Wednesday, October 29, 2014

QE3 R.I.P.

Today the Federal Reserve confirmed what the bond market has been expecting for many months: QE3 has ended, effective this week. What we don't know yet, however, is whether the end of QE3 will lead to another round of economic and financial market distress like we saw after the end of QE1 and QE2. I think we'll be OK this time around, because several key indicators today look a lot healthier than they did when QE1 and QE2 ended.

But first, let me point out once again that the real purpose of QE was not to print money or stimulate growth. It was to "transmogrify" notes and bonds into T-bill substitutes (aka bank reserves). QE boils down to the Fed simply swapping bank reserves for notes and bonds. Banks have been happy to hold most of the extra reserves as "excess" reserves, which means that they didn't use their reserves to collateralize a huge increase in lending. There was a huge demand for reserves qua reserves, and QE simply satisfied that demand. Without QE there would have been a critical shortage of safe, risk-free assets, and that would have threatened financial stability.

This time around, it looks like the demand for safe assets like bank reserves has declined and there's more financial stability, which means there is no longer a need for QE. Here's a quick look at the evidence:

The chart above shows the history of QE and 10-yr Treasury yields. Most of the Fed's quantitative easing efforts were focused on longer-term Treasuries, in a professed attempt to artificially depress yields and thus stimulate lending and the economy. But as the chart demonstrates, 10-yr yields rose over the course of each episode of QE. Moreover, 10-yr yields were unchanged during the period of Operation Twist, in spite of the fact that OT placed extra emphasis on bringing down 10-yr yields by buying long bonds and selling short bonds. In short, QE never achieved its intended result. I think that's because monetary policy is incapable of artificially manipulating long-term Treasury yields. Those yields are not determined by the size of Fed bond purchases, but rather by the bond market's perception of underlying economic and inflation fundamentals. Yields rose despite QE purchases because QE addressed a fundamental problem—a shortage of risk-free assets—and thus QE improved the outlook for growth.

It may sound strange, but despite the Fed's massive purchases of notes and bonds (totaling over $3 trillion), the Fed today holds about the same percentage of outstanding Treasuries as it did 10 years ago (see chart above). To be fair, a good portion of the Fed's holdings of Treasuries prior to the Great Recession were T-bills (a little over 30%). The Fed sold almost all of its T-bills in the first half of 2008 as it tried to respond to the market's desperate desire for risk-free assets. But it wasn't enough, and that is one of the reasons the Fed decided to embark on QE1.

In any event, after all those purchases of notes and bonds, 10-yr yields today are right around the same level as they were when QE1 was first launched. As the chart also shows, the correlation between big changes in the Fed's bond holdings and the level of 10-yr yields is not what we were told to expect. Big increases in Fed bond purchases (i.e., periods in which the blue line rose) were supposed to produce big declines in yields, because lots of Fed bond buying would push bond prices up. More often than not, however, the reverse occurred (i.e., both the red and blue lines moved together).

The chart above compares the S&P 500 index to the Euro Stoxx index. Both suffered serious corrections following the (largely unexpected) end of QE1 and QE2. The market is understandably concerned that this might happen again, now that QE3 has ended. Note, however, that equity valuations are significantly better today than they were at the end of QE1 and QE2, even though the end of QE3 has been known with reasonable certainty for many months. QE1 and QE2 were never tapered, by the way, they just ended all of a sudden. In contrast, the Fed has been tapering QE3 for the past 10 months. The end of QE3 cannot be a surprise or a disappointment to anyone at this point. If anything, it's a relief to know that it's over.

The chart above compares the level of 2-yr swap spreads—excellent coincident and leading indicators of systemic risk and economic and financial market health (see longer explanation here)—in both the U.S. and the Eurozone. Note that swap spreads rose significantly in advance of the recession and declined significantly in advance of the beginning of recovery. Note that they also rose following the end of QE1 and QE2. The major source of risk in the past four years has been the Eurozone, which has struggled with sovereign default risk and a double-dip recession. Eurozone banks were desperate to shore up their balance sheets throughout, thus creating significant demand for risk-free assets. Eurozone swap spreads were already elevated—symptomatic of rising systemic risk—in the runup to both QE1 and QE2, and they widened further after they ended. U.S. swap spreads rose in sympathy with Eurozone spreads, but to a lesser degree and starting from a lower base. Today, swap spreads in both regions are comfortably within "normal" territory. This, along with higher equity prices, suggests that the end of QE3 will not be painful.

The chart above compares the price of gold to the price of 5-yr TIPS, using the inverse of their real yield as a proxy for their price. Both of these are unique types of risk-free assets. Gold, because it is a classic refuge from political and monetary risk, and TIPS, because they are protected against inflation, they pay a government guaranteed real yield, and they are relatively short-term in nature. As the chart shows, the prices of both of these risk-free assets have been declining for the past few years. In other words, the market's demand for risk-free assets is lower (and falling on the margin) than it was when QE1 and QE2 were terminated. Again, this suggests that the end of QE3 should not result in tears. The market is no longer in need of more risk-free securities.

QE3, R.I.P.


Benjamin Cole said...

Per usual, very thoughful blogging.

QE both stimulates the economy and lowers interest rates relative to what they would have been without QE.

Also, we do not know what fraction of the QE spending ended up as bank reserves and what fraction ended up as spending in the economy or investments in other assets, like equities.

true bank reserves increased during QE, but that could also reflect in flows from savers and even foreign flight capital.

Perhaps it is time for central banks to consider QE as conventional policy to be applied whenever inflation is too low or growth too weak.

Low inflation and tiny growth may be the new normal, but I sure do not like that reality.

Matthew Grech said...

Excellent blog, Scott. The insight expressed in this blog - points you've made over the last five years - has been truly inspired. It's incredible to me that even with the benefit of hindsight the vast majority of pundits don't accept that QE was an exercise in meeting the unusually large demand for safe assets.

Houston Advisor said...

Scott, I wish your insights would end up on the talking head shows. I really appreciate your work.

Flow5 said...

If the Fed didn't establish the Primary Dealers as their go-between in 1960 there would have been a V shaped recovery.

dis737 said...

"What will really make a difference, however, is a decision by the Japanese government to also adopt genuine fiscally-stimulative measures such as lower marginal tax rates, reduced government spending, and reduced regulatory burdens."

How is "reduced government spending" stimulative?

Scott Grannis said...

Government spending can never be stimulative, contrary to what Keynesians think. In order to spend more, the government must borrow more from the private sector (or raise taxes). So if the government spends more the private sector must perforce spend less. The real issue then is whether government spending is more efficient and more productive than private sector spending. In my experience the public sector is far more likely to spend money inefficiently and with more waste and fraud than the private sector.

If the public sector spends less, the private sector can spend more, and that spending will be better for the economy in the long run.

Charles said...

Do the banks have an enormous appetite for excess reserves? Or does the Treasury/Federal Reserve have a policy driven mania for financing the public debt with excess reserves rather than bills, notes, and bonds? I contend it is the second and the banks should be happy to hold whatever quantity of excess reserves the government wants them to hold as long as the interest paid is competitive. There are two players in any market and markets are cleared at a quantity and a price. The demand for gasoline is inelastic with respect to price. The demand to hold excess reserves should be almost infinitely elastic. Gasoline meets a concrete need. Excess reserves are entries on balance sheets, they don't do anything except generate an adequate interest income.

And while the fed was buying notes and bonds, the treasury was increasing the maturity of the public debt.

There is no evidence that QE2, Operation Twist and QE3 had any first order effect on the economy.