2% Forever? Rethinking the Inflation Target

Transcription

1 2% Forever? Rethinking the Inflation Target Frederic S. Mishkin Graduate School of Business, Columbia University OENB-BIS Conference, Central Banking in Times of Change Vienna, September 13-14, 2016

2 Key Lessons from the Financial Crisis Should the Long-Run Inflation Target be Raised to Above 2%? Should Central Banks be Afraid of Going Above the Inflation Target? Should Central Banks Try to Overshoot the Inflation Target? Concluding Remarks

3 1. The zero-lower-bound (ZLB) constraint on policy interest rates binds more often than expected. View before crisis that at 2% target, ZLB would be infrequent thoroughly discredited. ZLB in U.S. occurred in and then Clearly can be long-lived Two reasons: 1.Economy is non-linear ( over the cliff ). Shocks from financial disruptions can be huge.

4 2. Once the zero-lower-bound constraint occurs, it is much harder to stimulate the economy and raise inflation. Nonconventional tools less effective than hoped. Key dilemma for central banks in advance countries is that despite heroic measures, inflation remains below 2% and economies remain weak.

5 Prominent economists suggest target should be raised to 4%. Pros: With higher target, ZLB less likely to occur. With higher target, conventional monetary policy of lowering policy rate can ease more because higher expected inflation allows lower real rate.

6 Cons: More difficult to stabilize inflation at 4% than 2%. 4% not consistent with Greenspan definition of price stability where it is not big factor in economic decisions. Once inflation rises above Greenspan definition at 4%, then why not 6%, or 8% and so on. Exactly what happened in U.S. starting in 60s: View that 4-5% inflation could be tolerated to achieve lower unemployment, with outcome of Great Inflation and high cost to getting inflation back down. One of great successes of central banks over last 20 years is anchoring of inflation expectations around 2%: Raising target to 4% would jeopardize hard-won success of establishing a strong nominal anchor.

7 Cons: Although raising target might have short-run benefits, it produces distortions in long run. These costs may be small for any given year, but they add up over time.

8 Bottom line: Costs of raising inflation target to 4% outweigh the benefits. Answer to question above? NO

9 Central bankers often are afraid to exceed the 2% target because they worry that inflation above 2% will unhinge inflation expectations. Act as if target is ceiling, and don t worry enough about inflation being too low Saw this in Japan pre Kuroda: led to overly tight monetary policy and disastrous policy mistake of raising policy rate in ECB has this tendency too. One manifestation is asymmetric target, close to but below 2% Mistake of monetary tightening in 2011 when Fed was easing. Continually behind the curve in recent years.

10 2% as ceiling creates an asymmetric target that is highly undesirable. Because there are negative as well as positive shocks to inflation, having a 2% ceiling means that on average inflation and hence inflation expectations will be below 2%. Two problems result: Inflation expectations are not anchored at 2% target. Lower expected inflation makes ZLB problem worse.

11 Bottom line: Central bankers subject to inflation phobia : more afraid of going above target rather than below. Has led to overly tight monetary policy and decline of inflation expectations to below 2%. Answer to question above? NO

12 In recent years, Fed, ECB and BOJ have undershot the 2% target. Inflation targeting typically is not historydependent: treats bygones as bygones, so past undershoots do not affect policy. Woodford (2003) provides compelling theoretical argument for history-dependent target in which if the target has been undershot in recent past, the monetary policy should strive to overshoot in near future.

13 Price-level target is one form of historydependent policy and it produces less output variance. Negative demand shock results in price level falling below target path, say 2% growth, requires raising price level back to path, so inflation rises above 2% temporarily. Expected inflation rises above 2%, lowering real interest rate, thereby stimulating economic activity. History-dependent price-level target is automatic stabilizer: negative shock leads to stabilizing expectations Even more effective when ZLB is binding (Eggertsson and Woodford, 2003).

14 Another similar history-dependent policy is nominal GDP target. Eggertsson and Woodford argue for target criterion of output-adjusted price level. Because output-adjusted price level target hard for public and markets to understand, Woodford (2012) argues for a simpler criterion of nominal GDP path which grows at the inflation target plus the growth rate of potential GDP.

15 Formidable challenges to price-level and nominal GDP targets: Hard to explain aiming for a target that is rising over time: targeting a level of inflation easier to explain. When inflation temporarily rises above 2%, may be hard to explain that commitment is to long-run 2% target. Nominal GDP target has additional problem because it requires that central bank takes a stance on number for potential GDP, which is highly uncertain and can lead to policy mistakes circa Fed in 1970s. Explains why have not been adopted.

16 Another way to skin the cat and adopt a history-dependent policy that can be explained to public and markets. 2% target should be for an average over a particular period, say 5-years or over the business cycle, rather than for a particular date such as 2-years ahead. If inflation had been running at 1.5% for a several years, then monetary policy would aim for 2.5% for several years. Would be particularly effective with ZLB because it would raise inflation expectations temporarily, lowering real interest rate. Additional benefit of encouraging more expansionary policy in face of negative demand shocks.

17 This history-dependent policy has been implemented by the Reserve Bank of Australia, with 2-3 target on average over the business cycle. Economic performance in Australia has been excellent: Average inflation since 1995 is 2.7%--very close to 2.5% midpoint of target range--and no recession in 25 years. Curdia (2016) provides evidence that such a policy, which would imply a temporary overshooting of 2% target to 2.5%, would produce better outcomes in U.S.

18 Bottom line: With inflation below 2% for number of years in U.S., Eurozone and Japan, Fed, ECB and BOJ should commit to achieve inflation above 2% for several years. However this should be done with continuing commitment to to long-run target of 2%, but one that is modified to be for an average over a period such as the business cycle.

19 Global financial crisis has taught us that ZLB is more frequent and when it occurs, it is hard to raise inflation to the 2% target. Analysis comes to following answers to three questions. 1. Should the long-run inflation target be raised to above 2%? NO 2. Should central banks be afraid of going above the inflation target? NO 3. Should central banks try to overshoot the inflation target? YES, when previously have undershot the target.

20 Bottom Line: 2% Forever makes sense for long-run inflation target. In current economic environment, not only should central banks not be afraid of exceeding the 2% target, but should aim to exceed it for the next couple of years.