The impact of unconventional monetary policy wanes over time.
Without ongoing policy efforts, the average duration of the Fed’s securities portfolio is always in decline, putting upward pressure on rates.
Due to continuous easing since 2009, we have yet to see the effect of this decaying impact in the form of increasing Treasury yields.
Many things in life get better with age (wine, cheese, blue jeans), but many others do not (athletic ability, sushi, hangovers). Unfortunately for central banks, the effects of unconventional monetary policy probably fall in the latter category. Unlike traditional monetary policy—in which the central bank only sets short-term interest rates—the impact of unconventional policies likely decays over time. This means that it is not enough for the Federal Reserve to keep its current policies in place—it actually has to take additional action to maintain the same impact on interest rates and the economy.
There are two main reasons for this. First, as time passes, the point at which the Fed will start raising the funds rate gets closer and closer. All else being equal this will cause longer-term interest rates to drift higher, because longer-term rates imbed expectations about the future path for short-term rates. Chairman Bernanke actually made this point explicitly in his recent speech on bond yields: “long-term rates would be expected to rise gradually … this rise would occur as the market’s view of the expected date at which the Federal Reserve will begin the removal of policy accommodation draws nearer.”
Second, without ongoing policy efforts, the average duration of the Fed’s securities portfolio is always in decline. The theory and evidence on quantitative easing (QE) suggests that the policy works primarily by removing duration risk from private sector hands which in turn reduces interest rates. In effect, the aging of the Fed’s balance sheet slowly adds this back duration risk to the market, putting upward pressure on rates. These changes are quantitatively meaningful. For example, the Fed’s securities portfolio is expected to reach about $3.75 trillion by the end of this year. At that level, a one year decline in the portfolio’s average duration would reduce the Fed’s total duration risk by about $415bn 10-year equivalents (10-year equivalents measure the amount of current 10-year Treasuries that would total to the same level of overall duration risk). Empirical work suggests that reducing the duration of the Fed’s balance sheet by this amount would push up long-term interest by 15-20 basis points (bp).
Using market data we can roughly estimate the total time decay effect. As shown in the chart below, in recent years there has been a very tight relationship between the level of 10-year yields and the length of time markets expect the funds rate to remain at zero. (We estimate the latter from Fed funds futures contracts). Longer-term rates decline when the first rate hike appears further away, and longer-term rates rise when the first rate hike appears closer. The estimated slope of this relationship is -0.05—meaning that as long as the calendar date of the first rate hike remains unchanged, 10-year Treasury yields should creep higher by about 5bp per month.
An obvious question would be: if the impact of the unconventional easing decays over time, why have Treasury yields mostly declined since 2009? The reason is that the Fed never really stopped easing. After QE1 came QE2, two phases of operation twist, and now open-ended QE. The Fed’s verbal easing also turned more aggressive (“some time”, “extended period”, “mid-2013”, “late 2014”, “mid-2015”). We will only observe the aging effect of unconventional policies when the easing process finally ends.
The market implications of the time decay effect are somewhat nuanced. On the one hand, the impact on rates means that the carry returns to Treasuries look even less attractive. In fact, the estimated decay effect of 5bp per month is greater than the monthly carry return for a 10-year Treasury (roughly 3bp per month per unit of duration). On the other hand, the decay effect increases the odds of more monetary easing in the future. New easing steps by the Fed look unlikely at the moment, but the decay effect could conceivably be used as a justification for more action at some point.