## 4.1 Data on consumers’ expectations ## Source details **Canonical URL:** [4.1 Data on consumers’ expectations](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022205-print-pdf.pdf) ## Other formats - [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022205-print-pdf.pdf.md) - [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022205-print-pdf.pdf.json) --- ### Data sources and coverage - Source: Joint Harmonized European Union Programme of Business and Consumer Surveys (European Commission, Directorate-General for Economic and Financial Affairs). - Frequency and sample: - Conducted on a rotating basis and published monthly. - Covers 27 EU member states and some other European (candidate) countries. - Around 41 000 randomly chosen consumers across the euro area are asked every month to participate (telephone, online, face-to-face). - Response rate to Q61 is close to 75 percent for the euro area as a whole. - Data used in the analysis: - Average inflation expectations and standard deviation of inflation expectations among respondents for 11 euro-area member states during the period January 2004–June 2019. - Countries in sample: Austria, Belgium, Germany, France, Finland, Greece, Italy, Luxembourg, the Netherlands, Portugal and Spain. - Note: Ireland excluded for prolonged data unavailability; specific earlier gaps for France, Germany, Spain, Netherlands noted. ### Questionnaire items and measurement - Inflation expectations (Q61): - “By how many percent do you expect consumer prices go up/down change in the next 12 months? (Please give a single figure estimate) Consumer prices will increase by...,...%/ decrease by...,...%” - Q61 is asked only to respondents who indicated a perceived positive or negative change in Q6; respondents who indicated “stay about the same” have zero percent imputed for Q61. - The paper uses averages and standard deviations of the quantitative Q61 responses for consumers surveyed every month in euro-area countries. - Aggregated euro area-level inflation expectations and perceptions are calculated and published quarterly by the European Commission (EC), including breakdowns by age, education, gender and income. - Winsorisation and outliers: - Responses are winsorised at the country level at the 5th and 95th percentiles to mitigate outliers and “implausible replies.” - Note: winsorisation typically removes implausibly high inflation rates and very rarely some expectations of deflation in the sample. ### Dispersion (disagreement) measurement and aggregation - Measure of disagreement: - Computed as the standard deviation of answers from the winsorized sample at the country level. - Aggregation: - Figures show a weighted average of cross-section standard deviations of consumer inflation expectations across the eleven euro area countries. ### Expectations of unemployment (Q7) — measurement - Q7 wording: - “How do you expect the number of people unemployed in this country to change over the next 12 months? The number will...increase rapidly (++)/increase slightly (+)/remain the same (=)/fall slightly (-)/fall sharply (- -)/Don’t know (DN).” - Balance measure (bal): - bal = [ (PP + 1/2 · P) − (1/2 · M + MM) ], where bal ranges between −100 and +100; P and M stand for the percentage choosing “increase slightly” and “fall slightly”, PP denotes “increase rapidly” and MM denotes “fall sharply.” - Dispersion of unemployment expectations: - Computed following EC (2016) as: σ_U,t = √( r^+_t + r^-_t − ( r^+_t − r^-_t )^2 ), where r^+_t is the fraction indicating “increase” (sum of ++ and +) and r^-_t is the fraction indicating “fall” (sum of −− and −). ### Empirical patterns and illustrative findings - Co-movement and gaps: - Weighted-average headline inflation and median consumer inflation expectations in the euro area co-move, with expectations staying above actual inflation throughout 2004–2019. - The distance between median expectations and headline inflation was relatively large before the Great Recession; post-2010 saw a closing gap and reduced dispersion across countries, notably during 2017–2018. - Starting in 2019 some decoupling between actual inflation and inflation expectations occurred. - Post-GFC trends: - Average inflation expectations declined in the post-GFC period in most countries in the sample, but with considerable cross-country variation. - If consumer inflation is measured with an index capturing frequent-out-of-pocket purchases (FROOP), this measure is closer to median expectations. - Dispersion dynamics: - Dispersion declined in the post-2010 period, consistent with the reduced gap between median expectations and headline inflation. - Unemployment expectations: - Consumers typically expect unemployment to increase sharply during recessions; the bal variable can be viewed as a lagging indicator. - Figures present weighted-average country-level bal and σ_U,t for the euro area from January 2004. ### Data processing, sample size and summary statistics - Winsorisation: country-level at the 5th and 95th percentiles. - Summary statistics (winsorised Q61 country samples, January 2004 – May 2019): - Table (8) EC Consumer expectations – summary statistics: - Number of countries | Min. obs. per country | Max. obs. per country | Earliest obs. | Latest obs. | Total obs. - 118 | 5 | 469 | 2004m01 | 2019m04 | 3247947 - Table (9) EC Consumer expectations – full sample statistics: - EA | mean 6.09 | median 3.00 | stand.dev. 8.11 | skewness 1.65 | kurtosis 4.93 | obs. 3247947 - Inflation (HICP) < 0%: EA | mean 3.97 | median 0.00 | stand.dev. 6.86 | skewness 2.35 | kurtosis 8.27 | obs. 427492 - Inflation (HICP) ≥ 0%: EA | mean 6.25 | median 3.00 | stand.dev. 8.17 | skewness 1.61 | kurtosis 4.78 | obs. 3020455 ### Sample limitations and caveats - Quantitative inflation perceptions and expectations data were first gathered on an experimental basis and since May 2010 collected regularly; some country-periods are missing. - The EC Survey does not provide sample answers or a specific basket; questions are deliberately vague and respondents are not guided to think in terms of costs of living or specific baskets. - The Survey does not track the same individual respondents over time, which limits panel-based individual-level analyses for consumers. *Source: wpiea2022205-print-pdf - 4.1 Data on consumers’ expectations (https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022205-print-pdf.pdf)* ### 4.1 Data on consumers’ expectations . . . . . . . . . . . . . . . . . . . . . . . . . . . ### 4.1 Data on consumers’ expectations ### Major sections and page references - 4.1 Data on consumers’ expectations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 - 4.2 Data on professional forecasters’ expectations . . . . . . . . . . . . . . . . . . . . . . . . 14 - 4.3 Other data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 ### Subsequent chapter and section map - 5 Individual inflation expectations 16 - 6 Behavior of average expectations 18 ### Subsections under "Behavior of average expectations" - 6.1 Average forecast errors and forecast revisions . . . . . . . . . . . . . . . . . . . . . . . 18 - 6.2 Responses of inflation expectations to macroeconomic shocks . . . . . . . . . . . . 19 - 6.3 Results for inflation expectations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 - 6.4 Results for unemployment expectations . . . . . . . . . . . . . . . . . . . . . . . . . . 25 - 6.5 Robustness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 ### Later chapters - 7 Disagreement about future inflation 28 - 8 Expectations post-GFC 32 ### Subsection under "Expectations post-GFC" - 8.1 Individual inflation expectations post-GFC . . . . . . . . . . . . . . . . . . . . . . 32 *Source: wpiea2022205-print-pdf - 4.1 Data on consumers’ expectations (https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022205-print-pdf.pdf)* ### 8.2 Average inflation and unemployment expectations post-GFC . . . . . . . . . . . 33 ### 8.2 Average inflation and unemployment expectations post-GFC . . . . . . . . . . . 33 ### Introduction and scope - Full-information rational expectations (FIRE) assumptions are questioned; alternative models include informational frictions, over-extrapolation, cognitive discounting, and level-k thinking. - This paper provides new evidence on expectations of future inflation and unemployment using euro-area data including: - monthly surveys of consumers in 11 euro-area countries, - Consensus Economics professional forecasters (individual and cross-section average) available at monthly frequency and going back until mid-1990s, - Joint Harmonized European Union Program of Business and Consumer Surveys (one-year-ahead expectations) covering 27 EU member states; analysis uses country-level cross-section average and standard deviation for 11 euro-area members between January 2004 and June 2019. - Focus: changes in expectations behavior after the global financial crisis (GFC) but before COVID, a period associated with a decline in average inflation and with monetary policy at or close to the effective lower bound. ### Data and empirical approach - Data sources: - Consensus Economics: individual and cross-section average professional forecasts, monthly, available since mid-1990s. - European Commission consumer surveys: one-year-ahead expectations of inflation and unemployment, monthly, covering EU member states (11 euro-area countries used). - Empirical strategies follow literature: - Predictability of individual and cross-sectional average forecast errors from past forecast revisions (as in Coibion and Gorodnichenko (2015) and Bordalo et al. (2020)). - Reaction of average forecast errors to externally identified macroeconomic shocks (as in Coibion and Gorodnichenko (2012) and Angeletos et al. (2020)). - Also analyze responses of actual inflation, average expectations, and standard deviation of expectations to shocks. ### Main empirical findings (Stylized Facts) - Stylized Fact 1: Individual inflation expectations in the euro area overreact to news about inflation. - Regressions of individual forecast errors for professional forecasters’ short-term inflation forecasts show coefficients on past forecast revisions are negative and statistically significant, implying average overreaction at the individual level (consistent with Bordalo et al. (2020)). - Stylized Fact 2: The average of individual inflation expectations underreacts to news initially, but overreacts in the medium term. - Regressions of the forecast error of the cross-section average of professional forecasters show coefficients on past forecast revisions are positive and statistically significant (short-term underreaction). - Regressions of average inflation expectations on externally identified macroeconomic shocks show average expectations tend to overshoot actual outcomes one to two years after a shock (medium-term overreaction). - The medium-term overreaction is present for consumers’ average expectations of unemployment as well. - Pattern consistent with Angeletos et al. (2020) for the U.S.; differs from Coibion and Gorodnichenko (2012) who document underreaction in both short and medium term. Re-running their regressions on a shorter sample excluding late 1970s–early 1980s high U.S. inflation yields evidence of medium-term overreaction. - Stylized Fact 3: Disagreement about future inflation responds to news differently depending on current inflation level. - Consumers’ disagreement about future inflation increases in response to both inflationary and deflationary shocks when current inflation is high, but declines when current inflation is low. - This contrasts with literature expecting disagreement to increase after shocks regardless of current inflation (Mankiw et al., 2003). - Result driven by lower percentiles of the expectations distribution being less responsive to news than upper percentiles, suggesting presence of a zero lower bound (ZLB) on expectations (similar observation in Gorodnichenko and Sergeyev (2021) for U.S. surveys). - Stylized Fact 4: The magnitude of overreaction of individual forecasts to news increased after the GFC. - Stylized Fact 5: The reaction of average forecasts and of actual inflation to news became much more muted post-GFC in the euro area. - Underreaction of average inflation expectations in the short-term and overreaction in the medium-term became much weaker post-GFC. - No evidence of changes in responsiveness of average expectations (and of actual inflation) to news in the U.S. in the same period. ### Theoretical interpretation - Facts 1 and 2 support a combination of imperfect information and over-extrapolation (Angeletos et al. (2020)): - Imperfect information: individual agents overreact to news, but cross-sectional averaging initially underreacts due to noisy signals. - Over-extrapolation: as time passes, over-extrapolation dominates and average expectations overreact in the medium term. - Evidence opposes under-extrapolation mechanisms implied by cognitive discounting or level-k thinking. - Facts 3–5 suggest post-GFC decline in persistence of shocks’ impacts on inflation and expectations; potential drivers include: - presence of downward nominal price and wage rigidities together with reinforcing feedback between low inflation and weaker passthrough from wage increases to inflation (Consolo et al., 2021), - changes in firms’ pricing behavior post-GFC (Koester et al., 2021), - increased inattention to inflation developments by agents in a low inflation environment. ### Policy implications - Initial muted impact of shocks on average inflation expectations can mask delayed medium-term overreaction; policymakers should remain vigilant even if average expectations respond modestly to a surge in inflation. - Nature of the shock matters: - Overreaction of average inflation expectations is more persistent following a demand shock. - Overreaction driven by an oil supply shock fades more quickly; consumers tend to believe gas price inflation is slightly negatively autocorrelated (Binder 2018). - Nonlinear relationship between disagreement and current inflation challenges expectations anchoring in high inflation environments. - For low-inflation countries, presence of a zero lower bound on expectations implies deflation episodes might be associated with lower economic costs than predicted by models without this constraint. ### Simple model illustrating over- and under-reaction - Partial equilibrium framework (Angeletos et al. (2020)) incorporates imperfect information and over-extrapolation. - Inflation process x_t follows AR(1): x_t = ρ x_{t−1} + r ε_t, with ρ ∈ (0,1), ε_t ∼ N(0,1). - Agents observe noisy signals s_{i,t} = x_t + u_{i,t} / √ˆτ, where ˆτ measures signal precision and u_t ∼ N(0,1). - Agents may perceive process with different persistence: x_t = ˆρ x_{t−1} + r ε_t; ˆρ > ρ captures over-extrapolation; ˆρ < ρ captures under-extrapolation. - Individual one-period-ahead expectations: E_{i,t}[x_{t+1}] = ˆρ E_{i,t}[x_t] = ˆρ[(1−ˆg) ˆρ E_{i,t−1}[x_t] + ˆg s_{i,t}], with ˆg = 1 − ˆλ/ˆρ and ˆλ as defined in the text. ### Parameter calibration and simulation scenarios - Simulations use equations (1) and (4) to generate inflation x_t and 200 individual expectations E_{i,t}[x_{t+1}], then compute average expectations and forecast errors. - Table (1) Parameter Calibration (values by scenario FIRE / underreaction / overreaction): - ρ AR(1) coef. in inflation dynamic: 0.7 / 0.7 / 0.7 - r standard deviation of error term: 1 / 1 / 1 - ˆτ perceived precision: 10 / 36 / 10 - ˆρ perceived AR(1) coef. in inflation dynamic: 0.7 / 0.7 / 0.9 - σ_u standard deviation of error term: 0.01 / 0.01 / 0.01 - Simulation outcomes: - Under FIRE (ˆρ = ρ, ˆτ → ∞): expectations respond to shock as much as actual inflation; forecast error is zero. - With noisy signals (finite ˆτ): average inflation expectations underreact to shocks, implying a positive forecast error. - With over-extrapolation (ˆρ > ρ): agents overreact and average forecast error becomes negative. *International Monetary Fund — IMF Working Papers excerpt (Biases in survey inflation expectations).* ### 4.1 Data on consumers’ expectations ### 4.1 Data on consumers’ expectations ### Expectations of inflation - Data source: Joint Harmonized European Union (EU) Programme of Business and Consumer Surveys (European Commission, Directorate-General for Economic and Financial Affairs). - Survey coverage and frequency: - Conducted on a rotating basis and published monthly. - Covers 27 EU member states and some other European (candidate) countries. - Around 41 000 randomly chosen consumers across the euro area are asked every month to participate (telephone, online, face-to-face). - Response rate to Q61 is close to 75 percent for the euro area as a whole. - Relevant questionnaire item (Q61): - “By how many percent do you expect consumer prices go up/down change in the next 12 months? (Please give a single figure estimate) Consumer prices will increase by...,...%/ decrease by...,...%” - Q61 is asked only to respondents who indicated a perceived positive or negative change in Q6; respondents who indicated “stay about the same” have zero percent imputed for Q61. - Measurement and aggregation: - The paper uses averages and standard deviations of the quantitative Q61 responses for consumers surveyed every month in euro-area countries. - Aggregated euro area-level inflation expectations and perceptions are calculated and published quarterly by the European Commission (EC), including breakdowns by age, education, gender and income. - Data used: average inflation expectations and standard deviation of inflation expectations among respondents for 11 euro-area member states during the period January 2004–June 2019. - Countries in sample: Austria, Belgium, Germany, France, Finland, Greece, Italy, Luxembourg, the Netherlands, Portugal and Spain. (Ireland excluded for prolonged data unavailability; specific earlier gaps for France, Germany, Spain, Netherlands noted.) - Data processing for outliers: - Responses are winsorised at the country level at the 5th and 95th percentiles to mitigate outliers and “implausible replies.” - Note: winsorisation typically removes implausibly high inflation rates and very rarely some expectations of deflation in the sample. - Empirical patterns (illustrative findings from figures and text): - Weighted-average headline inflation and median consumer inflation expectations in the euro area co-move, with expectations staying above actual inflation throughout 2004–2019. - The distance between median expectations and headline inflation was relatively large before the Great Recession; post-2010 saw a closing gap and reduced dispersion across countries, notably during 2017–2018. - Starting in 2019 some decoupling between actual inflation and inflation expectations occurred. - Average inflation expectations declined in the post-GFC period in most countries in the sample, but with considerable cross-country variation. - If consumer inflation is measured with an index capturing frequent-out-of-pocket purchases (FROOP), this measure is closer to median expectations. ### Dispersion (disagreement) of inflation expectations - Measure of disagreement: - Computed as the standard deviation of answers from the winsorized sample at the country level. - Aggregation: - Figures show a weighted average of cross-section standard deviations of consumer inflation expectations across the eleven euro area countries. - Empirical pattern: - Dispersion declined in the post-2010 period, consistent with the reduced gap between median expectations and headline inflation. ### Expectations of unemployment - Survey question (Q7) wording: - “How do you expect the number of people unemployed in this country to change over the next 12 months? The number will...increase rapidly (++)/increase slightly (+)/remain the same (=)/fall slightly (-)/fall sharply (- -)/Don’t know (DN).” - Quantitative representation (balance): - A “balance” (bal) is computed as the difference between positive and negative answers, measured as percentage points of total answers. For Q7, the balance is calculated using the weighed formula: bal = [ (PP + 1/2 · P) − (1/2 · M + MM) ], where bal ranges between −100 and +100; P and M stand for the percentage of respondents choosing the positive (“increase slightly”) and negative (“fall slightly”) option, respectively; PP denotes “increase rapidly” and MM denotes “fall sharply.” - Dispersion of unemployment expectations: - Computed following EC (2016) as: σ_U,t = √( r^+_t + r^-_t − ( r^+_t − r^-_t )^2 ), where r^+_t is the fraction of responses indicating “increase” (sum of ++ and +) and r^-_t is the fraction indicating “fall” (sum of −− and −). - Empirical patterns: - Consumers typically expect unemployment to increase sharply during recessions; the bal variable can be viewed as a lagging indicator. - Figures present weighted-average country-level bal and σ_U,t for the euro area from January 2004. ### Sample limitations and notes - Quantitative inflation perceptions and expectations data were first gathered on an experimental basis and since May 2010 collected regularly; some country-periods are missing. - The EC Survey does not provide sample answers or a specific basket; questions are deliberately vague and respondents are not guided to think in terms of costs of living or specific baskets. - The Survey does not track the same individual respondents over time, which limits panel-based individual-level analyses for consumers. *Italic: Source: wpiea2022205-print-pdf - 4.1 Data on consumers’ expectations* ### 6.1 Average forecast errors and forecast revisions ### 6.1 Average forecast errors and forecast revisions ### Method and specification - Re-estimate equation (5) using average forecasts for current-year inflation from Consensus Economics: - π^j_{t+k,t} − ̄π^j_{t+k,t|t+l} = β_0 + β_1(̄π^j_{t+k,t|t+l} − ̄π^j_{t+k,t|t+l−m}) + ε^j_{t+k} - ̄π^j_{t+k,t|t+l} and ̄π^j_{t+k,t|t+l−m} are average forecasts for inflation between months t and t+k made in months t+l and t+l−m. - Sample: 10 countries, 1996m01–2019m06 (results also hold for 2004m01–2019m06). - Estimations include month dummies and country fixed effects; Figure 10 plots month-by-month β_1 controlling for country fixed effects. ### Main estimation results (Table 3) - Coefficients from regression (6), dependent variable: current calendar year average inflation forecast error from Consensus Forecasts: - 1-month revision: 1.189*** (standard error 0.125) - 3-month revision: 0.557*** (standard error 0.066) - Observations: - Column (1): 2,629 - Column (2): 2,151 - Country FE: YES - Month FE: YES - Robust standard errors in parentheses - Significance notation: ***p <0.01, **p <0.05, *p <0.1 ### Key finding - Coefficient β_1 is positive and statistically significant across specifications, indicating underreaction of average inflation forecasts to news about inflation. - The size of underreaction declines over time (β_1 becomes smaller), as shown in Figure 10. --- ### 6.2 Responses of inflation expectations to macroeconomic shocks ### Purpose and data - Study how average survey expectations respond to news over horizons longer than next few months. - Data: - EC Consumer Survey: average expectations of 12-months ahead inflation and unemployment (monthly). - Consensus Economics: 12-months ahead average inflation expectations of professional forecasters, computed as weighted average of current and next calendar year inflation forecasts. - Methodology builds on CG (2012) ARMA specification and uses local projections of Jordà (2005) for flexibility. ### Key regression specifications - ARMA-style test (CG 2012) for forecast errors to shocks: - π_{t,t−4} − ̄π_{t,t−4|t−4} = α + Σ_{k=1}^K β_k(π_{t−k,t−4−k} − ̄π_{t−k,t−4−k|t−4−k}) + Σ_{l=0}^L γ_l ε^s_{t−l} + ξ^s_t - Under FIRE: γ_0 = 0; underreaction: γ_0 > 0; overreaction: γ_0 < 0. - Local projections (Jordà 2005) used to estimate dynamic responses: - π^j_{t+h,t+h−12} − ̄π^j_{t+h,t+h−12|t+h−12} = α_{j,h} + Σ_{k=1}^K β^h_k(...) + Σ_{k=1}^K δ^h_k π^j_{t−k,t−k−12} + Σ_{l=0}^L γ^h_l ε^s_{t−l} + ξ^j,s_{t+h} - K = 6, L = 2, horizons h = 1,2,..,24. - Positive γ^h_0 indicates underreaction h months after a positive shock; negative indicates overreaction. - Also regressions for actual year-on-year inflation and average expectations (equation (9)). ### Timing convention - Responses of actual inflation and mean inflation expectations: analyze h = 1,2,..24 months after shock. - Forecast errors in equation (8): analyze h = 13,..24 months (to avoid counting expectations formed before shock materialized). ### Identification of shocks and sample coverage (Table 4) - Shocks (frequency | sample | avg. share of sample inflation volatility (in %)): - Kanzig (2021) oil supply news | monthly | 2004m01–2017m12 | 15.8 - Kilian (2009) global demand | monthly | 2004m01–2019m06 | 3.8 - Jarociński and Karadi (2020) monetary policy | monthly | 2004m01–2016m12 | 2.2 - Jarociński and Karadi (2020) monetary information | monthly | 2004m01–2016m12 | 1.1 - Bobeica and Jarociński (2019) global demand | quarterly | 2004Q1–2019Q2 | 4.3 - Bobeica and Jarociński (2019) global oil supply | quarterly | 2004Q1–2019Q2 | 1.9 - All shocks are either global or, for monetary shocks, common to all countries. - Inflation variance decomposition based on country-specific VAR(2) models estimated on 2004m01–2019m06; reported contribution is average across countries. --- ### 6.3 Results for inflation expectations ### Consumers (EC Consumer Survey) - Response to adverse Kanzig (2021) global oil supply shock and positive Kilian (2009) global demand shock (regression (9)): - Both actual inflation and 12-month ahead expected inflation increase and remain elevated for a prolonged period (of 15 to 24 months). (Figure 11) - Forecast errors (regression (8), months 13-24 after shock; Figure 12): - For both oil and demand shocks, forecast errors exhibit a statistically significant downward bias over most of the horizon considered. - Interpretation: EC Survey respondents overestimate the medium-term impact of those shocks on inflation. ### Monetary shocks (Jarociński and Karadi (2020)) - Two shocks: monetary policy shock (surprise policy easing increases inflation) and monetary information shock (reveals central bank’s assessment; complementary shock leading to downward pressure). - Responses (regression (9), Figure 13): - Monetary policy easing: inflationary effect on realized inflation and on inflation expectations on impact. - Monetary information shock: deflationary effect on impact; response of actual inflation significant initially and flips sign after 10 months. - Both shocks have more significant and persistent impact on inflation expectations than on actual inflation. - Forecast errors (regression (8), Figure 14): - Monetary policy shock: forecast errors show a statistically significant negative bias (consistent with overreaction of expectations to inflationary shock). - Monetary information shock: forecast errors show a statistically significant positive bias (consistent with overreaction to deflationary information shock). ### Professionals (Consensus Forecasts) - One-year ahead average forecast errors (Figure 15): - Overreaction observed for three out of four shocks: the oil supply shock and the two monetary shocks. - After the global demand shock, forecast errors are positive at months 13–14 (suggesting underreaction) and then become statistically insignificant. - Overall interpretation: - Professionals’ responses are broadly consistent with consumers’: agents’ inflation expectations move in response to macro shocks consistent with economic intuition, but both professionals and consumers tend to overestimate the medium-term impact of shocks on inflation. --- ### 6.4 Results for unemployment expectations ### Responses to shocks (EC Consumer Survey) - Directional responses (regression (9), Figure 16): - Adverse oil supply shock: increases expected unemployment. - Positive global demand shock: reduces unemployment expectations. - Easing monetary policy shock: reduces unemployment expectations. - Deflationary monetary information shock: increases expected unemployment. ### Forecast errors and over/underreaction (regression (8), months 13-24; Figure 17) - Forecast errors behavior: - Global demand and the two monetary policy shocks: suggest overreaction starting already in month 13. - Oil supply shock: overreaction of expectations is delayed and starts after 18 months. - Interpretation: - Households tend to overestimate the medium-term impact of shocks on unemployment. ### Stylized fact 2 - "Average expectations underreact to news initially, but overreact in the medium term." - Consistent with Angeletos et al. (2020) evidence for the U.S.; contrasts with CG (2012) who find continuous underreaction in CG’s ARMA framework. --- ### 6.5 Robustness ### Alternative shocks (quarterly from Bobeica and Jarociński (2019)) - Quarterly global oil supply shock and aggregate demand shock analyzed using regression (8) with 4 lags of dependent variable and realized inflation; focus on forecast error responses in 4–10 quarters after shock (Figure 18). - Results: - Average year-ahead inflation expectations overreact to both inflationary shocks, producing a statistically significant negative bias in average forecast errors for most of the horizon considered. - Similar responses obtain using Consensus Economics professional forecasts. ### Country-specific regressions and U.S. evidence - The text indicates verification that results hold in country-specific regressions and discussion of U.S. evidence, with consistent findings for overreaction of medium-term expectations (referenced but detailed figures/tables beyond the provided excerpt). *Source: wpiea2022205-print-pdf - 6.1 Average forecast errors and forecast revisions* ### Appendix A shows responses of average forecast errors from the EC Consumer Survey to Kanzig ### wpiea2022205-print-pdf - Appendix A: Responses of average forecast errors from the EC Consumer Survey to Kanzig (2021) oil and Kilian (2009) global demand shocks ### EU consumer and panel evidence on expectations responses to shocks - Responses of average forecast errors from the EC Consumer Survey to Kanzig (2021) oil and Kilian (2009) global demand shocks for Germany, France, Italy and Spain point to overreaction of expectations, consistent with panel regressions (although in a few instances less significant). - Results are robust to controlling for the global financial crisis (GFC) and adding a post-GFC time dummy. ### Evidence from the U.S. - Regressions (8) and (9) estimated using U.S. inflation expectations from the Michigan Consumer Survey and the Survey of Professional Forecasters. - Shocks used: Gali (1999) technology shock, Hamilton (1996) oil supply shock, and Kanzig (2021) oil news shock. - Sample: quarterly, between 1990Q1-2020Q1. - Figure 29 (Appendix A) shows responses of average forecast errors for the two surveys; results indicate overreaction to shocks, particularly for consumer expectations (estimates characterized by larger standard errors). - Reconciliation with CG (2012): - Using CG (2012) sample and shocks (1976Q1-2007Q3) produces underreaction at medium-term horizons, consistent with CG (2012). - When using the CG (2012) data but restricting the sample to start in 1985 (excluding very high inflation in late 1970s/early 1980s), impulse responses change: where statistically significant, 5 out of 6 cases show signs consistent with overreaction (only SPF survey with Hamilton (1996) shock shows underreaction). (See Figure 30 in Appendix.) ### Disagreement about future inflation (regressions and impulse responses) - Regression estimated (CG (2012) specification): σ(π^j_{t+h,t+h−12|t+h−12}) regressed on its lags, past inflation, and Lags of absolute shocks |s_{t−l}| with K=6 and L=2. - Monthly shocks analyzed: Kanzig (2021) oil supply shock and Kilian (2009) global demand shock. Quarterly shocks: Bobeica and Jarocinski (2019) oil and global demand shocks. - Main empirical patterns: - Surprisingly, disagreement (cross-sectional standard deviation of 12-month-ahead inflation forecasts) broadly declines following macroeconomic shocks for consumers. - Disagreement clearly declines among consumers following the Kanzig (2021) oil supply shock and the Kilian (2009) global demand shock. - For the quarterly BJ (2019) shocks: no statistically significant response of disagreement to the global demand shock; for the global oil supply shock, disagreement starts to decline after one year. - Footnote: Impulse responses to two monetary shocks also indicate a decline in disagreement among consumers. ### Nonlinear specification and interaction with current inflation - Extended specification (equation (11)) includes interaction terms between absolute shocks and current inflation: η^h_l (|s_{t−l}| × π^j_{t−l,t−l−12}). - Findings from Figure 20: - Standalone shock coefficient (absolute value) remains negative (blue lines). - Interaction term coefficient with current period inflation is positive and statistically significant immediately after shocks (black lines). - Interpretation: consumers’ disagreement about future inflation increases in response to both inflationary and deflationary shocks when current inflation is sufficiently high, but disagreement declines when current inflation is low. ### Percentile dynamics and implied zero lower bound of expectations - Behavior of percentiles of the inflation expectations distribution in the EC Consumer Survey: - If mean expectations fall to zero, lower percentiles (20th, 30th) stay at zero and do not fall further. This dynamic is robust across countries and over time. - Lower percentiles are much less responsive to shocks compared to upper percentiles (70th, 80th). - Reestimation of (9) using percentiles as dependent variables: Figure 21 shows results for 30th and 70th percentiles (black lines 70th percentile; blue lines 30th percentile). - Interpretation: evidence suggestive of a zero lower bound of inflation expectations, possibly consistent with downward price rigidities. ### Stylized fact 3 - Consumers’ disagreement about future inflation: - Increases in response to both inflationary and deflationary shocks when current inflation is high. - Declines when current inflation is low. - Lower percentiles of the expectations distribution are less responsive to news about inflation compared to upper percentiles, suggesting presence of a zero lower bound of expectations. ### Expectations post-GFC and the ZLB (zero lower bound) interaction - ZLB definition: monetary policy at ZLB when policy rate is below 1 percent (following Ehrman et al. (2019)). - For the euro area, ZLB periods: between May 2009 and April 2011, and since January 2012 onward. - Regressions (5), (6), (8), (9) augmented with ZLB dummy and interaction terms (examples in equations (12) and (13)). - Caveat: ZLB period covers a majority of the post-GFC sample, so the dummy may capture other effects as well. ### Individual inflation expectations post-GFC (Consensus Economics professionals) - Table 5: regression of individual inflation forecast errors on forecast revisions with ZLB interaction; country-forecaster and month fixed effects included. - Key coefficients and statistics (preserved exactly as reported): - Column (1) — 1-month revision specification: - 1-month revision: -0.221*** (0.010) - 1-month revision-ZLB interaction: -0.092*** (0.020) - ZLB: -0.067*** (0.010) - Observations: 30,471 - Column (2) — 3-month revision specification: - 3-month revision: -0.044*** (0.013) - 3-month revision-ZLB interaction: -0.097*** (0.017) - ZLB: -0.045*** (0.010) - Observations: 24,625 - Country-forecaster FE: YES - Month FE: YES - Robust standard errors in parentheses - Significance notation: ***p <0.01, **p <0.05, *p <0.1 - Interpretation: - The coefficient on the interaction between forecast revisions and the ZLB is negative and statistically significant for both revision horizons considered, indicating presence of the ZLB is associated with a larger overreaction of individual inflation expectations to news. - Month-by-month regressions (not shown) indicate this amplification impact is driven by forecasts made in the first six months of the calendar year. ### Stylized fact 4 - Overreaction of individual forecasts to news increased after the global financial crisis. *Source: Appendix A of wpiea2022205-print-pdf (IMF Working Paper).* ### 8.2 Average inflation and unemployment expectations post-GFC ### 8.2 Average inflation and unemployment expectations post-GFC ### Responsiveness of inflation expectations and actual inflation at the ZLB - Impulse-response evidence (EU Consumer Survey and Consensus Forecasts) to four monthly shocks (Kanzig (2021) global oil supply shock; Kilian (2009) global demand shock; Jarociński and Karadi (2020) euro area monetary policy shock; Jarociński and Karadi (2020) euro area monetary information shock) shows: - For two monetary shocks and the global demand shock, the response of average inflation expectations is more muted at the ZLB compared to periods when monetary policy is outside the ZLB. - For the oil supply shock, the response of expectations at the ZLB is initially more muted but more prolonged. - The impulse responses of actual inflation present very similar patterns as for average inflation expectations. - Average unemployment expectations: - Responses to the same four shocks are generally more muted initially at the ZLB, but can be more persistent at the ZLB (notably for the oil supply and aggregate demand shocks). - Overall interpretation: - Evidence is suggestive of more muted responsiveness of actual inflation and average inflation expectations to a range of economic shocks when the policy rate is close or at the ZLB. - Evidence for unemployment expectations is more mixed. ### Stylized fact 5 - Reaction of both average inflation expectations and actual inflation to news is (much) more muted post-GFC. ### Inflation forecast errors and ZLB interaction (econometric evidence) - Table (6) reports results from regression (12) of average inflation forecast errors (Consensus Forecasts) on forecast revisions with a ZLB interaction. - Column (1) (1-month revision): - 1-month revision: 1.385*** (0.12) - 1-month revision–ZLB interaction: -0.409** (0.16) - ZLB: 0.015 (0.03) - Observations: 2,629 - Country FE: YES - Month FE: YES - Column (2) (3-month revision): - 3-month revision: 0.570*** (0.08) - 3-month revision–ZLB interaction: 0.002 (0.05) - ZLB: 0.042 (0.4) - Observations: 2,151 - Country FE: YES - Month FE: YES - Standard errors in parentheses. Significance: ***p <0.01, **p <0.05, *p <0.1. - Interpretation: - Some evidence that the ZLB reduces the magnitude of the short-term underreaction of average inflation forecasts to news: the 1-month revision–ZLB interaction is negative and statistically significant. - No statistically significant ZLB interaction for 3-month revisions. ### Forecast errors from survey data at the ZLB (EU Consumer Survey) - Impulse responses of average forecast errors (EU Consumer Survey) to the four shocks show: - Consumers’ inflation forecast errors are smaller in the medium term for all four shocks when monetary policy is at the ZLB, and frequently the forecast errors are not statistically different from zero. - Professional forecasts (Consensus Economics) show a similar pattern of smaller forecast errors; exception: Bobeica and Jarocinski (2019) global demand shock, where professionals underreact at the ZLB in the medium term. - U.S. robustness check: - Defining the ZLB as periods with the Fed funds rate below 0.25 percent (broadly covering the years 2008-2016), the authors do not find evidence of more muted responses of forecast errors in the U.S. ### Unemployment forecast errors at the ZLB - Average forecast errors for unemployment expectations (EU Consumer Survey) become smaller in absolute terms at the ZLB, though they continue to show overreaction to shocks when the ZLB constraint binds. ### Mechanisms and simple-model calibration explaining post-GFC changes - Framework: partial-equilibrium model with imperfect information and over-extrapolation (based on Angeletos et al. (2020)); simulate inflation x_t and 200 individual expectations E_{i,t}[x_{t+1}] using equations (1) and (4). - Parameter calibration (Table (7)): - ρ AR(1) coef. in inflation dynamic: 0.7 - r standard deviation of error term: 1 - ˆτ perceived precision: 0.15 - ˆρ perceived AR(1) coef. in inflation dynamic: 0.82 - σ_u standard deviation of error term: 0.01 - Simulation results (Figure 26): - Agents on average overreact initially to an inflationary shock but underreact over the medium term; by design individuals overreact since ˆρ > ρ. - Hypotheses considered for post-GFC changes: 1. Individual perceptions of persistence, ˆρ, have increased. 2. Transparency about shocks increased (higher τ), lowering signal noise. 3. Persistence of shocks has declined (lower ρ). 4. Impact of shocks on inflation has declined (lower r). - Evaluation of hypotheses: - Hypothesis 1 can explain larger individual overreaction post-GFC but not the more muted response of average expectations and actual inflation. - Hypothesis 2 (greater transparency) cannot generate larger individual overreaction nor more muted actual inflation response. - A configuration with lower values of ρ and ˆρ, but an increased difference between ˆρ and ρ, can generate: - Larger individual-level overreaction post-GFC (stylized fact 4). - More muted reaction of both average inflation expectations and actual inflation to shocks (stylized fact 5). - Smaller initial underreaction and medium-term overreaction of average expectations. - For lower responsiveness of actual inflation and expectations to shocks on impact, r needs to decline as well. - Example post-GFC calibration used in simulation (Figure 27): - ˆρ = 0.73, ρ = 0.6, r = 0.8 - Caveats and open questions: - Not straightforward why post-GFC responsiveness of actual inflation to shocks would decline by more than responsiveness of individual expectations, producing larger individual forecast overshoots. - General equilibrium effects are not modeled and could be important. - The simple framework does not satisfactorily capture dynamics of the standard deviation of expectations; in simulations the standard deviation always increases after a shock (though by less under the “post-GFC” calibration). ### Consolidated conclusions (stylized facts documented) - Five novel facts for the euro area: 1. Individual inflation forecasts overreact to news about inflation. 2. Average expectations underreact to news initially, but overreact in the medium term; same pattern for unemployment expectations. 3. Disagreement about future inflation increases in response to news when current inflation is high, and declines when current inflation is low, suggesting a zero lower bound of expectations. 4. Overreaction of individual forecasts to news increases after the global financial crisis. 5. Post-GFC the reaction of both average expectations and actual inflation to news is (much) more muted. - Model implications: - Imperfect information plus over-extrapolation can account for stylized facts 1 and 2. - A decline in responsiveness to shocks and in persistence of inflation dynamics can account for stylized fact 5 post-GFC. - A more sophisticated theory is needed to fully rationalize all five stylized facts, including state-contingent dynamics of dispersion of inflation expectations (stylized fact 3) and changes in the strength of over-extrapolation bias at the individual level (stylized fact 4). *IMF Working Paper — Biases in survey inflation expectations (section 8.2).* ### REFERENCES ### REFERENCES ### Cited works (selection from the source) - Angeletos, Marios, Zhen Huo and Karthik Sastry (2020), ‘Imperfect macroeconomic expectations: Evidence and theory’, NBER Macroeconomics Annual 53. - Arioli, Rodolfo, Colm Bates, Heinz Christian Dieden, Iona Duca, Roberta Friz, Christian Gayer, Geoff Kenny, Aidan Meyler and Iskra Pavlova (2017), ‘Eu consumers’ quantitative inflation perceptions and expectations: and evaluation’, ECB Occasional paper Series (186). - Bachmann, Rüdiger, Tim O. Berg and Eric R Sims (2015), ‘Inflation expectations and readiness to spend: Cross-sectional evidence’, American Economic Journal: Economic Policy 7(1), 1–35. - Binder, Carola (2018), ‘Inflation expectations and the price at the pump’, Journal of Macroeconomics 58, 1–18. - Bobeica, Elena and Marek Jarociński (2019), ‘Missing disinflation and missing inflation: A var perspective’, International Journal of Central Banking 57. - Bordalo, Pedro, Nicola Gennaioli and Andrei Shleifer (2018), ‘Diagnostic expectations and credit cycles’, Journal of Finance 73(1), 199–227. - Bordalo, Pedro, Nicola Gennaioli, Yueran Ma and Andrei Shleifer (2020), ‘Overreaction in macroeconomic expectations’, American Economic Review 110(9), 2748–82. - Coibion, Olivier and Yuriy Gorodnichenko (2012), ‘What can survey forecasts tell us about information rigidities?’, Journal of Political Economy 120(1), 116–159. - Coibion, Olivier and Yuriy Gorodnichenko (2015), ‘Information rigidity and the expectations formation process: A simple framework and new facts’, American Economic Review 105(8), 2644–78. - Coibion, Olivier, Yuriy Gorodnichenko and Michael Weber (2021), ‘How inflation expectations affect households’ spending decisions’, Vox EU. URL: https://voxeu.org/article/how-inflation-expectations-affect-households-spending-decisions - Coibion, Olivier, Yuriy Gorodnichenko and Rupal Kamdar (2018), ‘The formation of expectations, inflation, and the phillips curve’, Journal of Economic Literature 56(4), 1447–91. - Coibion, Olivier, Yuriy Gorodnichenko, Saten Kumar and Mathieu Pedemonte (2020), ‘Inflation expectations as a policy tool?’, Journal of International Economics 124. - Consolo, Agostino, Gerrit Koester, Christiane Nickel, Mario Porqueddu and Frank Smets (2021), ‘The need for an inflation buffer in the ecb’s price stability objective – the role of nominal rigidities and inflation differentials’, ECB Working Paper No 279. - Crump, K. Richard, Stefano Eusepi, Andrea Tambalotti and Giorgio Topa (2021), ‘Subjective intertemporal substitution’, Journal of Monetary Economics. - Dover, Jonas, Ulrich Fritsch and Jiri Slacalek (2009), ‘Disagreement among forecasters in g7 countries’, ECB Working Paper Series (1082). - Drager, Lena and Giang Nghiem (2021), ‘Are consumers’ spending decisions in line with a euler equation?’, The Review of Economics and Statistics 103(3), 580–596. - Duca-Radu, Ioana, Geoff Kenny and Andreas Reuter (2020), ‘Inflation expectations, consumption and the lower bound: Micro evidence from a large multi-country survey’, Journal of Monetary Economics. - Duca-Radu, Ioana, Goeff Kenny and Andreas Reuter (2021), ‘Macroeconomic stabilisation, the lower bound, and inflation: Expectations matter’, Vox EU. URL: https://voxeu.org/article/macroeconomic-stabilisation-lower-bound-and-inflation-expectations-matter - D’Acunto, Francesco, Ulrike Malmendier, Juan Ospina and Michael Weber (2021), ‘Exposure to grocery prices and inflation expectations’, Journal of Political Economy 129(5), 1615–1639. - EABCN (2021), ‘Euro area business cycle dating’. URL: https://eabcn.org/dc/news - Ehrmann, Michael, Gaetano Gaballo, Peter Hoffmann and Georg Strasser (2019), ‘Can more public information raise uncertainty? The international evidence on forward guidance’, Journal of Monetary Economics 108(C), 93–112. - European Commission (2019), ‘European business cycle indicators: 1st quarter 2019’, Technical Paper 031. - European Commission (2021), ‘Business and consumer survey database’. URL: https://ec.europa.eu/info/business-economy-euro/indicators-statistics/economic-databases/business-and-consumer-surveys/download-business-and-consumer-survey-data/time-series - Eurostat (2021), ‘Database prices’. URL: https://ec.europa.eu/eurostat/data/database?nodecode=lfsq - Farhi, Emmanuel and Iván Werning (2019), ‘Monetary policy, bounded rationality, and incomplete markets’, American Economic Review 109(11), 3887–3928. - Forsells, Magnus and Geoff Kenny (2004), ‘Survey expectations, rationality and the dynamics of euro area inflation’, Journal of Business Cycle Measurement and Analysis 2004(1), 13–41. - Gabaix, Xavier (2020), ‘A behavioral new keynesian model’, American Economic Review 110(8), 2271–2327. - García-Schmidt, Mariana and Michael Woodford (2019), ‘Are low interest rates deflationary? a paradox of perfect-foresight analysis’, American Economic Review 109(1), 86–120. - Gorodnichenko, Yuriy and Dmitriy Sergeyev (2021), ‘Zero lower bound on inflation expectations’, NBER Working Paper No. 29496. - Hamilton, James D. (1996), ‘This is what happened to the oil price-macroeconomy relationship’, Journal of Monetary Economics 38(2), 215–220. - Ichiue, Hibiki and Shusaku Nishiguchi (2015), ‘Inflation expectations and consumer spending at the zero bound: Micro evidence’, Economic Inquiry 53(2), 1086–1107. - Jarociński, Marek and Peter Karadi (2020), ‘Deconstructing monetary policy surprises—the role of information shocks’, American Economic Journal: Macroeconomics 12(2), 1–43. - Jordà, Òscar (2005), ‘Estimation and inference of impulse responses by local projections’, American Economic Review 95(1), 161–182. - Kanzig, Diego (2021), ‘The macroeconomic effects of oil supply news: Evidence from OPEC announcements’, American Economic Review forthcoming. - Kilian, Lutz (2009), ‘Not all oil price shocks are alike: Disentangling demand and supply shocks in the crude oil market’, American Economic Review 99(3), 1053–69. - Koester, Gerrit, Eliza Lis, Christiane Nickel, Chiara Osbat and Frank Smets (2021), ‘Understanding low inflation in the euro area from 2013 to 2019: cyclical and structural drivers’, ECB Working Paper No 280. - Maćkowiak, Bartosz and Mirko Wiederholt (2009), ‘Optimal sticky prices under rational inattention’, American Economic Review 99(3), 769–803. - Maćkowiak, Bartosz and Mirko Wiederholt (2015), ‘Business cycle dynamics under rational inattention’, The Review of Economic Studies 82(4), 1502–1532. - Malmendier, Ulrike and Stefan Nagel (2016), ‘Learning from inflation experiences’, The Quarterly Journal of Economics 131(1), 53–86. - Mankiw, N. Gregory and Ricardo Reis (2002), ‘Sticky information versus sticky prices: A proposal to replace the new keynesian phillips curve’, The Quarterly Journal of Economics 117(4), 1295–1328. - Mankiw, N. Gregory, Ricardo Reis and Justin Wolfers (2003), ‘Disagreement about inflation expectations’, NBER Macroeconomics Annual 18, 209–248. - Sims, Christopher A. (2003), ‘Implications of rational inattention’, Journal of Monetary Economics 50(3), 665–690. - Woodford, Michael (2003), Imperfect common knowledge and the effects of monetary policy, in P. Aghion, R. Frydman, J. E. Stiglitz and M. Woodford, eds, ‘Knowledge, Information, and Expectations in Modern Macroeconomics: In Honor of Edmund Phelps’, Princeton Univ. Press, Princeton NJ. - Zekaite, Zivile (2020), ‘What drives consumers’ inflation perceptions in the euro area’, Central Bank of Ireland Economic Letter 2020(6). ### Appendix — Figures (descriptions) - Figure (28): Impulse responses of inflation forecast errors: country-level regressions. - Panels: Oil supply (left) and Global demand (right) for DEU, FRA, ITA, ESP. - Note: Plots coefficients on the i) oil supply shock (left panel) and ii) global demand shock (right panel) from regression (8) estimated at the country-level. Grey areas denote 90% confidence intervals. - Figure (29): Impulse responses of average inflation forecast errors: evidence from the U.S. - Panels compare Michigan Consumer Survey (MCS, left column) and Survey of Professional Forecasters (SPF, right column) responses to: - Hamilton (1996) oil shock (top row) - Kanzig (2021) oil shock (middle row) - Gali (1999) technology shock (bottom row) - Note: A positive impulse response denotes overreaction to deflationary shocks and underreaction to inflationary shocks. Grey areas denote 90% confidence intervals. - Figure (30): Impulse responses of average inflation forecast errors: evidence from the U.S. using CG(2012) data. - Panels compare MCS (left column) and SPF (right column) responses to externally identified shocks used in Coibion and Gorodnichenko (2012) (CG, 2012): - Barsky Sims (2011) news shock (top row) - Hamilton (1996) oil shock (middle row) - Gali (1999) technology shock (bottom row) - Note: A positive impulse response denotes overreaction to deflationary shocks and underreaction to inflationary shocks. Grey areas denote 90% confidence intervals. The sample is 1985Q1–2007Q3. ### Appendix — Summary statistics (tables and exact figures) - Table (8): EC Consumer expectations – summary statistics - Title: Inflation expectations of consumers – the entire ample - Columns: Number of countries | Min. obs. per country | Max. obs. per country | Earliest obs. | Latest obs. | Total obs. - Row: 118 | 5 | 469 | 2004m01 | 2019m04 | 3247947 - Note: Statistics are based on winsorised country samples (5%) for the question (Q61) in the Survey. Min/Max observations per country are based on the entire sample (January 2004 – May 2019). - Table (9): EC Consumer expectations – summary statistics - Section: Full sample - mean | median | stand.dev. | skewness | kurtosis | obs. - EA | 6.09 | 3.00 | 8.11 | 1.65 | 4.93 | 3247947 - Section: Inflation (HICP) < 0% - mean | median | stand.dev. | skewness | kurtosis | obs. - EA | 3.97 | 0.00 | 6.86 | 2.35 | 8.27 | 427492 - Section: Inflation (HICP) ≥ 0% - mean | median | stand.dev. | skewness | kurtosis | obs. - EA | 6.25 | 3.00 | 8.17 | 1.61 | 4.78 | 3020455 - Note: Statistics are based on winsorized country samples (5%) for the question (Q61) in the Survey over the entire sample (January 2004 – May 2019). Characteristics calculated across the entire sample are conditional upon Y-on-Y %-change in the headline inflation (HICP) in the particular month. - Table (10): Consensus Economics forecasts – summary statistics (sample) - Title: Inflation expectations of professionals – sample - Columns: Number of countries | Min. obs. per country | Max. obs. per country | Earliest obs. | Latest obs. | Total obs. - Row: 104 | 3 | 33 | 1996m1 | 2020m12 | 35228 - Note: Minimum and maximum number of observations per country refers to the number of forecasters in a single month. - Table (11): Consensus Economics forecasts – summary statistics - Section: Full sample - mean | median | stand.dev. | skewness | kurtosis | obs. - 1.53 | 1.6 | 0.97 | 0.05 | 3.58 | 35228 - Section: Inflation (HICP) < 0%: Four largest counries - mean | median | stand.dev. | skewness | kurtosis | obs. - 0.08 | 0.1 | 0.42 | -0.07 | 3.51 | 1224 - Section: Inflation (HICP) ≥ 0%: Four largest counries - mean | median | stand.dev. | skewness | kurtosis | obs. - 1.65 | 1.7 | 0.84 | 0.25 | 3.53 | 14128 - Note: Characteristics for the periods with the headline Y-o-Y inflation (HICP) above and below zero are calculated across the same sample as for consumer forecasts (January 2004 – June 2019). *Biases in Survey Inflation Expectations: Evidence from the Euro Area Working Paper No. WP/2022/205* --- _Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022205-print-pdf.pdf_