The main costs attributable to an STT have already been mentioned above. In particular, the tax can be expected to raise the cost of capital for enterprises seeking to raise funds by issuing the securities whose transfer is subject to the STT (and so distort the choice by an enterprise between different ways of raising capital, insofar as the STT is levied on some forms of security and not on others), and reduce the efficiency with which the existing stock of those securities is distributed between potential holders. In addition, an STT could be harmful to the development of the domestic financial industry by reducing socially desirable transfers of securities, driving transactions in the taxed securities offshore to competing markets, and inhibiting the development of domestic markets in new instruments (such as options to buy or sell the securities subject to the STT).

Incidence

It is very difficult to ascertain who ultimately bears the burden of an STT. To begin with, the formal responsibility for paying the tax (which different countries assign to the purchaser, the seller, or both—or to the broker who intermediates the sale) does not generally determine its ultimate incidence, because the market prices of the taxed securities are likely to adjust according to where the payment responsibility is assigned, which is largely dictated by considerations of convenience and administration rather than a decision on who should ultimately bear the burden of the tax.

Beyond that, the main factor influencing the incidence of an STT seems to be that it reduces savers’ returns from holding the taxed instruments and, hence, increases the cost of capital for investors who use those instruments to raise new funds. Because the prices of different assets can be expected to adjust in such a way as to equate expected (net) returns on different assets, these differential effects on returns from particular kinds of assets will tend to be generalized to all assets. Thus, the burden of an STT is likely to spread to all asset holders—including holders of assets that are not directly subject to the STT. The ultimate incidence of an STT is, thus, not transparent.

This lack of transparency may be seen as a disadvantage of an STT. However, it may sometimes also make the tax attractive as a simple and rough method of taxing those who earn income from capital when other, more precisely targeted taxes—such as a capital gains tax—are seen as difficult to implement effectively.

Revenue Potential

Even in countries with highly developed securities markets, STTs do not raise significant amounts of revenue. In the United Kingdom, the stamp duty on share transfers raised about 0.12 percent of GDP and about 0.35 percent of total tax revenue annually in the mid-1970s—when it was levied at the comparatively high rate of 2 percent. The rate was reduced to 1 percent in 1984 and reduced further to 0.5 percent in 1986 (when the duty was extended, in the form of a stamp duty reserve tax, to cover transactions in shares without an instrument of transfer). By the end of the 1990s, the total duty on share transfers in the United Kingdom was raising about 0.4 percent of GDP a year, or a little over 1 percent of total tax revenue. Estimates that were made in the United States in 1990 of the potential yield of an STT of 0.5 percent were in the range of 0.12-0.17 percent of GDP, or 0.67-0.95 percent of yearly tax revenue.

These figures do not take account of the ways an STT would reduce revenues from other taxes. In particular, in countries where capital gains on financial securities are subject to tax, an STT increases the cost basis of the securities (if the purchaser is liable for the tax) or reduces the net proceeds from the sales (if the seller is liable). If capital gains are taxed at 30 percent, only 70 percent of the proceeds of an STT will represent net revenues to the government—even if the tax has no effects on prices or volumes. Revenues from the capital gains tax are reduced further if the STT reduces securities prices (thus reducing gains on the securities) or the volume of securities transactions (thus increasing the average lag between the accrual of gains and their realization).

Administrative Aspects

STTs are usually simple and inexpensive to collect—and are becoming even more so as the technology employed in financial markets improves. In the United Kingdom, for example, the collection cost of stamp duties is about 0.12 percent of revenue collected,⁸ compared with an average of between 1 and 2 percent for other taxes. Because STTs are simple, compliance costs for taxpayers and their agents are also likely to below.

An advantage that has often been claimed for STTs—particularly in the form of stamp duties on transfer documents—is that the information they generate for the government about individual securities transactions can be valuable in assessing liabilities to other taxes, such as capital gains taxes and income taxes. This benefit depends, of course, on the extent to which effective use can be made of the information that is generated.

Finally, STTs vary quite widely in the ease with which they can be avoided. As noted earlier, taxes levied on transactions made through a stock exchange can be avoided (at some cost) by transacting outside the exchange. Many STTs can be avoided by moving the taxable transactions offshore. And taxes that are based on the registration of documents can be avoided in a variety of ways—by holding securities in bearer form or in the form of derivative instruments such as American Depository Receipts. There are, of course, things that a tax authority can do to counter these avoidance strategies: for example, some countries apply a much higher rate of STT to transactions that convert registered shares into bearer shares. Nevertheless, it may reasonably be supposed that the financial community will continue to seek new ways to minimize the impact of STTs.