The 1970’s are not a Good Proxy for a Bond Bear Market

A reader kindly forwarded me this interview from Tony Robbins in which he defends his “All Weather” portfolio against some of the criticism laid out against it.  Several people, myself included, have criticized the recommendation of this portfolio given that the portfolio, which is bond heavy, is unlikely to perform well in a low and rising interest rate environment.  In countering this claim, Robbins cites the 70’s as proof that the All Weather portfolio can perform well in a rising interest rate environment.   That’s true, but the 1970’s are not a good proxy for a rising interest rate environment relative to today.

The reason for this is simple – in the 1970’s the Fed Funds Rate averaged 7.15% and actually started the decade at 9%.  There was never a period where the overnight interest rate was 0% (as it is today) or got even close to that level.  So the 1970’s were a HIGH AND RISING interest rate environment and not a LOW AND RISING interest rate environment like we might experience going forward.  So Tony’s not comparing apples to apples there when he cites the 1970s.  And let me be clear – I don’t mean to pick on Robbins because I see this all the time from other finance professionals who cite the 70’s as a good proxy for a rising interest rate environment relative to today.  It’s just not a good example.

Importantly, there is a historical period that gives us some good insight here.  In the 1940’s long-term interest rates fell just below 2%, remained low for a while and steadily increased until the early 80’s when they peaked.  The 10 year T-Note is currently at 2.25% so we’re looking at a pretty close starting point.  Therefore, this 40 year period from 1940-1980 actually provides us with a much better historical understanding of how bonds might perform in a LOW AND RISING interest rate environment.

Now, the Robbins All Weather portfolio advocated a 40% position in long-term govt bonds, 15% position in intermediate govt bonds, 30% position in stocks, 7.5% position in gold and a 7.5% position in commodities.  To better understand how this portfolio might respond in a LOW AND RISING interest rate environment let’s take the 40 year period from 1940-1980 when US T-Bonds troughed and peaked and then also apply longer-term historical returns to the other asset classes.  This will give us a very fair understanding of how this portfolio might perform in a LOW AND RISING interest rate environment.

Not surprisingly, when you start from a very low interest rate the bonds tend to generate a low nominal return.  Over this 40 year period T-Bonds generated an average annual return of just 2.65%.  A 40/15 bond portfolio of long and intermediate durations generated a 2.5% return over this period.  So there’s your bond return in the period where interest rates were LOW AND RISING – a whopping 2.5%.  This is less than half of the average return of T-Bonds in the 1970s (5.58%).  

Now let’s apply the other components.  Stocks have averaged about a 11.75% annual nominal return, gold has generated a 11% return since 1971 (when it was unpegged) and commodities have generated returns in-line with the rate of inflation (roughly 3.5%).  When we allocate this portfolio according to the Robbins weightings you get a 5.95% average annual return when applying the LOW AND RISING rate environment.  That’s not terrible, but it’s nowhere close to the 10% figure that Robbins cites in the book.  More importantly, you took a decent amount of risk to generate that return.  The average standard deviation over this period was 14.55 for the Robbins portfolio.  Historically, the S&P 500 generates an annual standard deviation of about 20.  So, in a LOW AND RISING interest rate environment the Robbins All Weather portfolio takes 72% of the risk of the S&P 500 and generates about half of the return.  Robbins focuses on asymmetric returns in the book, but failed to put together a realistic analysis of the types of risk adjusted returns this portfolio might generate in a LOW AND RISING interest rate environment.  So the portfolio he provides for readers isn’t asymmetric and is likely to generate a much lower nominal return than he cites.

This is an important exercise to think about going forward.  Not only will bonds fail to generate the necessary nominal return to substantially contribute to a balanced portfolio, but they will also fail to significantly boost the risk adjusted returns of portfolios to the same degree that they did during the period from 1980-2014 when rates were a one way bet on a historic bull market.