Guest Comment: Caveat emptor

6 min read

In January, the relationship in which total returns have exceeded interest rate duration hedged returns looked as if it was at risk of breaking down. Total returns for credit also seemed as if they could end the year in negative territory for the first time since the start of the century.

Fast forward six months and a quick review of credit market performance numbers for the first half of the year confirms the change in dynamic. Both the US dollar-denominated and GBP-denominated broad investment-grade credit indices posted negative total returns of –3.6% and –1.1% respectively over the first six months of 2013.

The euro-denominated broad investment-grade index just about hung on to a positive return at 0.1%. Excess swap returns on the other hand remain firmly positive across all three major indices, ranging from 1.3% to 2.2%.

The sell-off in rates has investors watching their government debt portfolios anxiously. Until now, German government debt has held in better than the Treasury and Gilts markets because investors have taken notice of the underlying fundamental and policy differences between the US and the eurozone.

Decoupling?

Following the Fed’s taper talk and the ECB’s soft forward guidance, should European fixed-income investors worry about a similar sell-off in German government bond markets or are European and American markets decoupling?

Correlation and spread action are the two main variables to check when analysing the probability of a decoupling between the US and Europe. The two graphs below plot the correlation and spread between the 10-year and two-year German and US benchmarks over the past 20 years.

correlation rates

Source: ECM Asset Management

Correlation and spread between 10 y US and German government benchmarks

rates correlation treasuries

Source: ECM Asset Management

Correlation and spread between 2y US and German government benchmarks

Over the past 20 years, there have been few episodes where yields have decoupled for long periods of time. The first period was in 1994–1995, when the US economy was emerging from a big recession following the savings and loan crisis of the late 1980s.

Fed chairman Alan Greenspan surprised markets by beginning to tighten monetary policy and Treasuries plunged as interest rates rose throughout the year.

Meanwhile, the Bundesbank was in the middle of a multi-year process of cutting rates from very high levels. This period of decoupling can be observed on both graphs via the very low or negative correlations during that time period.

The second period happened in 2004–2005 and was due to the weakening in economic data in the euro area, with industrial production in particular slowing down substantially from June 2004. The Fed’s first rate hike came in June 2004 but euro rates rallied until July 2005 even though the ECB was on hold until December 2005, when it hiked 25bp. This period can be seen on Figure 2 as the dip in correlation and the firmly negative spread between the two securities.

How should an investor interpret the current episode of diverging interest rate movements?

Now that the ECB has embarked on a new path by dropping its 14-year old policy to “never pre-commit”, the probability of decoupling has certainly increased in the short end of the curve, where central banks’ influence is largest.

Even looking at the 10-year forward Bund/Treasury spread on Figure 1, we observe that the forward market is pricing in a period of a couple of years of further modest widening in the spread before a mild convergence back to current levels around 2016.

Another insight can be gained from looking at the correlation between Japanese and US rates. Figure 3 shows the JGB market was relatively unaffected by US yield movements between 1994 and 2004 with frequent periods of negative correlations.

Although it is understood that Japan is not as closely integrated with Western markets as the US and the EU are with each other, perhaps one could conclude that the divergence or convergence of macroeconomic conditions between the US and Europe will be the ultimate driver of decoupling.

correlation rates

Figure 3: Regimes in long term correlations between 10 year government benchmark securities
Source: Citigroup

It is too early to say whether the ECB’s and the BoE’s attempts to decouple European yields from the Fed’s tightening grip will be successful.

Recent historical evidence suggests that decoupling will be hard to achieve but even if the current attempts are successful in the short term, how long until mean reversion, a concept so dear to financial markets, sets in again?

Running a European fixed-income portfolio based on the view that rates will not rise also seems quite risky from the point of view of the tiny break even spread widening buffer.

Unless exposure to duration is required for asset-liability matching purposes, a safer risk adjusted alternative for fixed-income investors is to reduce exposure to interest rate fluctuations.

In a world where European households, banks and governments are trying to delever at the same, investors can get exposure to that deleveraging trend by investing in short duration or duration hedged credit instruments.

The benefits of doing so are likely to be substantial over the coming years, just as they have been during the first six months of 2013.

Correlation and spread between 10 y US and German government benchmarks
Correlation and spread between 2y US and German government benchmarks
Regimes in long term correlations between 10 year government benchmark securities