## _wp15230app ## Source details **Canonical URL:** [_wp15230app](https://www.imf.org/-/media/websites/imf/imported-publications/external/pubs/ft/wp/2015/_wp15230app.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-publications/external/pubs/ft/wp/2015/_wp15230app.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-publications/external/pubs/ft/wp/2015/_wp15230app.pdf.json) --- ### Classification of Analyzed Recessions (Background Table) - Total Cases: 23187728 - Recession episodes listed by country and quarter ranges with four indicator columns: - "With Financial Crisis 1/" — 1 if recession coincided with financial crisis as defined in Laeven and Valencia (2013) - "With Increasing Oil Price 2/" — 1 if recession coincided with 1974 or 1979 oil price increases - "With Increasing Inflation 3/" — 1 if the average inflation during the year before the start of the recession was below the average inflation during the recession - "With Intentional Disinflation 4/" — 1 if recession was followed by important decreases in inflation as well as coincided with important increases of monetary policy rates - Example entries (preserving exact formatting from source): - Australia-1971Q3-1972Q10010 - Australia-1975Q2-1975Q40100 - Australia-1977Q2-1977Q40000 - Australia-1981Q3-1983Q20011 - ... - United States-1973Q4-1975Q10110 - United States-1980Q1-1980Q30001 - United States-1981Q3-1982Q10001 - United States-1990Q3-1991Q10000 - United States-2007Q4-2009Q21010 ### Time-series Diagnostics by Country (Figure 8 / Figure 8A) - Panels plotted for each country include: - Unemployment Rate (black line) and Natural Rate (red line, dotted blue line +/- 1 standard deviation) - Anchoring of Inflation to Long Term Expectations (λ) (dotted blue line +/- 1 standard deviation) - Slope of Phillips Curve (Θ) (dotted blue line +/- 1 standard deviation) - Importance of Import-Price Inflation (μ) (dotted blue line +/- 1 standard deviation) - Countries shown across Figures 8 and 8A: United States, Germany, Japan, United Kingdom, France, Italy, Canada, Australia, Spain, Netherlands, Switzerland, Sweden, Belgium, Norway, Austria, Denmark, Greece, Ireland, Portugal, New Zealand. - Time axis tick examples preserved as in source figures: - 61 65 69 73 77 81 85 89 93 97 01 05 09 13 - 68 72 76 80 84 88 92 96 00 04 08 12 - 70 74 78 82 86 90 94 98 02 06 10 14 - Y-axis tick examples preserved from source panels include: 0, 0.1, 0.2, ... up to 1 or higher; 0, 2, 4, ... up to 14; 0, 0.005, 0.01, ... ### Table 6.A1 — Slope of the Phillips Curve (using u*) - Table note: regression results correspond to equation 2 in the text, where: - PIE is headline CPI inflation (defined as quarterly inflation, annualized) - u_gap is the difference between unemployment rate and the natural unemployment rate (Kalman Filter estimated) - PIE_LTE is long term inflation expectations (the average of the last four quarterly inflation rates) - PIE_M is import price inflation relative to headline inflation - Robust standard errors are in parentheses, and ***, **, and * denote significance level at 1, 5, and 10 percent, respectively. - Selected regression coefficient blocks (1990-2014 / 2007-2014 where applicable): - United States: - PIE_LTE: 0.987*** (0.0371) ; 0.968*** (0.128) - u_gap: -0.285*** (0.0718) ; -0.264* (0.132) - PIE_M: 0.182*** (0.0251) ; 0.196*** (0.0336) - Observations: 10032 / 10032 - R-squared: 0.890 / 0.884 - Germany: - PIE_LTE: 0.649*** (0.139) ; 1.089** (0.441) - u_gap: -1.086*** (0.173) ; -2.373*** (0.638) - PIE_M: 0.0164 (0.0102) ; 0.0315*** (0.0111) - Observations: 10032 / 10032 - R-squared: 0.338 / 0.433 - Japan: - PIE_LTE: 1.031*** (0.0920) ; 1.024*** (0.131) - u_gap: -0.111 (0.0892) ; 0.145 (0.202) - PIE_M: 0.0105 (0.0114) ; 0.0394*** (0.0113) - Observations: 10032 / 10032 - R-squared: 0.672 / 0.765 - United Kingdom: - PIE_LTE: 0.943*** (0.0745) ; 1.134*** (0.131) - u_gap: 0.0438 (0.111) ; -0.0389 (0.298) - PIE_M: 0.00893 (0.0180) ; 0.0307** (0.0120) - Observations: 10032 / 10032 - R-squared: 0.717 / 0.790 - France: - PIE_LTE: 0.981*** (0.0688) ; 0.878*** (0.112) - u_gap: -0.486*** (0.139) ; -0.631*** (0.203) - PIE_M: 0.0109 (0.00906) ; 0.0347** (0.0134) - Observations: 10032 / 10032 - R-squared: 0.780 / 0.747 - Italy: - PIE_LTE: 1.189*** (0.0421) ; 0.906*** (0.135) - u_gap: -0.124 (0.0870) ; -0.296*** (0.0966) - PIE_M: 0.0121 (0.00884) ; 0.0320** (0.0129) - Observations: 10032 / 10032 - R-squared: 0.891 / 0.724 - Canada: - PIE_LTE: 0.970*** (0.0963) ; 0.773*** (0.141) - u_gap: -0.321** (0.161) ; -0.516 (0.450) - PIE_M: 0.0429 (0.0486) ; 0.101* (0.0499) - Observations: 10032 / 10032 - R-squared: 0.596 / 0.622 - Australia: - PIE_LTE: 0.937*** (0.0748) ; 0.754*** (0.116) - u_gap: -0.683*** (0.135) ; -0.780** (0.380) - PIE_M: 0.0324* (0.0190) ; 0.0419*** (0.0114) - Observations: 10032 / 10032 - R-squared: 0.670 / 0.805 - Spain: - PIE_LTE: 0.989*** (0.0424) ; 0.892*** (0.164) - u_gap: -0.0809** (0.0360) ; -0.0917* (0.0529) - PIE_M: 0.0252** (0.0115) ; 0.0591*** (0.0180) - Observations: 10032 / 10032 - R-squared: 0.840 / 0.671 - Netherlands: - PIE_LTE: 0.994*** (0.0573) ; 1.064*** (0.139) - u_gap: -0.402*** (0.119) ; 0.00873 (0.202) - PIE_M: 0.00753 (0.00831) ; 0.0154 (0.0132) - Observations: 10032 / 10032 - R-squared: 0.761 / 0.667 - Overall empirical highlights from Table 6.A1: - PIE_LTE coefficients are often positive and statistically significant (many entries show ***, **, or *) across countries and periods. - u_gap coefficients are often negative and statistically significant in many country-period regressions, consistent with a downward-sloping Phillips curve when using the unemployment gap (u*). - PIE_M (importance of import-price inflation) coefficients vary by country and period; several are positive and significant at conventional levels. ### Table 6.A2 — Slope of the Phillips Curve (using u_bar) - Table note: regression results correspond to equation 2 in the text, where PIE is headline CPI inflation (defined as quarterly inflation, annualized), U_gap_bar is the difference between the unemployment rate and the natural unemployment rate (estimated as the average unemployment during 1990-2014), PIE_LTE is long term inflation expectations (the average of the last four quarterly inflation rates), and PIE_M is import price inflation relative to headline inflation. Robust standard errors are in parentheses, and ***, **, and * denote significance level at 1, 5, and 10 percent, respectively. - Selected panel and column-block summaries (coefficients and statistics preserved exactly): - Panel 1 (first column block shown) - PIE_LTE 1.000*** (0.0378) - u_gap_bar -0.248*** (0.0652) - PIE_M 0.181*** (0.0253) - Observations 10032 - R-squared 0.888 - Panel 2 (second column block shown) - PIE_LTE 0.972*** (0.131) - u_gap_bar -0.242* (0.122) - PIE_M 0.196*** (0.0337) - Observations 10032 - R-squared 0.884 - Panel 3 (third column block shown) - PIE_LTE 0.834*** (0.153) - u_gap_bar -0.502*** (0.0883) - PIE_M 0.0153 (0.0103) - Observations 10032 - R-squared 0.312 - Panel 4 (fourth column block shown) - PIE_LTE 2.619*** (0.809) - u_gap_bar (blank) - - PIE_M 0.0318*** (0.0102) - Observations 10032 - R-squared 0.473 - Additional selected column blocks: - PIE_LTE 1.148*** (0.143); u_gap_bar -1.541*** (0.425); PIE_M 0.0108 (0.0111); Observations 10032; R-squared 0.678 - PIE_LTE 0.879*** (0.139); u_gap_bar -0.119 (0.0723); PIE_M 0.0393*** (0.0115); Observations 10032; R-squared 0.766 - PIE_LTE 0.939*** (0.0750); u_gap_bar 0.112 (0.138); PIE_M 0.00881 (0.0179); Observations 10032; R-squared 0.717 - PIE_LTE 1.140*** (0.159); u_gap_bar 0.0177 (0.0975); PIE_M 0.0308** (0.0120); Observations 10032; R-squared 0.790 - PIE_LTE 1.240*** (0.122); u_gap_bar -0.0401 (0.279); PIE_M 0.0115 (0.00858); Observations 10032; R-squared 0.767 - PIE_LTE 1.568*** (0.282); u_gap_bar -0.316*** (0.0811); PIE_M 0.0342** (0.0136); Observations 10032; R-squared 0.752 - Further selected blocks: - PIE_LTE 1.288*** (0.0561); u_gap_bar -0.154*** (0.0582); PIE_M 0.0121 (0.00822); Observations 10032; R-squared 0.896 - PIE_LTE 1.192*** (0.177); u_gap_bar (blank) -; PIE_M 0.0317** (0.0131); Observations 10032; R-squared 0.727 - PIE_LTE 0.995*** (0.101); u_gap_bar -0.248*** (0.0780); PIE_M 0.0426 (0.0488); Observations 10032; R-squared 0.592 - PIE_LTE 0.737*** (0.142); u_gap_bar -0.205 (0.125); PIE_M 0.102** (0.0498); Observations 10032; R-squared 0.623 - Selected later examples: - PIE_LTE 1.508*** (0.101); u_gap_bar -0.643*** (0.0764); PIE_M 0.00954 (0.00810); Observations 10032; R-squared 0.716 - PIE_LTE 1.585 (1.308); u_gap_bar -0.645 (0.595); PIE_M 0.0337** (0.0131); Observations 10032; R-squared 0.244 - PIE_LTE 1.260*** (0.157); u_gap_bar -0.479*** (0.0934); PIE_M 0.0110 (0.0139); Observations 10032; R-squared 0.681 - PIE_LTE 2.071*** (0.365); u_gap_bar (blank) -; PIE_M 0.0437*** (0.0108); Observations 10032; R-squared 0.628 - Selected country/period specific examples from lower table section: - PIE_LTE 1.006*** (0.0640); u_gap_bar -0.133 (0.103); PIE_M 0.00933 (0.00743); Observations 8315; R-squared 0.764 - PIE_LTE -0.893** (0.407); u_gap_bar 1.529*** (0.442); PIE_M 0.0197 (0.0181); Observations 10032; R-squared 0.728 - PIE_LTE 1.019*** (0.0692); u_gap_bar -0.210*** (0.0339); PIE_M -0.00262 (0.0111); Observations 3291; R-squared 0.687 - PIE_LTE 0.603*** (0.181); u_gap_bar (blank) -; PIE_M 0.0129 (0.0188); Observations 3210; R-squared 0.294 - Empirical highlights from Table 6.A2: - PIE_LTE coefficients are frequently positive and statistically significant across many column blocks. - u_gap_bar coefficients are often negative and statistically significant in numerous column blocks, though some cells are blank (no coefficient reported). - PIE_M coefficients vary in magnitude and significance across column blocks and samples. *Source: _wp15230app - Section 1; _wp15230app - Section 2 (table and note as provided).* ### Section 1 ### _wp15230app - Section 1 ### Classification of Analyzed Recessions (Background Table) - Total Cases: 23187728 - Recession episodes listed by country and quarter ranges with four indicator columns: - "With Financial Crisis 1/" — 1 if recession coincided with financial crisis as defined in Laeven and Valencia (2013) - "With Increasing Oil Price 2/" — 1 if recession coincided with 1974 or 1979 oil price increases - "With Increasing Inflation 3/" — 1 if the average inflation during the year before the start of the recession was below the average inflation during the recession - "With Intentional Disinflation 4/" — 1 if recession was followed by important decreases in inflation as well as coincided with important increases of monetary policy rates - Example entries (preserving exact formatting from source): - Australia-1971Q3-1972Q10010 - Australia-1975Q2-1975Q40100 - Australia-1977Q2-1977Q40000 - Australia-1981Q3-1983Q20011 - ... - United States-1973Q4-1975Q10110 - United States-1980Q1-1980Q30001 - United States-1981Q3-1982Q10001 - United States-1990Q3-1991Q10000 - United States-2007Q4-2009Q21010 ### Figure 8 / Figure 8A — Time-series Diagnostics by Country - Figures plot (for each country): - Unemployment Rate (black line) and Natural Rate (red line, dotted blue line +/- 1 standard deviation) - Anchoring of Inflation to Long Term Expectations (λ) (dotted blue line +/- 1 standard deviation) - Slope of Phillips Curve (Θ) (dotted blue line +/- 1 standard deviation) - Importance of Import-Price Inflation (μ) (dotted blue line +/- 1 standard deviation) - Countries shown across Figures 8 and 8A include (as labeled in figures): United States, Germany, Japan, United Kingdom, France, Italy, Canada, Australia, Spain, Netherlands, Switzerland, Sweden, Belgium, Norway, Austria, Denmark, Greece, Ireland, Portugal, New Zealand. - Time axes and tick labels are preserved as in source figures (examples): - 61 65 69 73 77 81 85 89 93 97 01 05 09 13 - 68 72 76 80 84 88 92 96 00 04 08 12 - 70 74 78 82 86 90 94 98 02 06 10 14 - Y-axis scales vary by panel and country; example tick values preserved from source panels include: 0, 0.1, 0.2, ... up to 1 or higher; 0, 2, 4, ... up to 14; 0, 0.005, 0.01, ... etc. ### Table 6.A1 — Slope of the Phillips Curve (using u*) - Table note: regression results correspond to equation 2 in the text, where: - PIE is headline CPI inflation (defined as quarterly inflation, annualized) - u_gap is the difference between unemployment rate and the natural unemployment rate (Kalman Filter estimated) - PIE_LTE is long term inflation expectations (the average of the last four quarterly inflation rates) - PIE_M is import price inflation relative to headline inflation - Robust standard errors are in parentheses, and ***, **, and * denote significance level at 1, 5, and 10 percent, respectively. - Regression coefficient highlights (preserved exactly, grouped by country columns as presented): - United States (1990-2014 / 2007-2014): - PIE_LTE: 0.987*** (0.0371) ; 0.968*** (0.128) - u_gap: -0.285*** (0.0718) ; -0.264* (0.132) - PIE_M: 0.182*** (0.0251) ; 0.196*** (0.0336) - Observations: 10032 / 10032 - R-squared: 0.890 / 0.884 - Germany: - PIE_LTE: 0.649*** (0.139) ; 1.089** (0.441) - u_gap: -1.086*** (0.173) ; -2.373*** (0.638) - PIE_M: 0.0164 (0.0102) ; 0.0315*** (0.0111) - Observations: 10032 / 10032 - R-squared: 0.338 / 0.433 - Japan: - PIE_LTE: 1.031*** (0.0920) ; 1.024*** (0.131) - u_gap: -0.111 (0.0892) ; 0.145 (0.202) - PIE_M: 0.0105 (0.0114) ; 0.0394*** (0.0113) - Observations: 10032 / 10032 - R-squared: 0.672 / 0.765 - United Kingdom: - PIE_LTE: 0.943*** (0.0745) ; 1.134*** (0.131) - u_gap: 0.0438 (0.111) ; -0.0389 (0.298) - PIE_M: 0.00893 (0.0180) ; 0.0307** (0.0120) - Observations: 10032 / 10032 - R-squared: 0.717 / 0.790 - France: - PIE_LTE: 0.981*** (0.0688) ; 0.878*** (0.112) - u_gap: -0.486*** (0.139) ; -0.631*** (0.203) - PIE_M: 0.0109 (0.00906) ; 0.0347** (0.0134) - Observations: 10032 / 10032 - R-squared: 0.780 / 0.747 - Additional country blocks (columns and rows preserved in table) include Italy, Canada, Australia, Spain, Netherlands, with coefficients and standard errors reported for 1990-2014 and 2007-2014 samples. Example extracts: - Italy (1990-2014 / 2007-2014): - PIE_LTE: 1.189*** (0.0421) ; 0.906*** (0.135) - u_gap: -0.124 (0.0870) ; -0.296*** (0.0966) - PIE_M: 0.0121 (0.00884) ; 0.0320** (0.0129) - Observations: 10032 / 10032 - R-squared: 0.891 / 0.724 - Canada: - PIE_LTE: 0.970*** (0.0963) ; 0.773*** (0.141) - u_gap: -0.321** (0.161) ; -0.516 (0.450) - PIE_M: 0.0429 (0.0486) ; 0.101* (0.0499) - Observations: 10032 / 10032 - R-squared: 0.596 / 0.622 - Australia: - PIE_LTE: 0.937*** (0.0748) ; 0.754*** (0.116) - u_gap: -0.683*** (0.135) ; -0.780** (0.380) - PIE_M: 0.0324* (0.0190) ; 0.0419*** (0.0114) - Observations: 10032 / 10032 - R-squared: 0.670 / 0.805 - Spain: - PIE_LTE: 0.989*** (0.0424) ; 0.892*** (0.164) - u_gap: -0.0809** (0.0360) ; -0.0917* (0.0529) - PIE_M: 0.0252** (0.0115) ; 0.0591*** (0.0180) - Observations: 10032 / 10032 - R-squared: 0.840 / 0.671 - Netherlands: - PIE_LTE: 0.994*** (0.0573) ; 1.064*** (0.139) - u_gap: -0.402*** (0.119) ; 0.00873 (0.202) - PIE_M: 0.00753 (0.00831) ; 0.0154 (0.0132) - Observations: 10032 / 10032 - R-squared: 0.761 / 0.667 - Further country blocks in the table report comparable sets of coefficients, observations, and R-squared values for periods 1990-2014 and 2007-2014 (including United States, Japan, Germany, United Kingdom, France, Italy, Canada, Australia, Spain, Netherlands; and additional panels with Italy, Canada, Australia, Spain, Netherlands repeated in other layouts). Exact coefficients, standard errors, observations, and R-squared values are reported in the table as shown above and in the source. - Overall empirical findings preserved from table content: - PIE_LTE coefficients are often positive and statistically significant (many entries show ***, **, or *) across countries and periods. - u_gap coefficients are often negative and statistically significant in many country-period regressions, consistent with a downward-sloping Phillips curve when using the unemployment gap (u*). - PIE_M (importance of import-price inflation) coefficients vary by country and period; several are positive and significant at conventional levels. *Source: _wp15230app - Section 1* ### Section 2 ### _wp15230app - Section 2 ### Table 6.A2 – Slope of the Phillips Curve (using u_bar): selected regression coefficients and statistics - Panel 1 (first column block shown) - PIE_LTE 1.000*** (0.0378) - u_gap_bar -0.248*** (0.0652) - PIE_M 0.181*** (0.0253) - Observations 10032 - R-squared 0.888 - Panel 2 (second column block shown) - PIE_LTE 0.972*** (0.131) - u_gap_bar -0.242* (0.122) - PIE_M 0.196*** (0.0337) - Observations 10032 - R-squared 0.884 - Panel 3 (third column block shown) - PIE_LTE 0.834*** (0.153) - u_gap_bar -0.502*** (0.0883) - PIE_M 0.0153 (0.0103) - Observations 10032 - R-squared 0.312 - Panel 4 (fourth column block shown) - PIE_LTE 2.619*** (0.809) - u_gap_bar (blank) - (no coefficient reported in that cell) - PIE_M 0.0318*** (0.0102) - Observations 10032 - R-squared 0.473 - Additional example blocks (selected further columns) - PIE_LTE 1.148*** (0.143); u_gap_bar -1.541*** (0.425); PIE_M 0.0108 (0.0111); Observations 10032; R-squared 0.678 - PIE_LTE 0.879*** (0.139); u_gap_bar -0.119 (0.0723); PIE_M 0.0393*** (0.0115); Observations 10032; R-squared 0.766 - PIE_LTE 0.939*** (0.0750); u_gap_bar 0.112 (0.138); PIE_M 0.00881 (0.0179); Observations 10032; R-squared 0.717 - PIE_LTE 1.140*** (0.159); u_gap_bar 0.0177 (0.0975); PIE_M 0.0308** (0.0120); Observations 10032; R-squared 0.790 - PIE_LTE 1.240*** (0.122); u_gap_bar -0.0401 (0.279); PIE_M 0.0115 (0.00858); Observations 10032; R-squared 0.767 - PIE_LTE 1.568*** (0.282); u_gap_bar -0.316*** (0.0811); PIE_M 0.0342** (0.0136); Observations 10032; R-squared 0.752 - Further example column blocks (additional selected coefficients later in the table) - PIE_LTE 1.288*** (0.0561); u_gap_bar -0.154*** (0.0582); PIE_M 0.0121 (0.00822); Observations 10032; R-squared 0.896 - PIE_LTE 1.192*** (0.177); u_gap_bar (blank) - (no coefficient reported in that cell); PIE_M 0.0317** (0.0131); Observations 10032; R-squared 0.727 - PIE_LTE 0.995*** (0.101); u_gap_bar -0.248*** (0.0780); PIE_M 0.0426 (0.0488); Observations 10032; R-squared 0.592 - PIE_LTE 0.737*** (0.142); u_gap_bar -0.205 (0.125); PIE_M 0.102** (0.0498); Observations 10032; R-squared 0.623 - Additional selected blocks (from later table sections) - PIE_LTE 1.508*** (0.101); u_gap_bar -0.643*** (0.0764); PIE_M 0.00954 (0.00810); Observations 10032; R-squared 0.716 - PIE_LTE 1.585 (1.308); u_gap_bar -0.645 (0.595); PIE_M 0.0337** (0.0131); Observations 10032; R-squared 0.244 - PIE_LTE 1.260*** (0.157); u_gap_bar -0.479*** (0.0934); PIE_M 0.0110 (0.0139); Observations 10032; R-squared 0.681 - PIE_LTE 2.071*** (0.365); u_gap_bar (blank) - (no coefficient reported in that cell); PIE_M 0.0437*** (0.0108); Observations 10032; R-squared 0.628 - Selected country/period specific examples from lower table section - PIE_LTE 1.006*** (0.0640); u_gap_bar -0.133 (0.103); PIE_M 0.00933 (0.00743); Observations 8315; R-squared 0.764 - PIE_LTE -0.893** (0.407); u_gap_bar 1.529*** (0.442); PIE_M 0.0197 (0.0181); Observations 10032; R-squared 0.728 - PIE_LTE 1.019*** (0.0692); u_gap_bar -0.210*** (0.0339); PIE_M -0.00262 (0.0111); Observations 3291; R-squared 0.687 - PIE_LTE 0.603*** (0.181); u_gap_bar (blank) - (no coefficient reported in that cell); PIE_M 0.0129 (0.0188); Observations 3210; R-squared 0.294 ### Note (as reported in the source) - Regression results corresponding to equation 2 in the text, where PIE is headline CPI inflation (defined as quarterly inflation, annualized), U_gap_bar is the difference between the unemployment rate and the natural unemployment rate (estimated as the average unemployment during 1990-2014), PIE_LTE is long term inflation expectations (the average of the last four quarterly inflation rates), and PIE_M is import price inflation relative to headline inflation. Robust standard errors are in parentheses, and ***, **, and * denote significance level at 1, 5, and 10 percent, respectively. *Source: _wp15230app - Section 2 (table and note as provided).* --- _Source: https://www.imf.org/-/media/websites/imf/imported-publications/external/pubs/ft/wp/2015/_wp15230app.pdf_