QE through the looking glass

By G.I. | WASHINGTON, D.C.

Even if you follow central banking, you may not have heard much about Jeremy Stein. He's a highly regarded finance economist from Harvard University who became a governor this year, and since then he has given only one speech. But judging by the contents of that speech, Mr Stein's profile seems bound to grow.

The subject of the speech was the Fed’s large scale asset purchases (LSAPs), colloquially known as quantitative easing (QE). It had none of the tantalizing hints about coming Fed actions that attract headlines. Rather, it mucked about in dark, abstruse corners of corporate finance, devoting three of its 19 pages to academic references.

But buried in those pages were striking insights into monetary policy. Some were the subject of David Wessel's column today in The Wall Street Journal, others of the Free Exchange column in this week's Economist. He will discuss the speech at a panel in Boston tomorrow.

One insight in particular caught my eye. Mr Stein sought to explore precisely how bond purchases stimulate economic growth. Start by asking what determines a bond’s yield. Assume investors expect treasury bill rates to average 2% this year and 4% next year. That means a two-year bond would yield 3% for its expected return to equal T-bills'. In practice, the bond will typically yield more: say, 3.5%, because investors usually want to be compensated for locking up their money, and borrowers will pay more to avoid refinancing risk. This difference is called the term premium.

QE can reduce bond yields two ways. First, it can signal that the Fed is serious about keeping short-term rates lower for longer. In our above example, investors may expect T-bills to yield 2% next year, instead of 4%. This should lower bond yields. Second, by reducing the supply of long-term bonds, it can force investors who want long-duration assets to accept a lower return on the smaller supply of those still available. This reduces the term premium. Investors have indeed moved out the date at which they expect the Fed to tighten, but that's mostly because the Fed says it won't raise rates until 2015 at the earliest. QE's much more noticeable effect has been on the term premium, which Fed staff reckon has fallen to minus80 basis points, near an all-time low.

Until I read Mr Stein’s speech, I didn’t think any of this mattered much. Whether yields were low because of sanguine expectations of Fed tightening or because of a negative term premium, the impact on borrowing and consumption was the same. And indeed, the Fed's own macroeconomic model, FRB/US, makes the same assumption. Mr Stein demolishes that illusion. Bear with him as he explains:

A risk-neutral firm faces a rate on its 10-year bonds of 2 percent. At the same time, it expects that the sequence of rolled-over short-term rates over the next 10 years will average 3 percent. Hence, there is a term premium of minus 1 percent. What should the firm do? Clearly, it should take advantage of the cheap long-term debt by issuing bonds. But it is less obvious that the bargain 2 percent rate on these bonds should exert any influence on its capital spending plans. After all, it can take the proceeds of the bond issue and use these to pay down short-term debt, repurchase stock, or buy short-term securities. These capital-structure adjustments all yield an effective return of 3 percent. As a result, the hurdle rate for new investment should remain pinned at 3 percent. In other words, the negative term premium matters a lot for financing behavior, but in this stylized world, investment spending is decoupled from the term premium and is determined instead by the expected future path of short rates.

This reasoning suggests why one might expect future rounds of LSAPs to have diminishing returns. As noted earlier, the data make clear that past rounds of LSAPs have pushed down interest rates and term premiums. But the further the term premium is driven into negative territory, the more the previous logic comes into play, and hence the weaker is likely to be the response of aggregate spending to further downward pressure on long-term rates.

If the term premium starts out significantly positive, so that long-term rates are well above expected future short rates, an initial reduction may indeed spur further investment among financially constrained firms who need to issue long-term bonds to fund new investment. However, once the term premium becomes negative, we hit a corner where the financial constraint no longer binds and the relevant opportunity cost becomes the option to invest in short-term securities or repurchase shares.

If that were the end of the story, Mr Stein’s conclusion would be quite troublesome for advocates of QE. It would reinforce Michael Woodford’s Jackson Hole critique of QE, i.e. that it only works if it changes expectations.

Fortunately, it’s not the end of the story. The above example may be true of a company with diverse financing options, but it’s less likely to be true of a household. Faced with a big drop in the mortgage rate, a consumer is unlikely to borrow $200,000 and plough it into a money market fund. He may pay off some short-term debt, or refinance an existing mortgage. But there’s a good chance he will buy a house (assuming he qualifies for the loan) irrespective of his expectations of short-term rates. This strengthens the case for QE to be conducted through purchases of mortgage-backed securities rather than Treasuries.

If Mr Stein’s story is right, we should expect to see corporations exploiting the drop in long-term rates to refinance short-term debt and repurchase stock but not boost capital spending, and individuals exploiting the drop in mortgage rates to refinance and to buy houses. That, of course, is precisely what is happening.

Mr Stein also argues that even if QE only induces businesses to fiddle with the composition of their balance sheets, that has some benefit. When businesses reduce their reliance on short-term debt, they are less vulnerable to a surge in financial system stress that cuts off their access to credit, making financial crises less likely.

This observation dovetails neatly with a theme of Mr Stein’s pre-Fed research. In a previous paper, (written up in The Economist here) Mr Stein and his co-authors argued that investors and other market participants have a large appetite for financial assets with the attributes of money: ultra short maturity, and minimal credit risk owing to the high quality of the collateral or sterling reputation of the issuer. Demand for this so-called private money rises when the relative supply of treasury bills and other riskless government paper shrinks. But this makes the financial system vulnerable to panic if holders lose confidence in their private money and try to exchange it for the real thing.

Their theory is well supported by history, as Gary Gorton shows in his new book, Misunderstanding Financial Crises (reviewed here in this week’s issue of The Economist). In the early 1800s, banks issued their own notes and during a panic, customers would seek to convert them to specie. That source of instability was eliminated in the 1860s when national bank notes replaced private bank notes. But then, a new source of instability arose. Banks began letting customers write checks on their deposits (Mr Gorton calls checking accounts the shadow banking system of their day). If customers lost confidence in banks, they would rush to convert deposits to bank notes. But the supply of bank notes, like the supply of specie in the pre-national bank era, was not large enough to meet all the redemption demands, and banks failed.

The creation of the Federal Reserve was meant to eliminate panics forever by supplying an elastic currency, i.e. unlimited bank notes to meet spikes in demand for cash. The Fed failed at that job in the early 1930s for complex reasons, but between 1934 and 2007 America was largely spared large-scale panics thanks to deposit insurance and a central bank acting as lender of last resort.

Then in recent decades, financial innovation once again allowed private money to flourish, in the form of repo loans, asset-backed commercial paper, and money market mutual fund shares, none of which benefited from deposit insurance or a lender of last resort. The latest crisis, Mr Gorton maintains, stemmed from a run on this money. Arguably the Fed's most important and most controversial actions during the crisis were to extend its lender of last resort authority to nonbanks, in effect creating actual money to redeem the private money no one wanted any longer.

While QE, as Mr Stein says, reduces the supply of private money, it doesn’t alleviate the demand for private money. That problem has only grown since the crisis. Regulators now require banks to hold much larger liquidity buffers, and money market funds to shorten the duration and raise the credit quality of their holdings. Sovereign wealth funds, European banks and retirees around the world are all more desperate for safety and security.

As the private demand for money-like assets rises but the private supply falls, what will bridge the gap? Mr Stein’s answer is that the Treasury should issue more short-term debt. But this conflicts with Treasury’s desire to reduce its own exposure to refinancing risk, although floating-rate notes are a promising compromise.

Little noticed has been the role of QE in bridging the gap. To buy all those bonds, the Fed has “printed” $1.5 trillion in additional money, i.e. reserves. By cramming so much money down the throats of the banks, QE has made them virtually impervious to a liquidity crisis. This raises an intriguing possibility. When the need for QE ends, it may not be wise for the Fed to allow all those excess reserves to disappear: it could leave the financial system scrambling for some, alternative money-like asset, planting the seeds for the next cycle of private money creation and panic.