Marvin Goodfriend | Macro Paper Warehouse

Central banking in the credit turmoil: An assessment of Federal Reserve practice

Written in the aftermath of the Federal Reserve’s extraordinary balance-sheet expansion during the 2007-2009 credit turmoil – reserves rose from roughly $10 billion in early September 2008 to over $1 trillion, and the balance sheet grew from about $900 billion in mid-2007 to more than $2 trillion by April 2009 – this paper argues that understanding central banking during the crisis requires separating three distinct kinds of initiative: monetary policy (open-market purchases or sales of Treasury securities that change the aggregate quantity of bank reserves and currency), credit policy (shifting the composition of the central bank’s asset portfolio between Treasuries and non-Treasury credit, holding the size of the balance sheet fixed), and interest-on-reserves policy (varying the interest paid on bank reserves, holding both of the others fixed). The paper’s central argument is that each of these has different fiscal implications. Monetary policy conducted under a strict “Treasuries only” acquisition rule is fiscally neutral, because the central bank returns all the interest it earns on its Treasury holdings to the fiscal authorities, so expansionary monetary policy simply hands revenue to the Treasury to allocate as it sees fit. Credit policy is fundamentally different: because it finances loans or non-Treasury security purchases by selling Treasuries (or, in combination with monetary policy, by creating new reserves), “the result is just as if the Treasury financed the loans or purchases by borrowing from the public” – credit policy is debt-financed fiscal policy that commits future tax revenue to particular borrowers and exposes both the central bank and taxpayers to credit losses and allocative controversy. Interest-on-reserves policy, similarly, uses public funds to pay banks and so also has fiscal features, but its chief practical virtue in the crisis was that it let the Fed fund expansive credit initiatives with newly created reserves without abandoning control of the federal funds rate. Goodfriend argues that monetary policy can be conducted independently of the fiscal authorities because its goals are clear and “Treasuries only” leaves fiscal allocation entirely to Congress and the Treasury, but that credit policy cannot claim the same independence, because its objectives have never been clearly circumscribed and it inherently allocates public funds. Reviewing five 2007-2009 episodes – the Term Auction Facility, the Fed’s financing of JPMorgan Chase’s acquisition of Bear Stearns via Maiden Lane I, the Fed’s $85 billion loan to AIG, the Fed’s push for emergency authority to pay interest on reserves, and the March 2009 Treasury-Fed joint statement – the paper argues that an ambiguous boundary of fiscal responsibility between the Fed and the Treasury contributed to the panic and economic collapse of fall 2008, particularly around the AIG episode, where the paper concludes flatly that “an independent central bank cannot be responsible for delivering or deciding upon the delivery of fiscal support for the financial system.” Drawing an explicit analogy to the 1951 Treasury-Fed Accord that established Fed independence over interest-rate policy, the paper proposes three principles for a parallel “Accord” on credit policy: a sustained departure from “Treasuries only” is incompatible with Fed independence; the Fed should adhere to “Treasuries only” except for occasional, temporary, well-collateralized last-resort lending to solvent depositories; and any broader credit initiative should require the fiscal authorities’ prior agreement and be structured as a bridge loan with a take-out arranged and guaranteed in advance by those authorities. The paper also argues the Fed should not be made the economy’s “pinnacle” systemic-risk regulator, since granting or denying fiscal support for troubled firms is inherently a fiscal decision that would politicize an independent central bank – consistent, in the author’s view, with Dodd-Frank’s choice to place that authority in a Treasury-chaired Financial Stability Oversight Council instead. Finally, the paper proposes that the fiscal authorities enlarge the Fed’s surplus capital account so the Fed can pay interest on reserves confidently under any future inflation or deflation scenario without first needing to shrink a balance sheet that may include substantial long-term securities acquired to fight deflation at the zero bound – a step the author argues would carry no fiscal cost as long as the Fed does not draw on the enlarged account, while substantially improving the Fed’s flexibility to tighten policy when needed.

Interest Rate Policy and the Inflation Scare Problem: 1979-1992

Using the 30-year bond rate as a real-time signal of the public’s long-run inflation expectations, this narrative study of Fed federal funds rate policy from 1979 to 1992 argues that the central challenge of the disinflation era was managing repeated “inflation scares” – sudden jumps in the long rate even without loose policy – and that delays in responding to them, more than any single decision, explain why acquiring disinflationary credibility took as long and cost as much as it did. Treating the federal funds rate (not the monetary base) as the Fed’s actual policy instrument, the paper develops a decomposition of the long-term bond rate into a component anchored by the current funds rate target (via arbitrage across maturities) and a component reflecting the public’s expected long-run inflation rate, and classifies funds-rate/long-rate co-movements into purely cyclical actions, changes in the long-run inflation trend, and aggressive disinflationary or stimulative actions. Walking chronologically through the October 1979 switch to nonborrowed-reserve targeting, the March 1980 credit-control interruption, the 1981-82 disinflation, the 1983-84 and 1987 inflation scares, and the 1990-92 easing, the paper documents that aggressive tightenings pulled the long rate in the same direction as the funds rate (not the opposite, as one might expect), that long-rate volatility was unusually high until 1988, and that the funds rate peaked in October 1981 – a full two years after the disinflation began – in part because a temporary Fed hesitation in early 1980, the March 1980 credit controls, and automatic funds-rate declines under the nonborrowed-reserve operating procedure each interrupted the tightening. The paper concludes that the Fed’s disinflationary credibility remained fragile throughout the 1980s – a scare could recur even years after inflation had stabilized, as in 1983-84 and 1987 – and argues, by comparison with the Bundesbank’s and Bank of Japan’s stronger price-stability mandates, that a congressional price-stability mandate could reduce the frequency of costly inflation scares and thereby give the funds rate more room to respond to unemployment in the short run.

The Phases of U.S. Monetary Policy: 1987 to 2001

Dividing 1987-2001 into six phases, this narrative traces how the Federal Reserve pursued the same four objectives – credibility for low inflation, accommodating productivity-driven growth, containing financial-market distress, and stimulus when needed – through remarkably varied circumstances. Phase 1 (October 1987-July 1990) covers the Fed’s liquidity response to the stock market crash and the inflation scare and entrenched inflation that followed its slow, delayed tightening response; Phase 2 (August 1990-January 1994) covers the Gulf War recession and the gradual disinflation that followed. Phase 3 (February 1994-February 1995) is the paper’s central success story: a preemptive tightening from 3 to 6 percent undertaken while inflation was stable at 2.5-3 percent, which “succeeded in its main purpose: to hold the line on inflation without creating unemployment” and laid the foundation for the long boom, even as the public came to misattribute the resulting low inflation to an independent “death of inflation” rather than to the Fed’s own preemptive action. Phase 4 (January 1996-May 1999) covers the long boom, in which the Fed had to learn to operate with newly won “near full credibility” for low inflation – itself a complication, since both the Fed and the public tend to overestimate noninflationary potential output once credibility is secured – compounded by genuinely rising but hard-to-measure trend productivity growth. Phase 5 covers the 1999-2000 tightening against overheating, and Phase 6 covers the collapse of business investment and the 2001 recession, in which the Fed cut the funds rate by 4.75 percentage points in real terms without triggering an inflation scare, “because of the near full credibility for low inflation” built up over the preceding decade. The paper’s overall argument is that despite the surface variety of the problems – financial crises, two wars, a productivity boom, an investment bust – the Fed’s policy actions throughout can be understood as consistently serving the same small set of objectives, with the 1994 episode standing as the clearest illustration of preemptive, credibility-building policy in action.