The Neo Fisher Effect and Exiting a Liquidity Trap

Transcription

1 The Neo Fisher Effect and Exiting a Liquidity Trap Stephanie Schmitt-Grohé and Martín Uribe Columbia University European Central Bank Conference on Monetary Policy Frankfurt am Main, October 29-3, 218 1

2 This talk is based on the following 4 papers: Uribe, The Neo-Fisher Effect: Econometric Evidence from Empirical and Optimizing Models, NBER WP 2589, September 218. Schmitt-Grohé and Uribe, Liquidity Traps and Jobless Recoveries, AEJ: Macroeconomics, 217. Schmitt-Grohé and Uribe, Liquidity Traps: An Interest-Rate-Based Exit Strategy, The Manchester School, 214. Benhabib, Schmitt-Grohé, and Uribe, The Perils of Taylor Rules, Journal of Economic Theory, 21. 2

3 Does setting nominal rates at zero for an extended period of time raise inflation or inflationary expectations? 3

4 Japan has had near zero rates ever since 1995 Japan, Call rate, 1989Q1 217Q1 6 5 Percent per year Vertical lines: Cabinet office recession dates, 1991Q1, 1993Q4, 1997Q2, 1999Q1, 2Q4, 22Q1, 28Q1, 29Q1, 212Q2, and 212Q4. 4

5 ... yet inflation has been below target throughout. Japan, Inflation, GDP deflator, yoy, 1989Q1 217Q2 3 2 percent per year Vertical lines: Cabinet office recession dates, 1991Q1, 1993Q4, 1997Q2, 1999Q1, 2Q4, 22Q1, 28Q1, 29Q1, 212Q2, and 212Q4. Horizontal line: 2% inflation target. 5

6 The U.S. had zero rates until 215Q4 but then got out. United States, Federal Funds Rate, 2Q1 218Q3 6 5 percent Vertical solid lines: NBER recession dates, 27Q4 and 29Q2. Vertical broken line: end of liquidity trap, 215Q4. 6

7 ... and the exit coincided with an inflection point for inflation. United States, Inflation, GDP deflator, yoy, 2Q1 218Q percent per year Vertical solid lines: NBER recession dates, 27Q4 and 29Q2. Vertical broken line: end of liquidity trap, 215Q4. Dotted horizontal line, 2% inflation target. 7

8 Downward revisions of long-run inflation expectations in the United States? Options-implied Inflation Probability Density Functions. Source: FRB Minneapolis, Prior to exit in December 215, long-run inflation expectations fell. 8

9 The Euro area has had near zero rates since Euro area, Interest Rate, Eonia, 2:1 218:6 5 4 percent per year Vertical lines: CEPR business cycles dates, 28Q1, 29Q2, 211Q3, 213Q1. 9

10 ... yet, inflation remains below target... Euro area, Inflation, HICP ex energy and unp. food, yoy, 2:1 218: percent per year Vertical lines: CEPR business cycles dates, 28Q1, 29Q2, 211Q3, 213Q1. 1

11 ... and long-run inflation expectations are low. HICP over next 1 years: Large increase and large decrease probabilities 4 <.75%pa >3.25%pa 3 Probability in percent Data source, Vogt, 218. Twenty-day moving averages of daily options-implied inflation probabilities, Oct 6, 29 to Aug 31, 218. Vertical lines: CEPR business cycle dates, 28Q1, 29Q2, 211Q3, 213Q1. 11

12 Standard way to rationalize the joint occurrence of near zero rates for an extended period of time and inflation well below target 12

13 Curdia (215) shows for a standard model to explain this pattern requires that the economy is continuously surprised by yet another negative natural rate shock: Source: Curdia, FRBSF EL

14 An Alternative View: Inflationary expectations fall. Mr. Draghi and his peers are afraid that consumers and investors will increasingly see low inflation as the new normal, creating a self-fulfilling prophecy. NYT, page B7, November 22,

15 A Brief Exposition of the Perils of Taylor Rules, BSU 21 The Taylor Rule: 1 + i t = max {1,1 + i + α π (π t π )} The Euler Equation: U (C t ) = β(1 + i t )E t U (C t+1 ) π t+1 In a steady state they become, respectively, 1+i 1 + i = max {1,1 + i + α π (π π )} and 1 + i = β 1 π 1+i 1 Solid Line: Taylor rule, 1+i = max {1,1 + i + α π (π π )} Broken Line: Euler equation 1 + i = β 1 π π L π π Two inflation steady states: The intended steady state (π ) and the Liquidity Trap (π L ) 15

16 Dynamics in a Flexible-Price Endowment Economy π t+1 π β π L π π t Solid line: π t+1 = max {β, π + βα π (π t π )} Broken line: 45-degree line Comment: Similar results obtain in sticky-price/wage economies (BSU 21, SGU 217) and also under time-consistent policy (Nakata & Schmidt, 217). 16

17 How to exit a persistent liquidity trap? To exit the liquidity trap the central bank should raise nominal interest rates. Doing so will increase inflation not only in the long run but also in the short run (the neo Fisher effect). Consider a model with nominal rigidities (on the next slide downward nominal wage rigidity). 17

18 Some more details of the SGU 217 model... Labor Demand by Firms Production function: Y t = X t F(h t ), where Y t = output X t = total factor productivity (TFP), assumed to be exogenous h t = hours X t /X t 1 = µ > 1, gross growth rate of TFP 18

19 W t P t X t F (h t ) Labor demand: W t = X t F (h t ), P t where W t = nominal wage rate P t = price level h t

20 Labor Demand: The Labor Market W t P t = X t F (h t ) Inelastic Labor Supply: h t h Unemployment: u t = h h t Downward Wage Rigidity: W t γ(u t )W t 1 W t P γ( h h t ) W t 1 t π t P t 1 W t P t X t F (h t ) B γ( h h t ) π L W t 1 P t 1 If π t = π, then the equilibrium is at point A. A γ( h h t ) π W t 1 P t 1 If π t = π L < π, then the equilibrium is at point B. h L h h t 19

21 Discussions of how monetary policy can lift an economy out of chronic below-target inflation are almost always based on the logic of how transitory interest-rate shocks affect real and nominal variables. Within this logic, a central bank trying to reflate a low-inflation economy will tend to set interest rates as low as possible. In the context of the SGU model, economies following this strategy can find themselves with zero nominal rates and with the lowinflation problem not going away. So, what to do? 2

22 The proposed exit strategy: Once the economy has been at the zero lower bound for some time, the central bank gradually raises the policy rate to the target level in steps of 25 basis points per quarter. Once rates are back to normal level, the central bank follows again a Taylor rule. In the context of the model, such a strategy raises long-run inflation expectations. The next slide illustrate how this exist strategy plays out in the model without capital and the slide thereafter in the model with capital. 21

23 Exiting a Chronic Liquidity Trap: Tightening is Easing Nominal Interest Rate Inflation i π 1 t π L 1 t Output Growth Rate Unemployment Rate u L µ 1 t 1 t 22

24 Exiting the Liquidity Trap: Tightening is easing also in the model with capital Nom. Int. Rate % pa Inflation, % pa Real Int Rate, % pa Emp Ratio, % Output.2 Investment.3 Real Wage.3 Consumption Taylor-Rule; Exit Strategy. Source: Schmitt-Grohé and Uribe (217). 23

25 Let s turn to data now, and ask whether the prediction of the model, namely, that a permanent increase in the nominal rate, raises inflation already in the short run, (the Neo Fisher effect), is consistent with empirical evidence. 24

26 The Fisher equation: where i = nominal interest rate r = real interest rate π = inflation rate i = r + π Effect of an increase in the nominal interest rate (i) on inflation (π) Effect on π in the long-run short-run Transitory increase in i Permanent increase in i Entry (2,1): The Fisher Effect Entry (2,2): The Neo-Fisher Effect 25

27 Cross-Country Evidence of the Fisher Effect Long-Run Average Inflation and Nominal Interest Rates 15 1 Average of it in percent Average of π t, in percent 25 OECD countries. Average sample period is 1989 to

28 Estimated Impulse Responses to a 1-percent Nominal Rate Increase United States, 1954Q4-218Q2 deviation from pre shock level percentage points per year Permanent Interest Rate Shock Response of the Interest Rate and Inflation.5 Interest Rate Inflation Inflation 95% band quarters after the shock deviation from pre shock level percentage points per year Temporary Interest Rate Shock Response of the Interest Rate and Inflation Interest Rate Inflation Inflation 95% band quarters after the shock percent deviation from pre shock level Permanent Interest Rate Shock Response of Output Output 95% band quarters after the shock percent deviation from pre shock level.5.5 Temporary Interest Rate Shock Response of Output 1 Output 95% band quarters after the shock Source: Uribe,

29 Response of the Real Interest Rate to Permanent and Transitory Interest-Rate Shocks 1.5 Permanent shock Transitory shock 1 Deviation from steady state percent per year quarters after the shock Source: Uribe, 218. Posterior mean estimates. The real interest rate is defined as i t E t π t+1. 28

30 Summary Models with nominal rigidities are prone to self-perpetuating liquidity traps. This holds for Taylor rules as well as for optimal policy under discretion. In such circumstances, models with nominal rigidities predict that a permanent increase in nominal interest rates can raise inflation already in the short run (Neo Fisher Effect) and thereby stimulate employment. This neo-fisherian prediction of the model is consistent with empirical evidence on the short-run effects of permanent interest rate shocks from Uribe (218). 29

31 Extras 3

32 The empirical model of Uribe, 218. all slides that follow are taken from Uribe

33 The Empirical Model y t π t i t log of real output inflation nominal interest rate ŷ t ˆπ t î t = B(L) X m t X n t z m t z n t = Λ ŷ t 1 ˆπ t 1 î t 1 ; X m t 1 X n t 1 z m t 1 z n t 1 + C ŷ t ˆπ t î t + Γ X m t X n t z m t z n t ɛ 1 t ɛ 2 t ɛ 3 t ɛ 4 t y t X n t π t X m t i t X m t where Xt m =permanent monetary shock; Xt n =permanent nonmonetary shock; zt m =transitory monetary shock; and zt n =transitory nonmonetary shock. Innovations ɛ i t N(,1) iid, for i = 1,2,3,4.. 32

34 Three Observables y t, growth rate of real output per capita. r t i t π t, the interest-rate-inflation differential. i t i t i t 1, time difference of the nominal interest rate. We then have the following observation equations: y t = ŷ t ŷ t 1 + X n t r t = î t ˆπ t (1) i t = î t î t 1 + X m t 33

35 Identification Assumptions Output, y t, is cointegrated with the permanent nonmonetary shock, X n t. Inflation, π t, is cointegrated with the permanent monetary shock, X m t. The nominal interest rate, i t, is cointegrated with the permanent monetary shock, X m t. Transitory interest-rate shocks, zt m effect on inflation. have a nonpositive impact Transitory interest-rate shocks, zt m effect on output. have a nonpositive impact 34

36 Variance Decomposition: Empirical Model y t π t i t Permanent Monetary Shock, X m t Transitory Monetary Shock, z m t Permanent Non-Monetary Shock, X n t Transitory Non-Monetary Shock, z n t Note. Posterior means. The variables y t, π t, and i t denote output growth, the change in inflation, and the change in the nominal interest rate, respectively. 35

37 Variance Decomposition: New Keynesian Model y t π t i t Permanent Monetary Shock, g m t Transitory Monetary Shock, z m t Permanent Productivity Shock, g t Transitory Productivity Shock, z t Preference Shock, ξ t Labor-Supply Shock, θ t

38 Impulse Responses to Interest-Rate Shocks: New Keynesian Model deviation from pre shock level percentage points per year Permanent Interest Rate Shock Response of the Interest Rate and Inflation.5 Interest Rate Inflation Inflation 95% band quarters after the shock deviation from pre shock level percentage points per year Temporary Interest Rate Shock Response of the Interest Rate and Inflation Interest Rate Inflation Inflation 95% band quarters after the shock percent deviation from pre shock level Permanent Interest Rate Shock Response of Output Output 95% band quarters after the shock percent deviation from pre shock level.1.1 Temporary Interest Rate Shock Response of Output.2 Output 95% band quarters after the shock 37

39 Response of the Real Interest Rate to Permanent and Transitory Interest-Rate Shocks in the New-Keynesian Model 1.8 Permanent shock Transitory shock.6 Deviation from steady state percent per year quarters after the shock Notes. Posterior mean estimates. The real interest rate is defined as i t E t π t+1. 38