How Important Is Asset Allocation to an Investment Portfolio’s Performance? Less Than You’d Think

Yesterday we took a look at an overview of the 2012 Robert Haugen and Nardin Baker research “Low Risk Stocks Outperform within All Observable Markets of the World”, as explored by Financial Times columnist Terry Smith. The study essentially debunks the truism that anyone investing online into a portfolio will be familiar with: higher risk equals higher returns. However, as explained by Smith, the data compiled by Haugen and Baker, which covered the period between 1990 and 2011 and 21 developed equities markets, demonstrates that is not actually the case. Rather, with risk defined in terms of volatility (high volatility = high risk), the lowest risk stocks very clearly outperformed their higher risk peers.

Today we look at another investment truism and whether the data across markets really supports it. That is that an investment portfolio’s returns are, over the long term, largely determined by its asset allocation. In 1986, a study by Brinson, Hood and Beebower, ‘Determinants of Portfolio Performance’, examines the data around this assumption more forensically. One of the most widely quoted conclusions of this study is that ‘asset allocation is responsible for 91.5% of an investment portfolio’s returns’.

This was based on the analysis of a significant sample of the investment portfolios offered by major pension providers. However, what the study actually demonstrated was not that 91.5% of returns were the result of asset allocation but that 91.5% of the variability in returns. The result of this misinterpretation of the report’s conclusions is what Smith believes is behind the fixation of the investment industry on asset allocation as being of greater importance of the individual asset within a portfolio.

Another fallout of this asset allocation fixation has been, because spreading asset allocation across classes such as equities, bonds and commodities is held so dear, domestic bias towards the equities allocation in a portfolio. With a chunk of the portfolio’s value given over to other asset classes, the equities section is less diverse and financial advisors and asset managers tend to focus on domestic stocks. So UK investment professionals are overly tied to the FTSE 100 and wider London Stock Exchange. This, a 2010 study by index compiler MSCI argues, fails to take account of the fact that so many listed companies, especially the larger ones, derive their revenues internationally. For example, more than 75% of the revenues of the companies in the FTSE 100 does not come from the UK economy.

The MSCI study put forward an equities balance its data demonstrates both, counter-intuitively, reduced risk and increased returns. That was ignoring domestic bias entirely and categorising equities only by their listing being in either developed or emerging markets and if they are large, mid or small cap. Developed market small caps and emerging market equities both have lower liquidity. 

Source: FT

The chart above is based on data from the past five years and shows the difference in returns as part of a portfolio is shifted from the MSCI World Index, composed of developed market large and mid-caps, and into the MSCI World Small-Cap Index. Each dot indicates a 5% additional allocation over to the small-cap index. At 35% allocation to the Small Cap Index, returns become higher at the same risk profile.

The conclusion is that it may well not be necessary to add bonds, commodities and other asset classes into a portfolio after all. The same reduction of risk can be achieved from global equities with an allocation to small cap companies.