Discussion Global Dynamics at the Zero Lower Bound

Transcription

1 Discussion Global Dynamics at the Zero Lower Bound Brent Bundick Federal Reserve Bank of Kansas City April 4 The opinions expressed in this discussion are those of the author and do not reflect the views of the Federal Reserve Bank of Kansas City or the Federal Reserve System.

2 Global Dynamics at the Zero Lower Bound Very interesting and well-executed paper Contribution: Rigorous modeling of a zero lower bound economy Effects of recent downturn lasting longer than expected Rigorous modeling is key first step to explaining recent outcomes

3 Why can technological improvements be contractionary? Aggregate demand curve has positive slope at zero lower bound Use linearized AS-AD model to build intuition π t = βe t π t+ + ψ y y t ( η) ψz t ψa t ( ) y t = E t y t+ (r t r) E t π t+ + ( ρ a ) a t ) r t = max (, r + φ π π t + φ y y t

4 Aggregate Demand Away From Zero Lower Bound ( + φ y + ψ ) ( y y t = E t y t+ φ π ) ( ) π t + Shocks β β! AD AS Declines in π t offset with lower real rates which raises Y t Y

5 Technology Shock Away From Zero Lower Bound ( + φ y + ψ ) ( y y t = E t y t+ φ π ) ( ) π t + Shocks β β! AD AS Higher technology lowers π t & raises Y t with policy accommodation Y

6 Aggregate Demand At Zero Lower Bound ( + ψ ) y y t = E t y t+ + ( ) β β π t + Shocks! AD AS Declines in π t cannot be initially offset which raises real rates Y

7 Technology Shock At Zero Lower Bound ( + ψ ) y y t = E t y t+ + ( ) β β π t + Shocks! AD AS Higher technology can cause larger disinflation & output losses Y

8 Discussion: Focus on Expectations About Future ( + ψ ) y y t = E t y t+ + ( ) β β π t + Shocks! AD AD(Y t+ ) AS Expectations about future crucial in determining effects of shocks Y

9 in high technology states. Higher labor demand raises equilibrium hours, which mitigates the decline in the real wage. In short, a tension exists at the ZLB between the supply-side effects of technology and the demand-side effects of the real interest rate. When the central bank responds less aggressively to the steady-state output gap when the ZLB does not bind (i.e., a lower φ y ), the demand-side effects at the ZLB are weaker and both real and nominal variables are less volatile. Assumptions About Zero Lower Bound Duration.75.5 Adjusted Output (ŷ adj ) Inflation Rate (ˆπ) 5 5 Labor Hours (ˆn).6.4. Real Interest Rate ( r/e[π]) 5 5 Nominal Interest Rate (ˆr) Assume preference shock is constant in impulse responses.75.5 Steady State Scenario Real Wage Rate (ŵ).8.4 ZLB Scenario Agents expect zero lower bound exit within - periods Economy surprised by zero lower bound duration every period Comment: What is the right zero lower bound scenario?

10 Unexpected Long Zero Lower Bound Episode.6 Output Inflation 3 Nominal Interest Rate Real Interest Rate Preference Shock Technology Gavin et al (4) Unexpected ZLB Length

11 Expected Long Zero Lower Bound Episode Output Inflation 3 Nominal Interest Rate Real Interest Rate..4 Preference Shock Technology Gavin et al (4) Unexpected ZLB Length Large AR() Shock Expected ZLB Length

12 Central Bank Responds to Output Gap.7 Output Inflation 4 Nominal Interest Rate Real Interest Rate.. Preference Shock Technology Output Gap

13 Central Bank Responds to Steady State Output Output Inflation 4 Nominal Interest Rate Real Interest Rate Preference Shock Technology Output Gap Steady State Output

14 Output Gap Response - Longer Zero Lower Bound Episode Output Inflation 4 Nominal Interest Rate Real Interest Rate..4 Preference Shock Technology Output Gap Steady State Output Output Gap Longer ZLB Episode

15 Expectations About Current Zero Lower Bound Duration Figure 8: The Expected Path of Nominal Interest Rates Gust, Lopez-Saĺıdo, and Smith (3) Model-implied (Red) & Futures Markets (Gray) Financial Market Federal Funds Rate Expectations % confidence bands median Expectations as of 9:Q 3 4 Expectations as of :Q Expected duration Blue Chip increased 3 month markedly Treasury with Bill late-4 Expectations language 6 6

16 Comments Use an alternative zero lower bound scenario. Use expectations from data at a given point. Use average scenario from model simulations 3. Show robustness under variety of scenarios Decomposing differences between linear & nonlinear model. Role of quadratic adjustment costs. Precautionary labor supply Away from zero lower bound Additional downside risk implied by the zero lower bound Solve linear with nonlinear costs or nonlinear with quadratic utility Very nice and well-executed paper

17 Additional Details

18 Precautionary Labor Supply W t L S (λ t ) L D (K t, Z t ) N t

19 Precautionary Labor Supply Without Capital W t L S (! t ) L D ( ) L D (µ t ) N t