Banque Nationale De Belgique Sa · EURONEXT:BNB

Banque Nationale De Belgique Sa — Technical Presentation (April 2014)

43 slides from Banque Nationale De Belgique Sa's technical presentation, Apr 28, 2014

  1. Slide 1: Estimating Monetary Policy Rules When Nominal Interest Rates Are Stuck at Zero

    Estimating Monetary Policy Rules When Nominal Interest Rates Are Stuck at Zero Jinill Kim and Seth Pruitt (Korea University and Federal Reserve Board) at NBB and DNB on April 28—29, 2014

  2. Slide 2: Motivation

    MOTIVATION • Policy Rules at the Zero Lower Bound — “between 0 and 25 bp” since December 2008 — how to estimate a policy rule when LHS is censored — especially when we suspect a policy change around the ZLB — use survey information as “artificial” or “synthetic” data -1-

  3. Slide 3: Motivation: Personal Motivation

    Motivation • Policy Rules at the Zero Lower Bound — “between 0 and 25 bp” since December 2008 — how to estimate a policy rule when LHS is censored — especially when we suspect a policy change around the ZLB — use survey information as “artificial” or “synthetic” data • Personal Motivation — Policy R

  4. Slide 4: The Policy Outlook in an Uncertain Environment (Bluebook)

    Class I FOMC - Restricted Controlled (FR) 20 of 43 Chart 7 The Policy Outlook in an Uncertain Environment FRB/US Model Simulations of Information from Financial Markets Estimated Outcome-Based Rule Percent Percent 11 11 11 11 Current Bluebook Expectations from forward contracts 10 Previous Bluebook

  5. Slide 5: Motivation: Bernanke's 2010 Atlanta AEA speech

    Motivation • Policy Rules at the Zero Lower Bound — “between 0 and 25 bp” since December 2008 — how to estimate a policy rule when LHS is censored — especially when we suspect a policy change around the ZLB — use survey information as “artificial” or “synthetic” data • Personal Motivation — Policy R

  6. Slide 6: Monetary Policy and the Housing Bubble

    Monetary Policy and the Housing Bubble by Jane Dokko Brian Doyle Michael Kiley Jinill Kim Shane Sherlund Jae Sim Skander van den Heuvel This discussion represents the views of the author(s) and should not be interpreted as reflecting those of the Board of Governors of the Federal Reserve System or a

  7. Slide 7: Slide 7

  8. Slide 8: Slide 8

  9. Slide 9: The Target Rate and the Taylor Rule Prescriptions

    The Target Rate and the Taylor Rule Prescriptions 9 Using Real‐Time Inflation Forecasts 8 7 6 5 4 3 2 1 0 2000Q1 2001Q1 2002Q1 2003Q1 2004Q1 2005Q1 2006Q1 2007Q1 2008Q1 2009Q1 Target Rate Taylor Rule (output gap and headline CPI inflation as currently measured) Taylor Rule (output gap and forecast o

  10. Slide 10: Motivation: Blue Chip survey

    Motivation • Policy Rules at the Zero Lower Bound — “between 0 and 25 bp” since December 2008 — how to estimate a policy rule when LHS is censored — especially when we suspect a policy change around the ZLB — use survey information as “artificial” or “synthetic” data • Personal Motivation — Policy R

  11. Slide 11: Blue Chip Economic Indicators

    Blue Chip ® Economic Indicators Top Analysts’ Forecasts of the U.S. Economic Outlook for the Year Ahead Vol. 38, No. 9, September 10, 2013 Wolters Kluwer Law & Business

  12. Slide 12: Actual Policy Rate Chart

    10 9 Dec 2008 Aug 2011 8 7 6 Percent 5 4 3 2 1 it: Actual Policy 0 1986 1990 1995 2000 2005 2010

  13. Slide 13: Actual and Forecast Rates

    10 9 Dec 2008 Aug 2011 8 7 6 Percent 5 4 3 2 1 Actual One−Year−Ahead Forecast Long−Horizon Forecast 0 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013

  14. Slide 14: Figure 1: Intraday Treasury Yields

    Figure 1: Intraday Treasury Yields Percent Percent 0.50 > 25 FOMC | 0.45 + announcement 1 24 0.40 } | 0.35 } 1 10-year yield ] + h i 0.30 (right axis) J] 22 0.25 2-year yield 24 an | (left axis) 0.15 2.0 8/8/11 8/9/11 8/10/11 8/11/11 Note Trading data fom 9 30am to 4.00p m EDT. at five-minute interv

  15. Slide 15: Actual and Forecast Rates

    6 5 4 Percent 3 2 1 Actual One−Year−Ahead Forecast Long−Horizon Forecast 0 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013

  16. Slide 16: Actual and Forecast Rates

    11 Actual One−Year−Ahead Forecast 10 Long−Horizon Forecast 9 8 Percent 7 6 5 4 3 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013

  17. Slide 17: Estimating Policy Rules

    Estimating Policy Rules  = (1 − ) [(∗ + ∗ ) + (  −  ∗ ) + ( − ∗ )] + −1 (1) • Rules Using Forecasts Data — non-censored one-year-ahead forecasts and long-horizon forecasts  = (∗ + ∗ ) + (  −  ∗ ) + ( − ∗ ) (2) • Rules Using Historical Data  =  +  

  18. Slide 18: Table 1-A: OLS, Historical-Forecast Data Combination

    • Table 1-A: OLS, Historical-Forecast Data Combination Inflation Activity 1986—2008 1358 −1402 standard error (0186) (0149) 1986—2013 0926 −1014 standard error (0238) (0104) until ZLB 1358 −1402 standard error (0183) (0129) during ZLB −0880 −1414 standard error (0637) (0201) -6-

  19. Slide 19: Table 1-B: Tobit, Historical-Forecast Data Combination

    • Table 1-B: Tobit, Historical-Forecast Data Combination Inflation Activity 1986—2008 1358 −1402 standard error (0188) (0150) 1986—2013 1422 −1584 standard error (0346) (0193) until ZLB 1367 −1422 standard error (1167) (0872) during ZLB e0878 −4489 standard error (10219) (2557) -7-

  20. Slide 20: Table 1-C: Tobit, Only Forecast Data

    • Table 1-C: Tobit, Only Forecast Data Inflation Activity 1986—2008 1550 −0786 standard error (0186) (0138) 1986—2013 1286 −0883 standard error (0276) (0087) until ZLB 1546 −0779 standard error (0255) (0131) during ZLB −1354 −2119 standard error (1184) (0244) -8—

  21. Slide 21: Estimation Results: road map

    Estimation Results: road map • Break for Subsamples — August 2007, March 2008, September 2008, December 2008. • Comparing Results between Historical and Forecasts Data — pre-Crisis: Similar — post-Crisis: Different • short-rate residuals and measures of risk • “unconstrained” nominal interest rates -

  22. Slide 22: Table 1: Estimates from Historic Data Including the ZLB Period

    • Table 1: Estimates from Historic Data Including the ZLB Period pre-Crisis post-Crisis August 2007 Break Inflation response 1893 −0275 standard error (0108) (0237) Chow test  0001 Unemployment response −1257 −0568 standard error (0097) (0136) Chow test  0001 Constant 6018 1529 standar

  23. Slide 23: Skeptical for two reasons

    Skeptical for two reasons • Statistical — censoring problem • Economic ∗ =  + ( − 1) ∗ + ∗ (5) — implied ∗ in Figure 4 -8-

  24. Slide 24: Table 2-1: Estimates from Historical Data Excluding the ZLB Period

    • Table 2-1: Estimates from Historical Data Excluding the ZLB Period pre-Crisis post-Crisis August 2007 Break Inflation response 1893 −0723 standard error (0108) (0957) Chow test 0047 Unemployment response −1257 −1352 standard error (0097) (0308) Chow test 0689 Constant 6018 1108 standar

  25. Slide 25: Table 2-2: Estimates from Historical Data Excluding the ZLB Period

    • Table 2-2: Estimates from Historical Data Excluding the ZLB Period pre-Crisis post-Crisis March 2008 Break Inflation response 1908 1763 standard error (0107) (0331) Chow test 0973 Unemployment response −1243 −0375 standard error (0097) (0112) Chow test 0585 Constant 5864 −0517 standard

  26. Slide 26: Results with Forecast Data

    Results with Forecast Data • The fit is quite good. — RHS variables explain more than 75% of expected short rates. • Some concerns — negative inflation coefficients for post-Crisis with early breakpoints — much variation in unemployment coefficients post-Crisis — sizable variation even for pre-Crisis co

  27. Slide 27: Table 3-1: Estimates from Forecast Data

    • Table 3-1: Estimates from Forecast Data pre-Crisis post-Crisis August 2007 Break Inflation response 1968 −0894 standard error (0260) (0889) Chow test 0002 Unemployment response −0608 −1137 standard error (0112) (0170) Chow test  0001 2 0810 -12-

  28. Slide 28: Table 3-2: Estimates from Forecast Data

    • Table 3-2: Estimates from Forecast Data pre-Crisis post-Crisis March 2008 Break Inflation response 1954 −0770 standard error (0259) (0918) Chow test 0004 Unemployment response −0650 −111 standard error (0117) (0175) Chow test 0002 2 0805 -13-

  29. Slide 29: Table 3-3: Estimates from Forecast Data

    • Table 3-3: Estimates from Forecast Data pre-Crisis post-Crisis September 2008 Break Inflation response 1774 0832 standard error (0282) (0364) Chow test 0069 Unemployment response −0800 −0796 standard error (0162) (0091) Chow test 0972 2 0781 -14-

  30. Slide 30: Table 3-4: Estimates from Forecast Data

    • Table 3-4: Estimates from Forecast Data pre-Crisis post-Crisis December 2008 Break Inflation response 1745 0599 standard error (0277) (0441) Chow test 0111 Unemployment response −0775 −0831 standard error (0139) (0108) Chow test 0605 2 0780 -15-

  31. Slide 31: Pinning Down the Crisis Period

    • Pinning Down the Crisis Period — Instability in the coefficient estimates ∗ Figure 6: Forecasts data ∗ Figure 7: Residuals -16-

  32. Slide 32: Figure 6: Forecast Data Fit

    5 E−* 4 uE−u* 3 iE−r*−*, Data iE−r*−*, Aug 2007 Break Fit 2 iE−r*−*, Dec 2008 Break Fit 1 Percent 0 −1 −2 −3 −4 −5 2006 2007 2008 2009 2010 2011

  33. Slide 33: Figure 7: Residuals

    1.5 Residual, Aug 2007 Break Fit Residual, Dec 2008 Break Fit 1 0.5 0 Percent −0.5 −1 −1.5 −2 −2.5 2006 2007 2008 2009 2010 2011

  34. Slide 34: Pinning Down the Crisis Period

    • Pinning Down the Crisis Period — Instability in the coefficient estimates ∗ Figure 6: Forecasts data ∗ Figure 7: Residuals — Sensitivity ∗ Statistical fixes: WLS and median regression ∗ Economic fix: Exclude the chaotic 2007:09 to 2008:12 period. -17-

  35. Slide 35: Table 4-1: Forecast Data Excluding 2007:09 to 2008:12, OLS

    • Table 4-1: Estimates from Forecast Data, Excluding 2007:09 to 2008:12 pre-Crisis post-Crisis Ordinary Least Squares Inflation response 1968 0599 standard error (0260) (0441) Chow test 0091 Unemployment response −0608 −0831 standard error (0112) (0108) Chow test 0071 2 0828 -18-

  36. Slide 36: Table 4-2: Forecast Data Excluding 2007:09 to 2008:12, WLS

    • Table 4-2: Estimates from Forecast Data, Excluding 2007:09 to 2008:12 pre-Crisis post-Crisis Weighted Least Squares Inflation response 1982 0669 standard error (0104) (0584) Chow test 0086 Unemployment response −0585 −0814 standard error (0065) (0110) Chow test 0077 -19-

  37. Slide 37: Table 4-3: Forecast Data Excluding 2007:09 to 2008:12, Median Regression

    • Table 4-3: Estimates from Forecast Data, Excluding 2007:09 to 2008:12 pre-Crisis post-Crisis Median Regression Inflation response 1959 1218 standard error (0105) (0593) Chow test 0068 Unemployment response −0576 −0682 standard error (0056) (0108) Chow test 0059 -20-

  38. Slide 38: Understanding the residuals

    • Understanding the residuals — What accounts for residuals’ fluctuations? — Figure 8, based on OLS of Table 4 — Atkeson and Kehoe (2009) on risk ∗ proxy risk by term spread ∗ economically and statistically significant ∗ 100 bp increase in term spread to 25 bp decrease in rates -21-

  39. Slide 39: Figure 8: Residual Fitted to Term Spread

    2.5 Residual Residual Fitted to Term Spread 2 1.5 1 Percent 0.5 0 −0.5 −1 −1.5 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011

  40. Slide 40: Unconstrained Short Rates

    • Unconstrained Short Rates — out-of-sample analysis — What would the rates be without the ZLB? — Figure 9 — Size of LSAPs -22-

  41. Slide 41: Figure 9: Unconstrained Policy Rate

    6 5 4 3 2 1 Percent 0 −1 * −2 A− uA−u* −3 iA, post−Crisis rule −4 Dec 2008 iA, pre−Crisis rule iA, Data −5 2008 2009 2010 2011 2012 2013

  42. Slide 42: Conclusion

    • Conclusion — Estimating policy rules at the zero lower bound — Inflation response weaker than before — Unemployment response as strong as before -23-

  43. Slide 43: Percent

    6 E−* 5 uE−u* 4 iE−r*−* 3 2 1 Percent 0 −1 −2 −3 −4 −5 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010

Banque Nationale De Belgique — Technical Presentation (April 2014) on 6ix