The Case for Evolving the Quality Factor

Applying the Quality Strategy to an Expanding Menu of Goals


Any effective investment process typically seeks to remain aligned with its driving forces while delivering performance in changing macro environments. When investors begin to use an effective process in unanticipated ways, however, we believe it is especially noteworthy. In recent years, the quality factor has advanced along such a path.

The quality factor originally was designed as a tool to gain confidence in the financial strength of dividend-paying equities, but investors have steadily evolved how they use it in their portfolios. Instead of looking at dividend-paying equities as a means to meet income needs, quality dividend payers can also be looked at as a means to grow assets through capital appreciation. Similarly, combining the quality factor with yield may make sense from a total return perspective, since dividend yield can be looked at as a way to gain exposure to the value factor.

Yet, we realized that even though a quality dividend strategy can be used to help meet capital appreciation goals, its application has certain limitations. Key among these is the exclusion of non-dividend payers from the starting universe of investable equities.

In our continuing mission to develop products that meet investors’ real-world needs, we created the FlexShares Quality Large Cap Score and our newest exchange traded fund, FlexShares US Quality Large Cap Index Fund (QLC). Here’s our thought process.


Originally, when we explored applying our dividend quality score (DQS) methodology beyond common stocks, we determined that each asset type required distinct modifications to the template. Using a cookie-cutter approach across asset types would not work.

For example, we realized that the unique aspects of real estate investment trusts (REITs) required a different application methodology than was needed for common stocks. REITs, because of their tax requirement to pass on the majority of their earnings directly to their unit holders, did not produce the same financial data as common stocks, such as net income. So we reexamined the metrics driving the DQS score and determined that we needed to modify the methodology for the REIT space. We use the resultant real estate quality score (REQS) in our global quality real estate strategy.

Our methodology for evaluating the credit quality of U.S. corporate bonds also required a modified process. Using insights from our DQS and REQS processes, we developed a proprietary credit scoring methodology as a replacement for traditional Nationally Recognized Statistical Rating Organization ratings.

All of these processes rely on the same foundational principles, but recognize that each asset type/class is unique — so much so that, to fruitfully apply our quality approach, each quantitative implementation has to be modified.

Coming full circle, we found that the universe used to apply our quality approach to dividend-paying equities ended up being quite similar to the universe used for large cap stocks — but it, too, required certain “tweaks.”


Applying quality to a non-dividend-paying large cap universe is a natural extension of our product suite and one that we believe enables us to better address the specific goals of large cap equity investors.

We think the domestic large cap space is exceptionally fertile ground for using our proprietary multi-factor quality approach. The extensive depth and history of financial data for these companies makes research quite robust, which in turn boosts convictions about how different factors perform in various market cycles. And, liquidity and capacity issues, which may affect small caps or certain foreign markets, are generally not a concern.

This progression also makes logical sense when you consider the pillars that drive our FlexShares quality scoring process:

  • Management efficiency is a quantitative evaluation of a firm’s deployment of capital as well as its financing decisions. Firms that aggressively pursue capital expenditures and additional financing generally lose flexibility in both advantageous and challenging portions of the market cycle.
  • Profitability scores a firm’s relative competitive advantage across several different metrics. Firms with wider margins may be better positioned to grow compared to firms with slimmer margins.
  • Cash flow assesses the liquidity levels of the company. A firm that cannot meet its debt obligations and day-to-day liquidity needs will be poorly positioned to take advantage of future opportunities or enjoy a financial cushion during periods of distress.
  • We apply this methodology to companies on a sector basis in order to:

  • Provide an “apples-to-apples” comparison, since different sectors — such as utilities and technology companies — vary considerably in how they profile;
  • Find quality companies in every sector without concentrating in certain pockets of the market; and
  • Prevent unintended stock selection and weighting biases caused by a dramatic sector drift from the starting universe.
  • We believe applying these metrics is not only a great way to judge a company’s ability to maintain and grow its dividend, but also to gain insight into its overall financial health — and that holds true for dividend and non-dividend payers alike.


    The FlexShares Quality Dividend Fund (QDF) 1,250-firm universe provides a robust selection of quality companies that pay dividends. Therefore, it aligns with investors’ expectations about how to generate reliable income.

    But quality-focused investors who want a total return strategy that stacks up against other large cap options may find that the quality dividend strategy has too much mid-cap exposure, which could result in periods of performance drift versus large cap competitors. A large-cap-only quality strategy that includes non-dividend payers would access the full opportunity set and easily work with traditional investor allocation and risk frameworks.

    We realized that when the investment objective changes from a focus on yield to one of capital appreciation, it also changes the way the strategy seeks to capture the compensated risk factor, freeing up the design to examine other factors.

    For example:

  • Is there a better proxy for value?
  • Moreover, why not evaluate several different metrics in order to get a full perspective on a stock’s value exposure instead of relying on a single ratio?
  • Additionally, for tax-sensitive clients, maximizing dividend yield1 instead of value can be a less effective way to capture returns, because dividend income creates a capital gain that must be paid in the year realized. In contrast, capital appreciation represented by unrealized rising asset prices can be flexibly managed by the investor vis-à-vis timing. This is an important distinction for investors seeking total return but not necessarily income.


    While it may look deceptively simple, a lot of care went into determining the parent large cap universe. All risk controls and diversification constraints relate to this starting point. Our goal was to create an “investability” universe that tracks closely to legacy large cap benchmarks while also allowing the most robust opportunity set for our quality process.

    Why not just start with a legacy benchmark? We examined legacy benchmarks and proxies and believed we could do better. We thought that a more robust starting universe would be advantageous if it allowed more dispersion in returns between companies with attractive factor exposures. Also, it would help limit concentrations in sectors with fewer qualifying constituents, such as telecom, utilities and materials.

    We examined several different breakpoints for the starting universe. In the end, we chose the largest 600 companies that passed our investability screens. This starting universe provides the greatest number of securities for us to evaluate, while keeping tracking error2 , sector and style bets to a minimum.


    Combining Quality and Value

    While quality is a powerful factor on its own, we believe combining it with value makes each factor even stronger. By focusing on large caps and looking beyond dividend payers, a robust definition of value that combines several metrics can be used. Again, similarly to our quality process, we apply three lenses:

  • Contemporaneous metrics use the latest reported filing data to calculate the valuation metric. This provides a view on where the company stands today. Accessing a firm’s most recent earnings report to calculate price-to-earnings (P/E)3 would be an example of a contemporaneous metric.
  • Normalized metrics smooth out the data by calculating an average over several years. Normalized measures remove the effects of the economic cycle and aberrant events that may distort the valuation. Shiller’s CAPE (cyclically adjusted P/E)4 is an example of a normalized valuation calculation.
  • Forward-looking metrics employ future estimates to aid in understanding where the company is going and how future data may be reflected in its stock price. A P/E calculation that uses an analyst’s estimate of future earnings instead of the latest reported value would be an example of a forward-looking metric.
  • We believe the combination of these different lenses of valuation informs a stronger understanding of how a company’s stock price is valued today, in the future and with respect to historic norms.

    Combining Quality and Value with Momentum

    Targeting undervalued companies that exhibit financial strength also provides a strong foundation on which to build an investment strategy. However, by utilizing market-sensitive data (not just company-reported data), the model may be further improved to help avoid value traps. We use momentum to measure the market sentiment for the company as it is reflected in the stock’s price. We evaluate momentum in two ways:

  • Using the stock’s price history to capture a picture of the recent past; and
  • Using analyst outlooks to get an understanding of future sentiment regarding the company.
  • Combining All Three Factors

    While each of these three factors works well in isolation, the combination of the factors — and how they are combined — are just as important. We selected the three factors and determined the process for combining them quite deliberately. Each step has distinct and paired significance. Two factors that are lowly correlated5 or even negatively correlated can have a powerful effect on one another and their combination typically smooths the cycle for each factor, dampening the swings that each would experience in isolation. Also, each is additive in achieving the objective of the fund, something that is often overlooked.

    The goal of diversifying different factors is similar to that of diversifying a basket of stocks that are not highly correlated: both processes seek more desirable performance on a risk-return basis. Historically, value and momentum have negative correlation to one another.

    The fact that low correlations between the factors may be beneficial when combined informs the decision on how best to conjoin the factors. We believe that the most effective way to capture factor exposures is by applying all three factors to the starting universe at the same time. Because of their low individual correlations, applying the factors separately then combining the results may actually dilute their combined effectiveness (i.e., you could end up with some exposures to poor quality, overvalued and negative momentum holdings.)


    The LCQ measurement process creates an alpha6 score based on the three targeted factors, which maximizes the strategy’s exposure to quality, value and momentum. The strategy strives to harness each factor through its selection and weighting process. Additionally the strategy seeks to limit non-targeted factors by placing sector, security, style and market volatility bounds relative to the starting universe.

    This simple but powerful investment process allows the strategy to focus on those traits that we find the most attractive, while it minimizes unintended exposures that are not part of the design. It also has the potential to make capturing the factor more efficient, since the generated tracking error supports the pursuit of compensated risks rather than non-targeted, idiosyncratic risks.

    Lastly, the new FlexShares US Quality Large Cap Index Fund strategy can replace a legacy large cap strategy in an existing allocation model with few follow-on adjustments needed to the model’s sector, style and stock exposures. This ease of implementation may help minimize the time and expense associated with portfolio reconfiguration.

    FOR MORE INFORMATION visit us here .

    1 Dividend yield for an equity security is calculated using the (expected) annual dividend divided by the most recent closing price. To calculate annual dividend, we utilize the most recent dividend payment and multiply that payment by the expected frequency over a year. For example, if the most recent dividend for a company is $2.50 and dividends are expected to be paid quarterly (4 dividend payments/year), the annual expected dividend would be calculated as $2.50 x 4 dividend payments/year, or $10.
    2 Tracking error is a divergence between the price behavior of a position or a portfolio and the price behavior of a benchmark.
    3 Price-to-earnings (P/E) ratio is a ratio for valuation ratio calculated as a company’s current stock price divided by its earnings per share.
    4 Shiller’s CAPE (cyclically adjusted P/E) – also known as the P/E 10 ratio – is a valuation measure, generally applied to broad equity indices, that uses real per-share earnings over a 10-year period. The P/E 10 ratio uses smoothed real earnings to eliminate the fluctuations in net income caused by variations in profit margins over a typical business cycle. The ratio was popularized by Yale University professor Robert Shiller, who won the Nobel Prize in Economic Sciences in 2013. It attracted a great deal of attention after Shiller warned that the frenetic U.S. stock market rally of the late-1990s would turn out to be a bubble.
    5 Correlation is a statistical measure of how two securities move in relation to each other. Correlations are used in advanced portfolio management. Source: Investopedia
    6 Alpha is a measure of performance on a risk-adjusted basis. Alpha takes a fund’s volatility (price risk) and compares the risk-adjusted performance to its benchmark index. The fund's excess return relative to the benchmark's return is its alpha.