## _wp15133

## Source details

**Canonical URL:** [_wp15133](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15133.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15133.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15133.pdf.json)

---

### Introduction
- Historical context:
  - Late 1960s–1970s: food price shocks were widely credited with contributing importantly to large increases in inflation in advanced economies such as the United Kingdom and the United States.
  - Blinder (1982) and Blinder and Rudd (2008) argue that episodes in the 1970s—food price shocks, energy shocks, and removal of wage-price controls—played leading roles in the Great Stagflation.
- Proposed reasons for reduced sensitivity of inflation to food shocks after the 1970s:
  - Food accounts for a smaller share of consumption baskets.
  - Greater wage flexibility so wages absorb more of supply shocks (Blanchard and Gali, 2007).
  - Changes in monetary policy conduct and credibility (focus on core inflation, greater anti-inflation credibility; Batini and Nelson, 2005; Rogoff, 2003).
- Recent developments and motivation:
  - World food prices declined through the 1980s and 1990s but rose with spikes in 2004, 2008 and 2012.
  - Concern that renewed elevated food prices could spill over into higher overall inflation, especially in emerging economies where food shares are higher and monetary credibility may be lower.
- Key questions:
  - What role have global food price movements played in domestic inflation since the 1960s?
  - Do other advanced economies display the same reduced sensitivity of inflation to global food price movements found for U.S. data?
  - Did the impacts of the food price shocks of the 2000s differ across advanced and emerging economies and what channels influenced such differences?

### Data
- Annual data set:
  - Coverage: 44 countries.
  - Period: back to the 1960s.
  - Average T in baseline sample: 43 (finite sample bias of lagged dependent variable and fixed effects is in the order of 1/T).
- Monthly data set:
  - Coverage: 34 advanced economies and 50 emerging economies.
  - Period: 2000–13.
  - Sources: Haver Analytics and IMF Primary Commodity Prices.
- Illustrative correlations:
  - Annual data (1960–2012): modest positive association between average CPI inflation and average food inflation; positive association between average CPI inflation and average excess money growth (excess money = average money growth minus average real GDP growth).
  - Monthly 2000–13: positive association between average CPI inflation and food inflation for both advanced and emerging economies.
- Data quality notes:
  - Emerging market historical data are limited and pose quality issues; monthly data (2000s) improves coverage but some quality issues remain.

### Global food prices and inflation: channels and estimation method
- Headline CPI decomposition highlights three channels:
  - (i) scale of food price shocks;
  - (ii) food share weight in CPI basket;
  - (iii) monetary policy response and credibility relating to sticky prices (New Keynesian perspective).
- Estimation approach:
  - Local projections per Jorda (2005) to estimate impulse response functions (IRFs) directly.
  - Baseline regression (annual data) for horizons k = 0,..,3:
    - Dependent variable: domestic CPI inflation for country i in year t+k.
    - Key regressor: global food price inflation in year t, weighted by the share of food imports.
    - Controls: food share in domestic CPI; lagged dependent variables (l chosen = 2 in baseline); country fixed effects; country-specific time trends.
    - Interpretation: estimated beta_k measures impact of global food prices on domestic inflation at horizon k, relative to country-specific trends.
- Inference:
  - IRFs plotted with 90 percent confidence bands; t=0 denotes the year of the shock.
  - Finite sample bias from lagged dependent variable with fixed effects is of order 1/T; average T = 43 mitigates concern.

### Baseline results (annual data, 1960–2012)
- Main qualitative result:
  - Global food price shocks have positive and statistically significant effects on domestic CPI inflation and these effects are persistent.
- Quantitative magnitudes:
  - A 10 percent increase in global food price (weighted by food import share) typically increases domestic inflation by:
    - 0.35 percentage point in the very short term (the year of the food price shock).
    - about 0.4 percentage point in the medium term (3 years after the shock).
    - peak effect of about 0.7 percentage point 1 year after the shock.
  - Many food price shock episodes involve increases of 50 percent or more.
- Table 1. Baseline Estimates (k = 0,1,2,3) — selected reported coefficients:
  - First regressor (effect per 1 percentage point change in world food price inflation weighted by import share):
    - k=0: 0.036 (5.48)***
    - k=1: 0.074 (7.02)***
    - k=2: 0.063 (7.40)***
    - k=3: 0.040 (4.55)***
  - Persistence (lagged dependent variable):
    - k=0: 0.607 (4.23)***
    - k=1: 0.386 (3.74)***
    - k=2: 0.289 (3.68)***
    - k=3: 0.187 (3.22)***
  - Additional control/regressor (third reported row):
    - k=0: 0.050 (0.54)
    - k=1: 0.062 (0.83)
    - k=2: 0.038 (0.65)
    - k=3: 0.070 (1.27)
  - Sample sizes and fit:
    - N: 706, 684, 662, 640 for k = 0,1,2,3 respectively.
    - R2: 0.73, 0.65, 0.60, 0.55 for k = 0,1,2,3 respectively.
    - IPS-statistics: -17.700***, -13.745***, -12.407***, -10.720*** for k = 0,1,2,3 respectively.

### Robustness checks
- Addressing endogeneity and reverse causality:
  - Re-estimation without country fixed effects: differences small and not statistically significant.
  - Two-step GMM system estimator using up to four lags of domestic and global food price inflation as instruments for global food price inflation: estimates similar to baseline.
  - Re-estimation using (CPI inflation – food inflation) as dependent variable: consistent results.
  - Panel-VAR approach to purge lagged feedback effects from domestic inflation to global food price inflation: consistent results.
- Conclusion from robustness:
  - Positive, persistent impact of global food price shocks on domestic CPI inflation remains under alternative estimators and specifications.

### 3.4 Sub-sample differences
- Changes in pass-through over time:
  - 1960–1982: impact of global food prices on domestic CPI inflation was large, statistically significant, and long-lasting.
  - 1983–2012: impact was much more modest and short-lived.
- Food-share composition:
  - Muted effect in 1983–2012 holds when food share in domestic CPI is assumed constant at its time-average value, implying changes in food-share composition are not a key factor.
- Country examples:
  - United Kingdom and United States show much more muted impacts in 1983–2012 versus 1960–1982.
- Identification and sensitivity:
  - Panel-VAR uses Cholesky ordering with global food price inflation ordered first; lag length = 2.
  - Results are not sensitive to the exact cutoff date in the early 1980s.

### 3.5 Food shocks vs. monetary factors
- Inclusion of excess money growth (excess money = money growth − nominal GDP growth) as independent variable:
  - Over 1960–2012 both global food price shocks and domestic excess money shocks have statistically significant and long-lasting effects on domestic inflation.
  - Time variation:
    - 1960–1982: effect of global food price shocks on domestic inflation dominates.
    - 1983–2012: effect of excess money shocks on inflation is more persistent and more precisely estimated.
- Robustness:
  - Findings hold under previously described robustness checks.

### 3.6 Evidence for the 2000s from annual data
- Advanced economies (AEs) vs emerging market economies (EMs), 1999–2012:
  - Global food price shocks of the 2000s had statistically significant short-term effects for both AEs and EMs.
  - Effect on domestic inflation in EMs was slightly larger and more persistent than for AEs.
  - Figures present impact of a 1 percentage point change in world food price inflation on domestic CPI inflation with 90 percent confidence bands; t=0 denotes the year of the shock.

### 4. Evidence from monthly data for the 2000s

4.1 Pass-through of world food inflation into domestic food inflation
- Estimation: country-by-country monthly regressions of domestic food price inflation on current and 12 lags of international commodity price inflation, controlling for 12 lags of domestic food price inflation.
- Median long-term pass-through (sum of coefficients on world food inflation divided by 1 minus sum of coefficients on lagged domestic food inflation):
  - Advanced economies: 0.18 percent (median long-term pass-through of a 1 percent world food price shock to domestic food prices).
  - Emerging economies: 0.34 percent (median long-term pass-through of a 1 percent world food price shock to domestic food prices).
- Reasons for incomplete pass-through:
  - Local components in production and distribution, retail/distribution margins, excise taxes, customs duties.
  - Food subsidies.
  - Significant domestic food production and weather influences.
  - World commodity price indices may not match country-specific consumption bundle composition.

4.2 Pass-through from domestic food inflation to overall inflation
- Key determinants:
  - Share of food in the consumption basket.
  - Anchoring of inflation expectations.
  - Higher food shares raise direct contribution to headline inflation and may increase pass-through into nonfood inflation via wage demands.
  - Poor track record in controlling inflation can raise inflation expectations and thereby increase pass-through.
- Food-share statistics:
  - Advanced economies median food share: 17 percent.
  - Emerging economies median food share: 31 percent.
- Contributions to overall inflation:
  - In 2008:
    - Food prices contributed about 5 percentage points to overall inflation in emerging economies on average.
    - Food prices contributed about 1 percentage point to overall inflation in advanced economies on average.
  - Since 2008:
    - Contribution exceeded 2 percentage points for emerging developing economies.
    - Contribution about half a percentage point for advanced economies.
  - Cross-country variation: in some economies food prices raised headline inflation by about 10 percentage points in 2008 and 5 percentage points in the more recent period.
- Additional channel:
  - World indices composed of relatively unprocessed food commodities; in developed countries most consumed food is highly processed, reducing importance of raw-material price movements in CPI.

4.3 The role of inflation expectations
- Estimation:
  - Responsiveness of medium-term inflation expectations to an unexpected increase in current-year inflation using Consensus Economics survey data (spring and fall vintages) and methods based on Levin, Natalucci, and Piger (2004).
  - Unexpected change defined as the revision of expectations for inflation in year t made between spring and fall of year t.
  - Data span 20 advanced and 18 emerging and developing economies.
- Main results:
  - Expectations are generally less well anchored in emerging and developing economies than in advanced economies.
  - In emerging and developing economies:
    - A 1 standard deviation shock to current-year inflation expectations = 1.8 percentage points.
    - Medium-term expectations rise by 0.3 percentage points as far as five years into the future in response to such a shock.
  - In advanced economies:
    - A 1 standard deviation shock to current-year inflation expectations = 0.6 percentage point.
    - Medium-term expectations rise by 0.04 percentage point in response.
  - Quantified comparison:
    - Medium-term expectations change 2.5 times more in emerging and developing economies than in advanced economies following a given inflation surprise — computed as (0.3/1.8) divided by (0.04/0.6).
  - Symmetry:
    - Response of medium-term expectations is similar for positive and negative inflation surprises; coefficient on positive-surprise term in augmented specification is not statistically different from zero.
- Effect of inflation-targeting frameworks:
  - Emerging economies with an explicit inflation-targeting framework:
    - After a 1 standard deviation shock equal to 1.3 percentage points, inflation expectations five years out rise by only 0.07 percentage point (statistically indistinguishable from the response for advanced economies).
  - Emerging economies without inflation targeting:
    - Inflation expectations five years out rise by 0.5 percentage point following a 1 standard deviation surprise in current-year inflation.
  - Interpretation:
    - Benefits of inflation targeting may reflect the general quality of domestic monetary management and institutions in economies that adopt such a framework.

### 5. Conclusions
- Impact on advanced economies:
  - Baseline estimate: A 10 percent increase in global food inflation has a peak impact of raising domestic inflation in advanced economies by about ½ percentage points after a year.
  - The impact has declined over time and become less persistent:
    - Over the period since the 1980s, the impact fell to about ¼ percentage points.
    - This reduced impact occurred in the year of the increase of global food prices; after a year, there was essentially no impact on domestic inflation.
  - Possible explanations:
    - (i) absence of significant food shocks in the 1980s and the 1990s;
    - (ii) declining share of food in the consumption basket;
    - (iii) greater wage flexibility, which prevents a wage-price spiral;
    - (iv) increase in the credibility of monetary policy so that an unexpected increase in inflation does not lead to a change in inflation expectations.
- Recent global food price shocks (2000s) — advanced vs emerging economies:
  - Recent global food price shocks had a much bigger impact on emerging than on advanced economies.
  - Contributing factors:
    - Larger share of food in consumption baskets in emerging economies.
    - Inflation expectations are more anchored in advanced than in emerging economies.
- Institutional and classification notes:
  - Inflation-targeting framework (Roger (2010)) defined by four elements:
    - (1) explicit central bank mandate to pursue price stability as the primary objective and a high degree of operational autonomy;
    - (2) explicit quantitative targets for inflation;
    - (3) central bank accountability for performance in achieving the inflation objective through high-transparency requirements;
    - (4) a policy approach based on a forward-looking assessment of inflation pressures using a wide array of information.
  - Many emerging economies have pegged exchange rates, which can reduce ability to respond to domestic inflation shocks; however, weak anchoring of expectations in emerging economies is not driven solely by exchange rate regime.

*Source: _wp15133 - References; IMF Working Paper — Sections 3.4–5 and associated figures and tables as provided in the source content.*

### References .............................................................................................................

### _wp15133 - References .............................................................................................................

### Introduction
- Historical context:
  - Late 1960s–1970s: food price shocks were widely credited with contributing importantly to large increases in inflation in advanced economies such as the United Kingdom and the United States (see Figure 1).
  - Blinder (1982) and Blinder and Rudd (2008) argue that episodes in the 1970s—food price shocks, energy shocks, and removal of wage-price controls—played leading roles in the Great Stagflation, with money and aggregate demand playing smaller roles.
- Proposed reasons for reduced sensitivity of inflation to food shocks after the 1970s:
  - Food accounts for a smaller share of consumption baskets.
  - Greater wage flexibility so wages absorb more of supply shocks (Blanchard and Gali, 2007).
  - Changes in monetary policy conduct and credibility (focus on core inflation, greater anti-inflation credibility; Batini and Nelson, 2005; Rogoff, 2003).
- Recent developments:
  - World food prices declined through the 1980s and 1990s (Figure 2, Panel A) but rose with spikes in 2004, 2008 and 2012 (Figure 2, Panel B).
  - Concern that renewed elevated food prices could spill over into higher overall inflation, especially in emerging economies where food shares are higher and monetary credibility may be lower.
- Questions the paper addresses:
  - What role have global food price movements played in domestic inflation since the 1960s?
  - Do other advanced economies display the same reduced sensitivity of inflation to global food price movements found for U.S. data?
  - Did the impacts of the food price shocks of the 2000s differ across advanced and emerging economies and what channels influenced such differences?

### Data
- Two assembled data sets:
  - Annual data set:
    - Coverage: 44 countries (mainly advanced economies and a few emerging economies).
    - Period: back to the 1960s; used to assess how the impact of food prices on inflation changed over time.
    - Summary statistics presented in Tables A.1 and A.2.
    - Average T in baseline sample: 43 (finite sample bias of lagged dependent variable and fixed effects is in the order of 1/T).
  - Monthly data set:
    - Coverage: 34 advanced economies and 50 emerging economies.
    - Period: 2000–13.
    - Sources: Haver Analytics and IMF Primary Commodity Prices; economies listed in Table A.3.
    - Used to compare advanced vs. emerging economies in the 2000s and to explore transmission channels.
- Illustrative correlations:
  - Annual data (1960–2012): modest positive association between average CPI inflation and average food inflation; positive association between average CPI inflation and average excess money growth (excess money = average money growth minus average real GDP growth) (Figure 3).
  - Monthly 2000–13: positive association between average CPI inflation and food inflation for both advanced and emerging economies (Figure 4).
- Data quality notes:
  - Emerging market historical data are limited and pose quality issues; monthly data (2000s) improves coverage but some quality issues remain.

### Global food prices and inflation: Channels and estimation method
- Headline CPI decomposition (expressed in logs and first differences) highlights channels:
  - Headline inflation deviates from non-food (core) inflation because of shocks to real food prices.
  - Three channels of interest:
    - (i) scale of food price shocks;
    - (ii) food share weight in CPI basket;
    - (iii) monetary policy response and credibility relating to sticky prices (New Keynesian perspective; see Woodford, 2003).
- Estimation approach:
  - Local projections per Jorda (2005) to estimate impulse response functions (IRFs) directly; preferred for flexibility over VAR/ARDL approaches.
  - Baseline regression for each horizon k = 0,..,3 (annual data):
    - Dependent variable: domestic CPI inflation for country i in year t+k.
    - Key regressor: global food price inflation in year t, weighted by the share of food imports in each country.
    - Controls: food share in domestic CPI; lagged dependent variables (l chosen = 2 in baseline); country fixed effects; country-specific time trends.
    - Interpretation: estimated beta_k measures the impact of global food prices on domestic inflation at horizon k, relative to country-specific trends.
  - Inference:
    - IRFs plotted with 90 percent confidence bands (solid line = IRF; dotted lines = 90 percent bands); t=0 denotes the year of the shock.
    - Finite sample bias from lagged dependent variable with fixed effects is of order 1/T; average T = 43 mitigates concern.

### Baseline results (annual data, 1960–2012)
- Main qualitative result:
  - Global food price shocks have positive and statistically significant effects on domestic CPI inflation and these effects are persistent.
- Quantitative magnitudes reported in the paper:
  - A 10 percent increase in global food price (weighted by food import share) typically increases domestic inflation by:
    - 0.35 percentage point in the very short term (the year of the food price shock).
    - about 0.4 percentage point in the medium term (3 years after the shock).
    - peak effect of about 0.7 percentage point 1 year after the shock.
  - Many food price shock episodes involve increases of 50 percent or more, making these effects economically significant.
- Table 1. Baseline Estimates (k = 0,1,2,3)
  - Coefficient estimates and t-statistics (clustered robust standard errors) are reported; *** denotes significance at 1 percent level.
  - For the first reported regressor row (presumably effect per 1 percentage point change in world food price inflation weighted by import share):
    - k=0: 0.036 (5.48)***
    - k=1: 0.074 (7.02)***
    - k=2: 0.063 (7.40)***
    - k=3: 0.040 (4.55)***
  - For the persistence (lagged dependent variable) row:
    - k=0: 0.607 (4.23)***
    - k=1: 0.386 (3.74)***
    - k=2: 0.289 (3.68)***
    - k=3: 0.187 (3.22)***
  - For the third reported row (additional control/regressor):
    - k=0: 0.050 (0.54)
    - k=1: 0.062 (0.83)
    - k=2: 0.038 (0.65)
    - k=3: 0.070 (1.27)
  - Sample sizes and fit:
    - N: 706, 684, 662, 640 for k = 0,1,2,3 respectively.
    - R2: 0.73, 0.65, 0.60, 0.55 for k = 0,1,2,3 respectively.
    - IPS-statistics (Im-Pesaran-Shin test for unit root): -17.700***, -13.745***, -12.407***, -10.720*** for k = 0,1,2,3 respectively.

### Robustness checks
- Potential endogeneity concerns:
  - Bias from including country fixed effects with a lagged dependent variable (Nickell bias) addressed by re-estimating without country fixed effects.
    - Result: difference in point estimates small and not statistically significant (Panel A, Figure 6).
  - Reverse causality and omitted common factors:
    - Three alternative approaches used; estimates remain similar to baseline:
      - Two-step generalized-method-of-moments (GMM) system estimator using up to four lags of domestic and global food price inflation as instruments for global food price inflation (Figure 6B).
      - Re-estimation using the difference between domestic inflation and global food inflation (CPI inflation – food inflation) as the dependent variable (Figure 6C).
      - Panel-VAR approach to purge lagged feedback effects from domestic inflation to global food price inflation (Figure 6D).
- Conclusion from robustness checks:
  - Results are robust: the positive, persistent impact of global food price shocks on domestic CPI inflation remains under alternative estimators and specifications.

*Source: _wp15133 - References .............................................................................................................*

### 3.4 Sub-sample differences

### 3.4 Sub-sample differences

### Changes in pass-through over time
- Re-estimation of Equation 3 for 1960–1982 and 1983–2012 shows the impact of global food prices on domestic CPI inflation was:
  - large, statistically significant, and long-lasting in 1960–1982 (Figure 7, Panel A).
  - much more modest and short-lived in 1983–2012 (Figure 7, Panel B).
- The muted effect in 1983–2012 holds when the share of food in domestic CPI is assumed constant at its time-average value.
  - Comparison of Figure 8 panels indicates weaker impact in the latter period, implying changes in food-share composition are not a key factor explaining the lower effect in 1983–2012.
- Country examples:
  - United Kingdom (Figure 9, Panel A) and United States (Figure 9, Panel B) both show much more muted impacts in 1983–2012 versus 1960–1982.
- Identification and robustness notes:
  - The panel-VAR approach assumes a Cholesky ordering with global food price inflation ordered first; lag length chosen equal to 2.
  - Results are not sensitive to the exact cutoff date in the early 1980s.

*italicized source attribution: IMF Working Paper — Section 3.4 Sub-sample differences*

---

### 3.5 Food shocks vs. monetary factors

### Inclusion of excess money growth (aggregate demand proxy)
- Excess money growth is defined as the difference between money growth and nominal GDP growth and is added as an independent variable to Equation 3.
- Main findings (Figure 10):
  - Over the entire sample 1960–2012 both global food price shocks and domestic excess money shocks have statistically significant and long-lasting effects on domestic inflation.
  - The response to both shocks has changed over time:
    - In 1960–1982 the effect of global food price shocks on domestic inflation dominates.
    - In 1983–2012 the effect of excess money shocks on inflation is more persistent and more precisely estimated.
- Robustness:
  - These findings hold up to the various robustness checks discussed earlier.
  - Conclusion: results on the importance of global food shocks for inflation are robust to the inclusion of aggregate demand factors.

*italicized source attribution: IMF Working Paper — Section 3.5 Food shocks vs. monetary factors*

---

### 3.6 Evidence for the 2000s from annual data

### Advanced vs. emerging market economies (1999–2012)
- Estimation of Equation 3 separately for advanced economies (AEs) and emerging market economies (EMs) for the 1999–2012 period (Figure 11) yields:
  - Global food price shocks of the 2000s had statistically significant short-term effects for both AEs and EMs (Panel A).
  - The effect on domestic inflation in EMs (Panel C) was slightly larger and more persistent than for AEs (Panel B).
- The figure presents the impact of a 1 percentage point change in world food price inflation on domestic (CPI) inflation; solid lines are IRFs and dotted lines are 90 percent confidence bands; t=0 denotes the year of the shock.

*italicized source attribution: IMF Working Paper — Section 3.6 Evidence for the 2000s from annual data*

---

### 4. Evidence from monthly data for the 2000s

### 4.1 Pass-through of world food inflation into domestic food inflation
- Estimation approach:
  - Country-by-country regressions of monthly domestic food price inflation on current and 12 lags of monthly international commodity price inflation, controlling for 12 lags of domestic food price inflation (Equation (4)).
  - Long-term pass-through coefficient = sum of coefficients on international food price inflation divided by 1 minus sum of coefficients on lagged domestic food inflation.
- Median long-term pass-through results (Figure 12):
  - Advanced economies: 0.18 percent (median long-term pass-through of a 1 percent world food price shock to domestic food prices).
  - Emerging economies: 0.34 percent (median long-term pass-through of a 1 percent world food price shock to domestic food prices).
- Factors behind incomplete pass-through:
  - Significant local component in food production and distribution (retail/distribution margins, excise taxes, customs duties).
  - Food subsidies.
  - Significant domestic food production, making domestic agricultural and weather conditions influential.
  - World commodity price indices may not reflect country-specific consumption bundles (composition differences such as wheat, barley, rice).

*italicized source attribution: IMF Working Paper — Section 4.1 Pass-through of world food inflation into domestic food inflation*

---

### 4.2 Pass-through from domestic food inflation to overall inflation
- Key determinants of domestic food → overall CPI pass-through:
  - Share of food in the consumption basket.
  - Anchoring of inflation expectations.
  - Higher food shares raise the direct effect on headline inflation and possibly pass-through into nonfood inflation via wage demands.
  - Poor track record in controlling inflation can raise inflation expectations and thereby increase pass-through.
- Food-share statistics (Figure 13):
  - Advanced economies median food share: 17 percent.
  - Emerging economies median food share: 31 percent.
- Contributions to overall inflation (Figure 14):
  - In 2008:
    - Food prices contributed about 5 percentage points to overall inflation in emerging economies on average.
    - Food prices contributed about 1 percentage point to overall inflation in advanced economies on average.
  - Since 2008:
    - Contribution exceeded 2 percentage points for emerging developing economies.
    - Contribution about half a percentage point for advanced economies.
  - Cross-country variation: in some economies food prices raised headline inflation by about 10 percentage points in 2008 and 5 percentage points in the more recent period.
- Additional channel:
  - World indices composed of relatively unprocessed food commodities; in developed countries most consumed food is highly processed, reducing importance of raw-material price movements in CPI.

*italicized source attribution: IMF Working Paper — Section 4.2 Pass-through from domestic food inflation to overall inflation*

---

### 4.3 The role of inflation expectations
- Estimation approach:
  - Measure responsiveness of medium-term inflation expectations to an unexpected increase in current-year inflation using survey data from Consensus Economics (spring and fall vintages, 1990–2013 / 1990–2014 depending on context) and method based on Levin, Natalucci, and Piger (2004) (Equations (5) and augmented (6)).
  - Unexpected change defined as the revision of expectations for inflation in year t made between spring and fall of year t.
  - Data span 20 advanced and 18 emerging and developing economies.
- Main results:
  - Expectations are generally less well anchored in emerging and developing economies than in advanced economies.
  - In emerging and developing economies:
    - A 1 standard deviation shock to current-year inflation expectations = 1.8 percentage points.
    - Medium-term expectations rise by 0.3 percentage points as far as five years into the future in response to such a shock (Panel A, Figure 15).
  - In advanced economies:
    - A 1 standard deviation shock to current-year inflation expectations = 0.6 percentage point.
    - Medium-term expectations rise by 0.04 percentage point in response (negligible).
  - Quantified comparison:
    - Medium-term expectations change 2.5 times more in emerging and developing economies than in advanced economies following a given inflation surprise — computed as (0.3/1.8) divided by (0.04/0.6).
  - Symmetry:
    - Additional analysis shows the response of medium-term expectations is similar for positive and negative inflation surprises; coefficient on positive-surprise term in augmented specification is not statistically different from zero.
- Effect of inflation-targeting frameworks (Panel B, Figure 15):
  - Emerging economies with an explicit inflation-targeting framework:
    - After a 1 standard deviation shock equal to 1.3 percentage points, inflation expectations five years out rise by only 0.07 percentage point (statistically indistinguishable from the response for advanced economies).
  - Emerging economies without inflation targeting:
    - Inflation expectations five years out rise by 0.5 percentage point following a 1 standard deviation surprise in current-year inflation.
  - Interpretation:
    - Benefits of inflation targeting may reflect the general quality of domestic monetary management and institutions in economies that adopt such a framework.

*italicized source attribution: IMF Working Paper — Section 4.3 The role of inflation expectations*

### 5. CONCLUSIONS

### 5. CONCLUSIONS

### Impact of global food price fluctuations on domestic inflation — advanced economies
- Baseline estimate: A 10 percent increase in global food inflation has a peak impact of raising domestic inflation in advanced economies by about ½ percentage points after a year.
- The impact has declined over time and become less persistent:
  - Over the period since the 1980s, the impact fell to about ¼ percentage points.
  - This reduced impact occurred in the year of the increase of global food prices; after a year, there was essentially no impact on domestic inflation.
- Possible explanations for the diminished role (as noted by Blinder and Rudd (2008)):
  - (i) the absence of significant food shocks in the 1980s and the 1990s;
  - (ii) the declining share of food in the consumption basket;
  - (iii) other changes in the structure of economies such as greater wage flexibility, which prevents a wage-price spiral;
  - (iv) an increase in the credibility of monetary policy so that an unexpected increase in inflation—due to events such as food price shocks—does not lead to a change in inflation expectations.

### Impact of recent global food price shocks — advanced vs emerging economies
- The increase in global food prices since the 2000s, which included three spikes, provides additional evidence:
  - These more recent global food price shocks had a much bigger impact on emerging than on advanced economies.
- Factors contributing to the larger impact in emerging economies:
  - The larger share of food in the consumption baskets in emerging economies on average than in advanced economies.
  - Inflation expectations are more anchored in advanced than in emerging economies, which can contribute to a smaller impact in advanced economies from a sudden burst in inflation due to global food price shocks.

### Institutional and classification notes relevant to interpretation
- Inflation-targeting framework definition used to classify countries (Roger (2010)) includes four main elements:
  - (1) an explicit central bank mandate to pursue price stability as the primary objective of monetary policy and a high degree of operational autonomy;
  - (2) explicit quantitative targets for inflation;
  - (3) central bank accountability for performance in achieving the inflation objective, mainly through high-transparency requirements for policy strategy and implementation;
  - (4) a policy approach based on a forward-looking assessment of inflation pressures, taking into account a wide array of information.
- Many emerging economies in the sample have pegged exchange rates, which reduces their ability to respond to shocks to domestic inflation. However:
  - Additional analysis suggests that inflation expectations are just as weakly anchored in emerging economies that do not have an inflation-targeting framework and have floating exchange rates (according to the de facto classification compiled by Ilzetzki, Reinhart, and Rogoff, 2008).
  - Thus, the association between inflation targeting and the anchoring of expectations is not driven by the exchange rate regime.

*Source: 5. CONCLUSIONS, _wp15133 - 5. CONCLUSIONS*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp15133.pdf_