Cfr - Iraq Currency Regime

1

Should Iraq dollarize, adopt a currency board or let its currency float? A policy analysis1 Nouriel Roubini Stern School of Business New York University and Brad Setser Council on Foreign Relations May 15, 2003

1

The authors’ emails are: [email protected] and [email protected].

2 The United States is reportedly shipping large quantities of small denomination dollar bills to Iraq. It needs to pay Iraq’s civil servants, and small denomination bills are necessary in a country where monthly salaries are often well under $50. Iraqi civil servants are soon to receive an interim payment of $20. 2 News reports also suggest that U.S. troops discovered substantial hidden stashes of hard currency in Baghdad – $600 million in $100 dollar notes in one place. These dollars, if they are real, also could conceivably be put into use in some way. There may be more pressing immediate needs in Iraq than deciding on a new currency and exchange rate regime. At the same time, some decisions cannot be put off for long. Steps taken to solve immediate, pressing problems – like the need to pay those Iraqis who show up for work – will strongly shape the choices that Iraq will subsequently be able to make. Facts on the ground have a way of becoming permanent. Once the interim authority starts paying people in dollars, the use of the dollar as a means of exchange is likely to increase rapidly. The supply of dollar bills – and twenty dollar bills – circulating inside Iraq is set to grow dramatically. If the interim authority starts contracting for services in dollars, the dollar will also become a standard unit of account in financial contracts. U.S. contractors hired to make immediate improvements in Iraq’s infrastructure are being paid by the U.S. government in dollars, and likely will want to sign contracts with their subcontractors that are denominated in dollars. The dollar has been used in Iraq for some time, no doubt, and the dollar circulates along side the local currency in many emerging economies. But informal dollarization has its own costs: it renders monetary policy less effective and makes it more difficult for borrowers to service their hard currency debts when the exchange rate depreciates. Consequently, it is increasingly urgent to start thinking about the right currency regime for a new Iraq, and how decisions made now will – or won’t – limit Iraq’s future policy options. Iraq’s transitional government in principle could decide to dollarize, to euroize, to “Swiss” dinarize, to adopt a currency board, to introduce a new currency that is pegged to the dollar, or to introduce a new currency that floats.3 In the near term, paying people in dollars is likely to be far easier than introducing a new currency. Indeed, Iraq’s interim authority is likely to have the opposite problem of most emerging market governments. Most emerging markets can print their own currency to solve immediate problems, but cannot print dollars. In contrast, Iraq’s transitional government initially will not be able to print the local currency but will have ready access to dollars. It may not be able to print dollars, but it should be able to borrow them from the Fed, using future aid receipts, frozen Iraqi bank accounts or oil revenue as collateral. Yet there is good reason to think that the dollar – or the euro, for that matter – is not an ideal currency for a major oil exporter. Dollarization, euroization or a strict currency board that eliminates any monetary policy autonomy can make sense when the country that dollarizes or adopts the currency board is subject to the same shocks as the anchor currency. Iraq, a major oil exporter, is likely to be subject to different shocks than net oil importers like the United States or the “Euro” zone. In particular, an oil supply shock would help Iraq but hurt oil importers. Dollarization or a rigid currency board also complicates the process of adjusting to volatility in 2

Bob Davis, Cummins and Kerr (2003). See also Slevin (2003). Some commentators, namely Hanke (2003), are in favor of Iraq giving up an independent monetary policy and either euroizing or adopting a strict currency board; the Wall Street Journal editorial board, while generally sympathetic to euroization/dollarization, advocated a tight currency board. 3

3 oil revenues: the real depreciation needed in the face of a negative oil price shock would require a – often painful – fall in domestic wages and prices since the “dinar/ dollar” or “dinar/euro” exchange rate could not change. Iraq’s government, like the governments of most oil states, is likely to be more dependent on oil than the overall economy, and a rigid currency regime would make it more, not less, difficult for the government to maintain its fiscal stability in the face of volatility in oil revenues. Even informal or “liability” dollarization – using the dollar to denominate local bank deposits and local financial contracts – would carry important costs, because it complicates the real exchange rate adjustment needed in the face of major oil price shocks. Consequently, a core policy objective should be to avoid the temptation to use dollars to solve all immediate problems and to take steps that preserve the option of eventually moving to a freely floating currency. This likely requires two things: •

•

Moving relatively quickly to introduce a new currency. In the very near term, there may be no choice but to use dollars to make some payments. But as soon as possible, the government should start to make payments in a local currency. The new currency initially could be loosely linked to the dollar to establish its underlying value. But it is not necessary to have a formal currency board to assure its stability.4 So long as the government avoids excessive monetary creation and government spending does not exceed the revenues that the government obtains from donor grants and oil sales, the new currency should retain its value. The physical circulation of a new currency and the ability to pay salaries in this currency would help to prevent the widespread use of the dollar as a means of settling domestic bills.5 Limiting the use of the dollar in financial contracts. To the extent possible contracts for local goods and services should be denominated in an Iraqi currency rather than dollars. One of the lessons to be learned from recent crises in emerging market economies is that informal dollarization can make it much more difficult to respond to external shocks.

It needs to be noted that the absence of full dollarization or a currency board does not imply that Iraq’s currency will float freely, without any government intervention. A managed float is more likely. Most emerging markets do intervene in the currency markets on occasion. Moreover, the Iraqi government almost necessarily will be the largest player in the foreign exchange market, as a large share of the revenue from Iraq’s oil exports will be intermediated through the 4

In a currency board, every “dinar” in circulation is fully backed by a “dollar” or other reserve asset. The need for full reserve backing completely constrains monetary autonomy. 5 It is possible for the symbols and faces on the new currency to evolve over time without a major currency reform. Old notes can be withdrawn from circulation and replaced with new notes. The easiest way to introduce the new currency is to start using it to pay government salaries. But there will also be a need to replace the existing “Saddam” dinars still in circulation through a currency exchange. Right now so-called “Swiss” dinars circulate in the Kurdish areas in the north and “Saddam” dinars circulate in the areas formerly under the control of the regime. Ideally, these two different currencies would be unified, and a new dinar would replace both the “Saddam” and the “Swiss” dinar. But, for a while, one could also continue using the “Swiss” or “Saddam” dinar issuing new amounts of either one as needed to satisfy the demand for currency as economic activity recovers. In practice, since the Saddam dinar is discredited for political reasons (i.e. it is undesirable to print new notes with the face of “Saddam” on it) and its value is way down and swinging sharply, issuing newly printed “Swiss” dinars may do the job while a new credible currency is being designed and created.

4 government. If the government draws on its hard currency savings to cushion the impact of low oil prices, the net effect will be to supply hard currency to the foreign exchange market. This analysis will focus on the economic arguments associated with various possible currency arrangements. A common currency is an important symbol of national sovereignty and is part of the political glue that helps to hold many countries together. However, the political value of a currency is not central to this analysis. Rather, this analysis will focus on the reasons why the long-term use of the dollar – or the adoption of a strict currency board – would complicate the economic management of a major oil exporter. Managing oil dependence With the world’s second largest proven oil reserves and a relatively small population (though not as small as many Gulf states), Iraq’s future economy will be dominated by the production of oil and the consumption of oil rents. Over time, its oil production is likely to rise – according to energy analysts – from 2.5 million barrels a day to around 6 million barrels a day, only a bit less than Saudi Arabia. Consequently, oil exports will account for the lion’s share – probably well over 75% – of Iraq’s exports and a very significant share of Iraq’s GDP. Petroleum accounts for 90% of Saudi Arabia’s exports and almost half of Saudi Arabia’s GDP. Petroluem accounts for 80% of Venezuela’s exports and about a third of its GDP. Russia’s economy is much more diversified than Iraq, and oil and gas still account for over 50% of Russia’s exports and a fifth of Russia’s GDP.6 The core economic policy challenge for an “oil state” like Iraq is managing the risks associated with heavy dependence on a single commodity whose world market price has been quite volatile. While there is a range of possible currency regimes, it makes sense to focus on three in particular. •

•

•

6

Dollarization. “Dollarization” is the outright adoption of another country’s currency as the basic unit of account. The currency that is actually selected need not be the dollar. It would also be possible to “Euroize.” Conceptually, Iraq could also adopt the currency of a net oil exporter, though there is not an obvious candidate. Venezuelan bolivarization is not an attractive option. A currency board. In a currency board, the national currency is backed one for one with reserves, so that for every “dinar” in circulation, the central bank holds an equal amount of reserves. A currency board makes the adoption of the dollar – or euro – more politically palatable by putting an Iraqi face on the currency. Economically, however, it is similar to dollarization, as the government gives up the ability to devalue or revalue the currency and instead promises to automatically tighten monetary policy in the face of pressure to devalue. A managed float. The currency’s value relative to major currencies is set in the market, with occasional intervention to smooth excessive currency movements. The government does not commit to maintain a given exchange rate. Many important commodity

Weafer (2003) provides data on Russia’s oil production. Additional data from www.cia.gov/cia/publications/factbook/geos.

5 exporters – including Canada, Australia, South Africa and Russia – have currencies that float against the dollar. In principle, a “soft” peg – a peg that lacks the institutional backing of a currency board and that can be adjusted – provides a fourth option. However, recent experience in emerging economies had demonstrated the weaknesses of such soft pegs. These weaknesses are large enough that a soft peg does not offer, in our judgment, a viable long-range option – though a soft peg could be part of a transitional arrangement. In our view, a managed float offers three significant advantages over a more rigid currency board or outright dollarization.7 • • •

The monetary policy autonomy made possible by floating would help Iraq adjust to an oil supply shock, which would create divergent economic conditions in oil exporting and oil importing countries. Floating facilitates the real adjustment that can be needed in the face of oil price volatility. It is easier for the exchange rate to adjust than for domestic prices to adjust. Floating helps the government manage the mismatch between its volatile oil revenues and its fixed domestic costs, and thus can help assure the government’s fiscal stability.

Responding to oil supply shocks Dollarization or a dollar-based currency board means adopting the monetary policy of the United States. Euroization or a euro-based currency board means adopting the monetary policy of the European Central Bank. Economic theory indicates that if the shocks hitting a common currency area are common to all the economies in the area, rather than being idiosyncratic national shocks, the monetary policy of the anchor country is likely to be appropriate for the “dollarized” economy as well. The degree of synchronization of the business cycle, in turn depends on factors such as the degree of trade integration and the similarity in production structures. Two countries that produce the same good are more likely to be exposed to similar shocks, as are two countries that trade extensively with each other. Since Iraq will export basically one commodity, oil prices and oil supply/demand cycles will determine the degree of synchronization of the business cycle between Iraq and the home country of its anchor currency far more than the degree of trade integration. The viable anchor currencies for Iraq – the dollar and the euro – are both currencies of net oil importers. Is the monetary policy of an oil importer likely to correspond with the monetary policy an oil exporter needs? The basic answer is yes during oil demand shocks, but no during oil supply shocks. If strong global demand for oil is pushing oil prices up, economic conditions in an oil exporting country are likely to coincide with economic conditions in major oil importing countries. A recession in either the U.S. or Europe similarly will push oil demand and oil prices lower, given the size of both economic blocks. However, the opposite is true if a supply disruption leads to a spike in the price of oil. A disruption in supply outside of Iraq will increase Iraq’s real income while slowing real growth in the U.S. or Europe. Booming Iraq risks 7

See Roubini (2001) for a more detailed analysis of the criteria to assess whether a country is ready for dollarization.

6 overheating and needs tight money, while slumping oil importers need loose money. Of course, protracted oil supply shocks eventually tend to produce a global slowdown, a fall in demand for oil and ultimately a fall in the oil price that brings the economic cycles of oil exporters and oil importers back into synch, but this convergence occurs slowly. It seems likely that “geopolitical” oil supply shocks will continue to buffet the global economy as they have over the past thirty years (1973-1974 Yom Kippur War; 1979 Iranian Revolution; 1990-1991 Iraqi invasion of Kuwait; 2003 Iraq and Venezuela). Consequently, one cost of dollarization or euroization is that it would lock Iraq into a pro-cyclical monetary policy during oil supply shocks. 8 It is worth noting that the ability of many emerging market economies to pursue counter-cyclical monetary policies – the key theoretical benefit of floating – remains a subject of debate. Studies have suggested that some small open economies with a history of high inflation and high exchange rate volatility often are unable to use monetary policy in a counter-cyclical way. A combination of informal dollarization, lack of policy credibility and wage indexation may render monetary policy ineffective. Worse, it has been argued that in some countries monetary policy may be pro-cyclical rather than counter-cyclical. Negative external shocks – be it an oil price shock and a reduction in the availability of international financing for emerging economies – may force monetary authorities to tighten monetary policy during a recession. The gains from exchange rate flexibility would be reduced and dollarization or a currency board would become a more attractive option if Iraq proves unable to use monetary policy to provide counter-cyclical output stabilization. But there is no reason to believe that the new Iraqi government should be fundamentally incapable of creating the credible economic institutions needed to pursue a counter-cyclical monetary policy. Facilitating adjustment to oil prices shocks (terms of trade shocks) Countries whose exports are concentrated in a narrow range of goods, often primary commodities like oil, are vulnerable to terms of trade shocks. If the country is small and a price taker in export and import markets, no currency regime can protect the country from this structural vulnerability. If the price of the country’s primary export – say oil – falls, and the global price of the basket of goods the country wants to import is fixed, the country will be able to buy fewer imports. If Iraq can buy four bushels of wheat for every barrel of oil it exports when oil is at $20 a barrel, Iraq can only buy two bushels of wheat when oil is at $10 a barrel. No matter how you cut it, falling oil prices reduce Iraq’s real income. The key question for an economy exposed to large terms of trade shocks is whether a given currency regime – dollarization, a currency board or a managed float – can help the country adjust to unavoidable terms of trade shocks. A country can adjust to a fall in its terms of trade in many different ways. First, it can sell more of its “exported” good (oil) at home, as foreign demand declines. However, in the near term, 8

This analysis assumes that an oil exporter such as Iraq is small enough to be a price taker in global markets. This may not be exactly true as Iraq has the second largest known reserves of oil and Iraq, like Saudi Arabia, could potentially use its oil production to help shape global oil prices and indeed to try to dampen volatility in oil prices. But if Iraq were to strategically use its market power to affect oil prices (either up or down) the correlation between Iraq’s business cycle and the business cycles of major oil importers would be even further reduced.

7 demand for oil in Iraq is likely to be fixed, so this is not a practical strategy. In any case, many oil producers keep domestic prices below the world market price. Second, the level of oil production could adjust. Classic economic theory suggests that a fall in the relative price of oil would tend to discourage its production. This is no doubt true at a global level. But it may not be true for a low cost oil producer with substantial market power. A low cost producer like Iraq still can make money at low oil prices – it just earns smaller economic rents on each barrel of oil. It could respond to lower prices by increasing production to try to increase its income. But this strategy only works if other members of the oil cartel don’t pursue the same strategy. Alternatively, a member of a cartel of oil exporters could respond to a fall in the price of oil by cutting production to try to stabilize the price.9 Third, a country can try to limit its need to adjust to a temporary shock in terms of trade by borrowing or drawing down reserves to allow it to continue to import at the higher level. Fourth, an increase in the relative price of imports – in the example above, a bushel of wheat costs ½ a barrel of oil when the oil prices falls instead of ¼ of a barrel of oil – can lead to a shift in demand toward domestic goods, both domestic goods that compete with imports and non-traded goods. It is worth walking through the economic processes that bring about this adjustment in relative prices in different currency regimes. A fall in the price of oil will reduce the export earnings and the dollar GDP of a country like Iraq. Since the price of a small country’s imports is set in the global market, the “dollar” price of imports does not change following a fall in the country’s oil income. The country either has to import less, or use more of its income to buy the same quantity of imports. In a dollarized economy, a fall in the country’s real income leads to pressure on the domestic price of “non-tradeables.” The local barber realizes that the local oilman now has to spend more of his oil revenue buying imported wheat, and has less to spend on a haircut. The oilman can either buy fewer haircuts, or the local barber can reduce his prices. This implies an absolute fall in local wages and prices – deflation – as nominal and real wages need to fall to maintain full employment in the face of the real income shock. The fall in local prices needed to adjust to a fall in oil prices, however, may trigger unemployment and otherwise reduce real income and output more than necessary if there are rigidities in the economy. These rigidities could take the form of sticky nominal wages, sticky domestic prices or the costs of transferring sector specific factors such as labor and capital from the traded sector to the non-traded sector (following the fall in the relative price of traded goods).10 The simplest case that leads to avoidable unemployment is the case of rigid nominal wages. As the international price of the traded exportable good falls, nominal wages should fall in the export sector to avoid a fall in demand for labor that would create unemployment. Otherwise, the fall in the international price of the exported good will reduce demand for labor as nominal stickiness increases real wages in the export sector. Similarly, if nominal wages are sticky, real wages in the non-traded sector would also remain too high to sustain full employment, since full employment requires a fall in real wages following a shock to the 9

See Krugman (1999) for a model of the oil market based on earlier work by Cremer and Salehi-Isfahani that allows for “perverse supply responses” and multiple equilibriums. 10 Traded good sector could be defined to include those employed “spending” oil rents as well as those directly employed in the production of oil. In many Gulf states where production costs are low, relatively few people are directly employed in the production of oil, while many more are employed “spending” the country’s oil rents. Often these oil rents are captured by the state, and maybe state employees whose salaries are paid for by the profits from oil are effectively employed in the “tradeable” or oil sector.

8 country’s real income. Without adjustment in wages, the fall in labor demand in both sectors causes unemployment. The adjustment process can occur somewhat differently if the currency is allowed to float. A fall in the price of oil will tend to lead the currency to depreciate, and the currency depreciation will lead automatically to a reduction of the real wage in the traded exportable sector and the nontraded sector so long as nominal wages stay constant.11 The country’s real income still falls in the face of an adverse terms of trade shock: a barrel of oil still buys less wheat. But the fall in the value of a dinar both increases the price of imported goods and reduces domestic real wages, making domestic non-traded goods more competitive with imports and supporting employment in the export sector. Consequently, a nominal depreciation of the currency can help to avoid unemployment if domestic wages are sticky. Nominal exchange rate flexibility can also prevent the emergence of unemployment if there are other domestic price rigidities or difficulties transferring economic resources from one sector to the other. 12 Empirically, there is little doubt that diversified economies that are also large commodity exporters such as Canada, Australia, New Zealand, South Africa and more recently Chile and Russia have successfully used a flexible currency to adjust to external terms of trade shocks. When commodity prices fall (either due to a fall in global demand or a glut in supply) the currencies of these countries tend to depreciate, helping to absorb the effect of the shock on the domestic economy. These economies produce many of the goods that they import, so they can respond to a fall in the world price of their commodity exports by, generally speaking, buying more domestic goods and fewer imports. Several of these countries – notably Mexico and Canada – are quite closely integrated into the U.S. economy and thus score well on the traditional criteria for dollarization, yet they have opted for a floating currency. Indeed, it is the small Central American countries (El Salvador, Panama) that lack natural resource wealth and the associated vulnerability to terms of trade shock that have tended to adopt the dollar. It is true that there are fewer opportunities for a shift in relative prices to lead to a substitution of domestic production for imports in less diversified economies. However, even undiversified economies could still benefit from exchange rate flexibility if it facilitates the needed adjustment between the relative prices of imports and locally produced non-tradeable goods and more generally limits the impact of wage and price rigidities. This may be the reason why most small oil producers with little market power have not dollarized, euroized or adopted currency boards. For example, Norway has remained outside the Eurozone. Only Ecuador, among oil exporters, has dollarized and the jury on this economic experiment is still out. A managed float would differentiate Iraq from a number of other regional oil states, since most major oil exporters in the Middle East – notably Saudi Arabia, Kuwait, Libya and the Emirates – have pegged their currency to a basket of other major currencies rather than adopted a floating or

11

Import prices necessarily increase, both relative to exports and relative to local production, so constant nominal wages implies a falling real wage. 12 Broda (2001) found that flexible exchange rates were associated with immediate large real depreciations and smaller output losses than fixed exchange rate regimes in the face of terms of trade shocks in a sample of emerging market economies.

9 a managed float.13 But there are good reasons why Iraq should not follow the example of other oil states in the region. Libya is a command economy. The economy of Saudi Arabia has been stagnant, with significant unemployment. Saudi Arabia is rumored to have borrowed significantly and perhaps unsustainably during periods of low oil prices. It may face financial difficulties should oil prices fall after the windfall created by Iraq’s withdrawal from the world market comes to an end.14 Iran, which has indicated that it plans to move toward a managed float, may provide a more useful model. Helping the government manage its fiscal dependence on oil In most oil states, the government is even more dependent on oil revenue than the overall economy. The government usually finances itself by capturing a large share of the gap between production costs and the world market price, and government spending is the primary means for distributing the country’s oil rent to the country’s population. Since oil often accounts for 75% or more of the revenue of an oil state’s government, the volatility in the government’s revenue is larger than the volatility in overall economic activity associated with oil price shocks. 15 The government’s exposure to terms of trade shocks can create fiscal problems since government spending typically is not as volatile as oil revenue. 16 The army, police, judiciary and health service all need to be paid when oil prices are low. Government employees are typically paid fixed salaries, not a salary linked to the price of oil. Government investment in roads, ports, and other infrastructure can be reduced when government revenues fall, but such pro-cyclical adjustment in capital spending tends to augment the downturn associated with an oil price shock. In more technical terms, there is a mismatch between an oil state’s highly volatile revenue stream from oil, and its need to cover the stable fixed costs associated with operating a government. There are lots of ways of trying to manage this mismatch. The government can build up fiscal reserves, though in practice this requires resisting democratic pressure to increase spending during good times.17 The government can borrow to sustain spending in “bad times.” The government could engage in various hedging strategies – long-term sales contracts, forward sales – that would limit the volatility in its oil revenues. The government could, in theory, index government salaries to the price of oil.

13

It is worth noting that pegging to a basket of currencies does allow for some exchange rate adjustment since a basket peg will depreciate against the dollar and appreciate against the euro when the dollar rises against the euro. 14 Saudi Arabia increased production during the Gulf War to offset the fall in Iraq’s production. More generally, “geopolitical” risks have contributed to recent high oil prices during a time of relatively weak global demand, providing a windfall to the Saudis. 15 Oil accounted for 90% of the Government of Angola’s revenue, over 75% of the revenue of the governments of Oman, Nigeria, Saudi Arabia, Qatar, Algeria and Yemen, and over 60% of the revenue of the governments of the United Arab Emirates, Bahrain, Iran, Gabon, Libya and Kuwait. See Daniel (2001). Barnett and Ossowski, (2002) offer a more detailed analysis of the operational aspects of fiscal policy in oil-producing countries. They note that for every $1 change in the world price of oil the revenue of Venezuela’s public sector is reduced by roughly 1% of Venezuela’s GDP; public sector revenues from oil fell from 27% of GDP in 1996 to 12.5% of GDP in 1998. 16 Oil can be produced in some Gulf states for less than $5 a barrel and the gap between the low production cost and the world market price produces large economic “rents.” 17 See Fasano (2000) for a review of the mixed experience with oil stabilization funds.

10 Letting the exchange rate float also helps the government manage this mismatch. The government of an oil state wants the real price of the non-traded goods it buys to fall when the real value of its oil export revenues falls. This tends to happen if the country has a flexible exchange rate regime. In the face of a negative oil price shock, the nominal and real exchange rate will tend to depreciate. This tends to improve the fiscal balance of a government, as a quick real depreciation allows the government to buy more domestic non-traded goods and services for each barrel of oil the country exports even as the external purchasing power of each barrel of oil falls. For example, if the government’s revenues from oil fall from $20 a barrel to $10 and the dinar also falls from 1 to 2, the government’s dinar earnings stay constant. One barrel of oil buys 20 dinars before and after the oil price falls. This makes it easier for the government to sustain fixed domestic salaries. The government earns fewer dollars per barrel of oil, but each dollar can buy more at home. Of course, constant salaries in dinar and higher import prices make it harder for a government employee with a fixed dinar salary to afford imported goods. There is still a need for a real adjustment. The needed adjustments occur with more difficulty in a country that has dollarized or adopted a close proxy for dollarization such as a currency board. As the government’s dollar revenues from oil fall, it must reduce its spending. That means fewer policemen and hospital workers if police and hospital salaries are fixed, or lower salaries and the same number of workers. If salaries fall, the fall in domestic prices provides the same real adjustment (a fall in the price of non-traded goods) that would happen with a floating exchange rate regime, but the adjustment comes about through the fall in nominal salaries rather than the fall in the nominal exchange rate. Adjusting nominal salaries takes longer than letting the nominal exchange fall. Indeed, the difficulties associated with adjusting expenditure and cutting salaries is one reason why many oil states have encountered fiscal difficulties. In practice, it may not be optimal for the government to adjust fully to a temporary oil price shock. Building up an oil stabilization fund to act as a buffer against temporary oil shocks can make more sense than constant adjustments. When oil prices are temporarily high the government should run surpluses and accumulate assets; when oil prices are temporarily low the government should run deficits and run down its assets. This helps insulate both the government and the overall economy from volatility in the price of oil. However, oil stabilization funds have been hard for many countries to implement. There is a temptation to fail to set enough aside in good times, leaving too small a cushion against bad times. Alternatively, a government could also use its capacity to borrow internationally to help smooth out volatility in oil revenues. But this too has its practical difficulties. There is a tendency to borrow in bad times but not to pay down debt in good times, which eventually leads to debt problems. Moreover, an oil exporters’ access to financing tends to be correlated with the price of oil. When prices are high, financing is available at a low price. When oil prices are low, financing is both less available and more expensive. It may not be possible to rely entirely on fiscal reserves and government borrowing to smooth oil price shocks. Consequently, exchange rate flexibility may provide a useful and practical way to help the government manage its own exposure to oil price volatility. Potential objections to a floating Iraqi currency

11 There are a number of potential objections to the idea of a floating national currency for Iraq that are worth exploring. First, some argue that a country with a poor record of policy credibility needs a fixed exchange rate to discipline monetary and fiscal policy and to assure that the currency serves as a stable store of value. It is true that the new transitional government will not inherit monetary policy credibility. But it also will not inherit Saddam’s record of monetary mismanagement. Given the clear break in regime and policies, Iraq’s new institutions will start with a clean slate. The past history of the country should matter less, and the record of the new government will matter more. There is no reason that the creation of an independent central bank and a sound fiscal policy could not provide Iraq with a chance to earn policy credibility quickly. Indeed, Iraq’s new institutions potentially start with two important advantages. First, Iraq should inherit significant reserves, whether from frozen Iraqi assets, from funds recovered from Saddam’s personal accounts or from funds in the UN’s Oil for Food escrow account.18 Second, Iraq also will start with a relatively low level of liability dollarization (the use of the dollar rather than the local currency in local financial contracts). While the use of the dollar could increase rapidly if the U.S. or the transitional government use the dollar to pay salaries for an extended period of time, Iraq’s recent economic isolation seems to have limited the use of the dollar as a basic means of payments, store of value and unit of account by ordinary Iraqis.19 A low level of liability dollarization makes it easier to float, as the stimulus to domestic activity from a real depreciation is not partially offset by an increase in the real burden of dollar denominated debts. There is nothing in principle that prevents a national currency and a managed float to succeed if appropriate monetary and fiscal policies are in place. And if such policies are not in place, dollarization or a currency board cannot provide monetary or fiscal policy discipline per se. Argentina demonstrates that a rigid currency regime does not guarantee fiscal discipline, and how the fiscal and financial difficulties can lead to the creation of quasi-currencies that eventually undermine monetary discipline as well. Second, some argue that Ecuador’s experience indicates that dollarization can succeed even in countries that fail to meet the traditional criteria for dollarization, including business cycle correlation, labor market flexibility and a history of sound fiscal policy. They would also point out that Ecuador is an oil exporter, just like Iraq. However, in our judgment, the experience of Ecuador neither strengthens the case for dollarization nor suggests that oil exporters are optimal “dollarizers.” For one thing, it is somewhat premature to declare Ecuador a success. Its fiscal and debt problems are still severe, the sovereign spread of 14% suggests a continued high risk of default, the banking system is severely undercapitalized and it is still vulnerable to a run. Moreover, Ecuador’s dollarization has not been really tested by a major oil price shock – one analogous to the 1998 oil shock that contributed to Ecuador’s 1999 crisis. Oil prices generally have remained high following Ecuador’s decision to dollarize. Ecuador also dollarized at a very undervalued exchange rate, which provided a large initial “competitiveness buffer” that has

18

Iraq will also inherit substantial debts, which will need to be renegotiated. However, in the near term, there is no need to draw on Iraq’s existing assets to resume payments on its debts. 19 Regime insiders obviously had large hoards of foreign currency.

12 mitigated the impact of subsequent real appreciation. Finally, the country is still on IMF life support and it is likely to remain on such intensive care for a long time. It is not far fetched to think that one day Ecuador may de-dollarize if it is unable to clean its fiscal house. A creeping de-dollarization could occur if Ecuador need to raise money at a time when it has lost access to domestic financing. It might then start issuing a quasi-money zerointerest bonded debt – something like Argentina’s Patacones or more appropriately Sucretones – to pay public employee salaries and to finance other spending. If issuance of such Sucretones were to become widespread, an new currency trading below par with the dollar will be created. Eventually creeping de-dollarization will turn into full-scale formal de-dollarization. It is also quite easy to see how the need to issue Sucretones would be correlated with an oil price shock that both reduced government revenue and required domestic deflation.20 Third, many “post-conflict” countries have either adopted a currency board or dollarized/ euroized, often with some success. Bosnia and Kosovo adopted currency boards, East Timor dollarized and Montenegro euroized. Iraq does not superficially look very different from such post-conflict entities. However, the superficial similarities mask substantial differences. Above all else, these countries/entities are not major oil exporters and our analysis suggests that it is not optimal for major oil exporters to dollarize. But there are other differences as well. Most postconflict states emerged following the break-up of larger national states and lacked many of the basic institutions of a nation-state, including a national currency of their own or a central bank. Many of these economies hope to join the European Union and eventually EMU, and expect to exit from the currency board by joining the Euro. In contrast, it has been difficult to successfully exit from a currency board to a float. It also should be noted that many of these countries remain on the semi-permanent dole of the international community, and rely on the sustained transfer of large amounts of aid to maintain a minimum standard of living. Iraq should be different. It should not need permanent aid or transfers from abroad given its future oil wealth. It is not a rump state that is emerging from the breakup of a bigger state. It has had its own national currency (however demeaned). It is not liability dollarized – at least not yet. It is not on its way to becoming a member of the EU – or the 51st state of the United States of America. There is no reason to believe that all post-conflict countries should dollarize or adopt a currency board. The example of Serbia is instructive. It too is a post-conflict state and like Iraq, it also inherited significant debts. But it got substantial – though not total – debt relief, and it now has its own stable currency. It has been phasing out its capital controls and generally flexibilizing its exchange rate regime. Finally, some may argue that the absence of a banking system, let alone a well functioning financial system, limits effective capital mobility and this makes exchange rate flexibility rather meaningless. Money changers on the street are not quite the same as a real foreign exchange market. But this argument strengthens the case for a national currency and an eventual managed float. If Iraq maintains fiscal discipline and avoids excessive and inflationary monetary creation (i.e. abusive seignorage taxation), it should be able to maintain a stable currency. If, as seems likely, Iraq finances new investment in infrastructure and the rehabilitation of its oil production 20

To date, Ecuador’s domestic prices have increased more rapidly than U.S. prices, resulting in continued real appreciation.

13 capacity with some combination of current oil revenue, grant aid and borrowing from the official sector, it will have little immediate need for large inflows of private capital. Early pressures on the currency should be limited. As the institutions needed for effective capital mobility develop, a managed float could be able to absorb domestic and external shocks without being subject to excessive volatility. Conclusions Ultimately, the Iraqi government itself must determine what currency regime would best serve Iraq’s interest. Our analysis suggests that Iraq should keep a national currency rather than euroize/dollarize or lock itself into the straitjacket of a currency board. In the face of an oil supply shock, an oil exporter would want a different monetary policy than an oil importer. A floating exchange rate also makes it easier to adjust to negative terms of trade shocks, as the needed real exchange rate adjustment could come about more easily through a fall in the nominal exchange rate than through a fall in domestic wages and prices. The introduction of a new currency might be easier if the new currency is temporarily pegged to the dollar, though this is not strictly speaking necessary. But the goal should be to move quickly to a managed float and to ensure monetary stability with sound fiscal management and some form of inflation targeting. In the short-run, the dollar offers a convenient solution to many of the immediate problems that the transitional government is likely to face. However, a core policy objective should be to find ways of solving immediate problems without “locking” Iraq into a suboptimal monetary arrangement, since a managed float exchange rate regime would be preferable to outright dollarization or a currency board over the medium term. The more payments the U.S., other members of the coalition and the interim government make in dollars, the more difficult it will be to “de-dollarize” the domestic economy at a later stage. The more contracts that are signed in dollars now, the greater the chance that the dollar will be used in subsequent financial contracts as well. Introducing a new currency relatively quickly would make it easier to establish this currency, rather than the dollar, as the basic means of exchange. It also makes sense for Iraq not to follow the example of post-conflict countries that have relied on a currency board to assure currency stability. There are ways of providing a sound currency that do not require a permanent commitment to devote monetary policy to the maintenance of a fixed parity with an anchor currency.

14 References: Barnett, Steven A. and Rolando J. Ossowski, (2002) “Operational Aspects of Fiscal Policy in Oil-Producing Countries,” IMF Working Paper No. 02/177. Broda, Christian (2001) “Coping with Terms of Trade Shocks: Pegs versus Floats,” American Economic Review, Volume 91, no 2 (May 2001). Celasun, Oya (2003) “Exchange Rate Regime Considerations in an Oil Economy: The Case of the Islamic Republic of Iran,” IMF Working Paper WP/03/26. Daniel, James (2001) “Hedging Government Oil Price Risk,” IMF Working Paper WP/01/185. Davis, Bob, Chip Cummins and Simeon Kerr (2003) “U.S. Dollars Are Sent to Iraq to Replace Discredited Dinar,” Wall Street Journal, April 16, 2003. Fasano, Ugo (2000) “Review of the Experience with Oil Stabilization and Savings Funds in Selected Countries,” IMF Working Paper No. 00/112. Hanke, Steve (2003) “The euro could help Iraq's economic recovery,” Financial Times, April 17 2003. Krugman, Paul (1999) “The Energy Crisis Revisited,” unpublished, MIT January (http://web.mit.edu/krugman/www/opec.html). Roubini, Nouriel (2001) “Factors to be Considered in Assessing a Country’s Readiness for Dollarization,” Stern School of Business, New York University, November. (http://pages.stern.nyu.edu/~nroubini/dollarization.pdf) Slevin, Peter (2003) “A Few Kinks in the U.S. Money Pipeline,” Washington Post, April 24, 2003. Weafer, Chris (2003) “Too Much Oil Could Be Bad for Russia’s Health,” Financial Times, April 28, 2003.