Quantitative Easing: A Requiem

Quantitative Easing, Ben Bernanke, interest rates, financial crisis, credit supply
Horse-drawn carriage built by A. E. E Roberts Carriage Works, c. 1900. From the Queensland Museum, Queensland Australia. https://en.wikipedia.org/wiki/Hearse#/media/File:Hearse-r.jpg

Quantitative Easing, Ben Bernanke, interest rates, financial crisis, credit supplyWhen the Federal Open Market Committee (FOMC) meets in Washington next week, its members are widely expected to vote to raise interest rates for the first time since June 2006.  By doing so, they will move towards monetary policy normalization, after more than seven years of near-zero interest rates, and a vast expansion of the central bank’s balance sheet.

But how did monetary policy become so abnormal in the first place?  Were the Fed’s unconventional monetary policies a success?  And how smoothly will implementation of the Fed’s so-called “exit strategy” go?  These are among the questions addressed by Dan Thornton, a former vice president of the Federal Reserve Bank of St. Louis, in “Requiem for QE,” the latest Policy Analysis from Cato’s Center for Monetary and Financial Alternatives.

Thornton begins with an account of the Fed’s deliberations and actions during the early stages of the financial crisis.  What’s particularly striking, looking back, is how the Fed resisted letting its balance sheet grow, or otherwise departing from its conventional, funds-rate targeting procedure, until the Great Recession was well underway.

In fact, from August 2007, when BNP Paribas suspended redemption of three of its investment funds, to September 2008, when Lehman Brothers filed for bankruptcy, the Fed loaned banks and others more than $300 billion.  But these loans were sterilized—meaning that the Fed sold an equal amount of government securities.  As a result, the Fed’s balance sheet didn’t grow, and neither did total bank reserves or the monetary base.  For over a year, in other words, the Fed reacted to a growing liquidity crisis not by taking steps to boost credit in general, but rather by reallocating the existing supply of credit towards particular, troubled firms — a move George Selgin examined (and criticized) in detail in a recent post on this blog.

The Fed’s approach flew in the face of widely-acclaimed research presented by Milton Friedman and Anna Schwartz in their A Monetary History of the United States.  As Thornton points out, Friedman and Schwartz “connected substantial reductions in the nominal quantity of money and credit to correspondingly large declines in economic activity, and declared it the Fed’s duty to act quickly to prevent such reductions by expanding the monetary base.”  In 2007–08, however, no such action was forthcoming, and the crisis continued to intensify.

The sad part is that then-Fed chairman Ben Bernanke knew all this.  Back in 2002, he concluded a speech in honor of Milton Friedman’s 90th birthday by declaring: “I would like to say to Milton and Anna:  Regarding the Great Depression.  You’re right.  We did it.  We’re very sorry.  But thanks to you, we won’t do it again.”  And yet, as Thornton makes clear, “it” — failing to expand the balance sheet in order to prevent a collapse of credit — is “precisely what the Fed did in the months leading up to Lehman Brothers’ failure.”

Eventually, the Fed was forced to change tack.  Once Lehman Brothers collapsed, it found itself lending and buying assets on a scale that couldn’t possibly be sterilized via correspondingly large asset sales.  The Fed’s balance sheet grew, bank reserves began to pile up, and the federal funds rate dropped well below the FOMC’s target.  Unofficially, and almost by accident, quantitative easing (QE) began.

Quantitative easing was put on a formal footing in March 2009, when the FOMC announced the combined purchase of $750 billion of mortgage-backed securities and $300 billion of Treasuries.  From the very beginning, however, it was clear that the Fed was not belatedly adopting a Friedmanite approach.  The stated purpose of their asset purchases was not to boost the monetary base — that was just a side effect.  Instead, the Fed’s goal was to manipulate the yields on certain long-term assets, in the hope that this would spill over into broader markets, and the economy as a whole.

How was it meant to work?  Well, as Thornton points out, the rationale was “vague and highly uncertain,” and developed over time as QE was put into effect.  But the general idea was twofold.

First, by forcing down yields on the long-term assets it was purchasing, the Fed hoped other investors would be induced to substitute those assets for similar, but somewhat higher-yielding alternatives.  In turn, that shift in demand would push down yields on the alternative assets, encouraging investors who held them to rebalance their portfolios as well.  As this ripple effect spread through the financial markets, equilibrium interest rates would fall.

Second, the Fed hoped that QE announcements would signal to markets that monetary policy was going to be “persistently more accommodative” — Ben Bernanke’s words — than previously thought.  This would lower investor expectations for the path of short-term interest rates, and in so doing put additional downward pressure on long-term interest rates.

In short, then, the Fed expected QE to boost the economy through the interest rate channel of monetary policy: lower interest rates would boost equity prices and weaken the dollar; lower borrowing costs, higher wealth, and greater international competitiveness would drive spending, investment, and exports, and stimulate an ailing economy.

So did QE actually work in this way?  Thornton assesses these claims in detail (see, in particular, pp. 12–22 of his analysis), and finds little — either in economic theory or in empirical evidence — to support them.

For one thing, as large as the Fed’s asset purchases were relative to previous open market operations, they were modest compared to financial markets as a whole.  It is therefore “difficult to believe,” as Thornton puts it, “that the distributional effects of the FOMC’s quantitative easing policy could have had a significant effect on either relative yields or the level of the entire interest rate structure.”  What’s more, the event-study literature on QE announcements does not offer any convincing evidence of their effectiveness, once you account for the effect of other, simultaneous announcements (Fed statements typically contain a variety of news capable of influencing bond yields).

The claim that QE announcements changed expectations for the path of short-term interest rates is similarly hard to substantiate, not least because event studies only show the immediate impact of such announcements — and for QE to reduce long-term interest rates via this signaling channel, its effect on expectations must be persistent.  It is telling, moreover, that the Fed’s QE announcements did not even produce a consensus among FOMC participants about the expected path of the federal funds rate.  That being the case, it is hard to see how those same announcements could have had a significant impact on market expectations — especially when it is well-established that interest rates are, in Thornton’s words, “essentially unknowable beyond horizons of a few months.”

Of course, that QE did not work as the Fed hoped does not mean it had no effect at all.  As Thornton points out, “a monetary policy directed at keeping interest rates on low-risk securities near zero … is sure to cause people to seek higher yields by purchasing more risky assets.”  Thornton suggests that pension funds, in particular, have been forced to hold riskier portfolios to generate the returns necessary to meet their obligations.  As well as raising concerns about future financial crises, this distortion of investor behavior has had distributional effects: wealthy investors, who are better able to assume more risk, have benefited from booming equity prices, whereas as less well-off pensioners — who have good reason to be risk averse — have had to settle for miserly returns on their fixed-income portfolios.

What QE hasn’t done, however, is enhance the aggregate supply of credit to the market.  This is hardly surprising, given that the Fed began paying interest on bank reserves in October 2008 — a move designed to encourage banks to build up excess reserves, instead of increasing lending.  Indeed, when it came up at FOMC meetings, Ben Bernanke, Janet Yellen, and others expressly rejected the idea that QE would stimulate the economy by boosting bank lending.  Still, when you couple this absence of increased credit with the lack of evidence that QE worked the way the Fed thought it would, you would be forgiven for wondering whether QE had any significant impact on output and employment at all.

What now?  The Fed’s large-scale asset purchases are over, and the FOMC’s interest rate target looks set to gradually rise.  But the Fed’s “exit strategy” remains a work in progress.  Interest on reserves (IOR), for example, may yet prove a troublesome way of raising the federal funds rate.  As Thornton points out, “An IOR of 3 percent … would see the Fed paying nearly $80 billion a year in interest to the commercial banking sector — something that is unlikely to prove popular in Congress or among the general public.”

Meanwhile, the Fed’s preferred alternative, overnight reverse repurchase agreements, would have to be rolled over continuously to have any effect on the size of the Fed’s balance sheet.  A better approach, according to Thornton, would be to “begin the process of policy normalization by selling long-term securities slowly … If things go well, sales could subsequently be accelerated.”  But the Fed seems reluctant to consider outright asset sales at all.  This suggests that the bloated central bank balance sheet that QE created could be with us for some time to come — whether or not the FOMC votes next week to begin the process of policy normalization.

Yet the real legacy of QE may come in how the FOMC reacts to the next crisis.  One concern is that the Fed seems to have lost “all confidence in the ability of markets to heal themselves,” and now assumes that “only monetary or fiscal policy actions” can “restore the economy’s health.”  Another worry is mission creep: “If the Fed can support the mortgage and commercial papers markets … why shouldn’t it support the market for student loans — or any other market for that matter?”  It’s a dispiriting prospect, but both of these observations suggest an increasingly hyperactive monetary authority going forward.

Ultimately, says Thornton, we should prepare ourselves for more of the same: given the opportunity, “the Fed will continue to distort markets until, by some as-yet-unknown magic, those markets return to normal.”  One suspects that such magic may be a long time coming.