Federal Reserve Mischief and the Credit Trap.

• Author: Thornton, Daniel L.

The Federal Reserve's monetary policy response to the financial crisis is a disaster. The financial crisis began on August 9, 2007, when BNP Paribus suspended redemption of three investment funds. The Fed responded first by reducing its lending rate (the primary credit rate) and then by reducing its target for the federal funds rate from 5.25 percent on September 17, 2007, to 2 percent on May 1, 2008. Despite these actions the recession worsened and the financial crisis intensified (see Thornton 2014a).

Instead of increasing the monetary base (the Fed's contribution to the supply of credit) as it should have, and as Friedman and Schwartz (1963) recommended, the Fed sterilized the effect of its lending on the supply of credit by selling an equivalent amount of Treasury securities. This forced the market to reallocate credit to the institutions to which the Fed made loans (see Thornton 2009). It did this so it could continue to implement monetary policy with the federal funds rate. Had the Fed not sterilized its lending, the reserves created would have made it impossible to keep the funds rate much above zero. Indeed, the Fed attempted to sterilize its massive lending following Lehman's September 2008 announcement by having the Treasury issue supplemental financing bills and deposit the proceeds with the Fed (as much as $559 billion). Despite these efforts, the monetary base exploded, doubling in size from August to December 2008. The funds rate headed for zero, well below the Fed's 2 percent target and well in advance of the FOMC reducing the target to zero on December 15 (see Thornton 2015).

Then, instead of simply allowing its balance sheet to shrink back to its pre-Lehman level as the economy improved and lending was repaid, as would have been appropriate, the Fed engaged in a massive purchase of long-term securities--quantitative easing (QE)--in an attempt to stimulate economic growth by reducing long-term yields.

However, in this article I show why both QE and the FOMC's forward guidance policy actually had virtually no effect on long-term yields. On the other hand, the FOMC's 8-year low interest rate policy has caused long-term yields to be lower than they would have been otherwise. It is important to draw this distinction because it pinpoints the real source of the Fed's mischief.

The FOMC's low interest rate policy has resulted in excessive risk taking by citizens, pension funds, banks, and other financial institutions; a large and likely an unsustainable rise in asset prices; and an unprecedented increase in the money supply. The low interest rate policy has also distorted the allocation of economic resources in a variety of ways, many of which are impossible to determine, and may have effects that are impossible to predict. These results are the consequence of a policy that was ill conceived, motivated by fear, and lacking theoretical foundations. The last point applies to the FOMC's low interest rate policy as much as to QE and forward guidance, as it is predicated on a model that lacks empirical support and has a theoretically weak central premise. In short, the FOMC's low interest rate policy has not only failed to deliver but has also had terrible and potentially drastic consequences. An increasing number of analysts are beginning to realize this. Unfortunately, none of them are members of the FOMC.

In what follows, I will discuss the consequences of the FOMC's low interest rate policy and speculate on how it might end. I will conclude by arguing that the Fed's mischief is just the latest installment in the long history of the credit trap--a trap facilitated, if not caused, by the widespread acceptance of Keynesian economics. I will begin, however, by showing why QE and forward guidance should not have had--and, in fact, did not have--a significant effect on long-term yields.

Doomed to Fail

QE and forward guidance were doomed to fail because their theoretical foundations are made of sticks, not bricks. Bernanke (2012), Gagnon et al. (2011), and others argued that QE reduced long-term interest rates through the so-called portfolio-balance channel, and by reducing term premiums on long-term debt. However, their version of the portfolio-balance channel requires that markets are segmented beyond the degree determined by the risk characteristics of bonds and market participants' aversion to risk. Specifically, it requires the market to be segmented along the yield curve, which is at odds with over 50 years of finance theory. Bernanke and others suggest that such circumstances prevailed after Lehman Brothers' bankruptcy announcement. But even this seems unlikely: the collapse of the interbank and other markets was more likely due to the fact that most banks and other financial institutions had a large quantity of mortgage-back securities (MBS) and other derivatives on their balance sheets. These securities were "toxic"--that is, no one knew exactly what they owned and, by extension, the real value of those holdings. Banks did not know the credit risk of their own MBS, let alone those of other banks. Consequently, the interbank market froze up. Other markets were also impaired. Credit risk premiums skyrocketed. But only temporarily, as is shown in Figure 1, which displays the spread between the 3-month CD and T-bill rates weekly over the period January 2007 through December 2009. This spread, which had risen from 56 basis points the week before the financial crisis began to 119 basis points the week before Lehman's announcement, jumped to 344 basis points the week of that announcement. The spread peaked at 437 basis points for the week ending October 17. Yet just as economic theory would predict, financial markets soon healed and credit risk premiums declined. By the week ending January 9, 2009, the spread was down to 96 basis points--below the pre-Lehman level. By early May, the spread dropped well below its pre-financial crisis level and remained there.

Other credit-risk spreads behaved similarly, although some returned to their pre-Lehman level more slowly. For example, Figure 2 shows the bi-weekly spread between Baa and Aaa corporate bonds yields over the same time period. The spread increased after the onset of the financial crisis and exploded following Lehman's announcement. It peaked in late December at 350 basis points. Its decline was briefly interrupted when the FOMC announced that the Fed would purchase a total of $1.75 trillion in long-term debt, because Baa yields rose relative to Aaa yields. However, this spread returned to its pre-Lehman level, about 150 basis points, by the two-weeks ending August 12. It continued to fall as the Baa yield fell relative to the Aaa yield as both yields declined.

There is no reason to believe that markets were segmented along the term structure on May 18, 2009. By these spreads and other indicators, markets were functioning well by the time the FOMC launched its massive QE program. Consequently, there is no reason to believe that QE had any significant effect on long-term yields--or on the structure of rates generally--through the portfolio-balance channel. While large by historical standards, the Fed's purchases were simply too small to have any significant effect on the level of the overall structure of interest rates (Bauer and Rudebuseh 2014). Any such effect would be so small as to be undetectable.

The idea that QE reduced term premiums, which was hypothesized to occur because the Fed's purchases removed a large quantity of interest rate risk from the market, is equally fanciful. The term premium of a long-term bond relative to a short-term bond depends on two things; both are independent of the amount of bonds in the market. The first is the relative duration risk of the two bonds, which is determined solely by the relative durations of the bonds, which in turn rests on those bonds' characteristics (e.g., their relative maturities, call provisions, collateral, and so on). The second is the risk aversion of market participants, which is innate to each investor. In the case of default-risk-free Treasuries, this means the risk premium could decline only if the Fed's purchases caused the most risk-averse investors to leave the market. This seems unlikely: the most risk-averse investors are surely more likely to remain in the default-risk-free Treasury market, while the least risk-averse investors seek higher returns elsewhere. In that scenario, term premiums would rise, not fall. In any event, whether the term premium...