This Selected Issues paper analyzes macroeconomic developments and prospects for Japan during the 1990s. Following a surge in activity during 1996 and early 1997, the economy fell into recession in the second quarter of 1997. Real GDP fell by 3¾ percent during the four quarters ended March 1998; the unemployment rate reached historical highs; and deflationary pressures reemerged. The downturn was largely unexpected, and most forecasters had projected growth of about 2 percent in 1997. This paper also examines fiscal policy issues for Japan.

Abstract

This Selected Issues paper analyzes macroeconomic developments and prospects for Japan during the 1990s. Following a surge in activity during 1996 and early 1997, the economy fell into recession in the second quarter of 1997. Real GDP fell by 3¾ percent during the four quarters ended March 1998; the unemployment rate reached historical highs; and deflationary pressures reemerged. The downturn was largely unexpected, and most forecasters had projected growth of about 2 percent in 1997. This paper also examines fiscal policy issues for Japan.

III. Monetary Policy Issues1

A. Recent Developments

1. Monetary policy continued to be highly accommodative over the last year (Figure III.1). The official discount rate (ODR) has remained at a record-low of ½ percent since September 1995, and the operational policy target—the uncollateralized overnight call rate—has been held slightly below the ODR. Long-term interest rates on government bonds have fallen steadily since the economy started to slow in early 1997, and in June reached the lowest levels in recorded history (Chart III.1).

FIGURE III.1.
FIGURE III.1.

OFFICIAL INTEREST RATES

Citation: IMF Staff Country Reports 1998, 113; 10.5089/9781451820447.002.A003

CHART III.1
CHART III.1

JAPAN INTEREST RATES, MONEY, AND CREDIT, 1988–98

Citation: IMF Staff Country Reports 1998, 113; 10.5089/9781451820447.002.A003

Sources: Nikkei Telecom, WEFA, and staff estimates.1/ End ol period.2/ Period overage.

2. The Japanese interbank market has been subject to considerable turbulence since November 1997 (Figure III.2). The collapse of Sanyo Securities in the first week of November resulted in the first-ever default in the overnight call money market, and was followed in mid-month by the failures of Hokkaido Takushoku Bank (HTB) and Yamaichi Securities. Concerns about counter-party risk related to possible further financial failures drove up market interest rates (including the “Japan premium” in international markets) and limited interbank liquidity, particularly for weaker institutions and for transactions maturing after March 31, 1998 (the end of the financial year). Market pressures eased considerably after March, but intensified again in June on market concerns largely associated with the Long-Term Credit Bank (LTCB).

FIGURE III.2.
FIGURE III.2.

MARKET STRAINS

Citation: IMF Staff Country Reports 1998, 113; 10.5089/9781451820447.002.A003

3. To help curb the rise in short-term interest rates, and to stabilize market conditions, the Bank of Japan (BoJ) has supplied ample liquidity to the financial system, thereby relieving strains on financial institutions. As an immediate consequence of the November financial failures, the BoJ increased its loans under Article 38 (which allows the BoJ to provide financing to financial institutions in the process of being closed) by around ¥3½ trillion yen, mainly to HTB and Yamaichi Securities, and also provided liquidity through loans under Article 33 (which allows the BoJ to provide financing to solvent institutions).2 In addition, the BoJ injected longer-term liquidity (up to 3 months in duration) to the market by buying bills, commercial paper, and Japanese government bonds (JGBs) through repo operations.3 These “twist” operations4 helped to reduce the end-March hump in the yield curve. It also provided support for financial intermediation through the commercial paper market, thereby allowing banks facing liquidity shortages to maintain commercial ties with clients.

4. These operations led to a rapid increase in the BoJ’s balance sheet. Total assets ballooned by ¥29.1 trillion at the end of March 1998 compared to a year earlier, an increase of almost 50 percent. Short-term market operations were ¥15.4 trillion higher than a year earlier—¥6.1 trillion in JGB-repos, ¥5.2 trillion in additional bills (including commercial paper), and ¥4.1 trillion in additional loans.5 The remainder of the expansion in the BoJ’s balance sheet largely reflected double counting of repo operations (both the underlying asset and the cash collateral are booked on the asset side, with equivalent double counting for liabilities) and an increase in holdings of government bonds. Short-term market support has been substantially reduced since the end of the financial year, falling by over half between end-March and end-June, with Article 33 loans to solvent institutions being almost fully repaid (some of the decline in short-term operations was also seasonal in nature). Market reports suggest that the BoJ has again been injecting large amounts of liquidity in June and July in response to market disruptions associated with uncertainties about the LTCB, through repurchase operations using JGBs and commercial bills.

Balance Sheet of the Bank of Japan: Assets

(In trillions of yen; end month)

article image
Source: Bank of Japan.

5. While most of the asset expansion was sterilized through sales of BoJ bills, some of the increase leaked through into the monetary base, which accelerated modestly from a 12-month rate of increase of around 7 percent in early 1997 to over 10 percent in late 1997 and early 1998 (Chart III.1). This was accompanied by accelerations in the monetary aggregates—M1, M2+CDS, and M3+CDS. However, there was no appreciable impact on wider measures of monetary conditions, such as broadly defined liquidity which, if anything decelerated over the period. The lack of impact on wider measures of monetary conditions reflected a shift in portfolio preference toward more liquid assets in response to the low level of interest rates.

6. The new Bank of Japan Law, which became effective from April 1998, has increased the BoJ’s independence. Monetary policy is now decided by a nine-person Policy Board that meets twice-monthly, made up of the BoJ Governor, two Deputy Governors, and six non-BOJ officials, mainly business people and academics. The Government is allowed to send representatives to the meetings; they have no voting power, but can request (although not insist on) a delay in implementation of any policy changes. Any decisions with regard to monetary policy are announced immediately after the conclusion of the meeting, while minutes are published with a delay of approximately one month. The minutes indicate that the discussions have been lively, and that a wide range of considerations and views are currently contributing to the setting of policy. Decisions have not been unanimous recently, with some members of the Board apparently favoring lowering nominal interest rates further. So far, such a step has been resisted for several reasons, including a reluctance to utilize the last remaining scope for interest rate action.6

B. Has There Been A Credit Crunch?

7. Bank lending, which had been lackluster since late 1996, weakened significantly after the failures of Sanyo Securities, HTB and Yamaichi Securities in late 1997 (Chart III.1). By June 1998, outstanding bank loans hit a six-year low. This tightening of bank credit, which has occurred in spite of a highly accommodative monetary policy, occurred against a background of rising bankruptcies and negative assessments by businesses of banks’ willingness to lend. Many commentators have argued that Japan is experiencing a “credit crunch,” in which curbs on bank lending are forcing otherwise viable businesses into liquidation, weakening confidence and hurting activity.

8. Surveys of corporate sentiment confirm that credit conditions tightened in late 1997 and early 1998 (Chart III.2). Responses to the tankan survey question on banks’ willingness to lend have deteriorated rapidly for all categories of enterprise, to levels not seen since the bursting of the asset price bubble in the early 1990s; surveys that focus specifically on small- and medium-sized enterprises (SMEs) show an even more striking deterioration in sentiment. Rising bankruptcies provide further indications of the impact of tightening credit conditions. While the recent weakness of activity would imply some increase in bankruptcies, current bankruptcy rates are significantly above predictions from a simple indicator model.7

CHART III.2
CHART III.2

JAPAN INDICATORS OF CORPORATE DISTRESS, 1980–96

Citation: IMF Staff Country Reports 1998, 113; 10.5089/9781451820447.002.A003

Sources: Nikkei Telcom, WEFA, and staff estimatas.

9. Loans to the business sector from public financial institutions were increased to help alleviate the downturn in private sector lending (Figure III.3). By early 1998, loans to businesses from public sector financial intermediaries were rising at an annual rate of over 1 percent, after falling through 1996 and most of 1997. However, the increase in public sector credit was insufficient to significantly offset weakness in the private sector (loans from public sector financial intermediaries to business are a relatively small proportion of the equivalent loans from private banks).8

FIGURE III.3.
FIGURE III.3.

PUBLIC SECTOR LOANS TO BUSINESS

Citation: IMF Staff Country Reports 1998, 113; 10.5089/9781451820447.002.A003

10. The tightening of bank credit has been particularly severe for small-and medium-sized enterprises (Figure III.4). Bank loans play a relatively important role in financial intermediation in Japan.9 This is particularly true for SMEs, which receive over two-thirds of bank loans and have very limited access to alternative sources of funds.10 Previous staff work has linked the anemic performance of SMEs over the early- to mid-1990s to constraints on bank financing.11 These difficulties have continued—bank credit to SMEs decelerating significantly in 1997.

FIGURE III.4.
FIGURE III.4.

SMALL- AND MEDIUM-SIZED ENTERPRISE LOANS

Citation: IMF Staff Country Reports 1998, 113; 10.5089/9781451820447.002.A003

11. Tighter bank credit has coincided with cut-backs in investment plans, particularly at smaller companies. The June 1998 tankan survey indicates that fixed investment by small enterprises fell by 4½ percent in FY1997—4 percentage points more than for large companies. Spending cutbacks appear likely to continue. In the same survey, FY1998 spending by small companies is planned to be 19 percentage points lower than in the previous year, compared to a fall of 1¼ percentage points for their larger brethren.

12. Several factors suggest the tightening of availability of bank lending does not simply reflect a lack of funding capacity or limited demand for bank lending:

  • Bank deposits have continued to increase, and the gap between the change in bank loans and bank deposits has widened to record levels.

  • The amount of outstanding commercial paper and bills rose after November 1997, in spite of weak activity and higher interest costs.

13. Macroeconomic factors only partly to explain recent loan activity. To investigate the size of the autonomous decline in bank lending, a demand function was estimated relating bank loans to real GDP, the prime rate, and real land prices (to represent the impact of declines in the value of collateral). If banks were constraining their credit below “normal” demand, this equation should over predict actual behavior. A dynamic forecast since end-1996 indicates that by first quarter of 1998 predicted lending was about 4 percent higher than actually observed (Chart III.3).

CHART III.3
CHART III.3

JAPAN BANK LOANS, 1996–98

Citation: IMF Staff Country Reports 1998, 113; 10.5089/9781451820447.002.A003

Sources: Nikkei Telecom, WEFA, and staff estimates.1/ Projection from staff model reported in the text.

Bank Loan Equation

The equation was estimated over 1980Q1–1996Q4. The results (with t-statistics reported in parentheses) were:

Δlog(REALLOANS)=0.72(8.8)Δlog(REALLOANS)1+0.38(2.5)Δlog(REALGDP)0.45(1.8)ΔPRIMERATE0.05(2.8)log(LONGRUN)1+0.0016(1.0),

R2 = 0.63, DW = 2.41,

where LONG RUN = log(REAL LOANS) − 1.46* log(REAL GDP) + 4.65* PRIME RATE − 0.23*log(REAL LAND PRICES) + 8.98,

14. Increasing attention to loan quality, shortage of bank capital and financial turbulence also played important roles in limiting credit (Chart II.4):

  • Increasing attention to loan quality. The “big bang” financial reforms are causing banks and financial markets to become more aware of credit risks. (This has been clearly reflected in wider spreads for higher risk bonds—for example, the spread between 5-year Baa corporate bonds and the equivalent government security has almost doubled since September to 110 basis points.) Banks’ attention to loan quality has been heightened by the need for more stringent assessment of loan quality and provisioning requirements as part of the system of prompt corrective action (PCA) introduced on April 1, 1998. The shift away from the convoy system (in which stronger financial institutions were expected to assist their weaker brethren) and greater competition as market barriers are reduced have increased pressures to strengthen balance sheets.

  • Shortage of bank capital. At the same time as the rewards to strong capitalization have increased, bank capital has been eroded by declining equity prices and the increased need to provision against bad loans. With low equity valuations increasing the cost of raising capital on the market, banks have preferred to shrink their risk-weighted asset base to enhance capital adequacy, largely by cutting bank lending. The major banks reduced their risk-weighted assets by 4 percent during FY1997.

  • Financial turbulence. The collapse of HTB and Yamaichi Securities in November 1997 triggered sharp increases in bank CD rates and the Japan premium. Significant differences in funding costs between banks regarded by the markets as strong or weak have also become evident. To a significant extent, the interbank market ceased to provide a ready source of funds, despite large liquidity injections by the Bank of Japan. Higher funding costs for banks in turn put upward pressure on loan rates, even though the official discount rate remained at its record low of ½ percent.

CHART III.4
CHART III.4

JAPAN PRIVATE MARKET INTEREST RATE DEVELOPMENTS, 1997–98

Citation: IMF Staff Country Reports 1998, 113; 10.5089/9781451820447.002.A003

Sources: Nikkei Telecom, WEFA, and Bank of Japan.

15. In sum, the heart of the “credit crunch” lies in a combination of increased sensitivity to risk and short-term liquidity shortages for banks. The increased awareness of risk implied by higher spreads is in many respects a welcome development in Japanese financial markets, and over the longer term will improve the allocation of capital. It is unfortunate, however, that this reevaluation of credit risk coincided with a period of economic weakness, in which the ability of firms to obtain loans was already deteriorating along with overall economic conditions. The timing of banks’ short-term liquidity problems, reflecting the failure to deal with underlying banking sector problems, has been even more unfortunate from a macroeconomic perspective, as the resulting constraints on bank loans amplified existing economic weaknesses related to fiscal consolidation and the Asia crisis.

C. How Could Monetary Policy Further Support The Economy?

16. With short-term interest rates close to their floor of zero, can monetary policy be used to further stimulate activity? There are five main channels through which an expansionary monetary policy could, in principle, support the economy:12

  • The interest rate effect in which lower short-term interest rates stimulate interest-sensitive components of demand such as investment and consumer durables.

  • The exchange rate effect in which a lower differential between domestic and foreign interest rates leads to a depreciated real exchange rate and higher net exports.

  • The credit effect in which monetary policy encourages greater financial intermediation. This can occur directly through encouraging bank lending (for example, by increasing banks’ lending capacity), or indirectly through reducing frictions in credit markets caused by moral hazard and adverse selection (for example, by improving operating profits and collateral).

  • The wealth effect in which increases in outside money, liquidity, and asset prices raise private wealth, which then encourages higher private sector spending.

  • The expectations effect, in which expectations of higher future inflation reduces real interest rates and depreciates the exchange rate by lowering expectations of its future nominal value.

17. The recent weakening of the real exchange rate has provided some support for the economy, as shown by recent movements in the monetary conditions index (MCI), which is a weighted-average of the real interest rate and real exchange rate, and is used by the staff to measure monetary conditions (Chart III.5).13 However, the Asia crisis limits the usefulness of the exchange rate channel in further reviving the economy, as a significant devaluation of the yen could exacerbate the financial crisis in the rest of Asia, hurting regional activity and partially offsetting any direct benefits to Japan’s net exports.

CHART III.5
CHART III.5

JAPAN INDICATORS OF MONETARY AND FINANCIAL CONDITIONS, 1980–1998 1/

Citation: IMF Staff Country Reports 1998, 113; 10.5089/9781451820447.002.A003

Sources: Nikkei Telecom, WEFA, and staff estimates1/ An increase indicates a tightening of monetary and financial conditions. The MCI is a weighted average of changes in the real interest rate and the real exchange rote. The FCI also includes the change in the fiscal stance and the stock price index; this index is presented on an annual basis because quarterly fiscal data are not available.

18. The BoJ has used shifts in its asset portfolio to support the credit channel. The credit channel is normally viewed as an adjunct to interest rate policy, with lower official interest rates encouraging credit creation by widening banks’ spreads and improving borrowers’ profitability and borrowing capacity.14 However, with banks under pressure to prune their loan portfolios and improve loan quality in order to strengthen balance sheets, credit conditions have become a major constraint on the effectiveness of monetary policy. By buying large amounts of commercial paper from banks on a repo basis, the BoJ allowed banks to provide credit to clients while limiting the impact on their risk-weighted assets.15 Such direct support for credit channel is generally only useful at times of financial stress, when bank lending is being severely constrained by short-term factors.

19. The wealth effect may be of limited use given the low level of interest rates. Open market operations can be used to inject liquidity into the economy, either through banks or by direct purchases of real or financial assets, which may then be spent on goods or assets. With interest rates so low and the financial system weak, however, the distinction between money and other financial assets may well be blurring, with individuals becoming indifferent as to which they hold. In short, Japan may be approaching a liquidity trap, implying that expanding the monetary base by itself will not add significantly to aggregate demand.

20. The final policy option would be to use the expectations effect, lowering real interest rates by increasing expectations about the future level of inflation. The staff has previously argued that a positive inflation target could make monetary policy more effective and would help forestall deflationary pressures.16 More recently Professor Krugman has argued that this expectations channel is the only way to make monetary policy effective in an economy such as Japan that may be in a liquidity trap.17 Recent staff work using MULTIMOD indicates that expectations of higher inflation could provide a significant boost to real activity if the commitment to higher inflation could be made credible.18 The real interest rate would be reduced without any fall in nominal interest rates, while expectations of a higher future price level would trigger an immediate exchange rate depreciation, thereby lowering the real exchange rate.19

21. A difficulty with implementing such a policy is making the commitment to higher inflation credible when the conventional levers of monetary policy appear largely ineffective. Relying on public statements pledging the BoJ to reflation would be unlikely to be sufficient even when combined with more concrete actions, such as using the limited remaining room to lower nominal interest rates. A more tangible change in policy regime, involving the adoption of an explicit, above-zero, inflation target, would have more chance of success. Even in this case, however, the effectiveness of the strategy is open to doubt, because it would depend upon convincing markets that the BoJ would continue with a higher inflation target in the longer run. Success would bring its own difficulties, including the dilution of the Bank of Japan’s hard-won anti-inflationary credentials and, perhaps even more importantly in current circumstances, the danger that the associated exchange rate depreciation could further destabilize financial markets in the rest of Asia.

1

Prepared by Tamim Bayoumi.

2

The current Article 38 was called Article 25 before the new Bank of Japan Law became effective from April 1998. Similarly, Article 33 was earlier known as Article 20.

3

BoJ repo operations using JGBs were first introduced in October 1997, to assist in implementing open market operations.

4

So called because they aim to “twist” the yield curve.

5

Including loans under Article 38.

6

See, in particular, the minutes to the Policy Board meeting on June 12, 1998.

7

The model relates the number of bankruptcies to their own lagged value, the output gap, the prime rate, and a time trend (the regression is BANKRUPT = 0.74* BANKRUPT(−1) − 13241* OUTPUT GAP + 18379*PRIME RATE + 42* TIME − 70832). Similar results are also found using the value of bankruptcies, instead of the number.

8

Government loans to businesses were calculated as the sum of outstanding loans of the People’s Finance Corporation, Japan Finance Corporation for Small Businesses, Small Business Credit Insurance Corporation, Japan Development Bank, Hokkaido and Tohoku Development Corporation, and Environmental Sanitation Business Finance Corporation.

9

This is true in most non “Anglo-Saxon” financial systems. For a comparison of financial system across a wide range of advanced countries see Claudio Borio, “Credit Characteristics and the Monetary Transmission Mechanism in Fourteen Industrial Countries,” in K. Alders, K. Koedijk, C. Kool and C. Winder Monetary Policy in a Converging Europe (Amsterdam: Kluwer Academic Publishers, 1996).

10

See Mitsuhiro Fukao, “Japanese Financial Instability and Weakness in Corporate Governance Structure,” mimeo, Keio University (1998). While financial deregulation allowed large corporations greater access to bond markets since the late 1980s, there is no significant junk bond market to provide such funds for smaller firms.

11

Robert Westcott, “Assessing the Risks of a Credit Squeeze Among Small- and Medium-Sized Enterprises in Japan,” in Japan—Selected Issues, IMF Staff Country Report 96/114 (October 1996)

12

For more details see the papers contained in a symposium on the monetary transmission mechanism published in the Fall 1995 Journal of Economic Perspectives, Vol. 9:4.

13

See G. Lipworth and G. Meredith, “Indicators of Monetary and Financial Conditions: A Reexamination” in B. Aghevli, T. Bayoumi and G. Meredith (eds.), Structural Change in Japan: Macroeconomic Impact and Policy Challenges (Washington: International Monetary Fund, 1998).

14

B. Bernanke and M. Gertler “Inside the Black Box: The Credit Channel of Monetary Policy Transmission,” Journal of Economic Perspectives, Fall 1995, Vol. 9:4, pp. 27–48.

15

Some commentators have suggested that such operations could potentially be made more aggressive by buying assets outright, which would increase spending power by providing liquidity (and wealth) to the economy directly, at the cost of the authorities accepting the credit risk of the borrower. (The BoJ’s recent operations mainly involved purchasing financial assets on a repo basis, so that the credit risk remained with the banks).

16

G. Meredith, “Monetary Policy: A Summary of Staff Views,” in B. Aghevli, T. Bayoumi, and G. Meredith (eds.), Structural Change in Japan: Macroeconomic Impact and Policy Challenges, (Washington: International Monetary Fund, 1998).

17

Paul Krugman, “Japan’s Trap,” http:/web.mit.edu/krugman/www/japtrap.html. Briefly, the argument is that when equilibrium real interest rate and inflation expectations are both very low, the implied equilibrium nominal interest can be such that money and bonds become perfect substitutes. Increasing inflation expectations breaks this impasse.

18

The size of the boost to activity is quite sensitive to the baseline, however, as it depends on the length of time that monetary policy is projected to be significantly constrained by the floor of zero percent on nominal interests rates.

19

The rise in the expected future price level caused by higher expected inflation would tend to lower expectations about the future nominal value of the yen. With unchanged nominal interest rate differentials, such a future expected nominal depreciation would cause a depreciation of the current exchange rate.

Japan: Selected Issues
Author: International Monetary Fund