Is it smart to buy smart beta?

Much has been said and written about smart beta with the latest Economist’s Buttonwood being a recent example. Supporters are typically those launching products, managing quantitative or factor based portfolios or offering related (‘indexing’) services. There are also many voices of caution suggesting care is needed to really understand what is being bought before making allocations. There are also sceptics some of whom say market capitalisation weighted indices will ultimately, in the long run, be best and/or that turn over costs significantly impact smart beta performance and some even go as far as to suggest the concept is actually smart marketing and is designed to help mitigate management fee compression.

I think the best way to answer the question is to put yourself in the potential smart beta buyers position. They have a large number of funds/ETFs to choose from. Analysing them all will require a lot of time consuming due diligence. As wary or sceptical as they might be there is probably also the feeling that the work is worth doing, after all smart beta strategies have been gathering assets over the years. Indeed as shown on this chart of European listed ETF flows in H1 2019 by State Street, reported by ETFStream, — smart beta has enjoyed a good start to the year in terms of asset gathering.

Source: https://www.etfstream.com/news/8309_defensive-factors-in-favour-at-state-street/

However, as noted, this could simply be reflecting a desire to have defensive exposure as flows have been skewed to low volatility and quality factors and there has been limited interest in multi-factor products.

How can ALPIMA help the smart beta buyer? By utilising our “Digital Evidence-Based Investing” capabilities.

Rather than starting the product/fund selection process by meeting with the all firms involved, seeing and understanding their presentations and getting comfortable with the details of all the products wouldn’t it be simpler to generate a number of sample/model portfolios that proxy past allocations and switch in and out the smart beta products one by one and see how they have performed over time?

Of course the response to this might be that this is just switching one level of complexity for another. However the meetings and presentations will take hours or days while the analysis on ALPIMA’s system will take minutes or hours. We have done this analysis for a large number of the European listed smart beta ETFs, for the newer ones by creating proxies using their indices. What we find is that there are relatively few products that appear to be genuinely additive to model portfolios and these should be the ones that portfolio managers and product selectors should analyse in detail.

For example here a very simple analysis of a multi- factor smart beta ETF that does seem to be worth a further ‘deep dive’. We compared four bond equity portfolio using classic benchmark equity and bond products, a smart beta equity product and bonds, and allocated them 60/40 (rebalancing monthly) and allocated using a maximum diversification logic. Here are the results graphically and in table form, the product ticker has purposefully obscured:

Over the ~17 year period the portfolios that include the smart beta ETF, rather than the market cap product, appear to have better returns in both the 60/40 (1st and 3rd on the table) and max diversification (2nd and 4th) allocations.

In this case, historical evidence suggests it was smart indeed smart to buy this particular smart beta product as part of a long term bond/equity portfolio, especially if the allocation used is dynamic and risk-based rather than a fixed 60/40.

Whether it still makes sense going forward depends on a number of things including:

(i) whether the allocation framework appears sensible for today’s markets and adaptive enough for changing market conditions,

(ii) the actual composition of the strategy today versus the benchmark and

(iii) its sensitivity to key markets, which must be in line with the market views of the holders — which is easily verifiable on the ALPIMA platform.

As ever please let me of any of my colleagues know if you’d like more information.