This Background Paper on the United States examines the effect of fiscal deficit reduction in the context of the IMF’s multicountry simulation model, on the current account and the real exchange rate. The simulations suggest that, other things being equal, fiscal consolidation will tend to cause the real exchange rate to depreciate in the short term. The paper also estimates a long-term relationship between the real effective exchange rate for the U.S. dollar and a number of variables.

Abstract

This Background Paper on the United States examines the effect of fiscal deficit reduction in the context of the IMF’s multicountry simulation model, on the current account and the real exchange rate. The simulations suggest that, other things being equal, fiscal consolidation will tend to cause the real exchange rate to depreciate in the short term. The paper also estimates a long-term relationship between the real effective exchange rate for the U.S. dollar and a number of variables.

II. U.S. Fiscal Deficit Reduction and the U.S. Dollar 1/

There has been considerable discussion recently about the potential effects of U.S. fiscal policy on the value of the U.S. dollar. This chapter reviews the assumptions and predictions of traditional models and draws the conclusion that, all other things equal, the effect of deficit reduction in the short run would be to cause the dollar to depreciate. However, this statement should be qualified for three reasons.

First, the effect of deficit reduction will depend on whether market participants demand a risk premium on U.S. securities. In such circumstances, a major fiscal contraction could result in a significant portfolio shift toward U.S. assets. This would put upward pressure on the U.S. dollar and downward pressure on U.S. interest rates. While the possibility of such a shift in preferences cannot be ruled out, there is very little evidence available to suggest that risk premiums on U.S. securities have responded significantly to changes in U.S. fiscal policy.

Second, the magnitude of the exchange rate response will depend on the reaction of the monetary authorities. For example, if the Federal Reserve did not allow nominal interest rates to fall immediately in response to a fiscal contraction but instead used the fiscal contraction as an opportunity to reduce inflation, the nominal exchange rate could appreciate in the short run. Third, the short-run and medium-term effects of fiscal consolidation on the exchange rate are likely to be in the opposite direction. The long-run effect of fiscal contraction would be to lower U.S. net foreign liabilities and net debt-service payments abroad, which would permit a larger net flow of goods from the rest of the world. Thus, in the long run the dollar would have to be more appreciated than in the absence of deficit reduction in order to reduce the trade surplus to its new steady-state equilibrium.

Section 1 provides a very brief review of the short-run predictions of traditional flow-equilibrium models such as the Mundell-Fleming model, as well as the implications of more fully articulated intertemporal models. Section 2 considers the case where a reduction in government debt results in a lower risk premium on U.S. securities. Section 3 shows that the short-run effects on the exchange rate will depend on the reaction of the monetary authorities. Finally, in Section 4 simulations from the Fund’s multicountry simulation model (MULTIMOD) are used to focus the discussions on the medium-term and long-run implications of a reduction in government debt.

1. The short-run effects of fiscal policy in standard models

Traditional Keynesian models that focus on flow equilibrium considerations--such as the Mundell-Fleming model--predict a depreciation in the real value of the currency in response to a contractionary fiscal policy. 1/ This prediction is also an important characteristic of more fully articulated dynamic intertemporal models where stocks play a critical role in the equilibration process. This is true because in both of these models the real exchange rate is a fundamental price that (with the real rate of interest) moves to equilibrate aggregate demand and supply in the goods market. As long as the shock provides a contractionary impulse to domestic demand, the real exchange rate must depreciate in the short run in order to crowd in net foreign demand.

In flexible price models, where aggregate demand has to be equal to aggregate supply at all points in time, the real exchange rate must depreciate sufficiently to offset the contractionary effects of the fiscal policy shock on aggregate demand. This is also true in sticky price versions of these models except it may take some time to re -equilibrate aggregate demand and aggregate supply.

Table II-1 provides the results of an illustrative simulation of MULTIMOD that assumes a permanent reduction in government debt equal to 10 percent of baseline GOP. 2/ This is accomplished by raising taxes by approximately 1 percent of baseline GDP for ten years and then allowing taxes to adjust to stabilize the debt-to-GDP ratio at a lower level. 3/ Since the objective of this section is to study the short-run effects of the shock, the discussion focusses on the first three years of the simulation. As can be seen in the first line of Table II-1, real GDP falls by 0.7 percent in the first year in response to the fiscal contraction. However, after the first year output rises back toward baseline and by the third year is 0.1 percent above baseline.

Table II-1.

United States: The Effects of a Cut in the U.S. Fiscal Deficit

(Percentage deviation from baseline, unless otherwise noted)

article image

The estimates for the long run measure the permanent effects of the shock.

In percentage points.

Although the magnitude of the output responses reported in Table II-1 will depend on certain key policy and behavioral assumptions, the same qualitative result holds in most models where the monetary authorities are attempting to stabilize the inflation rate. In this particular example, the inflation rate is stabilized over a three-year horizon. Since the shock is a contractionary impulse to the system, monetary conditions--the combined effects of interest rates and exchange rates on aggregate demand--must ease in order to crowd in net exports and interest-rate sensitive components of aggregate demand. Indeed, if one examines the contribution to real GDP in the table it can seem that the fiscal shock results in higher real net exports and investment.

For this particular example, the value of the dollar declines by about 2.7 percent on an effective trade-weighted basis and long-term interest rates fall by about 30-40 basis points. This easing in monetary conditions is sufficient to close the output gap by the third year. It is important to note that these effects assume that the monetary authorities allow interest rates to fall and there are no major changes in the perceived risk of U.S. securities. 1/ These issues are considered in the following two sections.

2. The effect of a risk premium

The standard result reported above ignores possible links between deficit reduction and uncertainty about future taxation. For example, deficit reduction may reduce the risk of a default on government interest obligations (equivalent to imposing a 100 percent tax rate on interest income), or increase the perceived risk of an increase in tax rates or of a surprise inflation in the future. Rational market participants will attach positive probabilities to these outcomes if it is apparent that the fiscal policy process is potentially unstable. 2/ In traditional macro models this is sometimes modelled with an upward sloping supply of funds schedule. In other words, it is assumed that the rest of the world is unwilling to absorb increasing amounts of U.S. securities at the prevailing real interest rate.

This possibility is usually introduced in macroeconomic models by assuming that, at some point, market participants demand a risk premium in order to be compensated for greater uncertainty about the after-tax rate of return. For example, the risk premium in the interest parity equation can be made a function of the net foreign liability position of the country. In this case, a fiscal contraction may actually result in an appreciation in the real value of the currency since the reduced demand for loanable funds by the government is more than offset by an increase in private sector supply. 1/

While this argument may be applicable to other countries, it would be difficult to argue that this offsetting effect would be very large in the United States given the very low level of long-term interest rates relative to other countries. 2/ Indeed, the low levels of long-term interest rates in the United States may already embody an expectation of significant fiscal consolidation in the future.

Table II-2 reports the effects of reducing the risk premium in the interest parity equation by 50 basis points in MULTIMOD. 3/ In this case long-term interest rates in the United States fall by about 20 basis points and the nominal effective exchange rate appreciates by 2.2 percent. The increase in real GDP is sustained because the shock reduces the cost of capital in the United States and increases investment and potential output. The result of the appreciation is a worsening of the trade balance, an increase in the current account deficit, and a rise in net foreign liabilities.

Table II-2.

United States: The Effects of a Reduction in the Risk Premium on U.S. Securities

(Percentage deviation from baseline, unless otherwise noted)

article image

The estimates for the long run measure the permanent effects of the shock. In this case they are all zero because the shock only lasts for 20 years.

In percentage points.

Note, however, that if the results of the previous section were summed with these, the effect of fiscal contraction would still be a depreciation. This suggests that the decline in the risk premium would need to be relatively large to cause an appreciation in the face of deficit reduction.

3. The response of monetary policy to a fiscal contraction

It is important to recognize that the short-term effects of any shock--including fiscal shocks--will depend on how the Federal Reserve responds. If the Federal Reserve cuts interest rates in an attempt to minimize the effect on the inflation rate then the nominal exchange rate is likely to depreciate significantly in the short run. However, if the Federal Reserve delays an interest rate cut one would expect a different effect on the exchange rate, since this is equivalent to assuming that monetary policy tightens in response to the contractionary fiscal policy shock. In a world with sticky prices in the short run, this could result in an appreciation in both the nominal and real exchange rate.

Table II-3 reports some illustrative simulation results for the same fiscal shock described in Section 2, except that this time the Federal Reserve is assumed to hold the short-term interest rate fixed in the United States. In this case, the contractionary effects of the shock are large and the nominal exchange rate appreciates in the very short run. Indeed, in this case the contractionary effects of the shock are so large that inflation is about 2 percentage points lower over the first three years.

Table II-3.

United States: The Effects of a Cut in the U.S. Fiscal Deficit with Unchanged Nominal Interest Rates

(Percentage deviation from baseline, unless otherwise noted)

article image

In percentage points.

4. The medium- and long-term implications of government debt

As mentioned above, the prediction that a fiscal contraction results in a real depreciation in the short run carries over to a large class of models where the real exchange rate plays a fundamental role for re-equilibrating the economy; as long as the shock has a contractionary effect on aggregate demand the real exchange rate must depreciate in order to crowd in net foreign demand. However, since the Mundell-Fleming model only focusses on flow equilibrium considerations, it does not provide a very useful framework for examining the medium- and long-term implications of a reduction in the stock of government debt. In order to do this, one has to turn to models that look beyond the short-run effects of fiscal policy to models where stock-accumulation effects are fully accounted for. Indeed, as shown below, these models predict that the real exchange rate will appreciate in the long run in response to a permanent reduction in government debt.

MULTIMOD shares the qualitative predictions of standard, open-economy, general equilibrium growth models. In this class of models, countries with relatively impatient consumers are net debtors to the rest of the world, and countries with relatively patient consumers are their creditors. 1/ In a world with debtor and creditor countries, there will be a constant real resource transfer in the steady state from debtor countries to creditor countries. The interest payments of debtor countries to their creditors will have to be financed with a positive trade balance in the debtor countries. Conversely, creditor countries will be able to import more than they export and pay for this trade balance deficit with the interest that they receive on their loans to the rest of the world.

In order to discuss the effects of government debt in the long run, it is useful to provide some very basic analytics. In the steady state, the current account balance will be proportional to the net foreign asset position of the country. This can be represented by the following equation. 1/

Y=(g1+g)(NFAY)(1)

where: CA is the current account balance, Y is nominal GDP, NFA is the country’s net foreign asset position, and g is the growth rate of all nominal variables. Note that if all nominal variables in the economy are growing at S percent in the steady state and net foreign assets are equal to 100 percent of GDP then the current account surplus must be equal to about 5 percent of GDP. Conversely, if the country were a debtor country instead of a creditor country then the country would be running a current account deficit in the steady state. Basically, because of growth a country can only remain a debtor country or a creditor in the steady state if its nominal current account deficit or surplus grows at the same rate as nominal income.

The current account can also be expressed as the sum of the interest payments received from foreigners and the nominal trade balance (TB). For simplicity assume that interest payments are paid on a one-period bond that rolls over each period so that interest payments received from foreigners will be r NFAt-1. Then the current account equation becomes:

CAY=r1+gNFAY+TBY(2)

because NFAt-1 is equal to NFAt/(1+g). These two equations are useful for understanding the steady-state properties of any model even though they make no assumptions beyond the conditions of a balanced steady-state growth path. Note that if we substitute Equation 1 into Equation 3 we obtain:

TBY=(gr1+g)NFAY(3)

As long as the real interest rate is greater than the real growth rate of the economy then the nominal interest rate will be greater than the growth rate of nominal income. Under these conditions the term in parenthesis in Equation 3 will be negative. 1/ Countries that have positive net claims on the rest of the world will be running a trade deficit in the steady state, while countries that are debtor countries will have trade surpluses. In other words, net creditor countries will receive a permanent flow of interest payments from the rest of the world and thus be able to maintain a level of imports that is higher than exports.

For example, a fiscal contraction reduces the economy’s net consumption of foreign goods and services in the short to medium run. During this period, the economy’s net foreign liability position is reduced since there is a lesser need for foreign saving. By borrowing and consuming less in the nearer term, a larger amount of consumption and a larger trade deficit can be sustained in the long run. In order to equilibrate the demand for domestically produced goods with domestic production, an appreciation of the real exchange rate is required.

This result depends on the assumption that consumers have finite planning horizons. To illustrate the importance of this assumption, consider the effect of fiscal policy in a model where consumers have an infinite planning horizon--i.e., they live forever. For example, the government acts to lower the level of government debt by means of a temporary tax hike. Although taxes rise in the short run, there will be lower taxes in the new steady state, because the government’s interest rate obligations will decline with the reduction in government debt. If consumers live forever and there are perfect capital markets, the reduction in consumers’ disposable income during the temporary tax hike would equal the rise in disposable income thereafter in present-value terms. In this case, households would not adjust their consumption decisions either now or in the future in response to the change in fiscal policy. Instead private saving would fall to offset the rise in public savings, and consumption, real interest rate, the real exchange rate, and the current account would all remain unchanged.

In models like MULTIMOD, consumers have a finite planning horizon. As a result, the present value of the fall in disposable income during the temporary period with higher taxes is greater than the present value of the rise in disposable income thereafter. 2/ Households therefore reduce their consumption in the short run rather than running down their savings initially and then building them back up later. Since the United States is a net debtor to the rest of the world, the rise in total national savings would result in a reduction in the U.S. net foreign liability-to-income ratio. As can be seen from Equation 3 above, this eventually would imply lower net interest payments to foreigners and a smaller trade surplus with the rest of the world. In order to induce a smaller trade surplus (lower imports and higher exports) with the rest of the world, the real exchange rate must appreciate in the steady state.

Table II-1 reports some MULTIMOD simulation results for this type of shock. Again, government debt is assumed to be reduced by 10 percent of baseline GDP by imposing a one percent tax rate hike for ten years. Thereafter, tax rates are allowed to be lower in order to stabilize the debt-to-GDP ratio. Note, in this simulation the real exchange rate depreciates in the short run and then starts to appreciate by the third year. Indeed, by the fourteenth year the real value of the exchange rate is higher relative to its baseline value. 1/ Again, this appreciation is necessary to sustain a higher level of net imports; since interest obligations to foreigners are now reduced there will be a larger sustainable net flow of goods to U.S. consumers. And under normal assumptions about demand and supply curves, this can only come about if the relative price of these goods falls.

References

  • Aizerman, John,Country Risk, Incomplete Information and Taxes on Foreign Borrowing,Economic Journal, Vol. 99 (March 1989), pp. 14761.

    • Search Google Scholar
    • Export Citation
  • Alesina, Alberto, and Guido Tabellini,External Debt, Capital Flight and Political Risk,Journal of International Economics, Vol. 27 (November 1989), pp. 199220.

    • Search Google Scholar
    • Export Citation
  • Alesina, Alberto, A. Prati, and G. Tabellini,Public Confidence and Debt Management: A Model and a Case Study of Italy.In Public Debt Management: Theory and History,” ed. by R. Dornbusch and M. Draghi (New York: Cambridge University Press, 1990).

    • Search Google Scholar
    • Export Citation
  • Alesina, Alberto, M. De Broeck, A. Prati, and G. Tabellini,Default Risk on Government Debt in OECD Countries.Economic Policy: A European Forum (October 1993), pp. 42863.

    • Search Google Scholar
    • Export Citation
  • Bayoumi, T., M. Goldstein, and G. Woglom,Do Credit Markets Discipline Sovereign Borrowers? Evidence from the United States.” (Washington: International Monetary Fund, Research Department, 1994).

    • Search Google Scholar
    • Export Citation
  • Bayoumi, Tamim, and Douglas Laxton, “Government Deficits, Debt, and the Business Cycle,” in Deficit Reduction: What Pain. What Gain? (Ottawa: C.D. Howe Institute) 1994.

    • Search Google Scholar
    • Export Citation
  • Bhandari Jagdeep S., Nadeem Ul Haque, and Stephen J. Turnovsky,Growth, External Debt, and Sovereign Risk in a Small Open Economy,IMF Working Paper No. WP/89/54 (Washington: International Monetary Fund, June 1989).

    • Search Google Scholar
    • Export Citation
  • Blanchard, O. J.,Debt, Deficits, and Finite Lives,Journal of Political Economy. Vol. 93 (1985), pp. 22347.

  • Buiter, Willem H.,Fiscal Policy in Open, Interdependent Economies,Chapter 3 in Economic Policy in Theory and Practice, ed. by Assaf Razin and Efraim Sadka (New York: St. Martins Press, 1987), pp. 10144.

    • Search Google Scholar
    • Export Citation
  • Dornbusch, Rudiger and Stanley Fischer, Macroeconomics (New York: McGraw Hill, 1978).

  • Faruqee, Hamid, Douglas Laxton, and Steven Symansky, “Government Debt, Life-Cycle Income and Liquidity Constraints: Beyond Approximate Ricardian Equivalence” unpublished manuscript.

    • Search Google Scholar
    • Export Citation
  • Laxton, Douglas, and Robert Tetlow,Government Debt in an Open Economy,” (Ottawa: Bank of Canada Technical Report No. 58.) 1992.

  • Macklem, R.T.,Terms-of-Trade Disturbances and Fiscal Policy in a Small Open Economy.Economic Journal, Vol. 103 (1993), pp. 91636.

    • Search Google Scholar
    • Export Citation
  • Mundell, Robert,Flexible exchange rates and employment policy,Canadian Journal of Economics and Political Science, Vol. 27 No. 4 (November 1961), p. 509517.

    • Search Google Scholar
    • Export Citation
  • Weil, Phillip,Overlapping Families of Infinitely-lived Agents,Journal of Public Economics, Vol. 38 (March 1989), pp. 18398.

1/

Prepared by Douglas Laxton and Steven Symansky.

1/

The Mundell-Fleming model amends the familiar closed economy IS-LM model with a condition requiring a balance of payments equilibrium. See Dornbusch and Fischer (1978), pp. 627-629, or Mundell (1961) for a diagrammatical exposition of this simple model. While this model has insights for understanding the short-run effects of fiscal policy it does not provide a very adequate framework for understanding the: medium-term and long-term implications of fiscal policy. Indeed, as discussed below, a contractionary fiscal policy that permanently reduces government debt will result in a real exchange rate depreciation in the short ruin and a real exchange rate appreciation in the long run.

2/

The version of MULTIMOD used in this chapter is based I on recent work by Faruqee, Laxton, and Symansky (1995). The revised model, relaxes the assumption that the real interest rate equals the real growth rate of GDP in the steady state. In addition, the revisions make private consumption more responsive to changes in current disposable income.

3/

The same story would hold if government expenditures were reduced for 10 years although the quantitative effects would be somewhat different. In general because government expenditures usually have stronger effects on aggregate demand, monetary conditions have to adjust more in the short run.

1/

In addition, the simulation experiment assumes that the new fiscal plan is fully credible and tax rates adjust immediately. If the public was skeptical that the debt reduction was going to be permanent, the short-run contractionary impulse would be larger and this would necessitate a larger depreciation in the short run. On the other hand, there are cases when policies are announced but expenditure and tax parameters do not change until sometime in the future. When these changes are credible the exchange rate will tend to lead movements in actual fiscal instruments. Indeed, one interpretation of the recent weakness in the U.S. dollar and the decline in U.S. long-term interest rates is that market participants are now expecting large cuts in the U.S fiscal deficit.

1/

In the context of the Mundell-Fleming model, this would be equivalent to a large downward shift in the world interest rate.

2/

For a discussion of these issues for Canada, see Bayoumi and Laxton (1994) and Chapter 14 in SM/95/81.

3/

The shock is assumed to last for 20 years since in the long run the uncertainty regarding fiscal policy is assumed to be eliminated.

1/

The basic consumption theory behind MULTIMOD is based upon work by Blanchard (1985), Buiter (1987), Weil (1989), and Macklem (1993).

1/

Equation 1 follows from the balance of payments condition that the current account balance (CA) has to be equal to the change in net foreign assets (CA = ΔNFA = NFAt-NFAt-1). If we define g to be the steady-state growth rate of all nominal variables then (NFAt/NFAt-1-1) = g or NFAt = NFAt/(1+g). Substituting this last expression into the CA equation and dividing by nominal GDP produces Equation 1.

1/

Although it is true that the real interest rate sometimes falls below the real rate of growth, this is not likely to be sustainable in the steady state.

2/

The same steady-state results could also be achieved with a temporary cut in government expenditures.

1/

If the shock were induced by a change in government expenditures, instead of taxes, then the long-run appreciation in the real exchange rate would occur earlier. This will also be the case for tax shocks in models where consumption is tied more closely to changes in disposable income. See, for example, Laxton and Tetlow (1992).