There was an article in yesterdays’ Wall Street journal that questioned the idea that stocks are a better long term investment than bonds. Moreover, it pointed out the importance of other types of assets such as commodities. This is part of a broader debate on asset allocation, pushing away from stocks. Many of the ideas in this debate are consistent with our methodology, but we disagree that past performance will be a good predictor for the future.

It is true that in the last 2,5,10 years bond returns have been greater than stocks’. In particular if we compare the S&P500 vs. the (ex) Lehman Aggregate (now it is called Barclays Capital US Aggregate Bond Index, we shall call it BC Agg). If we compare the last 5 years, the S&P annualized returns were 0.1% while the BC Agg annualized returns were 5%.

The article cites another example that illustrates why bonds are better performers than stocks: if during the last 40 years an investor would have bought 20 year Treasuries and re-invested the coupons and amortizations, he would have outperformed the S&P500.

There are at least three important considerations that should be taken into account when reading through this debate:

  1. In the last decades we have witnessed the conquest of American inflation. This means that inflation has been reduced significantly from previous decades like the 80s. Falling inflation produces a very clear effect, a bias towards bonds as rates come off to respect the falling inflation expectations. Moreover, we believe that inflation levels in the next 5 years will reach 5-7% reversing the previous trend, therefore reducing the bias towards bonds.
  2. Both 2008 and the collapse of the Nasdaq in 2001/2002 generated a big bias against stocks. For example, a quick look at the past using different periods will provide very different returns. If we look at the last 40 years finishing on February 2009, the S&P500 yielded 5.3% whereas if we look at the last 40 years finishing on Feb 2008, the S&P500 yielded 7%. The last 40 years have been detrimental for stocks, given that there have been big drops in 1970,1974,1987, 2001-2002, and 2008. The decades between 1930 and 1970 were more stable for stocks.
  3. The developed economies tend to have secular growth, despite sporadic crisis. Stocks tend to be better performers than bonds, given the correlation between the structure of payments and the economic growth. Therefore, during long periods of time stocks will outperform bonds. The key to performance is to be able to determine when to increase or reduce the asset allocation given the global economic cycle. This is not easy to predict. It appears that avoiding investment excesses tends to be easier than timing the perfect moment.

We would like to spend some time on the “equity premium puzzle”, this is a term that came to life with the famous academic article by Mehra-Prescott (1995). Basically, that article shows that the difference in returns between bonds and stocks is too big to be explained by the traditional parameterizations of economic models. Since that article has been published, there have been several follow up studies focusing on the US stock market that show that equities in the last 100 years have returned more than 7% over bonds. The main reason that these studies were done for the US market is that there exists longer and more reliable data. After the crisis in the 2000 decade, there has been some debate that the “premium” does not really exist. Economic theory, in an effort to address this debate, produced some modifications in the model assumptions (average risk aversion from the rational agent, etc.), the impact of tax changes, and some other less structured explanations.

In summary, both the traditional and the more modern approaches tend to be risky over-simplifications of an asset allocation process. This process should be more dynamic and rigorous.

We, at Baffin Advisors not only believe and invest in other types of assets, like commodities and international stocks and bonds, but we also take into account the asset pricing implications of our views on the global macroeconomic landscape, in order to produce a more rigorous asset allocation process. Our portfolios have the following allocations at the moment: 30%stocks (1/3 is US), over 20% corporate bonds, over 5% inflation linked bonds, and over 15% commodities. Such diversification provides less volatility, as we have observed in the last weeks, given that there is more differentiation amongst the different regions and currencies. We produce custom made portfolios for each client, but the overall theme is in accordance with our world vision, regions, inflation, growth, currencies, etc.

For more information view our contact info