Tempting targets

Should Tajikistan fix its exchange rate or its volume of currency? Zafar Davronov points out the pitfalls

You know the apocryphal Chinese curse: May you live in interesting times. Last week adumbrated an interesting trend in the exchange rate of the tenge for the dollar: After strengthening for most of the year, in line with rising oil prices, the tenge has been falling steadily this month, roughly from 146.7 tenge per dollar to 147.3T. The most evident reason is that the International Monetary Fund (IMF) released a forecast of Kazakhstan’s economy that was less giddy than the government’s. Astana projects 7% real growth in gross domestic product (GDP) this year; the IMF, 4%. The tenge fell most steeply after June 8, when the IMF report appeared. For most of 2010, the central bank of Kazakhstan has let foreign exchange markets set the value of the tenge, but it must have been tempted to check its fall toward 148T per dollar.

This is one episode in a larger drama: When, if ever, should the central bank fix the exchange rate in a transition economy? This February, the National Bank of Kazakhstan put the exchange rate for the tenge in a corridor from 127.5T to 165T, or 15% and 10% variations respectively from the target of 150T. You could drive a freight train down a corridor that wide. Why doesn’t the Bank save itself all this trouble and just announce that it will let the market determine the exchange rate? With a 25% leeway, that's effectively what it's doing.

As it happens, another central bank in the region has wrestled with this question. Tajikistan, saddled with one of the weakest economies in a weakened global economy, embarked last year on a radical experiment: Flexible exchange rates amid the turmoil of global finance. The plight of the National Bank of Tajikistan may shed light on that of Kazakhstan’s.

Tajikistan consigned its somoni to foreign exchange markets when collapsing national income, and devaluations by major trading partners, made it hard for the central bank to defend its currency. In 2009, the somoni lost a fourth of its value relative to the United States dollar. Half of that loss occurred in May, when it fell to 3.79 somoni to the dollar. It weakened sharply again in May 2010, to 4.57 somoni. Since then, it has held steadily at 4.38 somoni.

When the world economy catches a cold, Tajikistan gets double pneumonia. Repatriated earnings by its emigrant workers, which normally account for half of national income, shrank by nearly a third in 2009 as Tajikistanis in Russia lost their jobs in the regional recession. The relative demand for legal exports from this small once-open economy has been shriveling for years. The GDP share of merchandise exports fell from virtually 100% in 2000 to less than 20% in 2008, according to the World Bank. Despite this trend, national income had grown steadily since 2000, with an average annual gain of about 2.5 billion current international dollars, due partly to migrant earnings in a regional boom. Suddenly, in 2009, Tajikistanis had to tighten their belts. Their rate of income growth halved, according to the National Bank.

The loss of funds put Tajikistani banks in double jeopardy: Households and firms withdrew their money from the banks just as foreign creditors to the banks were demanding payment. Like banks in Kazakhstan, those in Tajikistan had relied heavily on foreign currencies. In 2007, these comprised four fifths of the banks’ deposits and two thirds of their loans.

The central bank’s sudden return to a flexible exchange rate added to bankers’ woes. In a free currency-market, the somoni was heading into a free fall. This would eviscerate the dollar value of bank assets expressed in somoni. And unhedged domestic borrowers of dollars and euros would have trouble repaying bank loans, since their wealth was in somoni, too. Bank defaults would become more likely. Since commercial banks were the key creditors in Tajikistan, firms would shelve expansion plans. Without this spending, the recession would deepen.

Collateral damage

In using the exchange rate as a policy tool, the central bank faces a trade-off. Both fixed and flexible exchange rates are risky. Which is the least risky?

To answer such questions, an American economist, Martin Weitzman at Harvard University developed a general analysis that tries to minimize the damages that uncertainty can wreak. An example from the news may help convey his approach.

To generate electricity, most power plants burn oil or coal, emitting residual carbon to the atmosphere. The accumulating carbon seals in terrestrial heat, threatening to create severe weather. To avert this threat, nations seek ways to reduce carbon emissions. One proposal is to compel any polluter to pay a tax per carbon ton emitted. The other proposal would compel him to buy a permit for each carbon ton emitted.

The two proposals may strike you as essentially the same, but they differ in the uncertainty that attaches to each. The carbon tax limits the environmental costs that a polluter would have to incur. But it doesn’t limit the volume of carbon emissions, since in principle the polluter can emit all that he wants, as long as he pays the tax. On the other hand, the scheme of carbon permits limits the volume of carbon emissions, since governments could agree to print no more than, say, 100 million permits. But the permit policy doesn’t limit the costs that a polluter would have to incur. If he must have a permit in order to operate, then he will be willing to pay any amount for it, up to his total expected profits.

In short, targeting the “price” of the right to pollute (that’s the tax) limits compliance costs. But it doesn’t mandate a given amount of compliance, so we may not get as much cooperation as we had expected. Targeting instead the volume of pollution (that’s the permit scheme) limits environmental damages. But it doesn’t limit how much that firms would have to pay for the right to wreak these damages, so we may find that containing global warming will cost us more jobs than we had expected.

Your choice between the permit and the tax will depend on what you worry most about – unexpectedly high environmental damages or unexpectedly high compliance costs. Since people seem to worry most about the former – melting ice caps, a flooded London, an Amsterdam in shambles – the permit scheme has won the most political support.

Now let’s return to our conversation about money. The same choice – between a price target and a volume target – faces the central bank of Tajikistan as faced environmental policymakers. In principle, the market for somoni determines the price of the currency (i.e., the exchange rate) as well as its quantity (the number of somoni in the world). If the central bank fixes either the price or the quantity of the somoni, then it must leave free the other variable for the market to determine, or no market can exist. (Yes, the bank can try to fix both price and supply. But it cannot predict changes in demand and supply perfectly, so its adjustments will sometimes err. In other words, the bank must live with some sort of market for somoni, whether it wants to or not. It can’t control speculators around the world.) Which variable should the bank fix? That depends on the type of uncertainty that worries it the most – uncertainty in the demand for somoni, or in their supply, suggest Chang and Velasco.

Suppose, for example, that the central bank worries mostly that the demand for somoni will collapse, perhaps because of a world depression. Then the bank may prefer to fix the exchange rate. Fixing the supply of somoni instead may prove to be a costly error, since the exchange rate may have to fall drastically in order to clear the market after an unexpectedly severe decline in demand.

On the other hand, suppose that the central bank is more vexed by the prospect that the commercial banks will call in loans of somoni, reducing their supply in circulation. Then the central bank may prefer to fix the supply of currency, perhaps by creating somoni to replace those that the private banks retire from circulation. The exchange rate would be left free to change until it clears the market. If the bank fixes this rate rather than supply, then it may have to live with disquieting changes in supply, thanks to arbitrage, until the market’s rate of exchange equaled the bank’s.

What now?

What should the central bank of Tajikistan worry about? A small economy that relies heavily on world trade, like Tajikistan’s, would benefit the most from a fixed exchange rate, since it may be buffeted by large and hard-to-predict changes in global demand for its exports, noted Kettell. The lack of diversity in Tajikistan’s exports strengthens the case for fixing the exchange rate, since this sameness magnifies potential changes in the demand for somoni. With respect to legal goods and materials, Tajikistan exports mainly cotton and aluminum, the export revenues of which have declined by 60 to 70 percent since last year. The undeveloped, inflexible nature of Tajikistan’s financial sector also intensifies the potential impact of changes in somoni demand, since few local opportunities for hedges exist. In sum, the central bank should not fix the supply of somoni, because it can’t be sure that people will want them.

Nevertheless, the central bank in 2009 adopted a flexible exchange rate. Why? Because it was running out of dollars with which to defend the somoni. The central bank would have needed to spend $234 million per month to defend the somoni’s high old rate of exchange, said the bank chairman, Sharif Rahimzoda. The country's total reserve of foreign currency was only $198 million in January 2009; it had almost halved in one year. At that rate, defending the somoni for little more than three weeks would have wiped out the entire reserve.

Also, two major trading partners of Tajikistan -- Russia and Kazakhstan -- had already devalued their currencies. Since this increased the relative cost of Tajikistani products, the National Bank was pressured to devalue as well.

In a sense, Tajikistan had to balance economic stability and autonomy. A fixed exchange rate – say, in terms of the U.S. dollar – would have turned over Tajikistan’s monetary policy to the Federal Reserve. That might have satisfied international investors, if not Tajikistanis. Given Tajikistan’s dire straits, a flexible exchange rate was more of an economic gamble. At some point, Tajikistan might have had to appeal to the International Monetary Fund for another stopgap loan; the IMF approved $116 million in April 2009. In exchange, the IMF might require the government to do more to balance its books – and the central bank to be more accurate in its reports.

Thanks to oil, Kazakhstan is not in the same predicament as Tajikistan, but it faces the same dilemma. How much autonomy is the government prepared to give up in order to placate international creditors? Interesting times, indeed. – Zafar Davronov and Leon Taylor

Zafar Davronov is a student in KIMEP’s MBA program. His interests include Central Asian economics and finance.

References

William J. Baumol and Wallace E. Oates. 1988. The theory of environmental policy. Second edition. Cambridge University Press. Applies Weitzman's principle to pollution taxes and permits.

Francesco Caramazza and Jahangir Aziz. 1998. Fixed or flexible? Getting the exchange rate right in the 1990s. International Monetary Fund Economic Issues, number 13.

Roberto Chang and Andres Velasco. 2000. Exchange-rate policy for developing countries. The American Economic Review 90: 2, pages 71-75.

International Monetary Fund. 2008. Republic of Tajikistan: Financial system stability assessment.

Brian Kettell. 2000. What drives currency markets: The signals you need to stay ahead of the game. Financial Times/Prentice Hall.

Martin Weitzman. 1974. Prices vs. quantities. Review of Economic Studies 41:4, October, pages 477-491.

Revised on September 23, 2010.