Back to all articles

23 October 2018

How to Avoid the Factor Masquerade

When analysing factor returns, how can we be sure that we are not just confusing factors with country or sector effects? During Q3 2018, we saw a wide disparity in returns across global equity markets and sectors. Our findings show that these sector and country impacts have masked how underlying factors have truly performed across market regions.

Country and Sector adjustments

Constructing factor portfolios using simply high absolute fundamental or market data values can often result in significant biases to sectors and countries. These biases are driven by systematic differences in the fundamentals of sectors and countries and even by cross-border accounting differences between countries. But professional investors understand that a stock is not necessarily a growth stock simply because it is in the technology sector, nor should it be designated as a value stock simply because it is in Russia. At Style Analytics, we ensure that factor analysis can also be considered on a country and sector relative basis to reveal the genuine factor impacts in equity markets, and for equity portfolios.

Below, we have taken examples from three different regions in Q3 2018 to illustrate material differences in outcomes when factor analysis is performed on a country and sector relative basis.

Global Developed Market (Q3 2018)

If factor performance of Developed Markets in Q3 were to be based on only unadjusted fundamentals, then a strong continuation of value weakness would have been reported across all value factors for that quarter. However, once country and sector relative adjustments are applied, most value factors were actually neutral with the exception of book-to-price which still showed continued weakness. Essentially, sector and country returns were masquerading as factor effects.

US (Q3 2018)

In the US, we saw another case of sector returns masquerading as factor returns. Constructing a value factor portfolio in Q3 based on the absolute values of sales-to-price across US equity stocks would show that this factor had underperformed. In fact, this could be mostly explained by the strong outperformance of the healthcare and technology sectors where the unadjusted factor portfolio was underweight. Using sector-relative sales-to-price actually revealed that value stocks within sectors actually outperformed slightly.

Emerging Markets (Q3 2018)

The need to avoid comparing country and sector returns with factor returns turned out to be vital in Q3 for Emerging Markets. In general, forming portfolios based only on absolute fundamental valuation metrics in Emerging Markets often leads to some quite strong active weights in sectors and countries. For example, an unadjusted value portfolio would have resulted in a heavy overweight in the energy sector and in Russia in Q3. However, using sector and country relative factors, we find that value did indeed outperform based on high cash flow yield, but not for EBITDA to Enterprise Value, sales-to-price, or earnings yield.

Genuine factor returns

Unadjusted factor analyzes based on absolute factor values are not technically incorrect, but as we have seen they can be potentially misleading. In the same way that investment analysts approach the assessment of a company’s fundamentals on a country or sector relative basis, Style Analytics also provides country and sector relative perspectives to reveal the true factor drivers of market returns.