Allocations within allocations

A successful asset allocation strategy goes further than just allocating assets to broad asset classes. Allocation within these classes is also vitally important. Unfortunately this is not as simple as following a quantitative model. It is of great importance to have a fundamental understanding of the current situation to start with in order to structure each component of the broader allocation correctly.

To illustrate this, we look at a common perception within currencies and commodities that some investors hold without assessing the fundamental backdrop, namely the “flight to safety” premise, where they believe Gold and the US$ gain in ‘risk off’ periods. The first chart shows the Yen relative to US$ over 8 bear markets since the mid 80’s. The Japanese Yen provides positive returns in these periods, which shows that it offered better protection than the US$. However, the Yen is not a perfect hedge, as when the IT bubble burst, it did not offer protection. Gold also only offered positive returns in 10 out of the 13 US bear markets since 1970.

The asset class where most investors attempt to get exposure to high Beta during periods of risk on and low Beta during periods of risk off is equity. Within this asset class there is the perception that some sectors clearly do the trick better than others. As above, this holds true to an extent on average, but the results of this strategy can be vastly improved with a deeper understanding of the current market. The bar charts below depict the average return (red bar) of each sector during all ‘risk off’ (bear) and ‘risk on’ (bull) markets over the past 20 years as a ratio to the general market as a proxy for its Beta. To give a sense of the reliability of this relationship we plot plus (green bar) and minus (blue bar) one standard deviation from the average. The smaller the range, the more reliable the average will be as a guide to action.

Whilst some average results are hardly surprising (such as staples, healthcare and energy offering downside protection), dispersion on many is quite wide. Tech stocks would have been a leveraged play on the downside at some stages, whilst being negatively correlated when markets rallied at other times. Also the consumer discretionary sector underperformed on the down and the up side at different periods. Industrials and materials are not biased one way or the other on average and financials does offer a leveraged play in general as anticipated, but this does not always work and disparities are large.

In a nutshell, this analysis suggests that quantitative analysis is not effective if not backed by fundamental reasoning, which can be different every time the wheel comes around