Globalization and Global Disinflation by Kenneth Rogoff, Economic Counselor and Director, Research Department, IMF

IMF News, August 29, 2003

Source details

Canonical URL: Globalization and Global Disinflation by Kenneth Rogoff, Economic Counselor and Director, Research Department, IMF

Other formats

Bibliographic details

Authors: Kenneth Rogoff
Published: August 29, 2003

Overview and central argument

Over the past ten years, global inflation has dropped "from 30% to 4%1."
Institutional improvements (central bank independence, anti-inflation-oriented appointees) matter, but Rogoff highlights increased competitiveness from globalization, deregulation, and reduced government roles as a major universal supporting factor for disinflation.
Increased competitiveness:
- Lowers monopoly pricing power and "quasi-rents."
- Makes prices more flexible, reducing the real output gains from unanticipated inflation.
- Strengthens central bank anti-inflation credibility and reduces trend inflation.

Evidence on the near-universal fall in inflation

Global and regional trends:
- World CPI inflation (percent per annum) by periods:
- 1980-84: World 14.1
- 1985-89: World 15.5
- 1990-94: World 30.4
- 1995-99: World 8.4
- 2000-03: World 4.1
- 2000-04: World 3.9
- 2003: World 3.9 (implied by table context)
- Industrial economies: 1980-84: 8.7; 1985-89: 3.8; 1990-94: 2.0; 1995-99: 1.8; 2000-03: 1.9.
- Developing countries: 1980-84: 31.4; 1985-89: 48.0; 1990-94: 53.2; 1995-99: 13.1; 2000-03: 5.7.
- Latin America: 1980-84: 82.4; 1985-89: 185.9; 1990-94: 232.6; 1995-99: 17.2; 2000-03: 8.2.
- Transition economies spike: 1990-94: 363.2; 1995-99: 53.9; 2000-03: 14.5.
Country-level extremes and changes:
- In the 1970s–1990s many high/hyperinflation episodes; examples of cumulative price increases since 1970 include Argentina (100 trillion times), Brazil (a quadrillion), Democratic Republic of the Congo (almost 10 quadrillion)2.
- Of the IMF's 184 member states, only 11 are projected to have inflation over 20% in 2003; only 6 over 30% in 2003.
- Using a 40% threshold: Myanmar (40%), Angola (over 75%) and Zimbabwe (over 400%) reach or exceed this level in 2003.
Persistence and anchoring:
- Augmented Dickey-Fuller unit root tests for 18 industrialized countries (monthly CPI 1960-2003):
- For the full sample, cannot reject a unit root in inflation for any G-7 country.
- Splitting sample around early 1980s: for 1981-2003 one can reject the unit root hypothesis for all G-7 countries at fairly high significance.
- Table 3 p-values (selected):
- United States: 1960.1-1981.12 = 0.4596; 1982.1-2003.4 = 0.0016.
- Canada: 0.7684; 0.0167.
- Japan: 0.0623; 0.0848.
- France: 0.6953; 0.0033.
- Germany: 0.3922; 0.0834.
- Italy: 0.1756; 0.0180.
- United Kingdom: 0.2656; 0.0176.
- Emerging markets (Chile, Israel, South Africa) show evidence consistent with mean-reverting inflation in recent periods.
Deflation concerns:
- Deflation arguably threatens more countries today than very high inflation (over 40%) once CPI upward biases are accounted for7.

Factors underpinning global disinflation

Central bank institutional change:
- Greater central bank independence and selection of anti-inflation-focused central bankers are important contributors.
- Central bank governor turnover used as a proxy (Table 4, fraction per five years): developing countries turnover dropped from 0.211 (1970-1989) to 0.165 (1990-2002) among other regional changes; regional values vary (table excerpted).
- Exchange rate regime analysis (Natural Classification) shows limited flexibility and freely floating currencies have the best inflation performance, but gaps across regimes are fairly narrow except for "freely falling."
Fiscal policy:
- Industrial countries averaged general government primary balances of 2.8% during 1990-2002 versus -0.1% for 1970-1989 (Table 5 summary).
- Latin America averaged primary surpluses of 1.3% in 1990s versus -0.13% earlier; Africa deficits improved from -3.4% (pre-1990) to -1.6% (1990-2003).
- Many countries show examples where rising deficits/debt coexisted with declining inflation (India, Japan).
- Overall: fiscal policy broadly supportive but not a universal or decisive explanation for global disinflation.
Productivity and technology:
- The productivity explanation fits the U.S. late-1990s experience but fails as a universal cause (e.g., Europe saw falling inflation despite slowing productivity growth).
- Productivity shocks tend to have temporary effects on inflation unless they affect the monopoly wedge k.
Globalization, deregulation, and declining monopoly power:
- Deregulation and globalization reduced monopoly rents and markups; IMF April 2003 WEO estimates markups about 0.4 on continental Europe vs. 0.15 for U.K./U.S.
- Emerging Asia accounts for almost 20% of world trade (China alone 5%), exerting downward pressure on goods prices and affecting intermediate goods and substitution patterns.
- Mechanisms through which globalization/deregulation reduce inflation:
- Reduces the wedge (k) between monopolistic and competitive equilibria, lowering gains to unanticipated inflation.
- Makes prices more flexible (lower μ), decreasing monetary policy potency on real output and increasing credibility of low-inflation policy.
- Openness amplifies exchange-rate-mediated costs of inflationary shocks, tempering incentives to inflate.

Simple mathematical formalization (stylized political-economy model)

Central bank objective (simple static formulation):
- (π - πe - k - z)2 + χ(π - π*)2
- π = inflation set by central bank.
- πe = private sector expected inflation.
- k = wedge inversely related to monopoly power.
- z = mean zero productivity shock.
- π* = central bank preferred inflation (example given: 2.5%).
- χ = relative weight on inflation stabilization.
- Equilibrium under rational expectations:
- πe = π* + k/χ
- π = π* + (k - z)/χ
Extension with price flexibility parameter μ:
- Objective: [μ(π - πe) - k - z]2 + χ(π - π*)2
- Equilibrium:
- πe = π* + μ k/χ
- π = π* + μ (k - z)/χ
- Interpretation:
- Smaller k (due to greater competition) and smaller μ (due to greater price flexibility) lower equilibrium and expected inflation.
- Productivity shocks (z) have temporary effects; long-run trend requires changes in k or μ.
Key conceptual takeaways:

1. Trend inflation depends on central bank incentives to inflate, not merely transitory relative price shifts. 2. Unexpected productivity shocks reduce inflation temporarily; deeper trend shifts arise from greater competition and price flexibility.

Reduced conflict and other supportive factors

Historical observation: war and civil conflict often drive very high inflation episodes; fewer large-economy conflicts in the 1990s likely supported disinflation.
Commodity price trends: global commodity prices were on a steady downward trend for much of the period since the 1990s, aiding importing countries' price reductions.

Risks of a return of inflation and policy implications

Potential triggers for renewed inflation:
- Reversals of globalization (historical precedent in early 20th century) or sustained trade frictions (e.g., extreme security measures disrupting supply chains).
- Reversal of deregulation leading to rising monopoly power, increased price inflexibility, and renewed incentives for monetary authorities to inflate.
- Fiscal pressures and demographic shocks (retirement bulges) that may push governments toward surprise de-indexing or tolerating inflationary policies.
- Weakening of central bank independence or shifts in political acceptance of low inflation as a central objective.
Policy implications and recommendations (implied across the analysis):
- Preserve and strengthen central bank independence and commitment to low-inflation objectives.
- Maintain openness, deregulation, and policies that sustain competitive markets to keep k and μ low.
- Monitor fiscal sustainability, particularly in the face of demographic pressures, to avoid fiscal-driven inflation incentives.
- Be vigilant about geopolitical or security developments that could materially reverse globalization and its disinflationary effects.

Selected key statistics and tabulated highlights

Global inflation averages:
- World: 14.1 (1980-84); 15.5 (1985-89); 30.4 (1990-94); 8.4 (1995-99); 4.1 (2000-03); 3.9 (2000-04).
- Industrial economies: 8.7 (1980-84); 3.8 (1985-89); 2.0 (1990-94); 1.8 (1995-99); 1.9 (2000-03).
- Developing countries: 31.4 (1980-84); 48.0 (1985-89); 53.2 (1990-94); 13.1 (1995-99); 5.7 (2000-03).
Regional 2000-03 inflation snapshots:
- Africa: 11.8
- Asia (developing): 2.3
- Latin America: 8.2
- Middle East: 15.3
- Transition economies: 14.5
Table 3 (ADF p-values for G7 split sample 1982.1-2003.4): United States 0.0016; Canada 0.0167; Japan 0.0848; France 0.0033; Germany 0.0834; Italy 0.0180; United Kingdom 0.0176.
Central bank governor turnover (fraction per five years, selected):
- Developing countries: 0.211 (1970-1989) -> 0.165 (1990-2002) (Table 4 noted regional variation).
Fiscal stance (primary balances as percent of GDP, summary):
- Industrial countries primary balances averaged 2.8% (1990-2002) versus -0.1% (1970-1989).

Prepared for the Federal Reserve Bank of Kansas City conference on "Monetary Policy and Uncertainty: Adapting to a Changing Economy", Jackson Hole, WY, August 29, 2003 — Kenneth Rogoff, Economic Counselor and Director, Research Department, International Monetary Fund.