by Stephen Grenville
This article first appeared in Voxeu.org 22 June 2013
Chairman Bernanke’s hints about the end of quantitative easing (QE) have produced volatility in financial markets. This column argues that financial markets were startled because an end to QE is likely to cause capital losses for bond holders since term premium is substantially negative. Bank regulators should be alert to the possibility. This fundamental explanation is teamed with widespread confusion among market participants about how quantitative easing actually works.
Fed Chairman Ben Bernanke’s prepared statement on 22 May was the epitome of even-handed non-committal drafting (Federal Reserve 2013b) but the mention of “stepping down” and “in the next few meetings” in the discussion sent a shiver through financial markets worldwide. Bond yields jumped just about everywhere; the Abenomics euphoria in Japan deflated; and capital flows to emerging markets reversed direction. Bernanke was just pointing out the obvious and he went on to say that
“we could either raise or lower our pace of purchases going forward”.
Why were financial markets so startled?
The problem is twofold: the first relates to the intrinsic nature of quantitative easing; the second reflects market confusion about how QE operates.
QE is not simply a continuation of conventional monetary policy by alternative means. It works by lowering the longer-term interest rates below what would be indicated by the profile of expected short-term rates. When conventional monetary policy is tightened in line with market expectations, there are no capital losses experienced by bond-holders. But QE has succeeded in getting the term premium not just negative (which would be unusual enough), but substantially so, as acknowledged by Bernanke in March (Federal Reserve 2013a).
At some stage during the unwinding of QE, bond-holders will suffer a painful capital loss as yields return to normality. Bernanke’s remarks on 22 May, innocuous though they were, reminded bond-holders that they need to be ahead of the pack when the moment comes to lighten the bond portfolio. With markets on tenterhooks, the adjustment could easily be sudden, and just as easily involve a price overshoot. No matter how careful Bernanke is, this adjustment is likely to look something like the 1994 bond shock because it has an intrinsic ‘tipping point’.
The capital losses on bonds are not the only tipping point ahead. The most prominent and consistent QE effects were on equity prices and exchange rates – effects which were not explicitly foreshadowed when QE began. While these effects might have been helpful in boosting economic activity, these distortions will reverse when QE is withdrawn. Equity markets and the international capital carry trade have become as dependent on QE as any drug addict.
If markets had a clear understanding of how QE operates, they might effectively anticipate the unwinding and soften the tipping point. Some (perhaps many) market participants have, however, misunderstood the nature of QE. They expected an automatic increase in lending through the credit multiplier and a boost in money supply (it was almost invariably referred to as ‘printing money’), hence pushing up inflation. In fact money just maintained its trend growth. Credit growth has been weak. Underlying inflation has actually fallen. None of this should have come as a surprise. There was never going to be CPI inflation while unemployment was so high and the slow cyclical recovery is enough to explain lethargic credit demand. But when outcomes have been far from market expectations, participants are disoriented.
In these confused circumstances, market responses have been based on expectations of policy changes rather than on fundamentals. Asset prices untethered to fundamentals will be volatile. If market participants don’t understand how we got here, they will be ready to panic at the thought of how we will exit. Markets were ready to seize on a spontaneous response in Bernanke’s testimony, rather than the finely honed text, which pointed to a careful and drawn-out unwinding.
Unwinding QE doesn’t have to be too traumatic. The Fed can shift short-term policy interest rates up as appropriate, while leaving the disposal of its abnormally large bond holdings to a later date. Even if it makes capital losses in the process, these can be absorbed without too much drama. Concerns about inflation from ‘printing money’ were always totally misplaced.
That said, some of the private-sector capital losses discussed above will fall on vulnerable sectors, especially the banking system. The prudential authorities need to be on top of these, and some carefully-modulated public discussion (and data) on bank balance sheets might be helpful.
We are currently witnessing a tug-of-war between financial markets that are pressuring the authorities to keep QE going, and those who see it as having served its purpose, losing its potency and storing up further problems in unwinding. Bernanke has always acknowledged that QE has a downside, but he has justified it on the basis of its support for economic activity. If it has reached the stage where the current dosage simply delays the inevitable painful withdrawal, it may be harder to resist the pressures to unwind.
If the balance of these forces results in QE being wound back soon, this will be occurring in an economy which is still much weaker than might have been hoped. As well, policymakers might have hoped for a financial sector which had a better understanding of how this process will work and thus more ability to make the transition with composure. There is still time to achieve this better understanding, but the flightiness of markets since Bernanke’s tentative attempt to prepare for the unwinding is a reminder of how difficult it is to have a sensible conversation with the financial sector.
Federal Reserve (2013a), “Long-Term Interest Rates”, speech, Ben S Bernanke, 1 March.
Federal Reserve (2013b), “The Economic Outlook”, speech, Ben S Bernanke, 22 May.
[iframe src=”http://econintersect.com/authors/author.htm?author=/home/aleta/public_html/authors/grenville_stephen.htm” width=”600″ height=”500″ frameborder=”0″ scrolling=”no”]