Bonds , Portfolio


There is a severe lack of long term perspective by most financial market participants these days. Interest rates, and in particular the potential for a dramatic rise in future rates, is just the most recent example. Most analyses of rising rates don’t go back far enough to be really useful. Going back to the last rate increase by the FED in 2004 to 2007 is not good enough. Even going back to the 1970s is not good enough. For a better perspective you need to go back further. The best analysis I’ve seen is here . In this post I’ll take a look even further back in history at the great rise in US interest rates from 1941 to 1981 and look at the role of short term bonds vs long term bonds in portfolios. I think this period is a good analog to study the current environment.

First, lets frame what we’re discussing. The US experienced a dramatic rise in interest rates from the end of 1940 through 1981. The chart below is a graph of the US 10 yr note from 1927 to 2012.

US interest rates 1927 to 2012

A few items to point out. After the Great Depression it took 11 years for the US 10 yr note to bottom out. In late 1940 the US 10 year bottomed at 2.01%. It took a further 16 years, until 1956, for the 10 yr note to reach the its level prior to the Great Depression. That’s 27 years for the 10 yr note to reach the same level it was in 1929. Keep that in mind next time you hear about rapidly rising rates. Rates can stay lower a lot longer than you think as shown by history. OK, on to the focus area. Once rates bottomed in 1940 they began a 40 year rise that didn’t end until 1981 when the US 10 yr hit 13.7%! Now lets take a look at what that meant for bond returns during that period.

For this analysis I’ll compare short term bonds, using US Tbills, versus US long term bonds, using the US 10 yr note. Obviously, such a period of rising rates was bad for all bonds. 30 year period bond returns from 1940 to 1951 still stand as the worst long term periods for bonds. 30 year period returns during this time were between 1.9% and 2.7% for the 10 year bond, and between 2.34% to 4.37% for US Tbills. In general short term bonds out performed long term bonds. Coupled with rising inflation bond returns during this period were mostly negative on a real basis as well. So, why hold bonds at all during a rising rate environment? Let me table that for a later paragraph. One thing about this 40 year period is that it hides a lot of subtleties. We need to dive deeper for a better view. The chart below shows rolling 3 year returns for US Tbills and the US 10 year note during this period.

Rolling 3 yr returns bonds vs bills dec 2014

The point to this chart is that during the initial rise in rates long term bonds performed better than short term bonds. It wasn’t until rates rose to higher levels and then proceeded even higher that short term bonds really shone, while long term bonds were hurt. Bond returns are a combination of the level of rates plus the change in rates. Look at the last few years on the chart in particular. From 1976 to 1981 rates doubled! Now let’s look at it on a total return basis.

total return bonds vs bills 2014

Same basic picture except it shows the late cycle difference between bonds and bills even more clearly. During the initial more subtle long term increase in rates long term bonds were better than short term bonds. But as the rate increases took hold and accelerated short term bonds more than caught up. Up through 1976 the total return form bonds and bills was about the same. In the last 5 years bonds returns lagged the return of bills by approximately 30%. I think the important point here is that you have plenty of time to react to rising rates and during the initial rise long term bonds are better than short term bonds. Or better yet just use a bond momentum model as I discussed here . A simple annual bond momentum model that chose between Tbills and US 10yr bonds would have outperformed either with lower drawdowns.

But why hold bonds at all? Given the level of current rates future bond returns (short or long) are unlikely to be very good especially on a real basis. The answer is we hold bonds mainly as portfoflio diversifiers. Over time bonds act as the best diversification for stock market risk. During the period from 1941 to 1981 there were only 2 years where both stocks and 10 yr bonds had negative returns and there were zero years when stocks and Tbills both had negative returns. This diversification coupled with rebalancing enhances portfolio returns. The other reason to hold bonds is to minimize negative returns and drawdowns. Again, even during this unprecedented rise in rates bond reduced negative returns and drawdowns vs 100% stock portfolios by about half. This is especially important for retirees who are trying to make their portfolios last and maximize their safe withdrawal rates. As usual, Vanguard has a great piece on the role of bonds in portfolios.

In short, bonds are a key piece of any diversified portfolio, even during a long term rise in interest rates. And as history has shown rates can stay lower a lot longer than most think. Also, the choice between short term and long term bonds during rising rates is not as obvious as it may seem at first glance. During the initial rise in rates long term bonds can perform better than short term bonds.

Tagged , ,