Going Down With the Dollar:
The Cost to Developing Countries of a Declining Dollar

 By Mark Weisbrot, David Rosnick, and Dean Baker1

September 20, 2004

Executive Summary

In the years since the East Asian financial crisis in 1997, many developing countries have sought to increase their holdings of foreign reserves, as a way to protect their currencies against financial instability. Since most countries hold most of their reserves in dollars, this build-up in reserves has led to a large accumulation of dollar holdings.

These large holdings of dollars could lead to substantial losses in wealth for developing countries. The current account deficit of the United States exceeds $600 billion at an annual rate, or more than 5.7 percent of GDP. This deficit is clearly unsustainable, and will almost certainly result in a large decline in the value of the dollar over the next decade. Such a decline will substantially reduce the value of the dollar reserves held by developing countries.

Based on a sample of reserve holdings among developing countries, this paper shows that the drop in the dollar will lead to a loss in the value of reserve holdings for the average developing country among the group examined of between 1.8 percent and 5.6 percent of GDP.

By comparison, the World Bank recently estimated that the complete liberalization of merchandise trade (both agricultural and non-agricultural) by rich countries, would lead to an increase in annual GDP in poor countries of 0.6 percent. This gain will only be fully realized after a period of 10 to 15 years. The losses that developing countries face due to the expected decline in the value of the dollar, likely exceed the gains from rich country trade liberalization over the next decade.

If the gains from trade liberalization by rich countries are viewed as being potentially important to the well-being of developing countries, then the potential losses due to a decline in the value of dollar reserves should also be an important concern.


Going Down With the Dollar: The Cost to Developing Countries of a Declining Dollar

In the last decade the amount of foreign exchange reserves held by central banks in developing countries has soared. This was in large part a response to the rash of world-wide financial instability that followed the East Asian financial crisis in 1997. The conventional wisdom in the post-crisis world has been that developing countries should hold as much reserves as possible, in order to reassure financial markets about the stability of their currency and their economy. Developing countries now hold an amount of reserves that is on average more than 10 percent of their GDP, and in many cases exceeds 20 percent of GDP2.

While countries generally expect to get a minimal return on money held as reserves, it is usually assumed that a reserve currency will maintain its value reasonably well through time. Unfortunately, this is not likely to be true with the dollar. The dollar is widely recognized to be seriously over-valued. This is apparent from the size of the United States current account deficit, which expanded to $660 billion in the second quarter of 2004, or 5.7 percent of GDP. A deficit of this size is clearly not sustainable3.

The only plausible way in which this deficit can be brought down to a manageable level is through a sharp decline in the value of dollar. The real value of the dollar will probably have to fall between 25 percent and 30 percent, measured against other major currencies, in order to restore the U.S. current account deficit to a sustainable level. While the exact timing of this decline is impossible to predict, it is virtually certain to occur some time in the next decade. Furthermore, the longer the adjustment is delayed, the larger it must eventually be. The United States is continuing to increase foreign debt at a rapid pace. A larger foreign debt implies higher future interest payments, which means that the United States will have to experience a larger adjustment in its trade deficit to bring its current account deficit down to sustainable level.

The prospect of such a sharp decline in the value of the dollar will be a serious blow to developing countries that hold large amounts of dollar reserves. In effect, these countries will lose a substantial amount of wealth as a result of the fact that their dollar reserves will be worth much less than what they paid for them. Table 1 shows the amount of money that a series of developing countries would lose based on their current holdings of reserves. It shows the losses expressed both in current dollars and as a share of GDP, under three different scenarios of dollar decline4.

Table 1

Developing Country Losses Due to a Declining Dollar

Country

Small 

Middle 

Large

 

Small 

Middle 

Large

Billions of dollars

 

Share of GDP

Argentina

$1.7

$2.9

$5.3

 

1.2%

1.9%

3.5%

Chile

$1.5

$2.5

$4.6

 

1.7%

2.8%

5.2%

Colombia

$1.1

$1.9

$3.4

 

1.3%

2.1%

3.8%

El Salvador

$0.2

$0.3

$0.5

 

1.2%

2.0%

3.6%

China,P.R.and Hong Kong

$56.4

$93.1

$171.6

 

3.2%

5.3%

9.8%

Kazakhstan

$0.6

$1.0

$1.9

 

1.9%

3.1%

5.6%

Kyrgyz Republic

$0.0

$0.1

$0.1

 

2.0%

3.3%

6.1%

Malaysia

$5.1

$8.5

$15.6

 

4.6%

7.7%

14.1%

Mexico

$5.8

$9.5

$17.5

 

0.9%

1.5%

2.7%

Peru

$1.0

$1.7

$3.1

 

1.6%

2.6%

4.7%

Philippines

$1.5

$2.5

$4.6

 

1.8%

3.0%

5.6%

South Africa

$2.7

$4.4

$8.1

 

1.5%

2.5%

4.6%

Thailand

$4.1

$6.8

$12.6

 

2.6%

4.2%

7.8%

Tunisia

$0.3

$0.6

$1.0

 

1.2%

1.9%

3.5%

Turkey

$3.2

$5.3

$9.8

 

1.0%

1.7%

3.2%

Source: International Monetary Fund, Time Series Data on International Reserves. Reserves data for China and Mexico were taken from The Economist, August 7, 2004, p82.


The losses implied from a decline in the dollar are substantial in every case. At the high end, countries with large reserve holdings, like China and Malaysia, could lose an amount equal to 10 percent of their GDP in the event of a sharp decline in the value of the dollar. In the middle scenario, both of these countries would still lose an amount that is larger than 5 percent of their GDP. The least serious loss shown in the table (for Mexico in the low dollar decline scenario) is still 0.9 percent of GDP. This would correspond to a loss to Mexico of $5.8 billion. The un-weighted average developing country losses from a declining dollar, expressed as a share of GDP, are 1.8 percent, 3.0 percent, and 5.6 percent, in the low, middle, and large decline scenarios, respectively.

For purposes of comparison, a recent World Bank study estimated that the static gains to the developing world from complete rich country liberalization of all merchandise trade (this includes trade in agricultural goods as well as manufacturing goods) as being equal to 0.6 percent of GDP5. While the gain from trade liberalization would be an annual return - as opposed to a one-time loss - countries would only realize this full 0.6 percent gain after a period of time. Typically, models assume that it takes 10 to 15 years to realize the full gains, as tariff and quota barriers are gradually phased out and industries respond to the new opportunities presented by liberalization.

Figure 1

            Source: Authors' calculations, see text.

Figure 1 shows the cumulative gain over a ten-year period to developing countries based on the World Bank model and the one-time loss to developing countries based on their average holdings of dollar reserves in the three scenarios shown in table16. As can be seen, the loss of 3.0 percent of GDP in the middle scenario is almost as large as the projected gain from rich country trade liberalization. The loss of 5.6 percent of GDP in the large dollar decline scenario is 1.7 times as large as the projected gains from trade liberalization. Even in the small dollar decline scenario the loss is more than half of the projected gains from rich country trade liberalization.

In short, even in the most optimistic case, developing countries stand to lose more than half as much money from the loss in value of their dollar reserves as they could potentially gain over the next decade from rich countries removing their subsidies and barriers to the exports of developing countries. In the more pessimistic scenario, they will lose 70 percent more from the loss in value of their dollar reserves as they stand to gain from trade liberalization by developed countries.

This comparison is striking because such a loss is so easily preventable. If developing countries simply traded the bulk of their dollar reserves for a currency more likely to maintain its value, such as the euro or the Japanese yen, they could insulate themselves from the effects of a falling dollar. While trade negotiations involve a long arduous process requiring tradeoffs and compromises, dollar reserves can be sold off in a matter of hours. The prospect of large losses on dollar reserves would seem to make the prompt disposal of these reserves a top priority for the financial authorities in developing countries. The decision of developing countries to hold large amounts of dollar reserves in a situation in which the dollar is virtually certain to decline will needlessly threaten their financial health, and risk imposing large costs on hundreds of millions of poor people throughout the world.


Footnotes:

1. Mark Weisbrot and Dean Baker are co-directors of the Center for Economic and Policy Research. David Rosnick is a research associate at the Center for Economic and Policy Research.

2. A second motive for holding reserves is to keep a country's currency down relative to the dollar. Some countries, most notably China, have been rapidly buying up dollars in order to maintain their currency at a below-market exchange rate.

3. A deficit of this size is explosive and very quickly would push the U.S. net international investment position (effectively the foreign debt) to implausible levels when measured as a share of GDP (see Baker, D. 2004, "Dangerous Trends: The Growth of Debt in the U.S. Economy," Center for Economic and Policy Research, [http://www.cepr.net/publications/debt_trends.htm]).

4. The low, middle, and high loss scenarios assume declines in the dollar of 13.8 percent, 22.8 percent, and 42.0 percent, respectively. The basis for these calculations can be found in Baker and Weisbrot, 2004. "Fool's Gold: Projections of the U.S. Import Market." Center for Economic and Policy Research, [http://www.cepr.net/Import_Projections.htm]. These scenarios understate the extent to which the dollar will decline since they worked from the trade deficit data available at the end of 2003. At that time, the U.S. trade deficit was equal to 4.5 percent of GDP. By the second quarter of 2004 it had risen to 5.1 percent of GDP. It is assumed that dollars comprise 69.1 percent of all reserves, which was the average ratio at the end of 2003 (Bank for International Settlements 2004, 74th Annual Report, p 89.)

5. World Bank, 2002. Global Economic Prospects and the Developing Countries 2002. Washington, D.C.: World Bank, table 6.1.

6. For purposes of calculating the gains from trade liberalization it is assumed that the benefits are phased in evenly so that in year 1 GDP is 0.06 percent higher due to trade liberalization, in year 2 it is 0.12 percent higher, in year 3 it is 0.18 percent higher etc. It is assumed that the full gains are realized by the tenth year. Thus the projected cumulative gain over a ten-year period is 3.3 percent of GDP as shown in Figure 1.