Chapter

3 The Yen: Past Movements and Future Prospects

Author(s):
Tamim Bayoumi, Guy Meredith, and Bijan Aghevli
Published Date:
June 1998
Share
  • ShareShare
Show Summary Details
Author(s)
Guy Meredith

The sharp swings in the yen in recent years have focused renewed attention on its underlying determinants. However, explanations in financial markets of the factors driving the exchange rate have, for the most part, been ex post rationalizations of the latest market movements that have been of little durable value in better understanding exchange rate behavior. In the event, a pattern has repeated itself whereby changes in market sentiment have driven the yen to levels that seem well out of line with past trends and fundamentals, followed by equally sharp reversals. Furthermore, the amplitude of these swings has not diminished over time, suggesting that markets are not converging on a better understanding of the determinants of the yen’s value.

These issues are of more than academic interest. Looking at the recent experience, the sharp rise in the yen since late 1992—culminating in its dramatic appreciation in early 1995—threatened to undermine a recovery in Japan that was already slowed by the headwinds of weak asset prices, balance-sheet adjustments, and industrial restructuring. The subsequent sharp depreciation has not only provided a desirable boost to activity, but also raised fears of the possible impact on inflation. Looking further back, the weakening of the yen in 1989-90 exacerbated excess demand pressures at the peak of the bubble period, helping to set the stage for the subsequent downturn in activity. Over the longer run, swings in the yen in both directions have led to concerns about the negative impact of exchange rate volatility on industrial planning and trade dislocation.

Macroeconomic instability resulting from yen volatility has also been reflected in large swings in Japan’s external surplus. While financial markets have viewed changes in the current account surplus as a “fundamental” that drives movements in the yen, at the same time, there have been strong lagged adjustments of the trade balance to exchange rate movements. The result of this feedback process has been self-reinforcing cycles in the yen and the current account that have, in turn, contributed to further confusion about the fundamentals. Large swings in the external balance have also perpetuated trade frictions and concerns about Japan’s alleged tendency to export its way out of recessions.

To shed light on the determinants of exchange rates, this chapter examines the forces governing the value of the yen in an overall macroeconomic framework. The second section looks at the historical experience, including long-run trends in the real value of the yen and recent deviations around these trends; explanations for such trends; and the relationships between movements in the yen, the current account, and international interest rate differentials. The third section presents a conceptual model of the equilibrium exchange rate, and discusses the meaning of misalignment in this context. The chapter then turns in the fourth section to a description of the past approach the staff has taken to implementing this definition of misalignment and its shortcomings. The fifth section presents the results of a richer approach based on simulations of a full macro-economic model, including an assessment of their sensitivity to alternative assumptions about key variables. The chapter concludes with a summary and directions for future research.

Past Movements in the Yen

The real value of the yen has exhibited several striking features over the postwar period (Figure 3.1).1 The first feature is the trend appreciation observed when nominal exchange rates are adjusted for movements in aggregate price indices, such as the consumer price index (CPI). The second feature is the wide divergence between trends in different measures of the real exchange rate, depending on whether it is constructed using consumer prices, wholesale prices, or export unit values. In particular, the upward trend in the wholesale price index (WPI)-based measure since the early 1970s has been only one-half that in the CPI-based index, while the export unit value measure has been broadly stable since the mid-1970s. Third, the deviations from trend in all three measures have been much more pronounced under floating exchange rates than under fixed rates, reflecting the sharp movements in nominal exchange rates during the post-Bretton Woods era. Finally, the swings in the export price measure are smaller than those using broader price indices, suggesting that movements in export prices have “buffered” some of the swings in nominal exchange rates.

Figure 3.1.Alternative Measures of the Real Effective Exchange Rate1

(Log scale, 1951 = 100)

Sources: IMF. International Financial Statistics, various issues; and IMF staff estimates.

1CPI-based rate employs an average of major industrial partner countries; WPI-based rate employs an average of major industrial partner countries for which WPIs are available: the United States, Germany, and the United Kingdom. Relative export unit values are based on major industrial partner countries.

The divergent trends in different measures of the real exchange rate are consistent with the Balassa-Samuelson hypothesis (Balassa, 1964; and Samuelson, 1964). In this framework, real exchange rate trends are explained by differences in sectoral productivity growth rates across countries—in particular, countries with rapid productivity growth in tradable versus nontradable sectors will experience real appreciations, as the relative price of nontradables rises over time without jeopardizing international competitiveness. There is strong evidence that this phenomenon applies to Japan: as shown in the upper panel of Figure 3.2, Japan’s CPI and WPI have both risen steadily relative to export prices since the early 1950s, while the ratios for trading partners have been much more stable. The higher weight of nontraded goods in the CPI than in the WPI explains the stronger upward trend in the CPI-based real exchange rate.

Figure 3.2.Evidence on the Balassa-Samuelson Hypothesis

Sources: IMF. International Financial Statistics, various issues; and IMF staff estimates.

While this hypothesis is consistent with some of the stylized facts for Japan, there is one important discrepancy. Specifically, productivity growth differentials are assumed to reflect productivity catchup in tradable sectors in lower-income countries to the levels in higher-income countries. In this case, international differences in productivity growth in the tradables sector—and for the aggregate economy—should gradually narrow, as productivity levels in tradable sectors converged across countries. Thus, trends in exchange rates owing to Balassa-Samuelson effects would decay over time.

Japan’s aggregate productivity growth has indeed slowed during the postwar period (Figure 3.2, lower panel). Yet, so has growth in the United States, the world’s leading major economy, leaving a gap between growth rates that has been relatively stable over the past three decades. Similarly, there are no obvious signs that the trends in relative prices or real exchange rates are diminishing. These observations suggest that differences in productivity growth do not reflect catch-up effects. Instead, productivity growth in the tradable and nontradable sectors appears to have slowed at roughly the same pace in both countries, leaving the gap unchanged.2 On this basis, there is little evidence that the forces underlying the real appreciation of the yen are diminishing through productivity convergence.3

Figure 3.1 also indicates how the sharp swings in the yen in recent years compare with past experience. The appreciation in 1993 took the exchange rate above trend, but not dramatically so by the standards of the 1970s and 1980s. With the further sharp rise in the yen in the spring of 1995, however, the gap reached about 20 percent on a CPI basis, approaching the previous peak in 1978. As was the case at that time, the surge in the yen was abruptly reversed; from mid-1995 to early 1997, the CPI-based measure fell by about 35 percent, taking the yen about 20 percent below the historical trend.

The relationship between movements in the yen and Japan’s current account surplus is also of interest. Figure 3.3 (upper panel) shows the deviations in the CPI-based real exchange rate from trend and the changes in the ratio of the current account to GDP since the mid-1970s. It is evident that appreciations (depreciations) in the real exchange rate have been followed, with a lag of about a year, by significant declines (increases) in the current account. It also appears that swings in the exchange rate have themselves followed shifts in the current account. Such a lagged feedback mechanism is reminiscent of the classic “cobweb” cycles in economics textbooks. Thus, the market view of the current account as a “fundamental” that drives the exchange rate appears to have been justified in the short run, if only because it has been self-fulfilling. But the lagged response of the current account to the yen has meant that the “fundamentals” have, in reality, been endogenous, although response lags have tended to mask this fact. The apparent confusion in markets about fundamentals underscores the need for a more robust understanding of the longer-run determinants of exchange rates.

The past relationship between exchange rate trends and interest rate differentials between Japanese and foreign assets also merits attention. If assets are highly substitutable and there are no barriers to capital mobility, real interest differentials should reflect expected movements in real exchange rates (as discussed in more detail below). Given that the yen has appreciated at an average rate of 3 percent a year over an extended period, this would suggest an equilibrium interest differential of a similar magnitude. As shown in the lower panel of Figure 3.3, however, the differential between Japanese and U.S. real interest rates has been much smaller than the trend rise in the yen.4 Indeed, a 10-year moving average of the real interest differential stood at less than ½ of 1 percentage point in 1995. As a result, measured in common currencies, the return on yen assets has substantially exceeded that on dollar assets. Two explanations could be advanced for this phenomenon: either yen and dollar assets have not been close substitutes, leading to an “equilibrium” yield spread that raised the ex ante return on yen assets or markets did not fully incorporate future yen appreciation in their assessment of expected yields.

Figure 3.3.External and Financial Market Variables1

Sources: IMF, International Financial Statistics, various issues; and IMF staff estimates.

1Financial data for 1997 are February 1997 values. Current account and inflation data are IMF staff projections.

2Measured as 10-year government-bond yield less CPI inflation.

This historical failure of “open interest parity” will be revisited below. Without passing judgment on the causes at this stage, it is interesting to note that (belated) recognition of the importance of exchange rate movements in determining relative asset yields may have contributed to the “bubble” in the yen in 1995. Subsequently, interest rates fell in Japan as monetary policy was eased, resulting in a significant widening of the uncovered yield gap on dollar and yen assets. The yen fell sharply in response. As of early 1997, the real long-term yield differential stood at about 2½ percentage points—much closer to the long-term real appreciation of the yen than in the past.

This discussion of past movements in the yen provides a background for some of the issues addressed later in the Chapter:

  • Will market forces tend to take the yen back to past trends over the medium term?

  • Beyond looking at historical trends, what analytical framework can be used to judge if the yen is misaligned?

  • How will Japan’s changing demographic structure—and its impact on output growth and saving rates—affect the future path of the yen?

  • What would be the implications for the exchange rate of policy initiatives such as the adoption of an ambitious fiscal consolidation plan?

  • What influence will structural factors, such as Balassa-Samuelson effects, have on the future path for the yen?

  • How will longer-run trends in the yen affect Japanese interest rates, and what effect will this have on saving-investment balances, the current account, and the equilibrium level of the exchange rate?

Theoretical Framework

Before turning to an empirical analysis of these questions, this section presents the conceptual framework for exchange rate determination that underlies the approach. The meaning of misalignment is then discussed.

A Stylized Model

In a conventional Mundell-Flemming framework, the determination of the exchange rate involves both intratemporal and intertemporal aspects. From an intratemporal perspective, the real exchange rate, representing the price of domestic relative to foreign goods, is a key determinant of the external balance. From an intertemporal viewpoint, the external balance must also be consistent with saving and investment decisions. Finally, capital flows introduce another intertemporal dimension—the external balance must be consistent with the capital account, which depends, inter alia, on the expected return on domestic versus foreign assets.

The interaction of these factors can be seen by writing out the three relationships referred to above. For simplicity, prices are initially assumed to be flexible, so that the economy is always at full employment:

where r is the domestic real interest rate,5rw is the world real interest rate, rer is the (log of the) real effective exchange rate, rere is the expected future level of rer,6 and the X vectors represent the influence of other factors. Equations (1) and (2) are straightforward: the saving-investment (SI) balance rises as the real interest rate increases, while the current account falls as the exchange rate appreciates.7 Equation (3) indicates that net capital inflows are positively related to the gap between domestic and foreign interest rates adjusted for expected changes in the real exchange rate, with the parameter y indicating their sensitivity to yield differentials. A special case of equation (3) is of interest—-when domestic and foreign assets are perfect substitutes and there are no barriers to capital mobility, Y is infinite, and the domestic real interest rate equals the world rate adjusted for expected exchange-rate movements:8

Through accounting identities, all three measures of the external balance must be equal, that is, SI = CA = KA. Equations (1) through (3) then determine three endogenous variables—-the external balance, the real interest rate, and the real exchange rate—conditional on the X vectors, the world interest rate, and the expected exchange rate. Under the assumption of perfect capital mobility, the solution can be shown in a two-dimensional diagram, as shown in Figure 3.4.9

Figure 3.4.Interest Rate and Exchange Rate Determination Under Perfect Capital Mobility

Curve 1, with slope -ß/α, indicates the combinations of r and rer that equate the current account balance and the saving-investment balance. A higher interest rate raises the SI balance, while a lower exchange rate raises the current account balance. Curve 2 reflects the interest parity condition, conditional on the world interest rate and the expected future exchange rate. A rise in the domestic interest rate causes the exchange rate to appreciate such that the expected future depreciation offsets the higher uncovered yield on domestic assets. The positions of the two curves—and thus the equilibrium interest and exchange rates—depend on the X vectors, the world real interest rate, and the expected exchange rate. For instance, a fiscal expansion would shift Curve 1 upward, causing r and rer to both increase. A rise in the world interest rate (or a drop in the expected exchange rate) shifts Curve 2 upward, causing the domestic interest rate to rise, while the exchange rate falls.

This model is, of course, not a complete description of interest rate and exchange rate determination. Allowing for price stickiness in the short run would introduce additional relationships for real output and prices. More fundamentally, the expected exchange rate cannot be considered exogenous—its future values will also be determined by equations (1) through (3), where the contemporaneous values of the X vectors are replaced by their expected future realizations. Substituting out for rere in equation (3) then implies a reduced-form model in which the current levels of the interest rate and the exchange rate are functions of both the current and expected future values of the X vectors (Xe):

In a long-run steady state, the real exchange rate will be stationary, and thus the domestic real interest rate will equal the world level (assuming open interest parity).10 This allows equations (1) and (2) to be solved recursively for the equilibrium exchange rate. Specifically, the domestic saving-investment balance is “tied down” by the world real interest rate, and the exchange rate is determined by the conditions that the current account equal the saving-investment balance. The dynamic relationship for the exchange rate represented by (3′) can then be thought of as an Euler equation, the solution to which must satisfy the transversality condition given by these long-run conditions.11

Meaning of Misalignment

At any given time, the observed values of rer and r must, by definition, reflect the intersection of the curves in Figure 3.4. Looked at from this static perspective, the only issue is how to “back solve” for the values of the X vectors, the expected future exchange rate, and the structural parameters that are consistent with these outturns. In this sense, the market exchange rate is always in equilibrium at a point in time. Any meaningful discussion of misalignment, then, must be framed in terms of the consistency of the current exchange rate with its expected future path—specifically, whether this path represents a full expectational equilibrium.

Put more concretely, the exchange rate is in equilibrium only when market expectations of its path are consistent with the predictions of the “true” model underlying equations (1) to (3), based on all available information. When these paths coincide, the current exchange rate is intertemporally consistent in the sense that there is no reason, ex ante, to expect future surprises in exchange markets. Ex post, of course, unanticipated shocks can occur that redefine the equilibrium and move the exchange rate away from the original path. But as long as such shocks are not predictable, the initial exchange rate is not misaligned.

Suppose, in contrast, that market expectations and the “true” equilibrium path do not coincide. As the future unfolds, even in the absence of unanticipated shocks, markets will discover that actual interest rates and exchange rates are higher or lower than expected. As a result, there will be unanticipated gains or losses on holding domestic versus foreign assets. Agents whose expectations are closer to the true values will make windfall gains, while others will make losses. Assuming that markets learn over time from their mistakes, expectations will converge on the true path, and the exchange rate will adjust toward its equilibrium level.

The dependence of the exchange rate path on interest rates implies that cyclical developments affect the equilibrium exchange rate. For instance, a downturn in aggregate demand will reduce interest rates and the exchange rate, while at the same time leading to an expected future appreciation. Similarly, policy initiatives can affect the exchange rate through changing either the real interest rate, the longer-term expected real exchange rate, or both. Consider, for example, a temporary fiscal stimulus. In general, both the real interest rate and the exchange rate will rise on impact. As the shock unwinds, both variables return toward their initial levels. Does the initial jump in the exchange rate indicate that it has become “misaligned”? In this framework, the answer is no, as long as the depreciation over the medium term is consistent with interest differentials.12

Past Staff Approach to Assessing Misalignment

To fully implement the above framework would require using a completely specified, forward-looking model consistent with generalized versions of equations (1) through (3). The IMF staff has not taken this approach in the past. Instead, rather than attempting to construct a full expectations-consistent path for the exchange rate, the level of the equilibrium exchange rate has been “tied down” at the end of a horizon over which output is expected to return to potential. In practice, this has been the end of the five-year forecast provided for the IMF’s World Economic Outlook (WEO) exercise. This choice allows one to abstract from cyclical factors in determining the medium-term level of the yen, which then depends only on the full-employment model implied by equations (1) through (3).

As far as the full-employment saving-investment balance (equation (1)) is concerned, the staff has used a relatively simple framework for Japan. The saving rate depends on fiscal variables, demographic factors, and asset stocks, while the investment rate is determined by the adjustment of the capital stock to a trend capital-to-output ratio. The investment rate also depends (conceptually) on the real interest rate. Neither the “structural” saving nor investment rate depends on the exchange rate, however, allowing equations (1) and (2) to be solved recursively: the structural SI balance ties down the medium-term current account; the exchange rate, in turn, is determined through inverting equation (2).13 In practice, the level of the exchange rate consistent with a given current account balance is derived from the staff’s trade model for Japan (see Chadha, 1996).

The observed exchange rate is compared with this medium-term equilibrium level, after adjusting for market expectations of exchange rate movements over the intervening period. In particular, the real interest rate differential on yen versus foreign assets is taken to be a useful proxy for expected movements in the yen.14 Thus, the historical failure of open-interest parity discussed above is assumed to primarily reflect expectational errors, rather than equilibrium yield differentials that persist into the future. If market expectations of the exchange rate path are consistent with hitting the equilibrium medium-term equilibrium level, then there is no evidence that the yen is misaligned. If, in contrast, the expected path does not hit the medium-term level, then the exchange rate is judged to be misaligned. In this case, if the exchange rate followed the path expected by markets, the economy would not achieve medium-term macroeconomic balance.

As an example of this approach, consider the situation in the spring of 1995, when the yen had risen to about ¥80—85 per dollar. If the yen had appreciated from this level at the rate suggested by interest rate differentials, the staff estimated that Japan’s current account balance would have fallen well below the structural SI balance over the medium term. Indeed, the real exchange rate would have had to experience a “jump” depreciation of about 25 percent to make it consistent with medium-term fundamentals, indicating a significant overvaluation of the yen. By early 1997, the yen had depreciated sharply, while yen interest rates had also fallen. The path for the yen implied (moderately) exceeded the structural SI balance by the end of the medium term, providing evidence of some degree of undervaluation.15

It should be noted that this test of misalignment is weaker than that associated with the full conceptual framework. In particular, the consistency of market expectations with medium-term equilibrium is a necessary—but not sufficient—condition for the exchange rate to be “correctly” aligned. Even if market expectations for the exchange rate path were consistent with medium-term equilibrium, these expectations could be based on implausible projections of interest rates. In this case, tensions would still arise in financial markets that would have implications for exchange rates.

In addition to this caveat, there are other drawbacks. When the exchange rate is judged to be misaligned, it is difficult to be specific about the magnitude, because market expectations of interest rates would change endogenously if the exchange rate were to move closer to equilibrium. More fundamentally, the procedure used to tie down the medium-term saving-investment balance is somewhat arbitrary, as it depends on ad hoc assumptions about the real interest rate that will be obtained at the end of the medium-term horizon. Conceptually, this rate should depend on the expected longer-term path of the real exchange rate, consistent with equation (3). In the absence of information about longer-term developments, though, it has been assumed that the historical real interest rate would prevail. Given this truncation of the analysis at the end of the medium term, many interesting questions about the impact of long-term developments—such as population aging—on the equilibrium level of the yen could not be addressed.

Empirical Analysis Using MULTIMOD

To deal with the weaknesses described above, we employ here a more complete model, consistent in spirit with the conceptual framework, to assess exchange rate fundamentals. Specifically, a variant of the Japanese block of MULTIMOD, the IMF’s multicountry macroeconomic model, has been used to construct long-run paths for the exchange rate that are fully consistent with the overall model solution.16 This section first summarizes some key features of MULTIMOD. The baseline path for the yen is then described, followed by an analysis of the sensitivity of the results to alternative assumptions.

Structure of MULTIMOD

MULTIMOD embodies neoclassical linkages for domestic spending and external trade, as well as disequilibrium relationships that determine how output and prices adjust to excess demand or supply conditions. It also reflects forward-looking behavior in financial markets, in that the predicted paths for the financial variables are “expectations consistent.” Specifically, paths for interest rates and the exchange rate are solved so that there are no unanticipated surprises in financial markets over the simulation horizon. For the exchange rate, this means that the open interest parity condition represented by equation (3′) holds for all simulation periods—with the important exception of the first period, when the exchange rate can “jump” to its equilibrium path. This jump indicates the extent to which the exchange rate is initially misaligned.

MULTIMOD has been extensively used within the IMF for simulation analysis. The present application, however, required constructing a forecasting version of the model that could be used to derive baseline projections for the key variables. Adapting it to this purpose required several modifications (as described in Appendix I). The most important changes involved reestimating the parameters in the trade and spending blocks to make them more “Japan specific”;17 modifying the definitions of human and capital wealth to allow for forward-looking behavior without making the associated expectations fully model consistent; and changing the policy reaction function of the monetary authorities such that the real short-term interest rate responds to deviations in nominal income from a target path.18

An additional issue that arose in constructing a baseline path was how to tie down the “terminal conditions” for the forward-looking variables.19 Forward-looking models are generally simulated over a finite horizon. Yet values for the forward-looking variables are needed beyond the simulation horizon to tie down their paths in the last periods of the simulation. In the case of the exchange rate, for instance, its value in the last simulation period depends on its expected value in first period beyond the simulation horizon. To solve this issue, a technique was used that has become increasingly popular in the solution of such models. Specifically, a steady-state version of the dynamic model was constructed that yielded long-run solutions for the endogenous variables for given levels and growth rates of the exogenous variables (as described in Appendix II). These steady-state values were then used as terminal conditions for the dynamic model.20

Exogenous Assumptions

The model was simulated using annual data for 1997—2070. The first step was to construct paths for the exogenous variables, including foreign activity, inflation, and interest rates; world oil and commodity prices; and domestic productivity growth, labor force growth, and population dependency ratios. For the external variables, WEO projections as of spring 1997 were used to the end of the medium-term horizon (i.e., the year 2002).21 Over this period, growth in foreign activity (defined as the weighted average of partner-country GDP) converges to 3½ percent a year; foreign inflation (average growth in partner-country export prices and GNP deflators, in U.S. dollars) converges to 1½ percent a year; and the foreign interest rate (the short-term interest rate on U.S. dollar assets) stabilizes at 5¾ percent. World oil and commodity prices remain broadly constant in real terms. Beyond 2002, we assume that foreign growth gradually declines to 1½ percent by 2050, as growth rates in developing Asian economies slow toward those in developed countries. The real interest rate on U.S. dollar assets also falls by about 1 percentage point to reflect the slowing in global growth.22

On the domestic side, exogenous variables include the rate of labor-augmented productivity growth and demographic factors. For productivity, growth is assumed to remain at the trend 1½ percent rate observed since the mid-1970s. The demographic assumptions are based on the latest projections for Japan released by the Ministry of Health and Welfare in January 1997, which cover the period to 2070. The total population, the population of labor-force age, and the dependency ratio come directly from these projections. To project the active labor force, the participation rate is assumed to rise gradually through 2002 and then to remain constant thereafter.

As for economic policies, the baseline assumption is that no medium-term fiscal consolidation measures are introduced beyond 1997, except those associated with the 1994 pension reform plan. On monetary policy, the nominal growth target equals the rate of potential output growth plus a fixed growth rate for the GNP deflator. The latter was set at ½ of 1 percent a year—-the rate at which inflation stabilizes in the medium-term WEO forecast.

After defining paths for the exogenous variables, it was necessary to adjust the model’s stochastic equations to yield reasonable predictions for the endogenous variables, especially over the near-term horizon. In particular, a completely “model-based” forecast, with no judgmental adjustments, yielded simulation values that in many cases were not plausible in light of current information. This is not surprising, given the highly aggregated nature of MULTIMOD and the fact that it was constructed as a policy simulation as opposed to a forecasting model.

To address this problem, “add factors” were introduced to some of the model’s stochastic equations such that they would replicate the WEO forecast over the 1997–2002 period if the values for the exchange rate and interest rates corresponded to their WEO assumptions.23 To better understand this approach, it is necessary to explain the basis of the WEO assumptions for financial variables. Real exchange rates remain unchanged over the forecast horizon, with the starting level being determined by observed market data at the beginning of the forecast period. For interest rates, paths are assumed consistent with output returning to potential over the medium term, conditional on the given level of the real exchange rate. Looked at in terms of the conceptual framework, the WEO projection for Japan first solves equation (2) for the current account based on the exogenous exchange rate assumption; equation (1) is then solved recursively for the interest rate that equates the SI balance to the current account.24 But no attempt is made to impose a consistent path for interest rates and the exchange rate (equation 3’).

Taking this WEO forecast as a starting point, MULTIMOD was used to impose consistency in financial markets via equation (3’), and then to assess the impact on the other endogenous variables. The model simulations then differ from the WEO projections for two reasons: (1) the paths for the financial variables are fully expectations-consistent through the medium-term projection period, as opposed to being arbitrarily imposed, and (2) the paths for the forward-looking variables are consistent with the model’s predictions beyond 2002, as opposed to relying on ad hoc assumptions.

Baseline Path

The baseline path for the real exchange rate is shown in Figure 3.5, while the simulation results for other variables are summarized in Table 3.1. The real effective exchange rate depreciates by about 3 percent in 1997 (the first simulation period) from its historical level in 1996, in spite of a real interest rate differential in 1996 in favor of U.S. dollar assets of about 3 percentage points.25 Taken together, this implies an “excess return” on dollar assets of slightly over 6 percent at the starting point of the simulation, which indicates the extent to which the yen was overvalued in 1996.26 In the event, the further depreciation of the yen through early 1997 took its real effective value about 10 percent below the 1996 average level. This decline exceeds the depreciation suggested by the model simulation by about 7 percent, suggesting that the yen had (modestly) undershot its equilibrium level as of the spring of 1997.

Figure 3.5.Simulated Long-Run Path for the Real Effective Exchange Rate

(Natural logarithm, 1990 = 0)

Source: IMF staff estimates and projections.

Table 3.1.Summary of Baseline Simulation Results(Growth rate in percent unless otherwise indicated)
1996199719981999200020012002200320042005202020402060
Real GDP3.62.33.53.93.02.42.41.02.11.90.30.41.0
Output gap (percentage point)-3.0-3.1-2.1-0.8-0.3-0.20.1-0.8-0.5-0.3-0.80.9-0.4
Potential output2.32.32.42.52.42.32.21.91.81.60.90.20.9
Real GNP4.02.63.53.82.92.42.41.02.01.80.30.51.0
Real disposable income5.10.53.33.82.42.42.80.01.51.4-0.52.80.1
Real absorption4.51.83.44.13.22.62.80.72.01.8-0.42.80.1
Consumption2.91.52.93.53.02.52.61.42.22.3-0.52.70.1
Investment7.45.45.96.23.32.22.70.92.21.8-0.63.70.0
Government spending6.8-1.62.03.23.73.63.6-1.70.9-0.20.12.20.2
Trade balance (percentage point contribution)-0.80.50.1-0.2-0.2-0.2-0.30.30.10.10.6-1.80.6
Real exports1.77.26.16.05.24.63.84.64.64.53.9-1.62.3
Real imports9.83.15.77.87.26.56.62.44.13.80.86.6-0.8
Total real wealth2.30.61.93.22.53.03.12.12.22.2-0.22.30.2
Of which: Human wealth0.5-0.51.43.42.43.02.92.12.22.5-0.52.70.0
Real capital stock2.83.23.53.83.83.63.53.23.13.00.90.70.8
GNP deflator0.20.50.60.70.90.80.50.2-0.2-0.31.00.70.2
Absorption deflator0.20.80.70.70.90.80.40.4-0.1-0.21.6-1.20.9
fix port price deflator7.1-1.1-6.1-4.0-2.8-2.3-2.3-2.7-3.5-3.8-0.80.00.4
Import price deflator10.91.5-5.5-3.4-2.1-1.8-1.8-1.5-2.5-2.90.5-3.51.2
Short-term interest rate (percentage point)0.30.91.11.92.32.32.61.51.10.94.30.44.4
Real short term rate (percentage point)-0.70.10.31.01.51.92.21.61.30.83.31.63.4
Long-term interest rate (percentage point)3.02.73.13.13.02.72.42.02.02.14.92.24.7
Real long-term rate (percentage point)2.12.02.42.62.72.62.21.81.51.34.23.43.9
Real effective exchange rate-15.2-3.03.33.32.72.11.81.21.92.3-0.13.9-1.4
Excess yield on yen assets (percentage point)-17.7-6.70.00.00.00.00.00.00.00.00.00.00.0
Total tax/GDP15.216.416.717.017.517.417.217.517.717.918.117.018.2
Total expenditure/GDP19.819.319.319.119.419.720.119.719.419.018.617.018.4
Debt/GDP16.018.520.321.522.624.226.428.329.530.127.024.627.2
Government deficit/GDP4.52.92.62.11.92.32.92.21.71.10.50.10.2
Current account surplus/GDP1.51.61.61.41.10.90.50.60.60.60.72.70.5
Trade surplus/GDP0.60.70.70.40.20.0-0.3-0.2-0.3-0.30.00.3-1.0
Real exports/GDP12.112.713.013.213.513.814.014.514.915.323.529.725.8
Real imports/GDP11.811.912.112.613.113.614.214.414.714.920.429.035.8
Net foreign assets/GDP17.418.818.718.418.017.717.116.916.516.112.042.028.3
Private investment/GDP20.621.021.321.621.421.121.020.820.620.416.915.316.2
Consumption/GDP59.859.859.759.759.860.060.260.761.061.665.967.667.7
Government spending/GDP19.018.518.318.318.518.919.218.718.618.317.316.917.1

The simulated level of the exchange rate in 1997 is about 15 percent below the trend implied by the historical experience. Two factors are at work. The first is the current cyclical position of the Japanese economy—the large output gap and the absence of inflationary pressures—which results in a low level of real interest rates in the initial years of the simulation. This, in turn, pushes down the initial level of the exchange rate, while at the same time increasing the rate of appreciation over the medium term. The second reason is the structural changes in Japan’s trade flows in recent years, which have boosted imports beyond the levels predicted by traditional trade equations. This shift in imports lowers the level of the exchange rate consistent with a given external balance, and thus lowers the equilibrium exchange rate for given determinants of saving and investment.

From its 1997 level, the real exchange rate appreciates at an average rate of about 2½ percent a year through the medium term, consistent with a continuing real interest differential between dollar and yen assets. While the yen appreciates at a rate similar to the historical trend, the level of the exchange rate remains well below trend over the medium term. This reflects, among other things, the assumed continuation of structural changes in Japan’s trade patterns referred to above.27

The level of real GDP gradually returns to potential over the medium term as cyclical factors unwind. The current account surplus, after rising slightly in 1997-98, narrows steadily to ½ of 1 percent of GDP by the end of the medium term. Given that output has returned to full employment, this external surplus corresponds to the “structural” saving-investment balance that has played a key role in the staff’s past approach to assessing exchange rate misalignment.

The value for the structural balance in the MULTIMOD results for 2002 is below that generated by the staff’s saving and investment equations referred to in the previous section. The main reason is the difference in real interest rates. In past work, the staff assumed that the real interest rate equaled its historical average level of 2½ percent. The simulation results, in contrast, determine the real interest rate based on the assumed yield on foreign assets, adjusted for the projected rate of yen appreciation beyond 2002. This approach generates a somewhat lower long-term real interest rate of about 2 percent of GDP. The lower interest rate, in turn, reduces the structural SI balance by about ½ of 1 percent of GDP.

Over the period to 2015, the yen continues to appreciate at just above 2 percent a year, while the current account averages slightly less than ½ of 1 percent of GDP. The fact that Japan’s current account remains in surplus into the next century stands in contrast to some other analyses, which have suggested that it would move into deficit as the aging population lowers japan’s saving rate.28 There are three main reasons why this result does not materialize here. The first reason is that the assumed effect of demographic changes on saving is smaller in this version of the model than some time-series evidence would suggest.29 The second reason is that private investment declines in relation to GDP as growth in the labor force and potential output slows. Finally, fiscal measures are assumed to be gradually implemented beyond 2002 that would prevent an unsustainable rise in government debt, so a rise in government saving partially offsets the drop in private saving over the longer term.

Over the period beyond 2015-2020, there are three distinct phases to the path of the yen. From 2020 to roughly 2030, the real value of the yen stabilizes, followed by a further sharp rise from about 2030 to 2045. Beyond 2045, the yen again stabilizes, and actually falls slightly in the last years of the simulation. Over the simulation period as a whole, the real value of the yen slightly more than doubles.30 As discussed in the next section, variations in the yen’s long-run growth rate are due primarily to demographic factors. In addition to a stabilization of these factors, the eventual leveling off of the yen’s value is also due to the assumption that the technological and other factors that have contributed to its trend increase in the past will decay over the future, consistent with the achievement of a steady-state growth path in the long run.

Effects of Alternative Assumptions

The above scenario is based on several key assumptions. For fiscal policy, for instance, no new policy initiatives toward consolidation during 1997-2002 are assumed beyond those contained in the FY (April-March) 1997 budget.31 It is also interesting to examine the role of structural factors, such as demographic developments, in determining the longer-run movements in the yen. Thus, several alternative simulations were performed. These involved changing the assumptions for medium-term fiscal policy; long-term demographic developments; long-term foreign growth rates; and the effects of time trends in the export price equation.

For fiscal policy, the alternative scenario assumes that medium-term consolidation plans are implemented to put the fiscal situation on a sustainable footing. As discussed in Okamura (Chapter 15, this volume), staff estimates suggest that measures amounting to about 4 percent of GDP would be needed to achieve this goal. To assess the impact of such a plan on the yen, both spending and taxes were adjusted in a way that divided the burden of deficit reduction roughly equally between these two components during 1998-2001.

The results for the exchange rate and the fiscal balance are shown in Figure 3.6. The initial depreciation of the yen of 8 percent in 1997 is sharper than in the baseline, as deficit reduction lowers aggregate demand and real interest rates. The subsequent appreciation is more rapid, however, as the effects of deficit reduction unwind and the exchange rate returns toward its baseline path.32 Over the medium term, the current account surplus stabilizes at 1½ percent of GDP, compared with ½ of 1 percent in the baseline. In the long run, with front-loaded fiscal consolidation, the exchange rate appreciates to a higher level. This is because a lower long-run level of public debt raises Japan’s net external assets and investment income, leading to a higher equilibrium level of the yen.

Figure 3.6.Medium-Term Fiscal Consolidation Scenario

Source: IMF staff estimates and projections.

To assess the impact of Japan’s long-run demographic changes on the baseline exchange rate path, two alternative scenarios were performed.33 The first assumes that there are no demographic changes beyond 2002—that is, both the level of the population and the dependency ratio are fixed at their 2002 levels. This is obviously unrealistic, but provides a useful basis for assessing the effects of the population changes that are actually anticipated. The other scenario is based on the older, 1992-based, population projections that the Ministry of Health and Welfare used in 1994 to assess the impact of demographic changes on public pension systems.34 These projections assumed a rapid rebound in the fertility rate, which has not materialized. Relative to the latest projections, the higher fertility rate raised the dependency ratio until about 2020 via a higher share of the young in the population: after 2020, the dependency ratio is lower, as the share of the working-age population is higher.

Before looking at the results, it is useful to summarize the channels through which demographic changes affect the exchange rate in the model. Consider the case of a permanent rise in the dependency ratio, holding the total population constant. Such a shift will (1) raise the level of consumption relative to income and (2) lower the size of the labor force and thus potential output. The increase in consumption puts upward pressure on the exchange rate in the near term, as higher spending raises the real interest rate, making domestic assets more attractive. Over the long run, however, higher spending lowers Japan’s stock of foreign assets and thus foreign asset income, which reduces both total spending and the equilibrium level of the yen. The reduction in the labor force and potential output causes the yen to appreciate at all horizons, as it reduces the supply of Japan’s output relative to foreign output. Given that Japanese and foreign output are imperfect substitutes in MULTIMOD, this “supply” effect drives up the relative price of Japanese output and thus the real exchange rate.

It is apparent from Figure 3.7 that, through these channels, demographic factors have an important impact on the exchange rate. In the absence of demographic changes beyond 2002, the large swings in the long-run growth rate of the yen would be almost entirely eliminated. At the same time, Japan would accumulate a huge stock of net foreign assets, amounting to about 140 percent of GDP. This accumulation occurs because fixing the population dependency ratio at its (relatively low) 2002 level would “freeze” Japan’s saving rate at a high level, leading to a continued accumulation of foreign assets. Notwithstanding this accumulation, the real exchange rate in this scenario is about 20 percent lower than in the baseline. This is due to the effect of unchanged demographics on the labor force and potential output—both would be about 40 percent higher by 2070 than under the baseline population projection, reducing the long-run level of the yen.

Figure 3.7.Alternative Demographic Scenarios

Source: IMF staff estimates and projections.

As expected, the results of using the old (1992) population projections are less dramatic. Until 2020, the exchange rate path is similar to the baseline, as the impact of a higher dependency ratio is offset by higher population growth. Beyond 2020, the exchange rate falls below baseline under the old population projections, as the dependency ratio declines, while the labor force rises about 20 percent above baseline.

In addition to demographic factors, two important secular factors in the model contribute to the long-run rise in the yen in the baseline. The first is the relatively high (though declining) rate of foreign output growth compared with Japanese growth, which reduces the relative price of foreign output given imperfect substitutability of national outputs. The second is the assumed continuation of the downward trend in Japan’s export prices relative to the GNP deflator, which allows the real value of the yen to rise over time without jeopardizing export competitiveness.35 Both of these factors are phased out gradually in the baseline—the first by 2050, and the second by 2030. To examine their role in contributing to the long-run rise in the yen, alternatives were run in which both trends were “turned off much earlier. Specifically, foreign output growth was set to 1½ percent from 2003 on, while the downward trend in export prices was also set to zero at this time.

The results of these simulations are shown in Figure 3.8. Not surprisingly, the yen rises less quickly when foreign output growth is reduced (top panel). In levels, foreign output ends up about 40 percent below baseline by the end of the long-term simulation, while the yen is about 20 percent below baseline. Similarly, turning off the downward trend in export prices lowers the path of the yen, although the effect is smaller (lower panel).

Figure 3.8.Alternative Assumptions About Foreign Growth and Export Price Time Trends

(Natural logarithm, 1990 = 0)

Source: IMF staff estimates and projections.

Looking at these simulation results and the demographic simulations together allows a decomposition of the factors that underlie the long-run appreciation of the yen in the baseline: slightly over one-third of the rise is due to the assumption of continued rapid growth in trade partners through the middle of the next century; about one-third is explained by the aging population in Japan and the associated drop in the labor force and potential output; and somewhat less than one-third is explained by a continued downward trend in Japan’s relative export price (i.e., Balassa-Samuelson effects).

Risk Premiums and Open Interest Parity

As noted in the discussion of the historical experience, open interest parity (OIP) has not typically held since Japan’s capital flows were liberalized in the early 1980s. While the real yield on yen and dollar assets has been similar, yen appreciation has meant that yen assets have provided a significantly higher return (roughly 2 percentage points on average) measured in a common currency.

There are two interpretations of this observation. One is that there has been an equilibrium “risk premium” on yen assets, making their ex ante expected return higher than on dollar assets.36 The other is that exchange market participants underestimated—or did not take into account at all—future trends in exchange rates in assessing the yields on assets denominated in different currencies. The assumption underlying the staff’s methodology, which uses the OIP condition as the “norm” for assessing the expectations-consistent path for the exchange rate, is that expectational errors are the main reason for the failure of OIP. (Indeed, the presumption that such errors can occur motivates the analytical assessment of the equilibrium exchange rate.)

Nevertheless, it is interesting to explore the effect of assuming a continuation of the historical excess return on yen assets. To answer this question, a simulation was performed where a constant “risk premium” was introduced into the exchange rate equation. This boosts the return on yen assets, adjusted for exchange rate movements, by 2 percentage points above that on dollar assets—a magnitude broadly consistent with the historical experience. The results are compared with the baseline in Figure 3.9. Not surprisingly, introducing a premium on yen assets lowers the initial level of the yen compared with the baseline, with the gap amounting to about 13 percent in the first year of the simulation. The lower yen boosts activity and prices in Japan, and raises real interest rates. As the initial increase in real interest rates is smaller than the 2 percentage point risk premium, the yen appreciates toward the baseline solution in the initial years of the simulation. Eventually, the risk premium becomes fully reflected in the real interest differential, and the yen stabilizes at a level slightly below the baseline.

Figure 3.9.Continuation of Historical “Risk Premium” on Yen Assets

Source: IMF staff estimates and projections.

Conclusions

This chapter has described a framework for assessing the equilibrium level of the yen and extended the staff’s past approach with a more sophisticated analysis based on long-run model simulations. Reassuringly, both approaches yield similar results for recent years. Specifically, the yen was somewhat overvalued in 1996 and significantly overvalued in 1995. The drop in the yen in late 1996 and early 1997 took it moderately below its estimated equilibrium level. These conclusions are based on the assumption of no further fiscal consolidation measures over the medium term. If, in contrast, a front-loaded consolidation plan is implemented, the equilibrium exchange rate would be lower in the short run, but more appreciated in the long run.

The simulations suggest that the upward trend in the real value of the yen will continue well into the next century, although at a more modest pace than in recent decades. The key factors explaining the continuing upward trend are (1) rapid growth in developing Asian economies; (2) continued Balassa-Samuelson effects; and (3) the effect of the aging population in lowering the labor force and potential output in Japan. As these factors all dissipate in the long run, the yen stabilizes. Two distinct phases of the aging population—early in the next century, and then toward the middle of the century—also lead to swings in the yen’s growth rate.

While the model-based results provide a useful check on the staff’s earlier, more partial, methodology, the use of full-scale models for this purpose is “work in progress.” As the sophistication of the approach increases, so do the information requirements and the sensitivity of the results to misspecification or inappropriate assumptions. Given these caveats, the present results should be considered suggestive. In terms of future directions for such work, an obvious and important extension would be to replicate the methodology for other countries to allow a global perspective on the impact of demographic changes and policy initiatives on exchange rates. Another refinement would be to revisit the specification of certain key blocks of MULTIMOD. The production sector, for instance, models Japan’s output as a single commodity—disaggregation would allow a richer analysis of factors such as deregulation and structural reform.

Finally, it would be interesting to incorporate “market learning” processes into the financial sector of MULTIMOD that would allow an analysis of how misalignments could be resolved over time. The current approach only describes the extent to which misalignments may exist, but not their dynamic implications for other variables in the model such as output, prices, and interest rates. This type of analysis might also provide the basis for examining how policies can best respond to misalignments.

Appendix I. Changes to MULTIMOD

The following changes were made to the Japan block in the simulation version of MULTIMOD described in Masson and others (1990) to adapt it to forecasting purposes:

  • Human and capital wealth: In the original version of the model, the (unobserved) levels of human and capital wealth are defined as the discounted values of future simulated earnings streams. It proved difficult to use this specification for baseline forecasting, as unique historical levels for these variables are not defined. Instead, wealth was defined using current flows for income streams adjusted for the expected rate at which they would return to their “potential” levels. This is similar to the approach used in an earlier version of MULTIMOD (see Masson and others, 1988).

  • Disaggregation of deflators: MULTIMOD has only one deflator for all components of domestic absorption. Yet there are significant historical trends in the growth rates of different components of spending—investment deflators, for instance, have tended to fall over time, relative to consumption prices. To deal with this issue, deflators for the components of absorption were disaggregated and time trends were introduced into the equations explaining their growth to allow for differences in secular growth rates.

  • Trade parameters and trends: The trade equations in the original model were estimated using pooled data for the major industrial countries. In some cases, the parameters differed significantly from those generated by Japan-specific trade models (see, for instance, Chadha, 1996). In addition, certain equations did not satisfy the long-run homogeneity properties required for a steady-state solution. To resolve this problem, the trade block was reestimated using only Japanese data from 1970-95, and long-run homogeneity was ensured by imposing appropriate parameter constraints.

  • Demographics and consumption: In the original version of the model, the population dependency ratio had a strong impact on consumption and saving, based on time-series estimation. More recent work, including that by Meredith (1995), suggests a smaller effect. Based on these results, the parameter on the demographic ratio in the consumption function was lowered such that a 1 percentage point rise in this ratio lowers the private saving rate by about ⅓ of 1 percentage point.

  • Reaction function for monetary policy: The reaction function in the original version of the model was modeled as a money supply rule, where nominal interest rates adjust in response to deviations in the money stock from target. The properties of this rule proved problematic for forecasting purposes, and it was replaced by a function that causes the real short-term interest rate to adjust in response to deviations in both the level and the growth rate of nominal GDP from a fixed target.

  • Inflation equation: The linear relationship between the output gap and inflationary pressures was replaced by the nonlinear relationship described in Laxton and others (1995). In this formulation, positive output gaps tend to raise inflation by more than negative gaps lower it, an asymmetry that seems consistent with the Japanese experience during the recent economic cycle.

Appendix II. A Steady-State Version of the Exchange Rate Model

As discussed in the main text, a steady-state version of the dynamic exchange rate model was constructed to derive “terminal conditions” for the forward-looking variables. This was done by transforming leads and lags on the variables in the dynamic model into their contemporaneous values, using the appropriate steady-state growth rate for each variable. These steady-state growth rates, in turn, were based on the growth rates assumed for the exogenous “driving” variables in the model.

As a stylized example of the procedure, consider the following dynamic equation for the endogenous variable x, where z is exogenous (both are measured in logarithms):

In the steady state, x will grow at the same rate as z, given the homogeneity imposed in the parameter values.37 As the variables are measured in logarithms, this growth rate corresponds to Zt—Zt-1 represented as λ. Then the lag and lead on x in the above equation can be substituted out as follows:

implying that:

In this way, the steady-state model can be expressed in terms of only the contemporaneous values of the variables and their steady-state growth rates. Simulating this static model then yielded the terminal conditions that were necessary for the dynamic simulations.

The steady-state growth rates themselves were determined as follows. Real growth for the Japanese variables is tied down in the steady state by the exogenous rate of labor-augmented productivity growth, which is assumed to equal 1½ percent a year, similar to the relatively constant growth rate that has been observed since the early 1980s. The inflation rate is determined by the assumed target for monetary policy, which has been set to ½ of 1 percent a year for the GDP deflator, equal to the rate at which inflation is projected to stabilize over the WEO medium-term horizon. Foreign real growth converges to the same 1½ percent rate assumed for Japan by the middle of the next century (a necessary condition for a steady-state to exist), while foreign inflation is constant at 1½ percent a year.

References

    BalassaBela1964“The Purchasing Power Parity Doctrine: A Reappraisal,”Journal of Political EconomyVol. 72.

    ChadhaBankim“External Adjustment in Japan: Recent Developments and the Medium-Term Outlook,” in Japan—Selected Issues IMF Staff Country Report 96/114 (Washington: International Monetary Fund) pp. 15072.

    ClarkPeter B.LeonardoBartoliniTamimBayoumi and StevenSymansky1994Exchange Rates and Economic Fundamentals: A Framework for Analysis IMF Occasional Paper 115 (Washington: International Monetary Fund).

    HoriokaCharles Yuji1990“Future Trends in Japan’s Saving Rate and the Implications Thereof for Japan’s External Imbalance,”Discussion Paper 182 (Osaka: Osaka University Institute of Social and Economic Research) pp. 155.

    IsardPeter and StevenSymansky1996“Long-Run Movements in Real Exchange Rates” in Exchange Rate Movements and Their Impact on Trade and Investment in the APEC Region IMF Occasional Paper 145 (Washington: International Monetary Fund).

    ItoTakatoshi1996“Japan and the Asian Economies: A ‘Miracle’ in Transition,” in Brookings Papers on Economic Activity: 2 (Washington: Brookings Institution).

    JorgensonDale W. and MasahiroKuroda1990“Productivity and International Competitiveness in Japan and the United States, 1960–1985,” in Productivity Growth in Japan and the United Statesed. by Charles R.Hulten (Chicago: University of Chicago Press).

    LaxtonDouglasGuyMeredith and DavidRose1995“Asymmetric Effects of Economic Activity on Inflation,”Staff Papers International Monetary FundxsVol. 42 (June) pp. 34474.

    LothianJames R.1990“Century Plus of Yen Exchange Rate Behavior,”Japan and the World EconomyVol. 2No. 1 pp. 4770.

    MassonPaulStevenSymanskyRichardHaas and MichaelDooley1988“MULTIMOD—A Multi-Region Econometric Model” in Staff Studies for the World Economic Outlook World Economic an Financial Survey Series (Washington: International Monetary Fund).

    MassonPaulStevenSymansky and GuyMeredith1990MULTIMOD Mark II: A Revised and Extended Model IMF Occasional Paper 71 (Washington: International Monetary Fund).

    MassonPaul and RalphTryon1990“Macroeconomic Effects of Projected Population Aging in Industrial Countries,”Staff PapersInternational Monetary FundVol. 37 (September) pp. 45385

    MeredithGuy1995“Demographic Change and Household Saving in Japan” in Saving Behavior and the Asset Price “Bubble” in Japan: Analytical Studies IMF Occasional Paper 124 (Washington: International Monetary Fund).

    Ministry of Health and Welfare Japan1992Population Projections for Japan: 1991-2090 institute for Population Problems Research Series 274 (Tokyo: Ministry of Health and Welfare).

    OkamuraKenji1998“Japan’s Medium- and Long-Term Fiscal Challenges,” in Structural Change in Japan: Macroeconomic Impact and Policy Challengesed. by BijanAghevliTamimBayoumi and GuyMeredith (Washington: International Monetary Fund) pp. 21539.

    SamuelsonPaul1964“Theoretical Notes on Trade Problems,”Review of Economics and StatisticsVol. 64.

Developments in the yen during the pre-1939 period are discussed in Lothian (1990).

More generally, the experience of a range of East Asian economies yields little evidence that productivity catch-up explains trends in real exchange rates (see Isard and Symansky, 1996).

Real interest rates are measured as long-term government bond yields less CPI inflation. The trend real appreciation of the yen versus the dollar has been similar in magnitude to the trade-weighted appreciation of 3 percent a year.

To simplify the exposition, there is assumed to be only one interest rate, the maturity of which corresponds to the time horizon over which exchange rate expectations are formed.

The determination of expected exchange rates is discussed below.

More generally, the current account will also depend on interest rates through investment income Rows, and the SI balance may be affected by the real exchange rate through terms-of-trade effects. Introducing these refinements would not change the thrust of the analysis.

This “real” interest parity condition can he derived by subtracting expected inflation from both sides of the nominal interest parity condition. In the more restrictive case of static real exchange rate expectations, rere = rer and the domestic interest rate equals the world rate.

The two-dimensional treatment is possible because the external balance drops out of equation (3΄) under perfect capital mobility. With the more general version of equation (3), a third dimension is necessary to show the solution, with the third axis representing the external balance.

More generally, Balassa-Samuelson effects could generate a trend in the long-run real exchange rate. This would not alter the thrust of the story—there would then be a fixed gap between the domestic and world real interest rates given by exogenous technological factors, and the long-run exchange rate would still be solved for recursively from equations (1) and (2).

In this sense, the current analysis introduces a dynamic element to the IMF’s earlier work on long-run equilibrium exchange rates (see Clark and others, 1994).

A permanent fiscal expansion would have more complicated effects: the interest differential would rise, while the long-term equilibrium exchange rate would fall due to rising public and external debt and thus higher debt-servicing burdens.

This level of the current account does not equal the staff’s WEO forecast. As discussed below, WEO forecasts are based on the exogenous assumption that real exchange rates remain unchanged at their observed market levels.

Nominal interest rates on five-year government bonds are adjusted for expectations of inflation based on published “consensus” forecasts.

Specifically, it was estimated that the current account would have reached about 1½ percent of GDP by 2002, while the structural SI balance was estimated at about 1 percent, suggesting the yen was undervalued by about 7 percent. The decline in market interest rates since 1995 tended to lower the estimated equilibrium level of the yen, providing an example of how this approach incorporates the effects of cyclical developments and monetary policy actions.

See Masson, Symansky, and Meredith (1990) for a description of MULTIMOD.

Most of the equations in the original version of MULTIMOD were estimated using pooled data with common parameters across the major industrial countries.

In contrast to the money-target rule used in the original version.

In addition to the exchange rate, these include prices and interest rates.

The validity of the steady-state assumption can be informally checked by inspecting the paths of the forward-looking variables for “anomalous” behavior toward the end of the simulation horizon. Inspection of the dynamic simulation results suggested that the model had indeed approached its steady-state solution by the end of the simulation.

The WEO baseline used in this analysis refers to the version prepared in March 1997.

This assumption about the external real interest rate is somewhat arbitrary. Yet, as it reflects the links between world capital markets and the Japanese economy, it is a key channel through which factors such as foreign saving and investment developments affect the path for the yen. As discussed below, extending the current framework to the rest of the world is an important direction for future research.

Beyond 2002, these add factors were specified to either gradually decay or were, in some cases, held constant, where it was judged that there was a permanent shift in the relationship compared with the historical estimates.

This is the exact reverse of the causation that underlies the staff’s past approach to measuring misalignment, as previously discussed.

This definition of the real exchange rate is based on relative GNP deflators. Historical trend growth is about 2½ percent a year.

Combined with an excess return on dollar assets of 17 percent in 1996 relative to the 1995 level, this implies that the exchange rate was overvalued by almost 25 percent in 1995.

See Chadha (1996) for a discussion of the staff’s outlook for Japan’s trade flows.

Horioka (1990) provides a useful survey of studies on this issue.

The effect assumed in this model is based on the analysis in Meredith (1995). This study compares the time-series evidence on the effect of demographics on saving with the results of a stylized life-cycle model for Japan, and explains why the lower effect found in the latter framework may provide a more appropriate benchmark.

Compared with the threefold increase that occurred from the 1950s to the 1990s.

Beyond 2002, however, MULTIMOD has a fiscal reaction function that forces taxes and spending to adjust such that the debt-to-GDP ratio converges to a fixed target level.

The effects of aggressive deficit reduction on aggregate demand unwind because, even in the baseline scenario, fiscal policy is forced to eventually act to prevent an explosive rise in public debt. In this sense, the fiscal consolidation scenario simply “front loads” the adjustments that would ultimately be needed.

For earlier work on demographics using MULTIMOD, see Masson and Tryon (1990).

The export deflator has declined at an average rate of 3½ percent a year relative to the GDP deflator over the historical period.

The term “risk premium” is used generically here to describe the impact of several possible factors on relative yields. For instance, yen and dollar assets could be imperfect substitutes due to regulatory restrictions on investors’ portfolio allocation or tax considerations.

As these homogeneity conditions did not universally apply in the original version of MULTIMOD, adjustments were necessary to a few relationships to ensure that a long-run steady-state solution existed.

    Other Resources Citing This Publication