This paper provides a brief overview of the causes of the poor economic performance of Japan in the 1990s, and a more detailed analysis of developments in the real sector during 1999. The paper highlights that the collapse of the asset price bubble in 1990–91 provided the trigger for the downturn in 1992, and compounded the economic problems thereafter through its effects on the banking system. This paper also analyzes the fiscal policy developments and the monetary developments in Japan.

Abstract

This paper provides a brief overview of the causes of the poor economic performance of Japan in the 1990s, and a more detailed analysis of developments in the real sector during 1999. The paper highlights that the collapse of the asset price bubble in 1990–91 provided the trigger for the downturn in 1992, and compounded the economic problems thereafter through its effects on the banking system. This paper also analyzes the fiscal policy developments and the monetary developments in Japan.

Japan: Selected Economic Indicators, 1992–2000

Nominal GDP: US$3,783 billion (1998)

Population: 126.3 million (1998)

GDP per capita: US$29,937 (1998)

Quota: SDR 8,241.5 million

article image
Sources: Nikkei Telecom; and staff estimates and projections as of July 12, 1999.

Contribution to GDP growth.

Includes provision for a FY 1998 Supplementary Budget.

Based on normalized unit labor costs; 1990=100.

I. REAL SECTOR DEVELOPMENTS1

A. Introduction

1. The Japanese economy is currently experiencing its worst post-war economic recession. Output contracted significantly for five consecutive quarters before growth turned positive in the first quarter of 1999. The recent declines in output have, moreover, followed on the heels of a lengthy period of slow growth—real GDP grew at an average rate of just about 1¼ percent per annum between 1991–98, in contrast to a growth rate of about 4 percent during the 1980s. Japanese economic performance during this decade has also fared unfavorably from an international perspective. Not only has output grown more slowly in Japan than in other major industrial countries, but the volatility of output in Japan rose during the 1990s, whereas it declined significantly in the other G7 countries (Figure I.1). This chapter provides a brief overview of the causes of the poor economic performance in the 1990s, and a more detailed analysis of recent developments in the real sector.

FIGURE I.1
FIGURE I.1

JAPAN: GROWTH AND VOLATILITY COMPARISONS

Citation: IMF Staff Country Reports 1999, 114; 10.5089/9781451820584.002.A001

Sources: IMF, World Economic Outlook database; and staff estimates.1/ For Germany, data start in 1991Q2 to exclude reunification effects.

2. The collapse of the asset price bubble in 1990–91 provided the trigger for the downturn in 1992, and compounded the economic problems thereafter through its effects on the banking system. The asset price collapse impacted on economic activity through a variety of channels, particularly the adverse impact on private investment from an increased cost of capital and lower net worth, although as discussed below, the adverse wealth effects on private consumption appear to have been small. The asset price collapse also created problems for the banking system through two distinct channels; not only did it impair loan collateral, but it also eroded bank capital directly as hidden profits on equity holdings were sharply reduced.2 The resulting fragile state of the banking system, left essentially unresolved for a number of years following the collapse of the asset price bubble, culminated in a full-fledged banking crisis during 1997–98, and forced the commitment of large-scale public funds to address the problem. The credit crunch and the financial distress that accompanied the banking crisis in this period, together with the contractionary stance of fiscal policy in 1997, contributed to the recent episode of sharply declining activity (Table I.1).

Table I.1.

Japan: Growth of Real GDP and Demand Components, 1993–99 1/

(Percent change from the previous period)

article image
Sources: Nikkei Telecom; WEFA; and staff estimates.

At 1990 prices.

Final private domestic demand is the sum of private consumption, residential investment, and business fixed investment. Final domestic demand is final private domestic demand plus government consumption and investment.

Contribution to real GDP growth.

Government consumption and public investment.

3. The lingering negative impact from the bubble period has also been related to the need to unwind the excess capital stock inherited from the years of “overinvestment” to more normal levels.3 As discussed further below, a brew of high growth expectations, easy access to credit, and a buoyant equity market had spurred business investment to unsustainable levels in the latter half of the 1980s. With these favorable conditions dissipating with the bursting of the asset price bubble, the negative impact of the increased cost of capital on business investment was compounded by the need to dispense with excess capacity.

4. A snapshot of real sector developments in the 1990s—captured in Figure I.2—reveals the following broad trends. The decline in private investment—both residential and non-residential—was precipitous following the collapse in asset prices. After what turned out to be a transient pick up in 1996, related in part to new opportunities created by deregulation, private investment declined sharply yet again. Public investment, which has served as the main ingredient of counter-cyclical fiscal policy in the nineties, filled in to a considerable extent the slack left by private investment. Private consumption grew more slowly in the aftermath of the collapse of asset prices; significant declines in the level of private consumption have, however, been a more recent phenomenon, occurring mainly during 1997–98. The external sector’s contribution to growth (both positive as well as negative) has been under 1 percentage point during the course of the 1990s, except in 1997, when it contributed about 1½ percentage points to growth.

FIGURE I.2
FIGURE I.2

JAPAN: GROSS DOMESTIC PRODUCT AT 1990 PRICES, 1990–99

Citation: IMF Staff Country Reports 1999, 114; 10.5089/9781451820584.002.A001

Sources: Nikkei Telecom; WEFA; and staff estimates.

5. The recession intensified in 1998, before activity picked up in the first quarter of 1999. GDP declined by almost 3 percent in 1998—the steepest annual decline registered in the post-war period—as all the main components of private domestic demand fell sharply. However, a surge of public investment starting in the fourth quarter of 1998, together with steps to ease financial constraints, resulted in activity reviving sharply by 1.2 percent in the first quarter of 1999. There has also been a recovery in confidence in the first part of 1999, reflecting a sharp rise in equity prices and improving indicators of corporate sentiment. Nevertheless, the staffs forecast for 1999 posits only moderate growth, as the stimulus from public investment is expected to wear off later in the year, private domestic demand is projected to be weak during the rest of this year, as continued efforts at corporate restructuring continue to weigh against a recovery.

B. Components of Aggregate Demand

Private Consumption

6. Private consumption has fallen sharply since the onset of the recession in April 1997. Despite the increase of 1¼ percent in the first quarter of 1999, the level of private consumption expenditure is still about 3¼ percent lower than it was in the first quarter of 1997. There are two main puzzles concerning trends in private consumption in Japan, Why did private consumption not fall steeply following the asset price collapse in 1991–92? Why has private consumption declined recently?

7. The relative stability of the private savings ratio in the aftermath of the asset price collapse in the early 1990s reflects in part the reluctance of households in Japan to build up debt during the boom years. Figure I.3 indicates that the liabilities of the household sector did not increase substantially in relation to personal disposable income during the latter half of the 1980s, and this ratio has continued to be stable during the nineties.4 The ratio of net worth to disposable income did fall significantly between 1990–92. However, to the extent that declines in net worth reflected the declining price of land, the impact on private consumption is likely to have been muted, as favorable price substitution effects are likely to offset the negative wealth effects, particularly in view of the relatively low levels of home ownership in the six largest cities in Japan (about 50 percent). Household direct holdings of equity were relatively low, and the impact of declining equity prices on bank balance sheets and the true value of pension and insurance funds was probably not fully perceived because of weak accounting standards. That is, the interaction between wealth effects and personal indebtedness which proved to be an important driving force of the instability in private consumption in a number of countries that experienced a boom-bust cycle in asset prices appears to have been much less of a negative force in the case of Japan.

FIGURE I.3
FIGURE I.3

JAPAN: TRENDS IN PRIVATE CONSUMPTION

Citation: IMF Staff Country Reports 1999, 114; 10.5089/9781451820584.002.A001

Sources: Nikkei Telecom; WEFA; and staff estimates.

8. The increase in the consumption tax was a proximate cause of the decline in private consumption in the second quarter of 1997, but the continued declines in private consumption thereafter reflected the effects of financial distress and uncertainties about job prospects. The increase in the consumption tax in April 1997 had been pre-announced, and the 5 percent decline in private consumption in the second quarter of 1997 reflected a significant degree of intertemporal substitution. For instance, private consumption expenditure increased by 3½ and 1½ percent in the preceding and subsequent quarters respectively. The declines in private consumption in the fourth quarter of 1997, and again through most of 1998 overlaps with a period of financial instability set in motion by the failure of some major banks and securities houses. The sharp increase in unemployment and the decline in wage earnings in 1998 (discussed below) engendered a climate of uncertainty about job prospects and also contributed to the cutting back of private consumption expenditure during this period.

9. The rebound in consumption in early 1999 may have reflected receding concerns about financial system instability following the commitment of public funds. However, it is too early to judge whether the recent increase in private consumption marks an effective break from previous trends. The continuing uncertainties about job prospects as corporate restructuring proceeds and the recent declines in earnings suggest that private consumption is likely to remain muted for a while to come.

Business Investment

10. The recent decline in business investment has been more precipitous than during any other episode in Japan’s post-war history. Business investment declined for 6 consecutive quarters before being interrupted by positive growth in the first quarter of 1999. Even with the recent increase, business investment declined by almost 20 percent between the first quarter of 1997 and the first quarter of 1999 (Figure I.4).

FIGURE I.4
FIGURE I.4

JAPAN: INDICATORS OF BUSINESS ACTIVITY AND INVESTMENT, 1985–99

Citation: IMF Staff Country Reports 1999, 114; 10.5089/9781451820584.002.A001

Sources: Nikkei Telecom; WEFA; and staff estimates.

11. The recent weakness of business investment has continued a trend decline in the rate of business investment in the 1990s, which in turn was driven by the effects of the unwinding of the capital stock overhang, the impact of the debt burden, and the collapse of equity prices. The share of business investment in GDP fell from a high of about 20 percent in 1990 to about 16 percent in 1998, as the expectations of high growth that had fueled the rapid build up of capacity in the latter half of the 1980s were revised down persistently during the 1990s. Many of the large firms that had earlier tapped into the buoyant equity market to finance their investments faced a drastic reduction in access to relatively cheap sources of funding when equity prices collapsed. The burden of high debt also impeded the capacity of many firms to undertake new investments, and the banking crisis in 1997–98 exacerbated these problems5.

12. The recent slump in business investment also reflected the effects of tighter credit conditions on the capital expenditures of small and medium sized enterprises (SMEs). While business investment declined across the board during 1997–98 in response to declining profitability and balance sheet fragilities, SMEs have borne the bulk of the credit crunch accompanying the banking crisis during this period. The authorities responded to the credit crunch facing small enterprises in October 1998 by allocating ¥ 20 trillion for guaranteeing the credit extended to small and medium-sized enterprises by financial institutions. The positive growth of business investment in the first quarter of 1999 seems to have reflected largely the capital expenditures undertaken by small and medium sized enterprises following their easier access to credit. Many large enterprises are, however, still in the process of dismantling excess capacity as part of the efforts to restructure the corporate sector, and recent monthly data on capital equipment orders and surveys of the capital expenditure plans of businesses indicate that investment is likely to decline in the months ahead.

Residential Investment

13. The decline in residential investment in recent years has been even more severe than the decline in business investment. By the fourth quarter of 1998, residential investment had fallen by about 35 percent from its peak in the fourth quarter of 1996. As in the case of private consumption, the sharp declines in residential investment during 1997–98, despite the decline in long-term yields during this period, were linked to the uncertainties engendered by the fragilities in the financial sector. Moreover, residential land prices have fallen continuously since early 1991, discouraging investment in housing (Figure I.5).

FIGURE I.5
FIGURE I.5

JAPAN: RESIDENTIAL INVESTMENT AND INVENTORY INVESTMENT

Citation: IMF Staff Country Reports 1999, 114; 10.5089/9781451820584.002.A001

Sources: Nikkei Telecom; and staff estimates.

14. The increase in residential investment by 1¼ percent in the first quarter of 1999 has been in part a response to measures in the budget for fiscal year 1999 to stimulate the housing market—tax credits for housing loans were extended from 6 to 15 years, and the ceiling on loans that could be qualified for tax credits was increased from ¥30 million to ¥50 million. There were also one-off factors driving the first quarter increase in residential investment this year, as housing starts were accelerated to take advantage of low lending rates available through March 1999—and which were expected to increase later in the year. However, housing starts declined in April–May, suggesting that this rebound may not be sustained.

Public Investment

15. Public investment has been an important ingredient of counter-cyclical fiscal policy in the 1990s, and has provided significant support to activity recently. As discussed in greater detail in the Selected Issues paper, each of the stimulus packages implemented since 1992 have included increased public works spending as a major component. As the impact of the September 1995 package wore off, however, public works spending turned sharply negative from mid 1996 and continued to decline through 1997, which had the effect of compounding the adverse effects on activity arising from the full-fledged breakout of the financial crisis in late 1997. The authorities responded to the recent weakening of activity by implementing fiscal stimulus packages of ¥16 trillion in April 1998, and ¥23 trillion in November 1998. The combined public works component in these two stimulus package was about ¥16 trillion (about 3 percent of GDP), and public investment has increased sharply by around 20 percent from the third quarter of 1998 to the first quarter of 1999. However, public works spending is likely to drop in the second half of FY 1999 once the bulk of the spending associated with the November 1998 fiscal package has fed through, unless additional measures are taken.

Inventories

16. Destocking has been a source of negative impulse to the economy over the last two years. After a rapid build up of inventories in 1996—which contributed almost ½ of a percentage point to growth in that year—destocking has proceeded essentially unabated, with inventories contributing negatively to growth for five consecutive quarters, before turning flat in the first quarter of 1999. As a result, the inventory-shipment ratio has now fallen below trend, suggesting that the down-phase of the inventory cycle may be coming close to having run its course (Figure I.5).

The External Sector

17. The external sector’s contribution to growth has turned negative following the sharp appreciation of the yen in the second half of 1998. Despite the crisis in Asia, the -external sector contributed about 1½ percentage points to growth in 1997, given the combination of a weak exchange rate and shrinking domestic demand (Figure I.6). The external sector continued to provide support to the economy through the third quarter of 1998, as the exchange rate continued to weaken, reaching a trough of ¥145/$ in August. Over this period, the impact of the downturn in Asia was offset to some extent by continued strength of demand elsewhere, particularly in North America.

FIGURE I.6
FIGURE I.6

JAPAN: EXTERNAL SECTOR DEVELOPMENTS, 1990–99

Citation: IMF Staff Country Reports 1999, 114; 10.5089/9781451820584.002.A001

Sources: Nikkei Telecom; WEFA; and staff estimates.

18. As the global financial turbulence in this period resulted in an abrupt unwinding of the yen-carry-trade positions that had been building up for some time, the exchange value of the yen appreciated sharply, reaching about ¥110/$ in early January of this year. Following the easing of monetary policy and exchange market intervention, the yen has subsequently moved into a trading range of about ¥118–122/$ in recent months. Thus, the external sector’s contribution to growth in the first quarter of 1999 was negative 0.4 percentage points, as imports registered the first increase in eight quarters and exports continued to fall.

C. Labor Market and Price Developments

The Labor Market

19. Despite the significant slowing of the Japanese economy in the 1990s, unemployment remained low until last year, when it rose sharply. The average rate of unemployment between 1991–97 was less than 3 percent, just slightly higher than the average unemployment rate during the boom years of the 1980s. However, unemployment started to rise sharply from early 1998 onwards, reaching a high of 4.8 percent in April 1999. Two questions arise naturally in this context. Why did unemployment remain low until last year? Why has it risen sharply recently?

20. The main reason for the low rates of unemployment between 1991–97 appears to be rooted in the reluctance of many firms to give up on life time-employment practices despite declining sales and profitability. As the state of the economy worsened with the banking crisis coming to a head in 1998, the need to change existing labor market practices began to be perceived, with large firms, which have been the main practitioners of life time-employment practices, shedding labor sharply in recent months (Figure I.7).6 The decline in the growth of the labor force also offers a partial explanation for the low rates of unemployment through much of the 1990s; however, the magnitude of the decline in the growth of the labor force—from an average of about 1.2 percent in the 1980s to about 0.9 percent in the 1990s—is small when viewed in relation to the magnitude of the slowdown in output growth.

FIGURE I.7
FIGURE I.7

JAPAN: LABOR MARKET CONDITIONS, 1990–99 1/

Citation: IMF Staff Country Reports 1999, 114; 10.5089/9781451820584.002.A001

Source: Nikkei Telecom and WEFA.1/ Seasonally adjusted data.

21. Employment trends suggest that the labor market is currently in a more severe state than is indicated by the recent increase in the unemployment rate. Employment declined by about 1¼ percent in the year to the first quarter of 1999, but the rise in unemployment in this period was capped to some extent by discouraged workers moving out of the labor force. The manufacturing sector has incurred the largest job losses recently, with employment in May 1999 being about 5 percent below that in January 1998. Employment in construction also declined in this period, though not as steeply as in manufacturing. Employment in services, in contrast, has been broadly stable since early 1998; but the stability of the overall figure masks structural shifts within the service sector (Figure I.7). There has also been a sharp increase in the proportion of part time employment in recent years, reflecting perhaps the nature of the new jobs being created in the service sector.

22. Along with the rise in unemployment, there is emerging evidence of potential mismatches in the labor market. For instance, while the ratio of job offers to job seekers has declined sharply from 0.68 in early 1998 to a record low of 0.46 in May 1999, the ratio of job offers in new activities to job seekers has been broadly stable during this period, indicating that employers in certain types of activities may be finding it difficult to recruit suitably qualified workers.

23. The deteriorating labor market situation has taken its toll on earnings. Earnings grew on average by about 1¾ percent during 1991–97, and decelerated modestly to about 1¼ percent on a year-on-year basis in 1998. However, the average growth of earnings in 1998 masks the considerable deterioration that has occurred recently, as earnings declined by 2¼ percent in the year to the fourth quarter of 1998, and by 3¾ percent in the first quarter of this year. The recent declines in earnings have been accounted for mainly by steep declines in overtime pay and bonuses (Figure I.7).

24. The authorities have recently proposed a number of measures to alleviate strains in the labor market. These measures include support for companies that take on new employees, the provision of training to deal with mismatches in the labor market, and efforts by the public sector to take on more employees, particularly in the fields of health and vocational training. While these measures should be useful in ameliorating conditions in the labor market to some extent, accelerated restructuring will inevitably create further strains in the labor market. Dealing with these problems would require further reforms to encourage labor mobility—for example tax changes to make pensions more portable and eliminations of restrictions on temporary employment—and steps to strengthen the social safety net—including allowing benefits to be drawn over longer periods, which would help in alleviating the social costs associated with restructuring.

Prices

25. Consumer price inflation fell sharply in the immediate aftermath of the collapse of the asset price bubble, but has remained broadly flat thereafter. Consumer price inflation fell from about 3¼ percent in 1991 to under 1 percent in 1994, and has been on average under 1 percent during the period 1992–98. Underlying inflation, which excludes the prices of perishable food and energy from the CPI, has broadly tracked the trends in overall CPI (Figure I.8). During the year after the sales tax hike in April 1997, the 12-month changes in the CPI (and to some extent the WPI) were distorted. However, price declines intensified once this one-off effect dropped out in early 1998. The wholesale price index plummeted by 4 percent in the year to the first quarter of 1999, after having fallen by about 3½ percent in the previous quarter. However, the recent sharp declines in the WPI appear to be largely on account of falling import prices, and the domestic WPI fell by about 2 percent in the year to the first quarter of 1999, not much different from the declines that took place during 1998. Both the CPI and the underlying CPI have been essentially unchanged on a 12-month basis in April 1999.

FIGURE I.8
FIGURE I.8

JAPAN: PRICE INDICATORS, 1990–99 1/

Citation: IMF Staff Country Reports 1999, 114; 10.5089/9781451820584.002.A001

Source: Nikkei Telecom and WEFA.1/ Seasonally adjusted data.

26. The absence of stronger deflationary pressures in Japan is a puzzle that is yet to be fully resolved. The rigidity of nominal wages provides a partial answer, though a less convincing one for recent months, given the recent decline in overall earnings. Product market rigidities, particularly in the services sector, also provide a partial explanation for why prices have not fallen more sharply recently. It is notable that private service sector prices have continued to rise even as product prices have fallen over the past year. Sticky inflation expectations have probably been a factor, with surveys of price expectations continuing to show positive inflation expectations despite the fact that prices have been falling.

27. It should be recognized that measurement biases in calculating the CPI imply that even consumer prices in Japan have been falling in recent years. A number of recent studies indicate that the consumer price index may be overstating inflation in many countries.7 This upward bias arises because the methodology for calculating the CPI does not adequately capture shifts in the pattern of spending by consumers when relative prices change, and does not fully account for the effects of changes in quality or the introduction of new products on prices. Moreover, consumers in many countries have recently been shifting their purchases away from established retailing channels toward discount stores. Shiratsuka (1999) has estimated that the consumer price index in Japan has been overstating inflation by about 1 percent.8 The outlet substitution bias, in particular, could be significant in Japan, as the methodology for calculating the CPI does not track purchases at discount stores in suburban areas; it misses special weekend-only sales; and it tracks only designated brands, missing lower prices on competing brands.

D. National Income Accounts

28. Interpreting economic developments in Japan is rendered more difficult on account of weaknesses in the National Accounts. There are, for instance, no “quick estimates” for the income side of the national accounts, with data being available only after a lag of about one year. Consequently, making judgements, for instance, about how private consumption will evolve during the latter half of this year is rendered more difficult than it otherwise would be, given that the data on private savings are available only up until the first quarter of 1998 (which is why the private savings ratio in Figure I.3 is truncated at that point). Moreover, the GDP data on some key components of demand, such as the breakdown of private consumption into durables and nondurables, and business investment into structures and equipment are also available only after a long lag. Data on general government activity—outside public investment—is also not available until nine months after the end of the fiscal year. The methodology used to construct quick estimates of the quarterly GDP from the underlying data is not well understood outside of the Economic Planning Agency (EPA).

29. The EPA has recently taken a number of measures to address some of the shortcomings in the National Accounts.9 Beginning in the first quarter of 1999, the EPA will, for a test period of about one year, release the “flash estimates” (calculated approximately a month after the end of the relevant quarter) together with the preliminary estimates for GDP—released usually about 70 days after the end of the relevant quarter—so that the public can be made aware of the differences in results. The flash estimates themselves will not be made available to the public at the time that they are first calculated during the test period. The EPA has also set up a committee to strengthen other aspects of the National Accounts, such as improving the statistical methodology used in computing the national accounts and enhancing the resources available to the SNA authorities, and has recently made efforts to increase public understanding of the methodology used in compiling the National Accounts.

References

  • Bayoumi, Tamim, “The Morning After: Explaining the Slowdown in Japanese Growth in the 1990s”, IMF Working Paper 99/13, 1999.

  • Economic Planning Agency, Handling of ‘Flash’ Estimates of Quarterly GDP, (Tokyo: EPA), May, 1999.

  • International Monetary Fund, World Economic Outlook, (Washington: International Monetary Fund), May 1997.

  • Ramaswamy, Ramana, and Christel Rendu, “Japan’s Stagnant Nineties: A Vector Autoregression Retrospective”, IMF Working Paper 99/45, 1999.

    • Search Google Scholar
    • Export Citation
  • Shiratsuka, Shigenori, “Measurement Errors and Quality Adjustment Methodology: Lessons from the Japanese CPI”, Economic Perspectives, Federal Reserve Bank of Chicago, Second Quarter, 1999.

    • Search Google Scholar
    • Export Citation
1

Prepared by Ramana Ramaswamy (ext. 38591).

2

Japanese banks have significant direct equity ownership in the corporate sector, and have counted part of the capital gains on equity ownership as bank capital.

3

A detailed analysis of Japanese economic performance during the 1990s is provided in Bayoumi (1999), and Ramaswamy and Rendu (1999).

4

The household debt ratio in Japan increased by about 25 percent during 1985–90. By contrast, the household debt ratio in Sweden—which had experienced an asset price boom broadly comparable to that of Japan during the latter half of the eighties—increased by about 40 percent over this period.

5

A more detailed discussion of the trends in business investment in the 1990s is provided in the Selected Issues Paper.

6

The accompanying Selected Issues Paper provides a more detailed discussion of corporate restructuring.

7

See, for instance, the World Economic Outlook, May 1997, for a review of studies on measurement biases in the CPI.

8

In the case of the United States, for instance, the U.S. Advisory Commission to study the consumer price index estimated that the CPI overstated inflation by 1.1 percent in 1996.

Japan: Economic and Policy Developments
Author: International Monetary Fund