Breaking Down Bond Yields

The yield on a bond can be broken down into three components (four if the bond is callable, but put that aside for the moment).

The first component is real interest rates. That is basically your opportunity cost of other investments. The more profitable other investment opportunities are, the higher the yield on a bond needs to be to entice capital into the bond market.

The second is inflation. If you are going to lock up your money for the long-term, you need compensation for both expected inflation and the risk that inflation is higher than one expects.

The final piece is credit risk. You need to be compensated for the chance that the borrower doesn't pay off bondholders in full.

These factors are as fundamental to bond yields as photosynthesis is to plants. No amount of Fed policy can change these basic elements, but they can influence them to be sure. The Fed's quantitative easing program, for instance, is reducing the amount of total bonds available for purchase. That alters how many bonds are out there in search of capital, which should lower the real interest rate portion.

These three factors also have correlations to the broader economy that must hold over any reasonable period of time. If economic growth accelerates, it should cause real interest rates to rise because it increases the competition for capital. At the same time, credit risk should decrease as economic growth improves.

We can see these correlations holding during past bond bear markets (which I've arbitrarily defined as periods where 10-year Treasury rates rose 100 basis points or more). The chart below shows movement in Inflation breakevens (red) during these bear markets. The breakeven is simply the yield gap between Treasuries and TIPS. Note that TIPS didn't exist before 1997. We can think of this as the pure measure of the "inflation" component mentioned above. We also look at credit spreads for investment-grade (green) and high-yield (purple). We can think of these as the "credit" component. In other words, looking at periods where interest rates in general rose at least 100 bps, what happened to the other factors that make up rates?

Source: Bloomberg

We can see that virtually every period, when interest rates in general rise, inflation is expected to be higher and credit risk is expected to be lower. Every time, that is, except the past eight weeks. In all other periods, the fundamental correlations described above held.

To be fair, it is entirely possible that in the period between September 2012 and April 2013, credit spreads moved beyond what was justified by the fundamentals, perhaps encouraged by Fed policy. I'm skeptical, but I will stipulate that it is plausible. Hence, we could see a short-term move where trading in inflation-protected and credit-risky bonds move the opposite way that we'd typically expect solely due to a mispricing.

Even so, such a move is limited. Interest rates can't keep rising without higher inflation or better economic growth. Credit spreads can't keep rising if the risk of default is falling, i.e., the economy is growing. It has never happened before, and it won't happen now.