IV Risks of Deflation

Taimur Baig, Jörg Decressin, Tarhan Feyzioglu, Manmohan Kumar, and Chris Faulkner-MacDonagh
Published Date:
June 2003
  • ShareShare
Show Summary Details

This section provides an assessment of conjunctural deflation risks in the world economy. It shows that deflationary pressures have risen in several countries, and that the susceptibility to deflation is high in some of them. Moreover, the global economic environment continues to be highly uncertain, not least owing to the significant declines in equity markets, large and possibly rising output gaps, and geopolitical concerns. These factors could increase vulnerabilities further. Nonetheless, the section shows that the deflationary pressures are not strong enough to lead to generalized global deflation. The assessment is based on an index of deflation vulnerability computed from a set of economic and financial indicators for 35 countries. The index is complemented by an analysis based on expectations augmented Phillips curves for the Group of Seven economies and a case study of China.

Global Deflationary Pressures

As Section II suggested, sustained deflation can be unanticipated even as inflation and interest rates fall close to zero. Spotting deflationary pressures is difficult since the aggregate demand shocks that push an economy into deflation may be confused with the “normal” business cycle.16 Moreover, the magnitude and effects of supply shocks may be unclear from an assessment of the price level alone. Thus a comprehensive set of indicators is required to identify deflationary pressures. The analysis in the previous sections also suggests that demand-driven deflation is more damaging than supply-driven deflation, so the indicators are primarily focused on demand shocks. Deflation driven by demand shocks may also be easier to detect. Demand-side variables are generally available at high frequencies, and they tend to move in the same direction as deflationary forces (falling prices are associated with falling output). In contrast, supply-driven deflation can be harder to discern, either because the economy appears to be improving (for example, output gaps may be closing or be positive, while prices are falling), or because supply-side data are available only with long lags and at relatively low frequencies (such as rates of return).17

The earlier analysis suggests four sets of indicators: (1) aggregate prices; (2) measures of excess capacity, or output gap; (3) asset markets; and (4) credit market and monetary indicators. In addition, structural characteristics, and the room for maneuver on the policy front, play a role in the overall assessment.

For each of the four sets of indicators, several variables were used to assess deflation vulnerability. The variables were computed from the early 1990s to the present for 35 of the largest industrial and emerging market economies. For each variable, an examination was undertaken as to whether the variable exceeded a predetermined threshold; if it did, the economy was given a score of 1 on that count; otherwise, 0 (see Appendix I for details). The thresholds were determined on the basis of a variety of considerations and were cross-validated with Japan’s experience in the mid-1990s.18 The scores were aggregated to compute the index. To account for the heterogeneity across countries in the importance of asset and credit markets, a weighted index was also computed. The discussion below first notes developments in each set of indicators and then looks at the construction of the aggregate index.

Price Indicators

Consumer and producer price inflation is at postwar lows in many countries. However, inflationary expectations generally remain well anchored (see Appendix Figure A7). According to both Consensus and IMF staff projections, very few countries are expected to register consumer price inflation of less than 1 percent: outside of the Asian region, these encompass only Germany and Switzerland. Although global oil prices have recently declined sharply, the lagged effects of the earlier higher prices could be reflected in an uptick in headline inflation rates in the near-term (Appendix Figure A8).19 However, higher oil prices are also likely to have had adverse effects on activity, which may increase deflationary expectations. Given the difficulty in predicting deflation and the measurement bias, price indicators with a low value (inflation of less than ½ percent) were taken as a worrying sign (see Appendix I).

Excess Capacity and Output Gap Indicators

Excess capacity may be acting as a drag on growth in several Group of Seven (G-7) and smaller industrial economies. GDP growth is below potential in several of these economies, and is expected to remain so (Figure 5). From a long-term perspective, if overcapacity were a significant drag, the rate of return on capital should fall. However, in the largest economies, only Japan is showing a sustained decline in the return on capital (corporate profits as a share of the business capital stock).

Figure 5.GDP Growth and Output Gap in the Group of Seven (G-7)

(Percent change from a year earlier, or as percent of potential GDP)

Source: IMF, World Economic Outlook.

In emerging market economies, particularly in Asia, evidence points to excess capacity in a number of economies, and in specific sectors. IMF staff calculations suggest that output is below potential in Hong Kong SAR, Singapore, Brazil, Poland, and Russia (Appendix Figures A9A10). Output gaps could be increasing in these countries, in particular in Singapore and Brazil. In China, surveys suggest high excess capacity in several state-dominated industries, which may be increasing given strong private and public investment.20

Asset Market Indicators

Since early 2000, equity markets have suffered exceptionally large corrections (Appendix Figures A11 and A12). Price-earning ratios have returned to close to historical averages in several G-7 economies. However, earnings projections in many cases may still be overoptimistic—exposing downside risks (Appendix Figure A13). Over the same period, house prices have risen considerably across several countries, sustaining household balance sheets. Conversely, the decline in house prices in Japan, and to a lesser extent Germany, has aggravated the adverse effects of declining stock prices. Stock markets in emerging market economies have also declined significantly: the MSCI index of emerging market equities has fallen by over 50 percent (in U.S. dollar terms) since early 2000. The assessment of the deflationary impulse from an equity price shock to household and corporate balance sheets took into account the magnitude of price declines, and the importance of equity markets in the economy.21

As in the case of asset price declines, an appreciation of the exchange rate has both income and balance sheet effects. An appreciated currency dampens activity and incomes in export and import competing sectors, and puts downward pressure on prices, both directly and through weaker activity.22 Given the differences in the extent and speed of exchange rate pass-through, a variety of measures were considered to proxy the effect on deflation via the exchange rate channel (see Appendix I). The measure utilized was based on the degree of appreciation of the effective exchange rate (Appendix Figure A14).

Credit and Money Indicators

Credit growth has slowed markedly in many industrial and emerging market economies, failing even to keep pace with nominal GDP growth (Appendix Figure A15). This may largely reflect weak demand, but in some cases could also be an indicator of elevated risk aversion by banks, or emerging stress in the financial sectors. For example, in Japan the ongoing contraction in credit reflects weak credit demand as firms seek to deleverage, as well as worsening credit quality of borrowers; it may also, however, reflect a restriction in lending by creditors given the difficulty in determining creditworthiness. Several measures of financial intermediation, including growth in broad monetary aggregates and credit, were used to assess the extent to which credit and money channels may be exerting deflationary pressures.23

Table 3.Index of Deflation Risk, with Latest Data (Weighted)1
Risk/Index<0.20.2 ≤ × ≤ 0.30.3 < × ≤ 0.5>0.5
ChileBrazilFinlandHong Kong SAR
DenmarkCanadaNorwayTaiwan Province of China
New ZealandFranceSingapore
South AfricaIndiaSwitzerland
United Kingdom
United States

Maximum index score is 1.

As explained in the text, this category includes countries such as Japan, which has deflation and where the policy interest rate is virtually zero, as well as Germany, which is at high risk of entering a period of mild deflation, but where the relevant policy interest rate is still well above zero.

Maximum index score is 1.

As explained in the text, this category includes countries such as Japan, which has deflation and where the policy interest rate is virtually zero, as well as Germany, which is at high risk of entering a period of mild deflation, but where the relevant policy interest rate is still well above zero.

Table 4.Index of Deflation Risk, with Latest Data (Equal Weight)1
Risk/Index<0.20.2 ≤ × ≤ 0.30.3 < × ≤ 0.5>0.5
CanadaBrazilFinlandHong Kong SAR
ChileChinaGermanyTaiwan Province of China
New ZealandMexicoSweden
SpainSouth AfricaThailand
United Kingdom
United States

Maximum index score is 1.

Maximum index score is 1.

Aggregation and Weighting

Two sets of indices were computed based on the raw scores: an unweighted index where the scores were aggregated and renormalized to one to yield the index; and a weighted index where the financial and credit indicators were weighted according to their relative importance in the economy. For the weighting procedure, data regarding market capitalization of the broader equity index as a share of GDP were obtained and countries were divided into three cohorts along the lines of relative importance of equity markets. In countries that have a relatively large equity market, changes in equity prices were given greater weight than where equity markets are smaller. A similar scheme was used to reflect the heterogeneity with regard to the financial sectors (see Appendix I.)24

Table 5.Index of Deflation Risk, 1995 (Weighted)1
Risk/Index<0.20.2 ≤ × ≤ 0.30.3 < × ≤ 0.5>0.5
GreeceUnited KingdomSweden
Hong Kong SARSwitzerland
New Zealand
South Africa
Taiwan Province of China
United States

Maximum index score is 1.

Maximum index score is 1.

Overall Assessment

The country rankings for both unweighted and weighted indices suggest the following (Tables 3–6):

  • There has been a clear increase in the vulnerability to deflation for a number of industrial and emerging market economies. However, at this juncture, apart from the countries already beset with deflation, only a few additional countries appear to be significantly at risk.
  • Japan ranks at the top of the list, and in Asia is followed by Hong Kong SAR and Taiwan Province of China—the three countries already beset by deflation. China, despite ongoing deflation, ranks rather low because, as noted above, the indicators used in the index have a demand-side focus. This underlines the largely supply-side nature of China’s deflation, which is fundamentally different from the mostly demand-contraction driven pressure on prices seen elsewhere.
  • The score for the United States indicates a relatively low risk of deflation. Nevertheless, there are considerable uncertainties with regard to the effects of equity price declines still in the pipeline, excess capacity in several major sectors, and the impact of security and geopolitical concerns.
  • Some European economies have an increased vulnerability. In the euro area, risks remain low in all the major countries, except for Germany. Were deflation to start in Germany and become sustained, pressures could spillover into other euro area countries fairly quickly, particularly through the financial system. However, considerable scope remains for easing monetary policy. Outside the euro area, Switzerland appears to have a moderate risk, but also has scope for policy response.

Although these indicators present a mixed picture, they do not support strong concerns of generalized global deflation. With flexible exchange rates, global deflation is likely only if there were to be a series of significant shocks. This conclusion is reinforced by developments in the major economies:

  • Corporate sector imbalances may be unwinding. Large corporate restructurings over the past two years have also helped facilitate adjustment in balance sheets.
  • Labor incomes have continued to grow, offsetting a fall in household wealth. Labor markets have so far been holding up better than during previous slowdowns, particularly in the United States and euro area.
  • Policymakers still have some room to maneuver. There is substantial scope for lower rates in the euro area, and officials in most major economies have indicated an awareness of risks.
Table 6.Index of Deflation Risk, 1995 (Equal Weight)1
Risk/Index<0.20.2 ≤ × ≤ 0.30.3 < × ≤ 0.5>0.5
Hong Kong SARSingaporeSwitzerland
IndiaUnited Kingdom
New Zealand
South Africa
Taiwan Province of China
United States

Maximum index score is 1.

Maximum index score is 1.

This is not to deny the significant downside risks in specific economies. It is far from clear that the full effects of the bursting of the equity price bubble have manifested themselves. There are also concerns about the effects of a possible correction in house prices. Furthermore, the expectations of a recovery, notably of investment, have been repeatedly disappointed. This suggests that excess capacity may be larger than suggested by the available data, and the profit outlook cloudier than previously thought. In addition, in several countries problems in the financial sector may be exacerbating risk aversion and provision of credit.25

Deflationary Pressures in the Major Economies

This section assesses in more detail the risks of deflation in the largest industrial economies. It augments the preceding analysis by presenting Phillips curve estimates of the changes required in the output gap and the unemployment gap to set off, or exacerbate, deflation. Taking the current estimates of the output gaps and the NAIRU as given, a modified form of the Phillips curve was estimated for each of the G-7 economies.26 The parameter estimates were then used to assess the likely output and unemployment gaps that could lead to declines in the aggregate price levels, defined here as either mild deflation, which would entail a modest decline in consumer prices for two to four quarters; or persistent deflation, which would involve at least four quarters of declines in consumer prices by more than 1 percent. Detailed assessment of conjunctural developments and risks for the United States, Germany, and Japan are provided in Boxes 13.

Box 1.United States: Economic Outlook and Deflation Risks

With strong financial headwinds from a large accumulation of debt during the late 1990s and a sharp decline in the U.S. equity markets, some observers have wondered if the United States could experience deflation. A few months of negative headline CPI inflation cannot be ruled out, but the likelihood of sustained deflation appears remote.

Aggregate Demand and the Output Gap

While there is uncertainty about the degree of excess capacity, the acceleration in productivity after 1995 means that the potential growth rate is substantially above recent growth, leading to a widening output gap of just over 2½ percent in 2003:Q1, and disinflationary forces. Domestic demand in the past two years has been buffeted by a series of shocks, including continuing equity price declines, accounting scandals, September 11th attacks, the war in Iraq, and broader geopolitical uncertainties. Consumption continued to grow, in part supported by house price increases, but the need to rebuild wealth could cause the saving rate to rise sharply, curtailing consumption. Weaker labor markets and the waning effects of tax cuts could exacerbate the effects on consumption. The recovery in investment has yet to become broad-based and firmly established. Nonetheless, with strong support from monetary and fiscal policies, and continuing impressive productivity performance, GDP growth is expected to begin strengthening in the latter half of 2003. The recent data on corporate earnings have been encouraging, and there are indications that consumer and business confidence may be recovering.


Corporate prices have fallen much more than consumer prices, cutting into corporate profits, but economy-wide aggregate price levels do not show signs of declining. Core CPI is down to 1¾ percent (year on year) in March 2003 (compared to 2¾ percent during 2001 and 2 percent during 2002), with headline inflation rising to around 3 percent. The GDP deflator is around 2 percent (year on year), and the deflator for the nonbusiness sector has fallen sharply (see the top panel of the Figure). However, given the policy stimulus in the pipeline and the expected pickup in activity, inflation expectations—as measured by surveys and spreads on inflation-indexed bonds—have been relatively stable.

Corporate Sector

With a sharp fall in industrial output, the U.S. corporate sector led the economy into recession. Industrial production declined 6 percent during 2001; but at the depth of the recession in 2001, GDP declined only ½ percent. When corporate profits came under pressure, firms embarked on an ambitious cost-cutting strategy, sharply reducing investment (see the middle panel of the Figure). While some have worried about overcapacity, rates of return on capital have held up, indicating that firms do not yet have difficulty using their capital stock efficiently.

United States: Economic Outlook and Deflation Risks

Monetary and Credit Conditions

Bank lending ground to a halt in early 2002, but it did not contract (on a year-on-year basis), and banks remain relatively healthy. Credit growth has subsequently recovered, and by March 2003 was growing by around 7½ percent (year on year). Corporate credit spreads, after widening in late 2001 and early 2002 reflecting Enron and other corporate accounting concerns, have recently narrowed.

Asset Prices

U.S. equity markets peaked in March 2000, and at the height of the technology bubble, household wealth peaked near $43½trillion, or 6.3 times disposable income (see the bottom panel of the Figure). Since then, net worth has fallen $4¼ trillion to 4.9 times disposable income. Normally, such a large decline would have sparked a significant decline in consumption, but well-timed tax cuts in 2001 cushioned the blow by helping to support disposable incomes. The housing market has responded to aggressive interest rate cuts, and housing wealth rose $1½ trillion during 2000–2002.


U.S. policymakers have shown remarkable flexibility in responding to the economic shocks, and the central bank appears ready to use unorthodox policy measures to prevent deflation. Monetary policy remains accommodative, and recent indications by Federal Reserve governors seem to be preparing the markets for unorthodox monetary policy if the target for the federal funds is near zero (Bernanke, 2002; and Greenspan, 2002). Fiscal surpluses achieved during the late 1990s have turned into deficits, but the stock of government debt is forecast to remain stable at under 40 percent of GDP for the remainder of the decade, leaving room for stimulus, if necessary.

United States

As suggested by the indicators analysis, current risks of deflation in the United States are low, but downside concerns remain because of weak activity. The underlying consumer price inflation has slowed considerably in the past several years, with an increasing number of industrial and service sectors experiencing price declines. Demand has been held back by the effects of a sharp decline in equity prices, a rise in unemployment, and excess capacity in many industries (see Box 1). The decline in equity markets to date is similar to the decline in Japan in the three years after the bursting of its asset price bubble (Appendix Figure A16). Nonetheless, strong productivity growth and tax cuts have helped support personal incomes and, combined with mortgage refinancings, have underpinned consumption. In addition, the underlying rate of return in the economy appears to have held up fairly well, owing to aggressive corporate restructuring over the past two decades. Accordingly, inflation expectations have been relatively stable, with the largest downside risks associated with the possibility that consumption will slow before investment has a chance to rebound.

Box 2.Deflationary Pressures in Germany

Many indicators suggest that deflationary forces are at work in Germany: (1) underlying inflation is running under 1 percent a year; (2) unit labor costs (ULCs) for the entire economy have grown at an average rate of about ¾ percent in the past five years; (3) equity prices have retreated by about 40 percent during 2002, to levels last seen since early 1997—the evidence on real estate prices is mixed, with some (BIS) indicators suggesting that prices have retreated over the past decade and others (CPI rent component) pointing to stability; (4) real domestic demand is contracting and GDP growth will fall well short of potential for the third year running in 2003, causing a widening of the output gap to nearly 2¾ percent of potential GDP; (5) the number of bankruptcies is reaching record levels; and (6) domestic credit almost ground to halt in mid-2002, as German banks are struggling to maintain their regulatory capital, while suffering asset price declines and loan losses.

Although the near-term outlook for the German economy is fairly bleak, most forecasters do not foresee a significant decline in inflation for 2003: the consensus forecast for 2003 is for 1.2 percent, with one-half of observers seeing a decline in inflation relative to 2002. However, the IMF task force’s analysis suggests that mild deflation is fairly likely to take hold even though the risks of pernicious deflation are low.

First, the estimated output gap is large and the (ILO) unemployment rate remains high—at some 7.7 percent in 2002:Q4—even if it is down from a peak of 9.3 percent in 1997. Balance sheet indicators for the ratio of after-tax profits to turnover display no structural decline but are likely to have taken a blow over the past two years; and OECD data for the business sector suggest that the capital-output ratio in 2001 still exceeded the post-1980s average by 2 percent, down from a peak excess of 4.3 percent in 1996. Notice that the size of the output gap is not unprecedented: it compares to that of 1982–84. However, during the early 1980s the unemployment rate and the capital-output ratio were lower and inflationary pressures higher.

Second, credit growth is low and bank profitability weak—owing to deteriorating asset markets, high competition, and a cost structure that does not compare favorably even against those in other continental European countries. Credit growth has fallen short of demand, as measured by a linear combination of GDP and interest rates, but the shortfall compares to that seen in earlier recessions. While a case for a credit crunch is hard to make at this stage, banks are facing, by their own admission, the most difficult business environment since World War II, suggesting that financial intermediation is unusually vulnerable to adverse shocks.

Third, although the external sector is serving as a floor on prices and output, that floor is unlikely to buffer the effects on GDP of a sudden, rapid deterioration in domestic demand. Germany has become more competitive, seeing its external current account balance improve from a deficit of about ½–1 percent of GDP during the 1990s, to an estimated surplus of 2½ percent of GDP in 2002, a level last reached just before unification. According to some estimates, a 2 percent of GDP surplus is in line with equilibrium estimates of Germany’s current account (see the Figure) and thus the potential for saving rates to rise has diminished. However, because the improvement in the external position was achieved largely through slow import and domestic demand growth—rather than higher productivity and export growth, as had been hoped following unification—downside risks remain.

Lastly, but perhaps most important, a major concern for Germany is the constraint on macroeconomic policies: monetary policy is geared to the euro area where price pressures are not considered to have subsided sufficiently. And Germany’s fiscal policy is widely seen as turning highly contractionary in the near term, to move the fiscal deficit back under the 3 percent Maastricht reference ratio. The automatic stabilizers, which—together with the ECB commitment to price stability—could offer some protection against sustained deflation, are thus not allowed to widen the fiscal deficit while growth continues to fall short of potential.

Deflationary Pressures in Germany

Sources: Deutsche Bundesbank; IMF, World Economic Outlook; and IMF staff calculations.

1 Data break in 1994.

2 In-sample forecast after 1999.

The Phillips curve estimates suggest that for mild deflation to take hold, the output gap would need to increase by around 3 percentage points, to over 5 percent (Table 7). GDP growth would have to remain around ¾ percent for the next five to six quarters, and the unemployment rate would need to increase from 6 percent (in 2002:Q4) to 8 percent, within four quarters. For persistent deflation, the output gap would have to increase by over 5 percentage points, and the unemployment rate would have to move to 10 percent.

Table 7.Deflation Risk: Incremental Changes in Output and Unemployment Gaps1
Deflation OnsetPersistent Deflation
Unemployment gapOutput gapUnemployment gapOutput gap
United States2–2½3–43½–45–6
United Kingdom2–2½2–32–34

Estimates of the increase in the output and unemployment gaps likely to lead to the onset or persistence of deflation.

Estimates of the increase in the output and unemployment gaps likely to lead to the onset or persistence of deflation.

While there is a small risk that mild deflation could occur—the IMF staff’s central projection is for the output gap to widen in the first half of 2003—the expected policy response makes the likelihood of persistent deflation remote. Policymakers appear both willing and able to respond preemptively to the risk of renewed weakness and of deflation, even though the target for the Federal Funds rate is at a very low level. For instance, Greenspan (2002) has emphasized that it is crucial to “ensure that any latent deflationary pressures are appropriately addressed well before they become a problem” (see also, Bernanke, 2002; and Reinhart, 2003a). Federal Reserve officials have also made public statements on the measures they would consider to forestall deflation, including purchase of long dated securities; and there is considerable fiscal stimulus in the pipeline and more on the horizon. Moreover, should activity fail to rebound, the U.S. dollar could depreciate further, providing a fillip to prices, exports, and corporate profits.

Box 3.Deflation in Japan

Japan’s experience with low inflation is hardly recent. In the 15 years prior to the onset of deflation in the mid-1990s, its annual core inflation (CPI excluding fresh food, and energy) averaged 2.1 percent. During the mid-1980s, core inflation fell close to zero, followed by several quarters of decline in the GDP deflator. As growth picked up sharply from late 1987 onward, coinciding with a tremendous run-up in land and equity prices, the economy overheated, operating at 2 to 3 percentage points over potential GDP in the late 1980s and early 1990s, but CPI rose only moderately. Core inflation peaked at slightly above 3 percent in early 1991, and then began trending down, with a tightening of monetary policy, and a collapse of asset prices (see Figure). Bernanke and Gertler (2001) argue that the Bank of Japan was behind the curve during this boom-bust cycle—the central bank waited too long before tightening monetary policy during the bubble period, and delayed in easing once the economy headed downward.

The sharp fall in land and equity prices was followed by real GDP growth coming to a crawl. A widening output gap, reflected in the sizable excess capacity in the manufacturing and construction sectors, exerted further downward pressure on prices. Banks, which had lent heavily to real estate and construction companies, struggled under a mountain of bad loans and rapidly declining profitability. They focused on consolidating their balance sheets and sharply curtailed further credit. Reflecting the loss of economic momentum, broad money (M2 + CDs) growth declined rapidly, from over 11 percent in 1990, to 0.6 percent in 1992.

With declining activity, inflationary pressures virtually dissipated. Additionally, prices in the tradable sector were affected by the further opening of the economy and the resulting competitive pressures.1 Prices in the non-tradable sector also faced some downward pressures owing to deregulation and innovations. In its efforts to stimulate demand and prices, the Bank of Japan eased monetary policy, lowering the uncollateralized overnight call rate from 8.5 percent in early-1991 to 0.5 percent by late-1995, but that proved to be insufficient in the face of an unrelenting decline in asset prices and resulting problems.

The GDP deflator began its near-continuous decline in 1995 and Core-CPI deflation materialized fully in 1998 with the onset of a recession. A short-lived economic recovery around the Y2K-related investment boom in the late 1990s did little to arrest deflation. Initiatives to recover the economy fell short, and consumption and investment remained weak. As a result, asset prices continued to decline, with both land and equity prices near two-decade lows at end-2002. Inflation expectations, which remained positive until the beginning of actual price declines, subsequently turned negative and became entrenched as a sustainable economic recovery proved to be elusive. Surveys indicated that households and firms expected prices to continue to fall in the foreseeable future. The Bank of Japan responded by gradually lowering the overnight call rate toward its floor in the late 1990s, but base money growth rates remained relatively muted until early 2001. A quantitative easing framework was then put in place, with banks’ excess reserves held at the Bank of Japan becoming the operating target. The Bank of Japan incrementally raised its target of purchasing government bonds as part of the quantitative easing policy, but so far there is little evidence that its actions have been sufficiently aggressive to dent deflationary pressures.

The ongoing deflationary episode has been strikingly broad-based in its nature. Very few items in the consumer price index have experienced increases or even price stability. Prices of major items, such as food, clothing and footwear, furniture, transportation and communication, housing rent, and durable goods, have all registered gradual declines. The general decline in price levels cannot be explained by a single factor, such as competitive pressure from abroad or excess capacity and deregulation at home. Rather, a combination of these and other factors, including banking sector difficulties, insufficiently loose monetary policy, and stagnant demand, have been instrumental.

On the policy front, clearly, an aggressive easing of monetary policy is essential in combating deflationary pressures. Studies of historical episodes of deflation have repeatedly shown that a large and sustained increase in liquidity is a prerequisite to generating inflationary expectations, and despite recent measures such as targeting banks’ excess reserves held at the Bank of Japan, Japan’s monetary policy has not been sufficiently loose (Baig, 2003). Additionally, a comprehensive approach to tackle the bank and corporate sector problems is needed to stimulate demand and prices on a sustainable basis and bring the economy out of its protracted stagnation.

1Kamada and Hirakata (2002), estimating a comparative advantage model for Japan, show that some of the downward pressure faced by consumer prices in the mid- and late 1990s can be explained by the increase in international competition.


With broadly stable prices over the past six months, the German economy faces substantial demand-driven drag to growth (Box 2).27 Credit growth, production, and incomes weakened considerably during 2001 and 2002, and the labor market has come under pressure. House prices have been falling; equity markets have adjusted more than in the other advanced economies (Appendix Figures A16 and A17); and corporate balance sheet adjustments still have some way to go. Furthermore, banks and insurance companies are, by their own admission, undergoing the most challenging period since World War II. Thus, the near-term outlook for a recovery in investment remains more clouded than in the other advanced economies.

Japan: Selected Economic Indicators

Sources: IMF, Information Notice System; CEIC database; and IMF staff estimates.

1 Excludes fresh food, fuel, water and light; seasonally adjusted. The two spikes around 1989 and 1997 are due to the introduction and subsequent increase of consumption tax.

The Phillips curve analysis for Germany suggests that for mild deflation to take hold, the output gap would need to rise 1 to 2 percentage points (Table 7). At the same time, the unemployment gap would have to increase by 2 percentage points. For persistent deflation, the output and employment gaps would have to increase by 2½ to 3½ percentage points. The IMF staff’s central projection sees a widening of the output gap on the order of 1 percentage point this year because real GDP growth would remain around ½ percent. However, the unemployment rate would increase relatively modestly.

Accordingly, the probability of mild deflation taking hold over the next year is considerable. Of course, as prices in Germany decline relative to its trading partners, particularly in the euro area, the improvement in its competitiveness will stimulate exports and investment. This may prevent expectations of falling prices from becoming entrenched. This mechanism is likely to operate only slowly, however, and can do little to remedy difficulties in financial intermediation that could develop rapidly in a deflationary environment. Unlike in other economies, the room for policy maneuver is constrained, if not absent altogether. Fiscal policy is set to become restrictive (cutting the cyclically adjusted deficit by ¾ percent of GDP this year)—in support of Germany’s commitments under the Maastricht Treaty. While monetary conditions have tightened given the euro appreciation, monetary policy may not ease significantly because of greater price pressures in Germany’s euro-area partners.28


In Japan, core CPI deflation continued for 41 consecutive months through March 2003, and there are only marginal signs of deflationary pressures waning. Unless policies play a more supporting role and the economy’s long-standing structural problems are resolved, the risk of continued deflation remains high. In the near term, with growth below potential, the output gap is expected to widen further, potentially increasing downward pressure on prices (Box 3). The Phillips curve analysis suggests that if the unemployment gap were to increase by 1 percentage point, or the output gap by 2 percent the pace of deflation would accelerate, which could heighten the risk of a deflationary spiral (Table 7).

An important issue relates to why deflation is not already worse given the continued large output gap, high unemployment, and record business sector bankruptcies.29 This could reflect in part sticky expectations. Both market and survey evidence suggests that deflation is expected to remain around –½ to –1 percent. In the Consensus forecast, market participants attach a 70 percent probability to deflation of around 0.8 percent in 2003. This expectation may reflect a relatively flat Phillips curve; that is, the decline in activity and output gap would have to be significantly greater than witnessed to date for there to be a substantial further increase in deflationary pressures. It is also possible that the monetary easing in recent years has played a helpful role. The cumulative growth in base money has not been aggressive enough to improve expectations, but it may well have played a role in preventing deflationary expectations from worsening.

Nonetheless, despite the recent uptick, the economy is vulnerable to further adverse shocks that could markedly worsen deflation. In particular, asset prices could come under further pressure if banking sector difficulties mount, and corporate deleveraging continues. In addition, investment could decline further in an uncertain environment.

Because of the risk that deflation could worsen, a more vigorous monetary policy response is needed even if structural impediments remain in place. Such an approach could include a more rapid expansion of the monetary base in tandem with a commitment to end deflation within a given time period. Backed up by a clear communications strategy, these measures could reduce deflationary expectations (see Section V).

Other Group of Seven Countries

The pace of activity in France and Italy has slowed markedly in recent months, but the risk of deflation appears to be appreciably lower than in Germany. Despite the recent downward revisions to projections, growth is projected to pick up in the second half of the year in both France and Italy, confidence is somewhat better, house prices are continuing to increase, and consumer prices are better sustained, with core inflation close to 2 percent.

In the United Kingdom and Canada, economic conditions do not raise immediate concerns about deflation. Robust household demand growth has continued in the United Kingdom—supported by a housing price boom, and GDP growth is expected to pick up in the second half of 2003 on the strength of public spending. Inflation and inflationary expectations seem well supported. However, there may well be a risk of an abrupt correction in house prices that could have a substantial adverse effect on domestic demand growth. In Canada, although some recent indicators have been mixed, the underlying pace of growth remains robust. For the second year in a row the economy is expected to attain the highest growth rate among the G-7 and is the only one in the group to have virtually no output gap. Inflation is well above the 2 percent target range.

The Phillips curve estimates for the above four countries suggest that the output and employment gaps would need to increase substantially to move them into mild deflation. These results complement the findings from the deflation vulnerability index, and suggest that in these economies deflation risks are very small.


In China, the mild deflation is linked mostly to transitory as well as longer-term supply-related factors. Transitory factors include lower commodity prices, WTO-related tariff cuts, and restraint in administrative price increases stemming from social concerns. Some of the longer-term factors include productivity gains from strong investment, state-owned enterprise (SOE) reform, and the adoption of new technologies—supported by a robust increase in foreign direct investment—and stronger market orientation. A less benign structural source of deflation is large excess capacity in some sectors, owing to the loss-making operations of some state-owned enterprises. Large surplus labor in rural areas is also acting to put downward pressure on costs and prices.

Risks of persistent deflation are small. Given the strong pace of activity, deflationary pressures should continue to ease over the course of this year as transitory factors fade (see Box 4). Policies are also supportive; the monetary stance is accommodative, with room left to maneuver, and fiscal policy has been expansionary. The recent depreciation of the U.S. dollar has also led to an effective depreciation of the yuan. Expectations of continued strong economic growth are reflected in buoyant private demand and increasing real estate prices. However, some longer-term factors, including continued expansion in capacity and excess labor, could continue to press prices down and prevent Balassa-Samuelson effects from taking hold. Further deflationary pressures arising from excess capacity could build if loss-making operations of state-owned enterprises are not closed down in a timely manner. There is also a risk that if the Severe Acute Respiratory Syndrome (SARS) epidemic is not contained quickly, its adverse impact on private spending could create deflationary pressures from the demand side.

China’s Trade Share and Transmission of Deflation

Supply-side factors in China are creating downward price pressures in a number of industrial sectors globally.30 Cheap and abundant labor and large production capacity have helped China in the last 10 years gain market share in the United States and Japan. By 2002, its share in total U.S. imports reached 11 percent, overtaking Japan as the largest exporter to the United States; its share in total Japanese imports reached 18 percent during the same period (Appendix Figures A18 and A19). Although actual imports from China account for a small proportion of consumption in most countries (in the United States, imports from China account for 1 percent of GDP), the potential imports appear to be much larger. However, cheaper imports entail terms of trade gains that, through their positive effects on aggregate income and demand growth, could mute the direct deflationary effects in individual sectors.

Apart from the direct impact on prices, there may be an indirect impact through activity on some competitor economies. The magnitude of the impact is illustrated by the increase in China’s share in the Asian region’s total exports to the world from under 15 percent to close to 30 percent over the last 10 years. However, this change in market share is not fully at the expense of other countries: there has been a sharp increase in intraregional trade, with a significant increase in imports by China from many of the regional economies, which is likely to help offset the adverse impact on activity and prices in these economies.31

Despite the increase in trade shares, there is insufficient evidence that China is exporting deflation, although there are marked price pressures in particular sectors. A vector auto regression (VAR) analysis on the linkage between prices in China and major trading partners suggests that price fluctuations in China have a moderate impact in a few Asian countries, and a discernable but small impact on prices in the United States and Japan (Appendix Figure A20). In the region, prices in Taiwan Province of China and Hong Kong SAR are affected the most, owing to their strong links to mainland China, and the price changes appear to be passed through fully in about two years. There is some evidence of transmission to Singapore, but in a more limited way. Prices in other economies in the region—Malaysia, Thailand, India, and Indonesia—on the other hand, appear not to be affected at all by changes in prices in China.

Box 4.China: Recent Deflation Episodes


The recent deflation episode is the second in the last four years. After peaking at over 25 percent in 1994, CPI inflation declined steadily as macroeconomic policies were tightened to cool the overheated economy. Prices began to fall in 1998 as the economy slowed further in the wake of the Asian crisis, and this lasted until 2000 (see the Figure). Deflation resurfaced during late 2001–end-2002, peaking in April 2002 at 1.3 percent. Falling goods prices more than accounted for this decline, as most service prices continued to increase moderately. During the same period, PPI declined also, with PPI deflation exceeding 4 percent in early 2002, but this decline ended by December 2002.

IMF staff analysis suggests that in the second deflationary episode, both transitory and longer-term supply shocks were at play. In the short-run, lower commodity prices and WTO-related tariff cuts put downward pressure on prices, and led to a mild deflation. Longer-term factors include: productivity gains from reforms to state-owned enterprises (SOEs), adoption of new technologies, and greater competition from more open markets. Among the less benign factors is high excess capacity in some sectors owing to difficulties in closing some SOEs.

Monetary and Credit Conditions

Strong foreign exchange inflows pushed the M2 growth rate to 18½ percent (year-on-year) in the first quarter of 2003. While overall lending was strong, lending by large state commercial banks remained cautious.

Policy Impediments

Despite improvements, shortcomings remain in China’s institutional framework, which reduce the effectiveness of indirect monetary instruments, and China faces medium-term fiscal pressures. Monetary policy faces significant headwinds from an underdeveloped interbank market. Furthermore, while fiscal policy was geared to support domestic demand in 2002, fiscal consolidation is needed in the medium term, in view of the large quasi-fiscal liabilities in the banking system and future costs of pension reform, infrastructure, and other social needs.

Nonetheless, the transmission effects appear to be growing over time. When a similar VAR analysis is undertaken for the late 1970s, the impulse response functions imply weaker links than the ones identified over 1993–2002. This change is not surprising because Chinese external trade began to pick up around 1993, once government initiatives to increase exports started to take hold. Nonetheless, it does suggest that with the expected continued integration of China into the global economy, the transmission effects could increase further.

    Other Resources Citing This Publication