Estimating and interpreting interest rate expectations.

• Author: Kloster, Arne

Expectations about future interest rates and inflation influence economic developments. For example, market expectations of higher inflation may themselves result in higher inflation, for instance through higher pay increases. Households' choice between consumption and saving is influenced by their expectations concerning future interest rates. A high level of short-term interest rates will probably have less of a contractionary effect on economic activity if the market believes this to be a transitory phenomenon than if it is expected to persist. Inflation expectations also reflect whether market participants are confident that economic policy will result in low inflation over time. One important source of information about these expectations is the market's pricing of interest-bearing securities with different maturities. This article describes the method used by Norges Bank for estimating interest rate expectations, and discusses how these estimates may be interpreted. In addition, the importance of various premia will be considered, and some alternative approaches for estimating interest rate expectations will be discussed.

Theories about the term structure of interest rates

There are several theories as to the determinants of the relationship between interest rates with varying maturities, often referred to as the term structure of interest rates. The expectations theory is probably the explanation most widely held. This theory is based on the assumption that market participants are risk-neutral and maximise the expected return without having specific preferences as to the maturity of their loans and investments. The interest rate on a long-term investment is therefore determined entirely by expected developments in short-term rates during the same period. If this were not the case, investors could achieve an expected excess return by raising long-term loans and reinvesting the funds in revolving short-term issues (or vice versa). Market participants' pursuit of this type of excess return ensures that the interest rate on long-term securities will always be an average of expected short-term rates. If the term structure of interest rates is determined by market expectations, these expectations may be inferred from the shape of the yield curve.

The liquidity preference theory is based on the assumption that market participants are averse to risk. The price of a bond with a long residual maturity will be more sensitive to interest rate changes than the price of a bond with a shorter residual maturity. The holding-period return on a bond is thus more uncertain the longer the residual maturity. Therefore, other things being equal, risk-averse investors will prefer to invest in short-term issues. To induce market participants to invest in long-term issues, they must be compensated in the form of a higher yield than the level implied by the expectations theory. This compensation, referred to as the term premium,(2) will increase with the term to maturity. Like the expectations theory, the liquidity preference theory implies that long rates are an average of expected short rates, but with the addition of a premium that depends on the term to maturity. Another complicating factor is that the term premium may not be constant over time.

The hedging pressure or preferred habitat theory assumes that market participants are highly averse to risk and that they wish to match the maturity of their investments to the maturity of their debt. Consequently, the market is split up into independent segments. Interest rates on securities of varying maturities are determined in-dependently by supply and demand in the various market segments. According to this theory, it is meaningless to calculate market participants' interest rate expectations on the basis of the term structure of interest rates.

Implied forward rates and their interpretation

Implied forward interest rates can be calculated on the basis of observed interest rates on issues of varying maturities, ie the yield curve. Implied forward rates are interest rates between two dates in the future derived from the yield curve. If the expectations theory or the liquidity preference theory holds true, implied forward rates will reflect market expectations about interest rates, possibly adjusted for term premia. For example, the expected three-month rate three months ahead may be calculated on the basis of observations of current three-month and six-month rates. The slope of the yield curve at maturities between three and six months indicates whether the three-month rate is expected to rise or to fall. Whereas the yield curve shows the average expected interest rate in the period up to various dates, the implied forward rate expresses the expected interest rate on those dates. The reason for calculating forward rates is thus their practical interpretation rather than because they contain other information than that contained in the yield curve.

There are a number of empirical studies of the expectations theory (for an overview see, for example, Browne and Manasse, 1990). Most studies reject the theory. This result is often attributed to the existence of a term premium which varies over time. In its pure form, the expectations theory poses the variation of long rates on a one-to-one basis with expected short-term rates rather than merely the existence of positive covariation between long-term and short-term rates. A number of studies find positive covariation, but must reject the theory even so. Rejection of the theory does not necessarily mean that the term structure of interest rates has no interest for monetary policy purposes. The variation in expected short rates may nevertheless explain a substantial part of the variation in long rates.

Interest rate expectations for the next few years will depend largely on the economic outlook and the market's perception of how the central bank sets its key rates. Expectations of growing pressures in the economy may, for example, generate expectations of higher interest rates -- both nominal and real -- if market participants are confident that the central bank will take steps to counter higher inflation. If such confidence is lacking, market participants may still expect higher nominal rates as a result of higher inflation expectations. Either way, this will be reflected in an upward sloping yield curve if the expectations theory holds true.

In order to be able to disentangle inflation expectations from nominal implied forward rates, it is necessary to assume that the expected nominal interest rate is approximately equal to the sum of the expected real rate and expected inflation (for further information see, for example, Froyland, 1997). In the short term, interest rate expectations will be influenced by the cyclical outlook. However, in the long term, say ten years ahead, it seems unlikely that market participants have specific expectations about the cyclical situation. On this horizon it may seem reasonable to interpret the implied forward rate as the sum of the expected equilibrium values of the real interest rate and inflation, plus any risk and/or term premia.

Mishkin (1990) analysed the term structure of interest rates in the US using a model which assumed that the nominal interest rate is determined by the expected real rate and expected inflation, and that market participants have rational expectations about inflation developments. Based on observations of interest rates between 1964 and 1986, he found that the term structure for maturities of up to six months did not contain information about future inflation developments, but did provide information about the term structure of real interest rates. In the area from nine to twelve months, however, the nominal term structure started to provide information about future inflation, as well as, to a lesser extent, information about the term structure of real interest rates.

Schich (1999) used Mishkin's model to study the relationship between the term structure of interest rates and expected future inflation in the US, Germany, Canada, the UK, France, Italy and Japan. Schich found a significant relationship for the first four countries. The most informative maturity segments were further out on the yield curve than in Mishkin's study. The relationship, however, varied both across countries and over time. The variation over time was primarily attributed to shifts in the monetary policy regimes. Intuitively, it seems reasonable to assume that the information content of implied forward rates is related to monetary policy and financial markets' confidence in this policy. If monetary policy is oriented towards low inflation and has a high degree of credibility, the term structure is likely to contain little information about future inflation other than that implied by the central bank's inflation target. In this case, implied forward rates may reflect expected developments in real interest rates. Changes in the monetary policy objective may lead to changes in the information content of implied forward rates, while changes in the structure of financial markets and the degree of regulation may also change the information provided by implied forward rates over time.

If the relationship between the term structure and future inflation is not stable over time, nominal implied forward rates will not be a reliable indicator of future inflation developments. Some countries have established markets for real rate bonds where the interest rate is linked to a price index. This makes it possible to assess expected inflation by comparing yields on bonds with nominal and real returns. By comparing implied forward rates based on the two types of bond, it is possible to estimate developments in inflation expectations. No market for index-linked bonds exists in Norway. However, in a world with free capital mobility, it may be assumed that the real interest rate in equilibrium must over...