68 - Scott Sumner on Fed Performance since the Great Recession

Episode Summary

Executive Summary: Scott Sumner argues the Fed turned a manageable 2008 downturn into the Great Recession by failing to act aggressively enough, especially at the zero lower bound. He favors nominal GDP or level targeting over inflation targeting, criticizes interest on reserves and reliance on interest rates as policy signals, and says long-run low rates reflect deeper global/supply-side forces more than Fed manipulation.

Main Topics: Fed policy errors in 2008-09 (Priority: 5/5): Sumner says the Fed focused too much on inflation fears and not enough on falling nominal spending, making policy too timid after Lehman and during the onset of the crisis. QE as a partial but insufficient tool (Priority: 4/5): Quantitative easing helped somewhat, but without the right monetary strategy and target, it could not fully stabilize spending or prevent the recession from deepening. Inflation targeting versus nominal GDP targeting (Priority: 5/5): He argues inflation targets are inferior because inflation is an ambiguous signal; nominal GDP better captures total demand and would have been clearer in the Great Recession. Low rates, r-star, and secular stagnation (Priority: 4/5): The discussion distinguishes short-run shocks from longer-run structural forces such as demographics, saving, and productivity slowdowns that keep natural rates low. Balance sheet policy and interest on reserves (Priority: 4/5): Sumner favors reducing the Fed's balance sheet and opposes interest on reserves, which he sees as contractionary and as reinforcing misleading interest-rate-based thinking. Institutional inertia and Fed appointments (Priority: 3/5): The conversation covers Obama-era vacancies, the limits of personnel changes at an inertial institution, and the political gridlock affecting Fed governance. Rules, generational bias, and policy discipline (Priority: 3/5): He supports rules like level targeting to reduce the impact of policymakers' lived-inflation memories and other cognitive biases on monetary policy.

Key Arguments: The Fed should have been more aggressive in late 2008 and 2009; merely making gestures with rate cuts and QE was not enough to hit inflation and spending goals. A temporary price-level level target at the zero bound would have been a modest but effective strategic shift without abandoning the 2% framework. Central banks should 'target the forecast' and act until market and macro forecasts indicate policy is likely to succeed. Inflation targeting is flawed because inflation can come from supply shocks, taxes, or commodity prices; nominal GDP is a cleaner indicator of aggregate demand. QE had positive effects, but it was only a tactic; without the right regime, it could not fully offset the downturn. The Fed persistently undershot its 2% inflation target, which raises credibility concerns and suggests the target may be suboptimal. Interest on reserves was a contractionary mistake introduced at the worst possible time and reinforces an overly interest-rate-centered view of policy. Long-term low rates are driven more by global demographics, saving patterns, and slower productivity growth than by Fed bond buying alone. Monetary policy is too blunt to manage asset bubbles sector by sector; housing and banking distortions are better handled through regulation. Level targeting can help reduce the influence of policymakers' generational memories and make policy more rule-like and disciplined.

Data Points: Fed inflation target: 2% - Explicitly adopted in 2012; discussed as the implicit benchmark for years before that. Average preferred inflation measure since June 2009: about 1.5% - Fed has persistently undershot its target during the recovery. Fed funds rate refusal after Lehman: did not cut below 2% - September 2008 meeting; Sumner says this reflected inappropriate inflation concern. Fed funds futures expectation: 3.5% one-year-ahead forecast in June 2008 - Market expected higher rates because Fed messaging signaled tighter policy. U.S. housing construction decline: from 2 million to 1 million annual rate - January 2006 to April 2008; housing fell sharply without a big rise in unemployment. Unemployment rate change during housing slowdown: 4.7% to 5.0% - Early housing bust period before the broader collapse. Unemployment rate change during broad recession: 5% to 10% - After April 2008, when weakness spread beyond housing to the whole economy. Natural rate estimate (Laubach-Williams example): fell from 2% to 0.34% - 2007 Q4 to one year later, cited as evidence of a sharp short-run drop in r-star. 30-year Treasury yield: about 3% - Used as market evidence for low expected nominal growth and a new normal of lower rates. ECB rate increase: 0.5 percentage points in 2011 - Cited as an example of a small hike having large contractionary effects. 1937 Fed tightening: about 0.25 percentage points - Used as historical evidence that small rate moves can trigger severe downturns. Obama-era Fed vacancies: 2 empty seats for about 12 months - Sumner argues the administration failed to prioritize Fed appointments early on.

Pivotal Quotes: "we're going to temporarily do level targeting of the price level where we have in mind a 2% increase over time in the price level" — Scott Sumner: His preferred policy response that would have been less radical than full NGDP targeting but more effective at the zero lower bound. "never reason from a price change" — Scott Sumner: His core critique of inflation targeting: inflation alone does not reveal whether shocks are demand- or supply-driven. "I think monetary policy is too blunt an instrument" — Scott Sumner: Explaining why the Fed should not try to manage housing bubbles directly through tighter money.

Implications: Listeners should expect future monetary debates to center on rules, target choice, and how to communicate policy credibly at low rates. The discussion favors NGDP/level targeting and warns against overreliance on inflation, interest rates, and balance-sheet optics.