Does a Tobin Tax Make Sense?


Published on

Lecture given at IDS on 11 March 2010 on the idea of a Tobin Tax. Includes analysiss of Robin Hood Tax campaign, evidence from previous tax models and idea of Inductance Tax.

  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide
  • The auction mechanism has intelligence in the sense that matches the buys and the sells let the pricing converging to the equilibrium point.
  • Hanke (2006) simulates 2 continuous double auction markets, denoted LEFT and RIGHT on which a foreign currency (Taler) can be traded for the home currency (Gulden). They analyse the effect of the imposition of the transaction tax (0.5% of the transaction value) on one and then on both markets. In order to capture long lasting effect of the imposition of a Tobin tax they also consider a scenario where the tax is abolished again after its introduction. Shi and Xu (2009). A main assumption is that informed traders's unconditional expectation of excess return depends on the noise component of the market, which does not affect noise traders' expectations, causing an “asymmetric expectation effect” on the gross benefits of entry. The noise component is the ratio on noise entrants to informed entrants, An increase in the noise component increase market volatility. Westerhoff (2003) and Westerhoff and Dieci (2006) developed a simulation model of heterogeneous interactive agents in which rational agents apply technical and fundamental analyses for trading in two different markets. The technical analysis is based on past price trends, whereas fundamental analysis predicts a convergence towards fundamentals. The agents are free to trade and so they have several options, which are chosen depending on their relative fitness, where the fitness is given as a weighted average of current and past profits. Ehrenstein (2002, 2005) augmented ZI agents model based on the percolation theory of Cont and Bochaud (2000) with the introduction of a Tobin Tax. The percolation theory describes the behavior of connected clusters in a random graph. Agents random form clusters. They take decision to buy, sell with probability (a), called “activity”, whose maximum value is calculated to be 0.5, or remain inactive with probability (1-2a), “inactivity”. Aggregate excess of demand, i.e. the sum of all the orders, is the only force influencing the determination of prices. Ehrestein (2005) used this model to evaluate the impact of Tobin tax on volatility, market distortions and government revenue, according to the introduction of different heights of the tax (from 0 to 1%), different curvatures of the price function (g=0, g=0.19, g=0.4) and a different percentage of producers in the market. Mannaro et al. (2008) propose a multi-agent simulation model for analyzing the effects of introducing a transaction tax on one and then on two related stock markets from a structural and behavioural perspective. The microstructure of the market model is composed of four kinds of traders (Raberto et al., 2003): Random traders, who trade at random; Fundamentalists, who pursue the “fundamental” value; Chartists, trend-followers who are divided into: Momentum traders, who follow the market trend; Contrarian traders, who follow the opposite of the market trend. Kaiser et al. (2007) gives an example of the game theoretical approach applied to asset market with or without the introduction of a Tobin tax. They used an experimental market inspired by a discretized double auction which is modified from the traditional Vernon Smith’s one, to reach a game – theoretical solution. The experiment was divided into 2 scenarios: untaxed and taxed, where the tax’s height varied in order to analyse the elasticity of volatility to the imposition of such a tax. They carried experiments on 96 subjects, mostly students from the University of Bonn. They analysed 6 sessions for the taxed scenario and 6 sessions for the untaxed one. Each session lasted 2 hours. Differently from the traditional measure of volatility, they use a variance which measures absolute difference between prices and their average. Bianconi et al. (2006) analyze theoretically the impact of the imposition of a transaction tax (Tobin tax) on the volatility of the exchange rate in a GCMG model. The first effect of the tax is to increase the profit threshold for speculators, discouraging them from trading because of the reduction of the strategy score. Furthermore, the effect of the imposition of a tax depends on the position of the market with respect to a critical zone where the information efficiency of the market is reached. If it is far to the left, the tax has mild effect on volatility and information efficiency. If the market is within the critical zone and volatility is high, only a sufficiently large tax will have impact on volatility. Moreover, the impact on volatility is found to be very dependent of the market size. Hence, since volatility decreases with the size of system, the effect of a Tobin tax would be much stronger in a small system than in bigger ones. Finally, in a market of evolving agents composition a tax can reduce volatility only if the change in the composition is slow.
  • In the same year, the Stamp duty reserve tax (SDRT) applies to transfers of beneficial ownership which were previously not subject to stamp duty, electronic, paperless transactions that are made through a stockbroker. In 1990, the government announced the abolition of the stamp duty, subject to the introduction of a new equity settlement system, TAURUS. However, its collapse leads stamp duty to remain alive.
  • Perhaps the problem is that we confound the term of holding (short-term vs long-term) with whether it is stabilising or destabilising. But short-term trades could be stabilising. What you want is not to tax trade – but to tax panic!
  • Similar ideas to the Inductance Tax have been proposed before. For example, Spahn has already proposed a two-tier Tobin Tax, where the tax rate would rise if market volatility rose above a certain threshold. But the Inductance Tax has the advantage that it would not require the specification of arbitrary thresholds between the lower and upper tier tax rates.
  • Does a Tobin Tax Make Sense?

    1. 1. Does a Tobin Tax Make Sense? Neil McCulloch Sussex Development Lecture Chichester lecture theatre 11 March 2010
    2. 2. Overview <ul><li>What is a Tobin Tax? </li></ul><ul><ul><ul><li>The characteristics of financial markets </li></ul></ul></ul><ul><li>Does a Tobin Tax reduce volatility? </li></ul><ul><ul><ul><li>In theory </li></ul></ul></ul><ul><ul><ul><li>Evidence from similar taxes </li></ul></ul></ul><ul><ul><ul><li>The Inductance Tax </li></ul></ul></ul><ul><li>Is a Tobin Tax feasible? </li></ul><ul><ul><ul><li>Substitution </li></ul></ul></ul><ul><ul><ul><li>Migration </li></ul></ul></ul><ul><li>How much money will a Tobin Tax raise? </li></ul><ul><ul><ul><li>and who will pay it? </li></ul></ul></ul><ul><li>Conclusion </li></ul>
    3. 3. 1. What is the Tobin Tax? <ul><li>“… an international uniform tax on all spot conversion of one currency into another, proportional to the size of the transaction.” </li></ul><ul><li>“ A Proposal for international monetary reform ”, James Tobin, 1978. </li></ul><ul><li>Suggested tax rate: 0.1 % </li></ul><ul><li>Main objective: penalizing short-term speculators but not long-term investors, in order to would stabilize markets. </li></ul>
    4. 4. Or perhaps the Keynes Tax? <ul><li>“ Speculators may do no harm as bubbles on the steady stream of enterprise. But the position is serious when enterprise becomes the bubble on a whirlpool of speculation” … </li></ul><ul><li>“ 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 enterprises in the United States.” </li></ul><ul><li>Keynes, J.M. The General Theory of Employment, Interest, and Money . New York, 1936 </li></ul>
    5. 5. Characteristics of Financial Markets <ul><ul><li>Excess liquidity (i.e. too much trading activity) due to the predominance of short-term speculation. </li></ul></ul><ul><ul><li>Excess price volatility </li></ul></ul><ul><ul><li>Long swings in asset prices and persistent deviations from fundamental equilibria. </li></ul></ul><ul><ul><li>Fat tailed distributions of returns and volatility clustering </li></ul></ul>
    6. 6. £/US$ Exchange rate
    7. 7. The much feared Black Swans
    8. 8. Fat tailed distribution
    9. 10. 2. Does a Tobin Tax reduce Volatility? <ul><li>Theoretical arguments: </li></ul><ul><ul><li>Traditional Approach </li></ul></ul><ul><ul><li>Heterogeneous Agents Models </li></ul></ul><ul><ul><li>Zero intelligence Agents Models </li></ul></ul>
    10. 11. Traditional Approach <ul><li>Traditional theoretical models assume that all actors in financial markets act to maximize their own welfare and that they have “rational expectations” about the future i.e. that people adjust their forecasts to be consistent with their experience so far. </li></ul><ul><li>However, such models do not explain many of the characteristics that are observed in real financial markets such as excess liquidity, excess price volatility, fat tailed distributions of returns and volatility clustering. </li></ul>
    11. 12. Heterogeneous Agents Models <ul><li>These models assume that market actors are not perfectly rational, but rather apply rules-of-thumb when making decisions to buy or sell, based on whatever information they have at their disposal. </li></ul><ul><li>They also assume that there are different types of market actors: </li></ul><ul><ul><li>“ fundamentalist” traders i.e. those that trade based on a view about the fundamental value of the assets, and “noise” traders i.e. speculators. </li></ul></ul><ul><li>The volatility of the market is therefore driven by what share of market traders are noise traders (who increase volatility) and what share are fundamentalists (who reduce it). </li></ul>
    12. 13. Zero intelligence Agents Models <ul><li>They assume that market traders in the aggregate, behave probabilistically rather than being driven by any intelligent maximizing behaviour. </li></ul><ul><li>The main assumption is that agents place orders to buy and sell at random, subject to constraints by current prices. They therefore have zero intelligence (although some minimal intelligence is needed for tracking prices). Only the institutions have some kind of intelligence, i.e. auction, which let the prices converge to equilibria. </li></ul><ul><li>Gode and Sunder(1993) claim that if students in an economics classroom are replaced in an experiment by zero intelligence agents with a budget constraint, the behavioural results are almost the same. </li></ul>
    13. 14. The Effect of a Tobin Tax on Volatility: Simulation Results AUTHOR(S) RESULTS Hanke (2006) Increase or decrease depending on market size Shi and Xu (2009) Increase or decrease depending on the effect on the number of noise traders Westerhoff (2003) and Westerhoff and Dieci (2006) Decrease Ehrenstein (2002, 2005) Decrease, as long as the tax rate is not too high to affect the liquidity Mannaro et al. (2008) Decrease, but only in presence of noise traders in the market Kaiser et al. (2007) Decrease Bianconi et al. (2006) Decrease but depending on market size
    14. 15. Volatility and Tobin Tax, THE SWEDISH EXPERIENCE <ul><ul><li>January 1, 1984 Introduction of a round-trip tax of 1% of the value of exchanged securities (i.e., taxes of 0.5% on both purchases and sales). </li></ul></ul><ul><ul><li>July 1, 1986 : The tax was increased to 2% </li></ul></ul><ul><ul><li>1991 , Removed. </li></ul></ul>
    15. 16. THE SWEDISH EXPERIENCE, cont. <ul><li>These results suggest no significant difference in the weekly variance </li></ul><ul><li>However, the daily variance is at its highest level during the 2% tax regime and the difference between the regimes are always statistically significant. </li></ul><ul><li>Umlauf (1993) compares the performance of the Swedish stock market under the no tax, 1% and 2% tax regimes. </li></ul>
    16. 17. <ul><li>Swedish shares traded in London are not subject to the tax. If the tax reduced volatility, we would expect the ratio of volatility in shares in London relative to those in Stockholm to increase </li></ul><ul><li>Umlauf looks at 11 key shares traded in both countries (e.g. Electrolux, Volvo, Ericsson etc) </li></ul><ul><li>Finding : The ratio falls or remains stable across the different tax regimes. </li></ul><ul><li>Final conclusion: The imposition and increase of a transaction tax increases volatility in the taxed market. </li></ul>Daily return standard deviations Weekly return standard deviations July 1986 to Dec. 1987 July 1986 to Dec. 1987 Number of the ratios that did not rise after the tax increase to 2% 9 5 Average change from the Jan 1984 - July 1986 period -0,06 -0,02
    17. 18. The Stamp Duty in UK <ul><li>The stamp duty is a tax on “change” in ownership which must be registered in UK. </li></ul><ul><li>Its rate changed over time: </li></ul><ul><ul><li>1694: introduction </li></ul></ul><ul><ul><li>August 1963: lowered from 2% to 1% </li></ul></ul><ul><ul><li>May 1974: increased from 1% to 2% </li></ul></ul><ul><ul><li>April 1984: lowered from 2% to 1% </li></ul></ul><ul><ul><li>October 1986: reduced to 0,5%. </li></ul></ul>
    18. 19. The Stamp Duty in UK effect on Volatility <ul><li>Saporta and Kan (1997) compare the performance of the UK stock market before and after the announcement of an increase or decrease of the Stamp duty. </li></ul><ul><li>Moreover, they compare the price of a sample underlying shares of UK-listed companies, (subject to stamp duty), with the price of their US-listed American Depositary Receipt (ADR) (not subject to stamp duty). </li></ul><ul><li>Hypothesis: similarly to Umlauf (1993), if the ratio ADR/UK diminishes with the increase of tax, it would suggest that transaction taxes increase volatilities. </li></ul><ul><li>Findings </li></ul><ul><li>no significant effect of UK Stamp duty imposition on volatility of equity prices. </li></ul><ul><li>Not much variation of volatility across stamp duty regimes. </li></ul>
    19. 20. Inductance tax <ul><li>The existence of spikes in a distribution caused by manias and panics reflects the presence of high frequency components in the time series. </li></ul><ul><li>Engineers routinely remove undesired high frequency elements using “low-pass filters”. </li></ul><ul><li>The resistance of the inductor is proportional to the rate of change of the current going through it – when the frequency is changing slowly resistance is minimal and the current passes through but when the frequency is high the current is blocked. </li></ul><ul><li>A tax with the same property as an electronic inductor would help to prevent spikes. </li></ul>
    20. 21. Inductance tax : Don’t tax trade, tax panic <ul><li>An inductance tax would tax transactions at a rate proportional to the rate of change of the aggregate market price . (Note that this is quite different from a Tobin Tax which would charge a fixed small tax rate). </li></ul><ul><li>With an inductance tax, sales and purchases would incur virtually no tax during normal times because the aggregate market movement is very small. But during crashes sales would face heavy penalties, as would purchases during booms. </li></ul><ul><li>Because market participants would know that they would face heavy taxation associated with buying during booms and selling during panics, they would be discouraged from making these trades thereby lessening the mania or panic. </li></ul>
    21. 22. Volatility: Conclusion <ul><li>Financial markets are very volatile. </li></ul><ul><li>There are periods of more volatility and periods of less </li></ul><ul><li>In theory, a Tobin like tax could reduce volatility </li></ul><ul><li>But existing theoretical and empirical evidence do not give a clear picture of the effect of a Tobin Tax on volatility </li></ul><ul><li>This may be because the underlying assumption (noise traders are bad) is wrong. </li></ul><ul><li>If so, an Inductance Tax might be more stabilising than a Tobin tax. </li></ul>
    22. 23. 3. Is the Tobin Tax feasible? <ul><li>The Tobin like tax “could be implemented relatively easily and cheaply, using existing market infrastructure and networks” (Kapoor, 2006) </li></ul>
    23. 24. Three challenges to implementation <ul><li>At what point in the system to levy the tax? </li></ul><ul><li>What instruments to tax? </li></ul><ul><li>How do you avoid migration to untaxed locations? </li></ul>
    24. 25. Structure of the Foreign Exchange market FOREIGN EXCHANGE MARKET REPORTING DEALERS 43% OTHER FINANCIAL INSTITUTIONS 40% NON FINANCIAL CUSTOMERS 17% mutual funds, pension funds, hedge funds, currency funds, money market funds, building societies, leasing companies, insurance companies, other financial subsidiaries of corporate firms and central banks smaller commercial banks and securities houses non-financial end users, such as corporates and governments. INTERBANK -WHOLESALE MARKET RETAIL MARKET Large commercial and investment banks; large securities houses
    25. 26. At what point in the system to levy the tax? <ul><ul><li>There are thousands of dealers and intermediaries across many countries </li></ul></ul><ul><ul><ul><li>… but the same is true of retailers in countries who still manage to charge sales tax/VAT </li></ul></ul></ul><ul><ul><li>However one might be able to levy at settlement </li></ul></ul><ul><ul><ul><li>The Continuous Linked Settlement Bank (CLS Bank) settles 55% of global foreign exchange transactions (rest through High Value Domestic Settlement systems of individual countries) </li></ul></ul></ul><ul><ul><ul><li>More than 60% of all sterling and euro transactions are settled in the CLS system. The majority of the remainder processed through the UK’s CHAPS and the ECB’s TARGET System for their respective currencies. </li></ul></ul></ul><ul><ul><ul><li>A single clearing system, SWIFT and its affilitates, is used for all large-value financial transactions, allowing tracking of transactions </li></ul></ul></ul>
    26. 27. What instruments to tax? <ul><ul><li>Foreign exchange spot? Forwards? Futures? Options? Swaps? All derivatives? </li></ul></ul><ul><ul><ul><li>They have dramatically different costs so you would have to tax at different rates (RHT propose 0.005% for forex and for bond and derivative markets; and 0.5% for equity) </li></ul></ul></ul><ul><ul><li>Financial markets are extremely innovative! </li></ul></ul><ul><ul><li>This imposes a cost on tax authorities keeping up with avoidance </li></ul></ul><ul><ul><li>But (almost) all foreign exchange contracts end in a spot transaction </li></ul></ul>
    27. 28. How do you avoid migration to untaxed locations? <ul><ul><li>Migration risk is real </li></ul></ul><ul><ul><ul><li>Almost 60% of the trading volume of the 11 most actively traded Swedish shares migrated to London because of the Swedish tax!) </li></ul></ul></ul><ul><ul><li>Long-run effects are larger than short-run effects </li></ul></ul><ul><ul><ul><li>Shifts in institutional capacity and expertise take time. </li></ul></ul></ul><ul><ul><ul><li>But </li></ul></ul></ul><ul><ul><li>Avoidance is harder than it used to be </li></ul></ul><ul><ul><ul><li>New centres have higher settlement risk (HERSTATT Risk) </li></ul></ul></ul><ul><ul><ul><li>They must follow Basel II rules and comply with money laundering regulations which is expensive </li></ul></ul></ul>
    28. 29. Click to add title Click to add text Only 27% to 28% of the Ericsson , Sweden’s most actively traded company, took place in Stockholm between 1988 and 1991. After the abolition of the tax it increases to 41%.
    29. 30. Bottom line on Implementation <ul><ul><li>Ideally </li></ul></ul><ul><ul><li>Get everyone in the world to agree! </li></ul></ul><ul><ul><li>Impose penalties on non-compliant states </li></ul></ul><ul><ul><li>In practice </li></ul></ul><ul><ul><li>if you tax immobile resources …and the big players complied then can still raise revenue </li></ul></ul><ul><ul><li>The UK’s Stamp Duty still raises £3,203 million each year (2008-09, National statistics, HM Treasury and Custom, September 2009). </li></ul></ul>
    30. 31. 4. How much money will a Tobin Tax raise? <ul><li>It depends on one’s assumptions about: </li></ul><ul><ul><li>Tax rate </li></ul></ul><ul><ul><li>Reductions in trade </li></ul></ul><ul><ul><li>Avoidance </li></ul></ul><ul><ul><li>Scope of application </li></ul></ul>
    31. 32. Author(s) Tax rate Year of Daily turnover Geographical coverage Elasticity Reduction of volume Total Annual Revenue ($ bn) Tobin 0,50% 1995 Worldwide 1500 Spahn (1995) 0,02% 1995 Worldwide none 50 Felix and Sau (1996) 0,25% 1995 Worldwide 0,5% to 1% 300.2 to 393.4 Frankel(1996) 0,10% 1995 Worldwide 0,32% 166 Spahn (2002) 0,10% 2001 EU and Switzerland (including the UK) none 15% 16 UN (General Assembly, 2001) 0.1% 2001 Worldwide 132 Paul and Walhberg (2002) 0,05% 2001 Worldwide but no US$ and other major currencies none 50% 38 Bruno Jetin and Lieven Denys (2006) 0,10% 2004 Worldwide 1% 67% 125 Spratt (2006) 0,005% 2005 UK (sterling) 0.11% 5% 3 Schmidt (2008) 0,005% 2007 Only dollar, Pound, Euro and Yen 0.43% 33.41
    32. 33. Who pays the bill?
    33. 34. 5. Conclusions <ul><li>The idea of a Tobin Tax is a good one – financial markets are excessively volatile, so there could be strong gains from greater stability </li></ul><ul><li>The evidence that a Tobin Tax would reduce volatility is weak … an Inductance Tax might do better </li></ul><ul><li>Implementing a Tobin Tax would be difficult … but not impossible </li></ul><ul><li>A Tobin Tax would raise significant revenue … but a lot less than its advocates claim </li></ul><ul><li>It seems likely that a significant share of the cost would be passed on eventually … but the revenue collected and any additional stability gained might still make it worth it </li></ul>