Normalizing Monetary Policy

By Martin Feldstein

The current focus of Federal Reserve policy is on “normalization” of monetary policy—that is, on increasing short-term interest rates and shrinking the size of the Fed’s balance sheet. Short-term interest rates are exceptionally low, and the Fed’s balance sheet has exploded from $800 billion in 2008 to $4.4 trillion now.

At the September 2017 meeting of the Federal Open Market Committee (FOMC), the Fed indicated its long-term goal for the federal funds rate and its near-term goal for shrinking the size of its balance sheet. In my judgment, the Fed’s plan is too little, too late. I also believe the Fed’s underlying goal is to increase inflation above its 2 percent target and that such a policy is wrong.

I think the Fed should have begun raising interest rates and reducing its balance sheet back in 2013 or 2014. I think the current goal of raising the federal funds rate from 1.4 percent now to 2.1 percent at the end of 2018 (when it would be a 0.1 percent real rate using the Fed’s median inflation forecast) is just too slow and will continue to encourage a dangerous bidding up of asset prices.

In this article, I will begin by discussing the Fed’s shift to the easy money policy. I will then turn to the adverse side effects of the quantitative easing (QE) policy, particularly the increases in asset prices that create a risk of financial instability. The next section considers the reasons that the FOMC members have continued to pursue the policy of excessively easy money. There is a brief concluding section.

The Shift to Quantitative Easing

As we all know, Ben Bernanke introduced quantitative easing as a way to stimulate economic activity at a time when the unemployment rate was very high and the recovery was very slow. Conventional monetary policy had failed to stimulate the economy even after the federal funds interest rate was cut to zero in 2008. The fiscal stimulus legislation enacted in 2009 also failed to raise real GDP growth because it was so badly designed.

Bernanke explained that a Fed policy of buying long-term bonds and promising to keep the federal funds rate low for a long time would cause a significant decline in long-term interest rates. That would raise the price of equities and of homes. The resulting increase in household wealth would then lead to increased consumer spending and therefore to faster GDP growth.

Bernanke explained that this would be reinforced by what he called the “asset substitution effect” in which the reduced availability of bonds in which to invest would cause households to shift their portfolios to equities.

Not everyone was convinced by Bernanke’s analysis. Why would investors buy equities that were made artificially high since they would know that those share prices would eventually decline? And how important could the asset substitution effect be when the household sector’s holding of Treasury bonds was less than 10 percent of its investment in equities?

Moreover, the federal government deficits were pouring substantially more bonds into the market than the Fed was buying. The Fed’s balance sheet grew by less than $2.5 trillion between the beginning of the large-scale asset purchase (LSAP) program and the end of 2011 while the government debt had grown nearly twice that amount.

The skeptics were initially correct. The value of equities owned by the household sector increased by less than 20 percent between 2009 and 2011. The unemployment rate continued to increase until October 2009 when it reached 10 percent and declined very little during the next two years to 8.8 percent in October 2011. Real GDP rose only 6 percent in the three years from the first quarter of 2009 to the first quarter of 2012.

But the volume of bond purchases then increased to $800 billion a year in 2013 and 2014 while the fiscal deficit fell to less than $700 billion in 2013 and less than $500 billion in 2014. At last the Fed was buying more long-term assets than the Treasury was creating.

The shift in the Fed’s net purchases and the strengthened commitment to keep short rates low caused equity prices to accelerate in 2012 and 2013. The value of equities owned by households rose 47 percent between 2011 and 2013. Overall household net worth increased nearly $10 trillion in 2013 alone.

Although Bernanke could feel vindicated by this increased value of net worth and the resulting 5.5 percent rise in real GDP between the beginning of 2013 and the beginning of 2015, he warned that the QE strategy could have adverse effects as well as the desired rise in aggregate demand. He identified those adverse effects of the extremely low interest rates as (1) excessive risk taking as investors and lenders reached for yield and (2) an unwanted acceleration of inflation.

The Adverse Effects of Quantitative Easing

The most obvious indicator of excessive risk taking is the rapid rise in share prices. The price-earnings ratio of the S&P 500 index rose from an average of 18.9 in the three years before the downturn to 25.6 now, an increase of 35 percent. The current price-earnings ratio is 63 percent higher than its historic average and higher than all but three years in the 20th century.

Robert Shiller’s cyclically adjusted price-earnings ratio, which is based on average inflation-adjusted earnings from the previous 10 years, is now 31.5 and therefore 87 percent above it historic average and at a level exceeded in the past 100 years only during the period from 1998 through 2001.

While higher share prices no doubt make investors happy and contribute to aggregate consumer spending, there is obviously now an increased risk that share prices will decline. If the price-earnings ratio declines to the historic average, the implied fall of 39 percent would reduce the value of household equities held directly and through mutual funds by $9.5 trillion.1 If every dollar of decline in wealth reduces spending by 4 cents, that would cut spending by $475 billion or more than 2 percent of GDP.

Bond prices are also out of line. With the consumer price index rising at about 2 percent and expected to continue at that pace, the yield on 10-year Treasury bonds would be expected to be about 4 percent or even a bit higher. In fact, it is now only about 2.25 percent. If the yield on a 10-year bond rises from 2.25 percent to 4 percent, the price of the bond would fall substantially. The lower price of bonds and other fixed income securities held by banks and insurance companies as well as by households could have a substantial destabilizing effect.

Commercial real estate prices have also risen in response to the lower cost of capital. Since these assets are often held in a significantly leveraged way, a decline in the asset prices would have a disproportionately large impact on the net value of the assets.

Banks and other lenders have also been responding to the very low level of interest rates by making riskier loans in order to get higher rates of return. Some of this reaching for yield involves lending to lower-quality borrowers. It also takes the form of lending with weaker conditions on the loans (i.e., covenant light loans). For both reasons, any weakness in the economy could be magnified by loan defaults that weaken the capital of the lending institutions.

Loans to consumers have also become more risky. The economic downturn that began in 2007 was increased by the widespread defaults on residential mortgages that happened as falling house prices caused loan-to-value ratios to rise well above 100 percent. After the downturn there was a general agreement that the loan-to-value ratio on new mortgages should be limited to 80 percent or less and that lenders should be required to keep a portion of the loans that they originate. Both of these conditions have been dropped in recent years as lenders seek ways to increase the yield on their portfolios.

In summary, the excessively easy monetary policy of the past decade has increased the fragility of the financial sector and therefore of the economy. Although the Fed has used its regulatory powers to strengthen the banks, it has not sought the power to limit the loan-to-value ratio on residential mortgages as other central banks have done.

More generally, Janet Yellen made it clear in a 2014 speech at the International Monetary Fund that the Federal Reserve does not believe that it should take into account the impact of its monetary policy on financial stability. She emphasized that the Fed has only two goals for monetary policy—price stability and maximum employment—and that macroprudential policies should be the responsibility of other agencies (Yellen 2014). Unfortunately it is not clear what those agencies are.

The excessively low interest rates that resulted from the Fed’s monetary policy also increase the risk that the inflation rate will rise rapidly at some point in the future. Although the inflation rate has remained surprisingly low until now, the current unemployment rate of 4.1 percent is much lower than the Fed’s own estimate of the sustainable level of unemployment. The September 2017 projections by the members of the Federal Reserve’s Open Market Committee point to a range for the unemployment rate in the long run of between 4.4 percent and 5.0 percent with a median projected value of 4.6 percent. If inflation starts to rise rapidly at some point in the future, the Fed will be forced to increase the fed funds rate more rapidly, potentially causing a downturn of asset prices and economic activity.

Why Does the Fed Continue with Excessively Easy Monetary Policy?

The Federal Reserve justifies its continued low interest policies by noting that the core personal consumption expenditure measure of inflation is still below its target of 2 percent and that the excessively high asset prices are not a responsibility of the Fed even if they contribute to financial instability.

The Fed goes further and argues that the “equilibrium real interest rate” that is consistent with stable inflation and employment has been declining over the past years. According to the “Economic Projections” released after the September 2017 FOMC meeting, the “appropriate” projected long-run federal funds rate has declined to 2.8 percent even though the projected inflation is 2.0 percent, implying an equilibrium real rate of just 0.8 percent (Federal Reserve 2017).

My judgment is that the reasons given for a decline in the equilibrium real rate are too weak to support changing the Fed’s target interest rate. I suspect that the claim of a declining equilibrium real rate is a reflection of the analysts’ preferences rather than of hard evidence.

The primary reason given for a decline of the equilibrium real interest rate is the assertion of a rise in the saving rate, lowering the interest rate at which savings can be absorbed at full employment. It is hard to see evidence to support a rising saving rate. Between 1960 and 1980 the personal saving rate varied above 10 percent of disposable income, reaching a peak of 15 percent in the second quarter of 1975. After that it has been drifting lower and is now at 3.1 percent.

During the same years, the federal budget has moved from near balance to very large deficits that absorb household saving and reduce the national saving rate.

China’s current account surplus adds to the funds available for investors worldwide. The current account surplus rose dramatically from 2000 to 2008, supporting Bernanke’s comments about a savings glut. But in the decade since then, China’s current account surplus has fallen sharply, from $420 billion in 2008 to $196 billion in 2016, back to its level in 2006. So the decline of China’s surplus means that China has not been a source of downward pressure on global interest rates.

The other major source of current account surpluses has come from the oil-producing countries. But the price of oil has fallen from over $100 a barrel between 2011 and 2014 to about $60 a barrel now. Even a very low cost producer like Saudi Arabia has gone from significant current account surpluses before 2015 to current account deficits in recent years.

Putting all these pieces together implies that there is no increase in savings either in the United States or in the international economy to cause a fall in the equilibrium rate of interest.

A simpler explanation for the low global level of interest rates is that all of the major central banks have been keeping rates low by a combination of asset purchases and open market operations. This includes not only the Fed but also the European Central Bank and the Bank of Japan.

There are several reasons why the Federal Reserve has moved so slowly to raise interest rates. Different FOMC members no doubt have different reasons for their reluctance to raise rates, but the following three reasons probably capture all of the different opinions.

First, a more rapid increase in the federal funds rate could increase the risk of a sharp decline in asset prices. An asset price correction could of course happen even without a rise in the interest rate caused by the central bank, but the Fed would be severely criticized if it is seen as raising rates less cautiously.

Second, continuing the low interest rate environment helps to increase employment. Although the current 4.1 percent unemployment rate is very low, those who want easy money to stimulate employment point to the decline in the labor force participation rate and the large number of workers who are involuntarily working less than full time as indicating that a stronger expansion supported by low interest rates could increase employment further.

There are also those FOMC members who would welcome a rise of the inflation rate, not just to the 2 percent target level but to an even higher level. Some members would justify this goal by noting that the 2 percent target is not a ceiling but the midpoint of a desirable range. Since the actual inflation has been below 2 percent for an extended period, they would justify an inflation rate that is temporarily above 2 percent as a way of demonstrating that the 2 percent target is to be interpreted as the midpoint of an acceptable range.

Some FOMC members want a higher inflation rate for a different reason. They worry that the Fed lacks the ability to cut interest rates significantly when the next downturn occurs. A higher rate of inflation would allow the Fed to raise the nominal federal funds rate by a substantial amount without raising the real federal funds rate. That would put the Fed in a position to cut rates if necessary to deal with the next downturn.

My reaction to this clever argument is that little ability to deal with the next recession would be gained by an increase in the inflation rate to 3 percent and that the public’s confidence in the Fed’s attachment to a 2 percent inflation goal would be substantially weakened if the Fed allowed the inflation rate to rise to three percent and to remain at that level. If the Fed allowed the inflation rate to rise to 5 percent and then increased the federal funds rate to 6 percent, it would have acquired the ability to deal with the next downturn. But a 5 percent inflation rate would greatly undermine the Fed’s credibility.

Conclusion

The Fed could reduce the risk of a financial correction by raising interest rates more quickly than it currently projects, reaching a federal funds rate of 4 percent by the end of 2019 or 2020 and aiming for a real federal funds rate of 2 percent. This could be achieved by increasing the pace of shrinking its balance sheet and by the way that it manages the overnight reverse repo policy.

References

Federal Reserve (2017) “Economic Projections” (September 20). Available at www.federalreserve.gov/monetarypolicy/files/fomcprojtabl20170920.pdf.

Yellen, J. (2014) Remarks on “Monetary Policy: Many Targets, Many Instruments. Where Do We Stand?” International Monetary Fund Conference on “Rethinking Macro Policy II,” Washington (April 16).

1The Flow of Funds data for the second quarter of 2017 put household sector ownership of equities at $24.9 trillion. The equity value of noncorporate businesses is estimated in the Flow of Funds data to be an additional $11.8 trillion.

Martin Feldstein is the George F. Baker Professor of Economics at Harvard University. These remarks were prepared for presentation at the Cato Institute’s 35th Annual Monetary Conference on November 16, 2017.