Smart Beta investing has been one of the more popular topics among investment professionals over the last few years. Its success has been built on the realization that when selecting stocks to include in a portfolio one can benefit from the accumulated research gathered over many years of study.

When filtering an eligible universe of stocks, the evidence does suggest that one can construct portfolios with attractive risk and return characteristics by focusing on a small number of factors, such as value, growth, momentum, income, and quality. On the back of this, interest in factor investing is going from strength to strength.

Given there are now over 1,300 ETFs listed on the London Stock Exchange, and the fact that ETFs are well suited to the process of asset allocation, there is an increasing level of interest in how one can apply a rules-based framework to identify which ETFs one should include in one’s portfolio?

At Twenty20 we have developed our own rules-based approach to ETF selection, where the rules, backed by empirical evidence, relate to an ETF’s target benchmark, and not to the stocks included in the index. These rules can be viewed as an extension of the concept factors, and in many ways aim to capture the extent to which an ETF’s return is driven by changes to market data, such as the level of interest rates or the level of the VIX, and to changes in macro-economic data such as the business confidence level or the monthly PMI print for any particular country. Moreover, the concept can be extended to all categories of ETFs, such as Fixed Income ETFs or ETFs offering exposure to Property companies like REITS.

In the first article in our series on Factor Investing we started off with a Fixed Weight Reference Growth portfolio, which acts as the benchmark from which any new ideas can be tested. This week we look at the impact of using a model to determine what the weights of each ETF should be on a monthly basis.

By using the ideas outlined above, we calculate a score for each ETF in our portfolio, and employ an optimization process to identify the portfolio’s efficient frontier and to determine the weight of each fund in the portfolio. This model has been applied to the list of ETFs in the Fixed Weight Reference Growth portfolio to produce a similar, but differently weighted, asset allocation process. If one compares the two backtests, the key benefit of dynamically adjusting the weights as the economy itself evolves is to smooth out the down side risk and consequently improve the overall annual return over the backtesting period from 8.5% to 12.8%. In both cases management fees of the ETFs, an estimate of trading costs and a discretionary management fee of 30 bps is deducted.

If you would like to learn more about rules-based ETF selection, contact us and we will be more than happy to provide you with further information.