Low volatility stocks for higher returns

 

I was going to write in my next blog entry about the challenges of value investing and outsmarting Mr Market but my attention was grabbed by the ‘low volatility anomaly’ after recently reading about it on the investment blog ‘Abnormal Returns’.

If you are wandering what I am talking about, several analyses of stock market data seems to show that low volatility stocks outperform those with high volatility (i.e. low beta beats high beta).

Further evidence is provided by AllianceBernstein who have released a report titled ‘The Paradox of Low Risk stocks’. They conclude stocks of low volatility beat market indices as these so called ‘steady eddies’ are less likely to get caught up in market booms and then subsequently don’t loose as much during market crashes (see blow). These differences compound over time resulting in a superior performance by the low volatility portfolio.

The reason as to why it is referred to as a ‘paradox’ or an ‘anomaly’ because it goes counter to the fundamental concept in finance that higher risk is rewarded by higher returns (although admitedly this is not widely accepted by value investors who argue a 'cheaper price' means lower risk but also higher returns e.g. 'margin of safety').

Anyhow, interestingly the low volatility strategy is also uncorrelated with other investment approaches such as value, small cap and momentum, hence, such stocks could be used to further diversify a portfolio. It also tends to work best when the market is pessimistic about the future, for example during the recent European debt crises.

I will not write anymore about this paper apart saying it is recommended reading as they go on to describe how they improve performance by combining it with ‘quality’ (no further details what that means I'm afraid) and how low volatility stocks can be added to a portfolio to significantly improve returns at a similar level of risk.

Perhaps , like me, you are now wondering how to apply this strategy and which markets it works with good affect. Well, this is where an article by Nardin L Baker and Robert A. Haughen provides the answer. With the the following title ‘Low Risk Stocks Outperform within All Oberservable Markets in the World, it certainly got my attention!
It sets the scene with the following introductory paragraph.

 

So, what  did they do? Every month from 1990 to 2001 they computed the volatility for each stock from 21 developed countries and 12 emerging markets. The companies were the ranked by volatility and formed into deciles. The difference in returns between the most and the least volatile stocks are compared below.




In all cases low volatility outperforms, although the effect is less pronounced in certain emerging markets e.g. China.

From my perspective it was very exciting to see Sweden included the study (12th on list; orange colour code) with a near 25% return difference between the two volatility deciles. I could not find any information on which universe of stocks was used (small and large cap? all stocks?), how many companies were exactly included and if the low volatility portfolio outperformed an index such as OMX30. It would be very interesting to backtest this strategy and see how it compares but I’ve yet to find a suitable database.

In addition, does the portfolio really need to be rebalanced every month? That would result in a huge costs drag for a small investor like myself. Clearly there are some finer details that I think are missing which would be required to implement this strategy in ones portfolio.

In the meantime, to satisfy my curiosity, I have made some assumptions and put together a portfolio of 25 Swedish stocks, shown in table below. This list has the lowest beta values (<0.5) from a universe of 252 companies with a market cap. >$25million (in effect they are the low volatility decile).




I'm sure Swedish investors will recognise the large cap, high dividend payers Axfood and Teliasonera. Swedish Match is a company whose stock seems to be attached to a rocket at the moment, it's up 33% the past year and 130% since 2009! It's not a compnay I'd normally be attracted to but I guess this is why the low beta strategy is so interesting. Others are completely new to me e.g. Alltele, NAXS and Selena.


Several of these companies reported no earning last quarter, hence the missing P/E values. Again this would not make them a stock I'd usually invest in. I do wonder though if filtering by P/E or P/B would enhance the returns of such a 'low beta' portfolio.

Anyway, I will put together the above list of companies as a 'model' low voloatility portfolio and track it's peformance from the start of this year (this assumes of course beta has not changed significantly over the past 6 months).
I will post the returns on this blog to keep readers up to date.

(More information about low volatility investing can be found on the following website  - thanks to Smarta Investingar for sending the link).

Editorial: I have freely interchanged volatility and beta is this article under the assumption they were the same or at least beta is a good surrogate measure for volatility. This is not quite true as described here http://www.indexuniverse.com/sections/blog/12033-parsing-low-vol-and-low-beta.html