Improving diversification with smart beta

Factor investing, also known as smart beta, is increasingly recognizing that investors can achieve factor-driven excess returns and diversified investments beyond market capitalization-weighted benchmarks through a simple, transparent, rules-based approach. It is becoming more and more popular.

Based on research affiliate research, six factors: value, low beta, profitability, low investment, momentum, and scale create a substantial diversification effect across multiple return drivers. It conveniently summarizes several factors that are labeled as indicators of a quality company, such as profitability and investment. The diverse aspects of combining factors with different risk and return characteristics and low correlations help investors “survive the storm” in adverse market conditions.

However, questions arise when deciding on a multi-factor portfolio. Which elements do you need to include? You will find that the right balance is achieved by making decisions based on the trade-offs between the effective harvest of Factor Premium and the low cost implementation.

Combination of smart beta factor strategies

Research Affiliates analyzed a six-element-based smart beta strategy by building a simple, long-term, investable portfolio. Except for small cap strategies, we will start with the world of large cap stocks of US stocks. For each, select the best strain based on the corresponding characteristics. For example, build a value portfolio by selecting the top 30% of books to market ratios. Within the portfolio, selected stocks are weighted by market capitalization, except for low beta, which is weighted by beta ranking.

The portfolio is rebalanced every July, while momentum and low beta are rebalanced quarterly.

As shown in Table 1, on average, six factor-based smart beta strategies provide improved returns with an average annual excess return of 1.86% over the study period from July 1973 to December 2018. I found out.

Table 1: Performance of long-term only factor-based smart beta strategy, USA July 1973-December 2018

Source: Research Affiliates, LLC using CRSP / Compustat data

The correlation of excess returns for the six factors is mostly or negative. However, we find that there is a more positive correlation between investment factors and value and low beta.

According to historical data, momentum and size factors appear to have a substantial diversification effect on multifactorial strategies. Momentum factors are negatively correlated with the other three factors (value, low beta, investment) and slightly positively correlated with profitability and size. The size factor has a relatively low correlation with other factors and seems to be a particularly strong diversification factor for profitability factors. SMEs seem to spend time growing rather than making a profit.

Implementation issues

It is very important to consider factor investment from a practitioner’s perspective rather than a scholar’s perspective. The difference is that research can find a relationship between a company’s financial characteristics and future returns, but investors are only interested in relationships that can result in excess returns in the actual portfolio.

Given the infamous reputation for high transaction costs associated with many factors, implementation requires careful analysis. The biggest offenders are momentum and small factors. Explicit implementation costs such as mediation are generally well managed. We focused on the market impact of transactions, the implicit factor of implementation costs that can be measured by fluctuations in the price of securities due to transactions.

We understood the importance of reducing implicit implementation costs and analyzed them in depth. We incorporate measures directly into the design to reduce the market impact of the strategy.

The impact of portfolio rebalancing on the market can be due to several factors. The first and best known is strategic sales. The more you trade, the higher the cost. However, you also need to consider the volume, liquidity, sales concentration, and slope of your strategic portfolio.

Portfolio volume is the sum of the median daily trading volumes of all stocks held in the portfolio. If the portfolio is small, that is, if the portfolio holds a large number of illiquid stocks, the implementation cost will increase.

Liquidity refers to how quickly and easily you can buy and sell securities. The stocks of large, well-known companies with large market capitalization are usually more liquid than the stocks of small, lesser-known companies with small market capitalization.

Sales concentration measures the spread of transactions across the securities held in the portfolio. When the transaction volume (transactions) focuses on only a small number of securities and is of a relatively large size within the portfolio, the transaction is usually more evenly distributed and generally more costly to execute than the smaller size. Will be charged. portfolio.

Finally, the slope measures the illiquidity of the portfolio compared to the most liquid portfolio possible. If your portfolio holds more illiquid stocks, it will cost more to trade, the portfolio will have a higher slope.

Intuitively, you can see that the portfolio is large, has a low slope, has a low turnover rate, and has a low turnover rate.
A centralized strategy has a low cost of impact on the market, a small portfolio, a large slope, high sales, and a high sales. A centralized strategy has a high cost of impact on the market.

Assuming that the average sales of the six strategies analyzed from 1973 to 2018 are 61.6% and the average estimated transaction costs of the strategies are US $ 10 billion (US $ 13.6 million) in assets under management (AUM). It is 127 basis points (bps). It is shown in Table 2.

Table 2: Implementation Costs of Long-Term Only Factor-Based Smart Beta Strategy, USA July 1973-December 2018

Source: Research Affiliates, LLC using CRSP / Compustat data

The cost of implementing a strategy with high turnover, high slope, or low portfolio volume can be very high. For example, the momentum strategy has a very high one-way sales of 159.5%, and despite the reasonable volume and slope of the portfolio, the relatively high transaction costs of 241 bps for a US $ 10 billion portfolio. Will occur. Given the lack of implementation, using a simply constructed momentum factor as a standalone investment strategy does not seem to give good results.

In contrast, strategies with large portfolios, low slopes, or low sales usually have lower implementation costs. For example, a profitability strategy with the lowest transaction costs of the six factors (17 bps in a US $ 10 billion portfolio) is characterized by all three low cost characteristics.

Value strategies also incur low transaction costs (56bps) due to their relatively low turnover and slope. Interestingly, small cap-focused size strategies have below-average transaction costs (91bps). The size factor tends to trade small, illiquid stocks, as evidenced by the relatively small volume of the portfolio and the large slope, but as of December 2018, 2,352 names were included. Therefore, the turnover rate is low. It almost offsets the transaction costs of SMEs.

Multi-factor of momentum and size

Momentum is very expensive to implement as a standalone strategy, and size appears to be risky in terms of volatility and tracking errors, but because these factors have a negative or low positive correlation with others. Factors that can be a good addition to a multi-factor strategy.

We also analyzed four portfolios to assess how the inclusion of both of these factors impacts the performance characteristics and implementation costs of a multi-factor strategy. Examine the four multi-factor portfolios that are evenly distributed to each set of named elements.
Portfolio 1: Value, low beta, profitability, and investment.
Portfolio 2: The four elements and momentum of Portfolio 1.
Portfolio 3: The four elements and size of Portfolio 1.When
Portfolio 4: Momentum and size, in addition to the four elements of Portfolio 1.

Creating Portfolio 2 by adding momentum to an evenly weighted 4-element portfolio reduces tracking error by 84 bps and reduces information ratio (IR) to 0.46 due to negative or low positive correlation of momentum with other factors. On the other hand, it improves to 0.57. The transaction costs of Portfolio 2 are surprisingly lower than the transaction costs of Portfolio 1, at the portfolio level of US $ 10 billion, 32bps vs. 33bps, respectively.

Gaining momentum in a multi-factor portfolio increases the volume of the portfolio, reduces the slope of the portfolio, and lowers transaction costs. Additional liquidity compensates for the increase in sales, as momentum is associated with more liquidity stocks. In addition, low or negative momentum correlations with other factors lead to transactions initiated by momentum rebalancing, canceling transactions initiated by value or other factors.

Adding a size factor to Portfolio 3 also benefits performance and implementation costs. Adding size to a four-element portfolio will increase revenue by 26bps (13.13% vs. 12.87%) and reduce the cost of the US $ 10 billion portfolio by 7bps. Indeed, the size strategy is the most volatile of the six factors in the analysis. Therefore, including size increases the volatility of the four-factor strategy by 0.6%, but poor correlation between size and other factors reduces tracking errors by 0.27% and improves IR.

In summary, diversified investments are made in both performance and trading activity, so adding momentum and size strategies to your multi-factor portfolio can lead to good results. However, it is important that the two strategies must be in the same portfolio so that offset trading can achieve lower implementation costs.

Get the right balance

As investors become more interested in multi-factor smart beta investments, understanding how to best combine the factors is important for the desired return on investment. You need to fully understand the correlation and implementation costs of each element. When properly constructed and blended with other elements, perhaps surprisingly, the momentum and size elements are useful components of a multi-factor smart beta strategy.

Adding momentum helps reduce tracking errors and improve IR due to negative or low positive correlation with other factors. These benefits are offset by offsetting trades across the factor strategy, and gaining momentum increases liquidity and is achieved without significant implementation costs.

The size factor has relatively wide coverage and low turnover, so it’s actually quite cheap to trade. Therefore, adding a size element to a combination of other factors will improve the performance of the multi-factor strategy, lower tracking errors, higher IR, and higher transaction costs if the size correlation with other factors is low.

Research affiliates strongly advocate the thoughtful design of multi-factor strategies. This requires a conscious and careful decision to find the most favorable balance between effective collection of element premiums and implementation costs.

Mike Aked is the Australian Research Director of Research Affiliates.

Improving diversification with smart beta

Source link Improving diversification with smart beta

Related Articles

Back to top button