The misguided attractions of foreign exchange controls - Reforming Globalization

For many, the liberalization of capital flows was one of the causes of recent unstable financial markets around the world. Seemingly successful exchange controls in Malaysia suggest that such regulations can limit instability, as has been argued in previous issues of Challenge. But in this piece, economist Robert Dunn presents a contrary case. He says exchange controls can be evaded too easily to be useful.

THE BELIEF THAT BALANCE-OF-PAYMENTS DEFICITS-CAN BE ELIMINATED through a set of strict administrative controls on international transactions has, after decades of disrepute among economists, regained a semipopularity Discussions of exchange controls have been encouraged by the modest success of temporary controls in Malaysia and in a few other countries. Most important, however, controls appear to offer a solution to an unpleasant dilemma faced by developing and transition economies. These countries want both a fixed exchange rate, which encourages domestic price stability in an open economy, and the ability to manage an independent national monetary policy, but it appears that they cannot have both.

It is obvious from both theory and experience that these two goals are not compatible with freedom for private agents to move capital internationally. An expansionary domestic monetary policy and lower interest rates, for example, will cause capital outflows and a balance-of-payments deficit, which, given limited foreign exchange reserves, will compel a devaluation of the local currency or the adoption of a floating rate that will depreciate. Either will result in a parallel increase in domestic prices of tradable goods and broader inflation. If, however, the balance-of-payments effects of an expansionary monetary policy could be blocked, through strict controls on capital flows and other international transactions, monetary policy independence could be maintained without balance-of-payments disequilibria that compelled undesirable exchange-rate adjustments. A set of exchange controls appears to offer a solution to an otherwise miserable choice--either give up national monetary policy autonomy or accept a vol atile exchange rate that creates parallel volatility in the domestic price level, with price instability becoming more serious as the economy becomes more open. Exchange controls appear to offer the possibility of both a stable exchange rate and price level, and an independent monetary policy. (1)

The attractions of such a system of exchange controls are obvious, but the disrepute in which this approach has been held by most economists remains justified. Exchange controls simply do not work very well, and the longer they are in place, the worse the outcomes they produce. They are easily evaded, and the combination of avarice and ingenuity, upon which modern economics rests, guarantees that evasion routes will be found and aggressively pursued. Graft and corruption often accompany exchange controls, as those caught in various evasion schemes offer bribes to customs or central bank officials to be allowed to pursue their goals. A few cynics have even suggested that the maintenance of such controls is sometimes encouraged by the desire of government officials to protect their opportunities to collect such payments.

During wartime, exchange controls may be more successful, particularly in industrialized countries, because a combination of appeals to patriotism and severe criminal sanctions for those caught cheating may encourage citizens to obey the law. Under more normal circumstances, however, the prospects for the success of such a system are poor, particularly in developing countries, where financial record-keeping is weak and where law enforcement systems are frequently inefficient and subject to corruption.

What follows is a road map of a variety of evasion techniques, along with indications of why they are so difficult to stop. First, however, it is necessary to describe how the typical system of exchange controls is designed and how it is to operate. These systems obviously vary in detail among countries, but they typically share certain common characteristics.

The Design of an Exchange-Control Regime

The first requirement for a set of capital controls is that the government or central bank gain control of all foreign-exchange receipts coming into the country. A law will therefore be passed requiring that any resident receiving export revenues, remittances, tourist income, capital inflows, or foreign funds from any other source sell such foreign exchange to the central bank within a brief period of time. Domestic residents will be prohibited from keeping funds in foreign bank accounts or from maintaining other foreign investments unless they have individual permits to do so, and such permits will very seldom be granted. If this law could be enforced, it would mean that all credit items in both the current and capital accounts of the balance of payments would produce prompt flows of the resulting foreign-exchange receipts to the central bank.

The central bank and the government then decide, on the basis of the volume of such inflows, what imports and other debit transactions can be allowed. The second part of the law allows domestic residents to purchase foreign exchange only if they have permits to do so, with separate permits being required for each transaction. Such permits must specify the purpose of the transaction and the amount of foreign exchange that can be purchased for its completion.

The volume of foreign exchange to be purchased under such permits must be regulated to match the inflow of foreign-exchange receipts, with the central bank or government distinguishing between "what matters much, and what matters most." Oil, food, raw materials, machinery, and other vital imports are allowed, with permits being promptly provided. Less critical imports are allowed only if foreign exchange receipts are sufficient, and permits are almost never provided for foreign investments. Domestic residents will also not be provided with permits to educate children abroad, and foreign travel will be allowed only with modest sums of money. Domestic residents wishing to start foreign businesses or to buy foreign securities will simply be refused the right to purchase the necessary foreign exchange, thereby, in theory, making such investments impossible.

The operation of this system will produce a large number of unhappy domestic residents who wish to purchase foreign exchange but who are not allowed by law to do so. Such people will probably be willing to pay more than the official exchange rate to purchase dollars in illegal transactions in an informal market. The result will be two exchange rates: the official rate, at which exporters are paid when they surrender funds and which importers with permits are allowed to use, and a second rate on the street, at which a variety of informal transactions will occur. Using India as an example, if the official exchange rate is 49 rupees per dollar, the street rate might be 55 or even 60 rupees per dollar. The gap between the two rates can sometimes become very wide. Recently in Turkmenistan, the legal rate was about 500 mannats per dollar, while 1,500 per dollar was readily available on the street from what are called "banka babushka"--elderly women who trade foreign exchange from the back seats of parked cars. (2)

Routes to Evasion: The Various Ways of Getting Money Out

The spread between the legal exchange rate and that prevailing in the informal street market produces strong incentives for the diversion of a variety of current-account receipts from legal channels to the parallel market. If the government is offering 49 rupees per dollar, while 55 or 60 are available on the street, Indian residents receiving dollars will certainly want to sell them at the higher rate. Such diversion may be easiest in the case of tourist receipts and remittances. Foreign visitors will be besieged by children offering to buy dollars for 52 or 53 rupees. Such entrepreneurial youngsters will profit handsomely if they can resell the dollars for 55 or 60 rupees. Indian families receiving dollars from relatives working abroad are supposed to sell such funds for 49 rupees but will obviously prefer 55 or 60. If the dollars arrive as currency. either through the mail or being brought into the country by returning workers, they can easily be sold in the street market.

Recorded tourist receipts and remittances in the current account of India's balance of payments will decline sharply, not because fewer foreigners are visiting the country or fewer Indians are working abroad, but merely because the resulting receipts are being diverted to the street market. The purchasers of the dollars from that market will find ways to get the currency out of India, where it can be used for the education of children, for travel, or for investments.