104

either with restraints on capital mobility and foreign exchange transactions or with free movement of capi- tal.

For any country, the ad- vantage of exchange con- trols and restrictions on

NEW ECONOMY

The present regime of the Group of Seven - really of the ERM (mainly Germany in practice), the USand Japan-is a bit dirty. m e governments and central banks sometimes agree on ranges for the dollar vis-a-vis the yen and/or the

"Vast resources of intelligence and

entelprise are wasted in financial speculation, essentially in playing

zero-sum games"

Speculators' tax International policy coordination

and national monetary autonomy JAMESToBIN

- why both are needed and how a transaction tax would help

Sterling Professor

Yale University. Nobel Eme,-jtusofkonomics,

prize Winner ,98,

107Q-3535/94/020104 + 06 SO8.OOIO (c) THE DRYDEN PRESS

SPECULATORS' TAX 105

own currency. Under the Bretton Woods sys- tem (1946-1971), for instance, panties could be, and frequently were, changed by the coun- try itself. This was almost always by devalu- ations of weak currencies rather than appre- ciations of strong ones.

A single currency with mobile capital

The most successful currency union has been that of the USA which was, of course, just one feature of a comprehensive union formed two hundred years ago. It is now inconceiv- able that the dollar will not always be the same in Texas as it is in New York State. Within the US, states and regions cannot run separate monetary and credit policies. Finan- cial markets are completely integrated.

This system, a single currency together with perfect capitalmobility, has great advan- tages: gains from trade, economies of scale and scope, and diversifications of risk.

The disadvantages are the vulnerability of regions and sectors of the economy to eco- nomic upheavals that reduce their competi- tiveness. When currency revaluations, capital controls and trade restrictions are ruled out, other remedies have to be in place.

The main adjustment process is simply the natural response of competitive markets. In the US there are always depressed areas, but their identities are always changing. Mobility of labour and capital prevents areas from be- ing permanently depressed. In addition, the federal government generally assists de- pressed regions and sectors, to ease the dis- tress of transitions resulting from the natural shocks of economic development. Responses of both kinds are, I think, essential to make a single currency system work.

A single currency differs dramatically - in kind not just in degree - from any monetary system that has separate currencies. I there- fore doubt that Europe can move to complete monetary union via progressively greater fixity of rates among separate nationalcurren- cies. A permanent single currency is what Americans call a 'whole new ball game'.

Adjustable pegs and floating rates Compromise systems, with fixed but adjust- able rates (or ranges of rates) are not as dif- ferent from floating rate systems as they are usually said to be. This is because the chosen panties are not irrevocable. There will be speculation against them whenever funds are allowed to move across currencies. In contrast, there is no currency speculation within a single-currency system.

However, speculation is not the same under a Bretton Woods or ERh4 system as under mar- ket-floating. In the former, speculation focuses not on what the currency market will do tomor- row, but on whether, when and by how much country X might devalue its currency. In float- ing-rate systems speculation concerns eco- nomic influences on rates as well as government interventions.

Under a semi-fixed parity systemchanges in exchange rates are usually very traumatic. They are called crises every time they happen. They draw much more attention than continuous and gradual market rate changes. Perhaps this is because changes in fixed rates are govern- ment policy decisions, whereas fluctuations in market-floating rates seem impersonal.

Many business managers are unhappy with floating rates because the exchange rates on which sales and profits seem to depend are unpredictable. But they are not predictable under an adjustable-peg systemeither. And large, discrete changes in exchange rates un- der government control are probably more disruptive of private business plans and transactions than the day-to-day fluctuations of floating-rate markets.

An endemic flaw in the Bretton Woods sys- tem, and to some extent the E M , was the asymmetry between countries running cur- rent account surpluses and those running deficits. When countries ran out of reserves, they had no choice but to devalue. But there was nothing beyond pious words to stop sur- plus counties from accumulating reserves in- definitely while maintaining comfortably un- dervalued currencies. This kind of asymme-

106 NEW ECONOMY

try is almost inevitable in any system of ad- justable pegs.

Coordinating international policy In a fixed exchange rate regime, in which funds are highly mobile across currencies and frontiers, the interest rates in any one country can differ very little from those in the countries to which its currency is pegged. A single country must therefore rely on fiscal measures for counter-cyclical policies. But even so a country may confront an impasse, a ‘fundamental disequilibrium’. Domestic full employment, given the fiscal policy re- quired to sustain it, may imply chronic trade deficits and continuous drains of intema- tional reserves.

One way out is devalu- ation - provided deprecia- tion of the nominal exchange rate is not cancelled out by inflation and therefore does succeed in restoring com- petitiveness. If devaluations are excluded, the country

its own national monetary policy, who is re- sponsible for the overall monetary policy of the group?

The several centres jointly deploy enough instruments to determine all the exchange rates among their currencies as well as the group’s overall macro-policy. But the latter will be an accidental by-product of the mone- tary strategies the several members of the group are using.

If a system is not run by a dominant power there has to be some coordination. The role of policy coordination is to prevent jockeying for macroeconomic advantage that, collectively, can be counterproductive. When national central banks focus on interest rate differen-

“The world average interest rate level may

become a12 accident

I of national game-playing”

must somehow contrive to have very flexible prices and wages and so adjust the economy to the exchange rate. Ex- perience suggests that such adjustment is often slow and painful.

Follow my leader In a fixed exchange regime overall intema- tional policy has generally been determined by one country acting as the decisive mone- tary policy-maker. It is the interest rate deter- mined by the hegemony of that country that other countries have to accept. The dominant country was Britain in the pre-1914 gold standard, the US in the Bretton Woods era, and Germany in the EMS.

In the absence of the hegemony of a single government or central bank, where does macroeconomic control and policy reside in a fixed exchange rate regime with substantial capital mobility? If there are several inde- pendent centres of policy - governments or central banks -and if each is trying to conduct

tials, as they must do in a world of capital mobility, the overall average of interest rates will be nobody’s busi- ness. It may end up too high for general prosperity or too low to keep inflation at bay. The world interest rate level, or the European interest rate level, requires international

coordination. Likewise, coordination is needed to define permissible national devia- tions from the agreed average.

Macroeconomic policies in a floating rate regime

Policy coordination is also necessary in a floating exchange rate regime. Experience of the past twenty years has refuted the extrava- gant claims of some advocates of floating rates that national monetary policies could proceed without reference to the outside world leaving the currency markets to recon- cile the policies and macroeconomic per- formances of the several economies.

National monetary policy is a more effec- tive tool for domestic demand management in floating systems than in fixed rate regimes. But it works by moving the exchange rate: depreciating it to shift demand from foreign to domestic goods or appreciating it to shift demand in the opposite direction. It can be a

SPECULATORS‘ TAX 107

beggar-thy-neighbour policy for employ- ment and output in the first case and an expo- nent of inflation in the second.

Differential interest rate movements, ac- tual and anticipated, are the mechanisms by which national monetary policies move ex- change rates. Funds move across currencies into money markets and bond markets where interest rates are higher. In a floating-rate re- gime these movements alter exchange rates: capital inflows, induced by high interest rates, appreciate the exchange rate and, in time, generate trade deficits. This scenario was dramatically played out for the US dollar in the early 1980s.

In the floating-rate system, as in the fixed- rate regime, the world average interest rate level may become an accident of national game-pla y ing.

However, floating rates add some degrees of freedom. Interest rate differentials can be sustained by compensating expectations of currency appreciations and depreciations. Say a country needs demand stimulus through interest rates lower than the rest of the world - like the United States in 1991 or Japan in 1994. It could, in principle, carry out this policy with an appropriately low ex- change rate, provided the markets expect its currency gradually to appreciate in future.

Policy coordination must pay attention, not only to short-run demand stabilisation, but also to trends in current accounts. Cur- rency values should be low but rising for countries with excessive current account defi- cits and high and falling for countries with excessive surpluses. To make these develop- ments possible fiscal policies should be tight in the former countries and easy in the latter.

Policy coordination of this kind is not easy either intellectually or politically. Govern- ments have not been notably successful at it. So far, ’coordination’ seems to have been di- rected less to policy than to one of the out- comes of policy, namely exchange rates.

Sand in the wheels Speculation on currencies is a serious prob-

lem in any regime short of a single currency. Greater freedom of capital movement brings important advantages but at the cost of en- hanced speculative opportunities, as the dra- matic collapse of the ERM in 1992-1993 illustrated. There is more scope for bubbles and for false signals from financial markets. Markets have whims which the avthorities might not, should not, always like to follow. The markets have developed extremely effi- cient technologies of transactions and infor- mation. It is so easy and so cheap to make financial transactions, and the amount of pri- vate funds that can quickly be mobilised to support fashionable market opinion is so large, that countervailing official interven- tions have difficulty controlling exchange rates and interest rates.

I take credit, or confess guilt, for having suggested sixteen years ago a radical pro- posal for international monetary reform. I proposed to put ’sand in the wheels’ of the excessively efficient currency markets. These markets are engines that work all too well technically but not that well economically, es- pecially macro-economically. The sand in the wheels would take the form of transactions taxes, which direct traders’ attention to long- run fundamentals and away from contagious but transient market sentiment.

Suppose you have to pay a one per cent tax every time you make spot transactions in for- eign exchange markets. If you intend to hold the asset purchased for 30 years, the tax is one per cent going in now and one per cent com- ing out 30 years later. The calculation of the rate of return that induced you to make the transaction is negligibly influenced by the tax. If it was a socially worthwhile allocation of saving, within or between national econo- mies, the tax would not interfere with it. But say you are a ’day trader’ - in this morning and out this afternoon -buying sterling with dollars and then reversing the transaction the same day. Then that one per cent tax each way eats up any gain in the value of the sterling pretty fast.

‘Sand in the wheels’ deters traders from

108 NEW ECONOMY

acting on short-run views of investments, either international or domestic. Transaction taxes would diminish excess volatdity. They would focus investors’ attention on longer- run fundamentals. Perhaps they could secure the benefits of increased mobility without some of the costs.

The intent of a tax on foreign exchange transactions is to slow down capital move- ments, not commodity trade. But even if it were not possible to exempt bona fide com- modity transactions, the tax would be too small to be a significant trade barrier. In any case, since it would have to be imposed by all countries, it would not affect export-import balances.

At the same time a tax on currency transac- tions would create room for greater national autonomy in monetary policy. A two per cent tax on a round trip to another currency wipes out an eight-point differential in annual interest rates on three-month bills. The tax thus sustains differences in lo- cal-currency interest rates that would otherwise be erased by arbitrage.

Let me be clear about sand in the wheels. There’s good sand and there’s bad sand. Saying that a little sand in the wheels may be a good thing

against speculators’ guesses of the short- range behaviour of other speculators. The same is true for the foreign exchange markets. Vast resources of intelligence and enterprise are wasted in financial speculation, essen- tially in playing zero-sum games. Transac- tions taxes might re-allocate some of these resources. If not, they will at least produce needed government revenues without bad side effects.

It is estimated that more than $100 billion gross foreign exchange transactions occur every business day in New York alone. Since the currency transactions tax would have to be international, the proceeds might appro- priately go to international institutions such as the United Nations or the World Bank.

A sense of community The international integration of financial

markets is a trend that can-

“Transaction taxes would diminish excess

volatility, focusing investors’ attention

on longer-run jiindamentals. Maybe

they can secure the benefits of increased

mobility without some of the costs.’’

does not mean that every mo-

not be stopped. Even if they wanted to governments would not be able to impose effective barriers on move- ments of funds across cur- rencies and borders. Goods and services and labour are becoming more mobile also, but at a much slower pace. Governments can still im- pose barriers on their move- ment, and they do.

nopolistic restriction of -entry into financial industries is justified. Nor does it mean that every fixing of interest rates by administrative decree is beneficial. Interven- tion that is impersonal, constant and automat- ic - like the fixed transactions tax I suggested - will do its job in efficiently compared with the usual ad hoc interferences with mobility of capital. The transactions tax might actually make it easier to clear up some counterpro- ductive monopolistic and restrictive financial practices.

In 1936 Keynes pointed out that a transac- tions tax could strengthen the weight of long- range fundamentals in stock market prices

The discrepancies be- tween speedily moving funds and slowly ad- justing trade create grave problems in macroeconomic management. They are not easily solved by any international monetary regime.

A single currency, superseding national currencies, has great merits. But these can be realised only with substantial help from other common economic, political, and social insti- tutions, for all of which a basic sense of com- munity is essential. Perhaps this can be achieved one day in Western Europe. Else- where, even for potential or actual ’free trade areas’ such as NAFTA, the outlook for cur-

SPECULATORS' TAX 109

rency union is much less promising. A system of fixed-but-adjustable exchange

parities among national currencies does not achieve the benefits of a single currency. In an adjustable-peg system defense of reserve po- sitions and parties becomes the primary con- cern of governments and central banks of defi- cit countries. No counterpart adjustment re- sponsibilities are felt by surplus countries. And, once the adjustable-peg system is loose from its anchor, the regime delivers neither the advantages of a permanent single cur- rency nor those of a flexible floating-rate sys- tem. Nowadays the funds at the disposal of the private sector are so enormous that it may be impossible for central banks to defend cur- rency parities against speculative attacks, as the break-up of the ERM showed.

A floating-rate systemby no means assures painless automatic adjustments to distur- bances in trade and capital transactions among countries. Nor does it guarantee easy

adjustment to idiosyncratic differences be- tween countries' economic policies and devel- opments. But it does, as we have seen, have important virtues.

The key point, however, is that interna- tional policy coordination is needed in any system that preserves national currencies and national monetary policies, whether adjust- able pegs or floating rates. The speculative opportunities created by the technologies of modern financial markets and financial insti- tutions are a threat to rational polices, both national and international.

A modest international transactions tax may be useful in focusing finanaal investors on long- run fundamental prospects rather than short- run gains, and in preserving some room for manoeuvre for national monetary authorities.

This is an updated, edited version ofa paper which appeared in the Greek Economic Review in March 1993.