Monetary Policy and Long-Term Real Rates (Digest Summary)

The authors assess why news on short-term nominal interest rates has a strong effect on longer-term real forward rates. To study these real interest rates, they investigate US TIPS (Treasury Inflation-Protected Securities) and similar UK data. The authors check whether changes in future real rates or changes in term premiums explain this relationship and conclude that it is the latter. By using a supply and demand model, they conclude that because of yield-oriented investors, short-term nominal rates can affect longer-term real rates. In particular, broker/dealers tend to act as arbitrageurs by adjusting their portfolio mix to changes in short-term interest rates.

Previous research in this area has produced varied results. Some conclude that short-term monetary shocks affect distant nominal rates, and others conclude that monetary shocks can affect inflation expectations but not real rates. Furthermore, other researchers have examined the interaction between short-term rates and real risk premiums, which is similar to the approach that the authors take.

New Keynesian macro models assume that prices are sticky in the long term, and hence, monetary policy has no impact on real prices in the longer term. The authors note that in practice there seems to be a relationship between short-term nominal rates and longer-term real rates. But because of limited data, which are only available from 1999 onward, the results may be flawed. The authors tackle this argument by showing that after 1987, US inflation expectations have been fairly constant, resulting in a high correlation between real and nominal interest rates. Hence, the dataset can be extended by using nominal rate data instead.

Another explanation could be that changes in short-term monetary policy may reflect other macro news or some of the Fed’s private information (also referred to as the reverse-causality hypothesis). To assess this argument, the results on Federal Open Market Committee (FOMC) announcement days are compared with nonannouncement days. The authors conclude that there is not sufficient evidence to support the reverse-causality hypothesis. Instead, they conclude that yield-oriented investors cause the pattern to arise. Particularly, broker/dealers tend to act as arbitrageurs who shift their portfolio depending on changes in short-term nominal rates.

This article is particularly interesting for professionals working in the bond industry and whose work is closely related to changes in short-term interest rates. It may also be an interesting article for professionals working in the pension and insurance industries who need to be aware of the impact of nominal policies on longer-term real rates.

The authors use the FOMC announcement dates from 1999 to February 2012 and exclude some announcement dates related to quantitative easing. The change in two-year Treasury yields on FOMC dates is considered the independent variable, and the change in real rates is the dependent variable. A two-day window is used to assess the impact of the change on short-term rates because it may take longer-term bonds more time to adjust to the news. Svensson’s (NBER 1994) six-parameter model is used to generate US forward curves, and spline-based techniques are used to create UK forward curves.

Using empirical data, the authors show that there is a relationship between short-term nominal forwards and longer-term real forwards. In addition, they show that using similar short-term instruments (e.g., one-year Treasury) results in a similar relationship. They then check whether this relationship is caused by the small TIPS’s liquidity premium and exchange the TIPS data for inflation swap yields. But the results are not statistically different.

Studying the change in old and current on-the-run Treasuries, they do not find a relationship between liquidity and monetary updates. Furthermore, they assess their hypothesis on both the US and UK markets and conclude that although the effects are smaller in the United Kingdom, the relationship is still similar. They then determine whether changes in expected future real rates or term premiums explain this causality. By applying regression analysis, the authors show for both 1987–2012 and 1999–2012 that term premiums explain the relationship.

Lastly, they assess why monetary policy influences real-term premiums. A consumption-based asset pricing model is insufficient because it is difficult to argue that changes in either the volatility of bond returns or the correlation between real bond returns and a stochastic discount factor seem probable. Alternatively, the hypothesis that markets are segmented and some investors are yield-oriented is more likely. Finally, there does not seem to be a statistical difference between the yield-oriented behavior of public and private banks.

Many financial professionals consider Fed announcements as a key indicator in their analysis and recommendation process. The effect of announcements on real rates, however, has not been straightforward because many factors come into play, such as inflation and the liquidity of financial instruments. The authors use an extensive set of empirical data to support their analysis and assess several hypotheses before stating their conclusions. They tend to be rather technical and use a lot of mathematics, which might make the article unsuitable for every financial professional. But even discounting the technical parts of the article, it provides very useful information for financial professionals that are analyzing bond markets.