Summer 2021

Investment Management Using Factors

The seventeenth century laid the foundation for the modern world. Using observations and consistent methods, Sir Isaac Newton explained gravity and the earth’s rotation around the sun, establishing a new world view which remains unchallenged today. It is easy to take Newton’s achievements for granted, but he presented new ways to see the world and a cohesive framework to explain how things worked. Newton’s numeric proofs provided order, an important consideration at a time when questioning scripture caused chaos.

Mark Webster
Director, Institutional & Advisory, Western Canada
Jul. 4, 2021

The seventeenth century laid the foundation for the modern world. Using observations and consistent methods, Sir Isaac Newton explained gravity and the earth’s rotation around the sun, establishing a new world view which remains unchallenged today. It is easy to take Newton’s achievements for granted, but he presented new ways to see the world and a cohesive framework to explain how things worked. Newton’s numeric proofs provided order, an important consideration at a time when questioning scripture caused chaos.

Investment management attempts to provide order to the random walk through Capital markets. Decades of research and analysis have helped investors to minimize subjectivities and to provide a robust framework to manage risks and returns. Just like the seventeenth century, observation and method provide new insights into what moves markets and determines outcomes.

The S&P 500 was conceived in 1956 and published in 1957 because the investing world required a foundation to distinguish skillful asset managers from those who were lucky. Originally envisaged solely as a benchmark, nobody imagined at the time that the S&P 500 would become a product, available through Futures, Swaps and Pooled funds or ETFs (SPY is the most widely traded security in Capital Markets). Its transition into a product occurred because analysis showed manager outperformance did not persist over time. The majority underperformed the Index return in a given year and those who outperformed did so when their style, sector or capitalization preferences were in sync with a point in the economic cycle. As Capital works its way through an economy, outperformance reverts to its Mean.

The famous Brinson, Hood & Beebower study in 1986 (Determinants of Portfolio Performance, The Financial Analysts Journal, July/​August 1986), showed that security selection and market timing, fascinating subjects though they may be, had minor influence in determining long-term performance. Shortly thereafter Eugene Fama and Kenneth French revealed several Factors which helped to explain sources of outperformance.

Continued study over several decades, both academic & industry, have demonstrated that part of what we refer to as Alpha (outperformance over Index benchmarks), may be more appropriately explained as identifiable, measurable and repeatable Factors which define outcomes. 

The Evolution of Indexing

The evolution of indexing


Source: MSCI Inc.

Factor versus Style – Is it Alpha or is it Beta?

The simple Alpha / Beta distinction was refined to distinguish between Growth and Value styles to explain manager performance cyclicality. Accounting for material differences in investment disciplines provided a much better perspective as to why managers performed better or worse as the economy moved through the cycle. Growth and Value benchmarks were created to evaluate managers against a more precise data set, acknowledging that management styles displayed strength or weakness at over time rotating in and out depending on the stage in the economic cycle. Style benchmarks, however, retained the same market capitalization methodology used in Broad Benchmarks. 

Fama & French’s pioneering work spurred additional scrutiny to explain return profiles, starting with a simple three Factor model which has expanded with additional research, facilitated through more robust data and greater computing power. 

Style benchmarks have now been supplanted by Factor Beta Indices which use alternative weighting methodologies to accentuate positive characteristics which have demonstrated an ability to add value across time in all markets. Institutions can use these Factor Beta indices to improve their ability to model outcomes. 

Risk Level


Factors may be one of the most important tools in parsing risk profiles and return sources, resulting in more tailored exposures to use in Asset-Liability models.

Attribution is the foundation for analysis and decision making. It separates the skillful from the lucky and it also shows which managers adhered to mandates and who strayed. Factor investing uses a consistent rules-based data set, without subjectivities or drift. 

Institutions manage liabilities, investing to achieve long-term compounded returns. It seems logical for them to integrate Factors into their asset allocation to exercise greater control through an economic cycle. 

Both Active management and Factor investing provide several refinements over broad Beta indexing:

  • Possibility to outperform;
  • Possibility to reduce risk;
  • Assumed high conviction portfolio;
  • Assumed high Active Share portfolio composition.

Investors accept cyclicality when they stray from market capitalization exposure, but they do so to enhance performance by either raising returns or by lowering risk. In their quest to escape market capitalization-driven outcomes and the concentration or valuation risk it may bring, investors must humbly accept that coveted Alpha can be positive or negative, and fickle or unpredictable.

It is reasonable to suggest that Factors provide a more comprehensive approach to asset allocation, enabling Institutions to use complementary Factors at different points in an economic or market cycle. Because Factors can be identified and isolated, they provide a more transparent and consistent exposure, a more disciplined data set upon which to model risk and return over time.

As the tables below demonstrate, Factor exposure can help Institutions to monitor and to manage the relative risks they assume over time. Even over short time periods, Low Volatility and Quality Factors have demonstrated a tenancy to reduce market risk while generating attractive returns. A focus can be maintained on enhancing returns, providing the opportunity to earn returns exceeding broad Beta Indices but, equally important, consistent data is available to measure relative risks.

Diversification effects over longer time periods

Longer time horizons → higher historical occurrence of outperformance, with diversification benefits becoming apparent over longer investment periods.

Diversification effects over longer time periods

Gross returns in USD from December 1999 to December 2020.
Source: MSCI inc. - Information Classification: GENERAL.
(Please refer to the material in the Appendix which shows how Factors can mitigate risk compared to Broad Beta Indices).

As the graphic below shows, Active mandates may demonstrate significant deviation at any point in time, something which may be fleeting or which may endure for an extended period. In this instance, the Size and Quality Factors have had greater importance than the Growth element which is supposed to define the mandate. Do such mandate drifts decrease or increase risk or return? What do such drifts indicate about exposures Institutions can expect in the future.

MSCI FaCS: Introducing the MSCI FactorBox

What is the fund’s current exposure? Is it consistent with its objective?

FactorBox and Style Box for the Intl Growth Fund as of 12.31.2020.
Source: MSCI inc. - Information Classification: GENERAL.

Assessing Factor Definition & Portfolio Construction

Worthy Factors must demonstrate advantages across all markets and across time. Unlike opaque Quantitative models, Factor Indices provide transparent, rules-based exposures, excellent material for Institutions to execute a thorough due diligence as they would with Active mandates.

When evaluating Factor methodologies, investors must examine the Factor integrity to determine if the investment thesis is developed using a rigorous and extensive data set. Once Factor relevance has been confirmed, it is equally important to review the portfolio weighting methodology to ensure it provides diversification benefits and does not pose unintended risks. 

Some Factor methodologies use shorter timeframes, others stick closely to sector weights in the parent Index and some are unconstrained. These are decisions which will have profound influences on the outcomes they generate. If a Factor measure provides desirable characteristics which can replicated throughout a cycle, the portfolio weighting methodology should not undermine the Factor’s desirability by introducing unintended or uncontrollable outcomes. 

BMO Exchange Traded Funds is the largest Factor Beta provider in Canada (Bloomberg June 2021), listing its own proprietary methodologies for Low Volatility and Dividend ETFs, MSCI for Enhanced Value and Quality listings and S&P Dow Jones for Size (S&P 400 Mid Cap and S&P 600 Small Cap), complemented by Equal Weight sector indices which also capture the Size Factor. 

We exercised an exhaustive due diligence to select the exposures, evaluating both the Factor definitions and the portfolio construction methods. 

Factor Overview:

Low Volatility – BMO Proprietary methodology:

  • Uses Beta as its risk measure. Competing methodologies use either Standard Deviation or Minimum Variance, both of which measure idiosyncratic risks. Beta provides a more stable measurement over time and also measures market sensitivity, which is better for modelling macro risks;
  • Sector and Security concentration limits are used to ensure the portfolio is diversified and does not assume any unintended, and potentially unrewarded risks;
  • Further analysis is done to determine Interest Sensitive exposure. The Low Volatility Factor may capture a higher than market exposure to Interest Sensitive sectors or companies. If this is an issue, the companies may be prorated to minimize unintended risk which may occur if rates rise.

Click here to learn more >

Dividend – BMO proprietary methodology:

  • Identify Sustainable Dividend Growers, companies which demonstrate an ability to raise their dividend and maintain it from Core Operational Earnings. This eliminates companies which use debt, investment income or acquisitions to pay dividends;
  • Sector and Security concentration limits are used to ensure the portfolio is diversified and does not assume any unintended, and potentially unrewarded risks;
  • Further analysis is done to determine Interest Sensitive exposure. The Dividend Factor may capture a higher than market exposure to Interest Sensitive sectors or companies. If this is an issue, the companies may be prorated to minimize unintended risk which may occur if rates rise.

Click here to learn more >

Quality – MSCI Factor methodology:

  • Identifies companies displaying High Return on Equity, Stable Earnings Growth and Low Debt-to-Equity (Leverage);
  • Compared to the parent Index, the Quality Factor Index reduces volatility while generating attractive returns.

Click here to learn more >

Both the Quality and Dividend Factors identify companies where management has demonstrated an ability to enhance investor Capital through either higher Return on Equity or a sustainable dividend policy.

Enhanced Value – MSCI Factor methodology:

  • Identifies companies with Low Price to Forward Earnings, Low Price-to-Book and Low Enterprise Value-to-Cash Flow from Operations. Forward Earnings are used to reveal companies which may be mispriced. Low enterprise Value-to-Cash Flow from Operations is used to minimize Value traps which may occur in highly leveraged companies;
  • Exposure is Sector-neutral and captures 30% of the Market Capitalization of the Parent Index (50% in Canada).

Click here to learn more >

Size – S&P Dow Jones for Small & Mid Capitalization Indices / SolActive for Equal Weight Sector Indices:

  • S&P Down Jones uses an Earnings screen to ensure all constituents in the Small & Mid Cap Indices have positive earnings over the previous four Quarter. This acts as a Quality screen, indicating management has exercised a positive influence on investor Capital;
  • SolActive is used for Equal Weight Sector exposures. Equal Weight is a worthy approach because it forces diversification across the portfolio, minimizing significant security and portfolio concentration risks which occur in market capitalization or market capped sector Indices.


This communication is for information purposes. The information contained herein is not, and should not be construed as, investment, tax or legal advice to any party. Particular investments and/​or trading strategies should be evaluated relative to the individual’s investment objectives and professional advice should be obtained with respect to any circumstance.

BMO Global Asset Management is a brand name that comprises BMO Asset Management Inc., BMO Investments Inc., BMO Asset Management Corp., BMO Asset Management Limited and BMO’s specialized investment management firms. 

®/™Registered trade-marks/trade-mark of Bank of Montreal, used under licence.


APPENDIX

One Year DataTickerYieldTotal ReturnSt. Dev.SharpeSortinoInfo RatioBeta
BMO S&P/TSX Capped Composite ETFZCN2.8228.0312.132.106.64n/​a1.00
BMO Canadian Dividend ETFZDV4.1134.4712.482.4522.451.360.94
BMO Low Volatility Canadian Equity ETFZLB2.4922.7513.651.574.42-0.680.93
BMO MSCI Canada Value ETFZVC2.6440.1315.212.3013.682.271.17
BMO S&P 500 ETF (CAD)ZSP1.3423.2812.141.784.10n/​a1.00
BMO US Dividend ETF (CAD)ZDY2.6623.1712.281.764.98-0.230.89
BMO Low Volatility US Equity ETF (CAD)ZLU1.9210.8711.210.962.09-1.640.69
BMO MSCI USA Value ETFZVU2.1732.0415.371.898.370.680.87
BMO MSCI USA High Quality ETFZUQ0.9219.5414.971.262.47-1.061.16
BMO MSCI EAFE ETFZEA2.5018.6813.301.353.69n/​a1.00
BMO International Dividend ETFZDI3.9320.3015.651.253.290.091.14
BMO Low Volatility International Eq ETFZLI2.5910.2712.890.811.66-2.110.90
BMO MSCI All Country World High Qual ETFZGQ0.9717.9810.551.322.47-0.811.04
Three Year DataTickerTotal ReturnSt. Dev.SharpeSortinoInfo RatioBeta
BMO S&P/TSX Capped Composite ETFZCN11.0917.092.100.90n/​a1.00
BMO Canadian Dividend ETFZDV9.2618.222.450.67-0.270.99
BMO Low Volatility Canadian Equity ETFZLB12.0014.451.571.130.140.74
BMO MSCI Canada Value ETFZVC6.3720.240.350.48-0.711.11
BMO S&P 500 ETF (CAD)ZSP14.8414.691.781.50n/​a1.00
BMO US Dividend ETF (CAD)ZDY5.7015.681.760.50-1.410.95
BMO Low Volatility US Equity ETF (CAD)ZLU10.0111.870.961.21-0.560.61
BMO MSCI USA Value ETFZVU6.2418.421.890.52-0.971.07
BMO MSCI USA High Quality ETFZUQ17.7914.781.261.910.580.95
BMO MSCI EAFE ETFZEA6.8212.731.350.72n/​a1.00
BMO International Dividend ETFZDI2.5315.981.250.23-0.861.16
BMO Low Volatility International Eq ETFZLI3.6910.750.810.41-0.690.75
BMO MSCI All Country World High Qual ETFZGQ16.7413.531.321.881.030.94
Five Year DataTickerTotal ReturnSt. Dev.SharpeSortinoInfo RatioBeta
BMO S&P/TSX Capped Composite ETFZCN9.6313.570.680.96n/​a1.00
BMO Canadian Dividend ETFZDV7.9714.640.540.70-0.331.00
BMO Low Volatility Canadian Equity ETFZLB9.0611.680.721.06-0.080.73
BMO MSCI Canada Value ETFZVC-n/​an/​an/​an/​an/​a
BMO S&P 500 ETF (CAD)ZSP15.6312.541.141.92n/​a1.00
BMO US Dividend ETF (CAD)ZDY7.9013.180.570.81-1.420.95
BMO Low Volatility US Equity ETF (CAD)ZLU8.9810.900.761.23-0.850.66
BMO MSCI USA Value ETFZVU-n/​an/​an/​an/​an/​a
BMO MSCI USA High Quality ETFZUQ18.1313.091.272.290.530.99
BMO MSCI EAFE ETFZEA7.8510.950.661.01n/​a1.00
BMO International Dividend ETFZDI4.8913.480.350.49-0.711.12
BMO Low Volatility International Eq ETFZLI4.8610.300.420.61-0.680.80
BMO MSCI All Country World High Qual ETFZGQ15.9611.901.232.140.770.99


Source: Bloomberg, as of September 302021.


Factor Valuation Metrics


Price Earnings Ratio
Price Book RatioAverage Price/​Cash FlowReturn on Common EquityTot Debt to Common EquityAverage Dividend Yield
ZCN21.292.1713.779.49151.162.53
ZDV17.791.8512.9810.34181.473.76
ZLB17.962.2510.8413.11131.282.62
ZVC19.091.7413.019.04123.882.74
ZSP31.084.8319.3913.82124.471.34
ZUQ28.789.3222.0331.8856.611.03
ZDY22.533.4715.0013.28163.872.61
ZLU23.123.4712.3112.65121.402.34
ZVU20.781.988.238.15142.822.55
ZEA22.961.809.736.94180.652.50
ZDI17.991.857.839.43155.893.65
ZLI25.862.3011.396.8685.922.64
ZEQ24.776.2016.1123.4668.462.36
ZEM16.481.9210.3213.0795.342.17
ZLE19.382.3421.4613.27119.482.53
ZGQ26.788.4120.2727.0144.301.32

Source: Bloomberg June 302021.


Canada - Historical Frequency of Low Volatility vs Broad Market

Rolling WindowLow VolatilityHigh Dividend
1Y83%77%
3Y94%76%
5Y100%78%
10Y100%96%
15Y100%100%
20Y100%100%

Time Range: January 1996 - June 2021

Simulated Returns: January 1996 - October 2011

Actual Returns: November 2011 - June 2021


US - Historical Frequency of Low Volatility vs Broad Market

Rolling WindowLow VolatilityHigh Dividend
1Y61%53%
3Y66%51%
5Y66%44%
10Y98%28%
15Y100%0%
20YN/AN/A

Time Range: January 2003 - June 2021

Simulated Returns: January 2003 - October 2011

Actual Returns: April 2013 - June 2021


INTL - Historical Frequency of Low Volatility vs Broad Market

Rolling WindowLow VolatilityHigh Dividend
1Y73%30%
3Y87%7%
5Y94%0%
10Y100%100%
15YN/AN/A
20YN/AN/A

Time Range: January 2010 - June 2021

Simulated Returns: January 2010 - September 2015

Actual Returns: October 2015 - June 2021