The yield curve

Summary

– The yield curve shows the interest rate that issuers have to pay for a given maturity at a given time.

– The short end of the curve is, in theory, more closely linked to changes in central bank policy rates. The long end depends on the credit risk/inflation risk associated with the issuer.

– However, the financial crisis and central bank policy have significantly altered the shape and structure of the yield curve.

– Changes in the shape of this curve are a good leading indicator of economic activity. The last eight US recessionary cycles were all preceded by a curve inversion.

The term structure of the yield curve

The yield curve of a financial bond is the function that, on a given date and for each maturity (its maturity date), indicates the associated interest rate level. It is represented on a graph with the x-axis corresponding to the maturity of the bond and the y-axis to the interest rate. It meets two demands on the financial markets since, on the one hand, it aggregates all the interest rates that an issuer will have to pay and, on the other hand, it informs investors of the yields on a security according to its maturity. The structure and evolution of the curve are therefore crucial information for the efficiency of bond markets.

We can separate the short-term maturities of the curve (from 3 months to 2 years) from the long-term maturities (beyond 2 years and up to 50 years). These are referred to as the short and long ends of the curve. The slope of the yield curve, calculated by the spread between short-term and long-term rates (10-year rate – 3-month rate or 2-year rate), is a frequently used indicator to determine the shape of the curve. In theory, this slope is supposed to be positive, since long-term interest rates are higher than short-term rates. According to J. R. Hicks (theory of expectations of the term structure of interest rates, 1939), risk-averse investors have a preference for liquidity, i.e., for securities that are traded in large quantities on the markets. The longer the maturity of a bond, the higher the return the issuer must offer investors to encourage them to hold less liquid assets. The risk premium, known as the term premium, increases over time but is marginally decreasing, which explains the theoretical upward-sloping and concave shape of the yield curve.

Several factors, both endogenous and exogenous to the issuer, alter the appearance of the yield curve. Changes in the central bank’s key interest rates will govern movements in the short end of the curve.

In the long term, two factors influence the curve. On the one hand, there is the risk premium linked to uncertainties about the issuer’s solvency and default. This credit risk is assessed by rating agencies using their rating scales. The higher a country’s rating, the lower its probability of default and therefore the lower the long-term interest rate required. As can be seen in the chart below, Eurozone countries with AAA ratings[i]by rating agencies benefit from lower rates across the entire curve than lower-rated sovereign issuers. Investors apply a low or even zero risk premium to these issuers, which are considered safe.

On the other hand, inflation risk determines the level of interest rates demanded by investors, as it erodes the yield on securities held. Investors who buy long-term bonds will therefore demand a higher yield if their inflation expectations are also higher.

However, with inflationary pressures currently low, credit risk alone explains the differences in the yield curve between countries.

Source: European Central Bank website

Contemporary determinants of the yield curve structure

Since the early 2000s, there has been a fall in long-term interest rates following strong growth in demand for these long-term securities, causing an imbalance and excess demand in the securities market.

International demand for long-term securities from central banks has a significant impact on the trajectory of long-term interest rates. Central banks, seeking to limit the volatility of their exchange rates, build up foreign currency reserves. Data from the International Monetary Fund (IMF) show very strong growth in foreign exchange reserves among exporting and oil-producing countries. In China, for example, dollar reserves rose from $286 billion to $3.2 trillion between 2002 and 2012. These huge amounts of liquidity are invested in long-term securities such as US sovereign bonds.[ii](in the case of dollar reserves) and highly rated European bonds such as French and German bonds (in the case of euro reserves).

In addition to this demand from political authorities, there is also demand for long-term savings linked to demographic aging. Faced with an increase in the number of retirees in developed countries, the latter are seeking to build up long-term savings reserves for their retirement in order to smooth their intertemporal income level. In fact, institutional investors (insurance companies, pension funds, etc.) have seen their assets under management increase over the last two decades, leading to massive purchases of long-term securities in order to meet customer needs. The preferred habitat theory (Modigliani & Sutch) therefore applies to institutional investors, since the long maturity of their resources encourages them to invest over long-term horizons.

Both monetary authorities and institutional investors therefore adjust their investment policy according to their asset/liability management. Given their long-term investment horizon, they naturally gravitate towards safe, long-term bonds.

These movements, which favor a flattening of the yield curve, can sometimes be so significant that they lead to a curve inversion, with short-term rates becoming higher than long-term rates. This is particularly the case when the central bank decides to curb inflation by raising its key rates, while the increase in structural demand, as mentioned above, pushes long-term rates down. This restrictive monetary policy leads to higher borrowing costs from banks, a fall in the money supply and , ultimately, a slowdown in economic activity in the medium term. The negative slope of the yield curve is therefore seen as a leading indicator of economic conditions.

Yield curve inversion: how predictive is it?

The Central Bank’s policy is a counter-cyclical policy that aims to smooth growth over time by limiting periods of overheating and stimulating activity during recessions.

In our analysis of economic cycles, we will use Juglar’s approach and his work on short economic cycles. According to Juglar, cycles last between 6 and 11 years and depend on investment, and therefore indirectly on interest rates. An inverted yield curve is therefore synonymous with overheating, as short-term rates rise to limit inflationary risks and the creation of bubbles, while low long-term rates stimulate overinvestment. This phase precedes a slowdown in economic activity.

The most striking example of a yield curve inversion is that of the United States in 2004. Alan Greenspan, then chairman of the Federal Reserve, called this phenomenon the  » conundrum. » In order to limit the growth of the real estate bubble, the Fed decided to raise its key rates several times between 2004 and 2007 (from 1% to 5.25%). In theory, a rise in short-term rates should be accompanied by an upward shift in long-term rates. However, this countercyclical policy pursued by the Fed was hampered by certain exogenous factors that limited the transmission of monetary policy. Strong growth in demand for long-term financial securities since the early 2000s (due to an aging population and an increase in foreign exchange reserves) pushed long-term rates down. The US bond yield curve thus inverted in August 2006, a year before the first signs of a slowdown in activity and a contraction in the US real estate market.

The yield curve therefore appears to be a good leading indicator of economic conditions. Indeed, a yield curve inversion tends to be accompanied, within four to six quarters, by a recession in both the US and Europe. As can be seen in the charts below, when the slope of the curve enters negative territory, this is accompanied by a recession. In fact, since 1960, there have been nine inversions of the US Treasury yield curve, eight of which were followed by a recession.

Source: FED and ECB websites

However, while the inversion of the yield curve has historically proven to be a reliable indicator for predicting recessions, it does not provide any quantitative forecasts of recessions.

It should be noted that the Fed publishes a monthly indicator of the probability of recession in the US market (over a 12-month horizon) based on the spread between short-term and long-term interest rates. This model indicated a 40% probability of recession in 2008. The same was true for 2001. Today, this model indicates a probability of recession for 2014 of only 4%. This is low enough to suggest that the US economy is in a phase of expansion.

Finally, even though statistically the vast majority of recessions have been preceded by a curve inversion, contemporary monetary policies aimed at ending the economic and financial crisis have had a direct impact on the structure and shape of the yield curve, thus calling into question its use as a leading indicator. Both in the United States (Quantitative Easing, Operation Twist) and in Europe (Outright Monetary Transactions[iii]), central banks have taken direct action to deliberately modify and flatten the yield curve. To cite just one example, Operation Twist, decided by the Fed in September 2011, aimed to sell short-term securities and thus bring about a mechanical rise in short-term rates (less than 3 years) while buying long-term securities (over 6 years). The dual challenge of such a policy (€667 billion) is therefore to finance the US government and economic activity by lowering long-term rates while leaving the money supply unchanged (no inflationary risk).

In fact, recent adjustments to the yield curve may limit its predictive capacity in the coming years.

Notes:

[1] AAA is the highest rating (see article on sovereign ratings and their implications), indicating a zero probability of default.

[2] The flattening of the yield curve: causes and implications in terms of economic policy, M. Collin, June 2007