Factor Spotlight
Factor University

Illuminating the Dynamics of Technology, Media, and Telecom

In last week’s Factor Spotlight, we discussed the merits of using sector models as an additional lens for understanding risk in your portfolio. Broad market models are very helpful to establish a baseline for risk in the portfolio relative to the overall market; however, sector models can ‘tune’ the factors more accurately to the unique market behavior of each sector. This week, we’re going to dive into the Wolfe Research QES US Technology, Media, and Telecom (“Wolfe US TMT”) risk model to see what blind spots we can uncover compared to using a broad market risk model.

Methodology for the Wolfe US TMT Risk Model

The Wolfe US TMT risk model is constructed using an estimation universe of US all-cap technology, media, and telecommunication services securities. The model includes fundamental factors relevant to the technology sector, such as Total Yield and Sales-to-EV, as well as technical, macro, and positioning factors that are common to other sector and broad market models, such as Volatility, Momentum, and Liquidity. Lastly, the model includes alternative factors, HF Crowding, Short Interest, and ETF Crowding, to identify crowding trends specific to the TMT space. While the model’s estimation universe is the technology sector, the model is able to cover all US equities due to the time series regression methodology used to estimate the security sensitivities to the factors. Please see the risk model factsheet for more information on the construction and factors used in the model.

For the broad market risk model, we’ll be using the Wolfe Research US Broad (“Wolfe US Broad”) risk model.

To run our analysis, we built a technology-focused market-neutral portfolio of names that are typical long & short names held by institutional investment managers. The long side of the portfolio is comprised of the top 100 securities, equal-weighted, based on exposure to the Hedge Fund Crowding factor within the Wolfe US TMT model. The short side of the portfolio applies a similar construction methodology, but uses the Short Interest factor within the Wolfe US TMT model. The portfolio is rebalanced on a monthly basis to update the constituents, reset the weights, and maintain market-neutrality.

Capturing Sector-Specific Style Behaviors

One benefit of sector models is that they view traditional factors using metrics that are relevant to the respective sector. For example, the level of value in a portfolio can be measured in different ways depending on what drives the relative value of securities in the sector. Value for technology stocks tends to be better measured by sales multiples rather than book value multiples. As such, the Wolfe US TMT risk model uses Sales-to-EV as the factor to quantify a tech stock’s value exposure, whereas the Wolfe US Broad risk model uses Book-to-Price. Looking at each of these factors for our technology portfolio, we see that the two measures behave rather distinctly.

The portfolio’s performance in 2020 attributable to value differs depending on whether we use the sector-specific definition or more generic definition of value. The chart below illustrates that when using a broad definition of value, such as Book-to-Price, this factor appears to help the technology portfolio’s 2020 performance by almost 2%.

Value Performance using Wolfe US Broad


However, when using a sector-specific definition of value, such as Sales-to-EV, the factor actually hurts the performance in 2020 by almost -1.5%.

Value Performance using Wolfe US TMT


By capturing the unique factor dynamics within the sector, a technology-specific risk model allows us to see through potential distortions created when viewing risk for our technology-focused portfolio using a broad market lens.

Isolate Granular Industry Effects

Another benefit of sector models is that granular industries are included as factors which can isolate particular sector dynamics. Broad market risk models typically combine industries together and this can confound the true industry level impacts in the portfolio. Using the sector risk model as an overlay in these cases allows us to more clearly distinguish the differing behaviors of specific industries.

In this example, we first look at the Wolfe US Broad risk model Software & Services factor, which shows that these industries combined contribute almost 5.7% to the overall portfolio return.

Software & Services Performance using Wolfe US Broad


However, when we isolate each of these industries individually using the Wolfe US TMT risk model, we see that the two factors actually perform quite differently within the portfolio. In particular, the IT Services factor contributes 6.1% to overall performance, while the Software factor detracts from the portfolio performance by close to -1%.

Isolated Software & Services Performance using Wolfe US TMT


Again, more distinct industry factors allow us to understand the factor dynamics within the sector models at a more precise level.

Using the TMT Model in Practice

As we’ve highlighted in the above examples, TMT managers are able to leverage the Wolfe US TMT risk model to better isolate the systematic drivers of volatility and performance within the technology universe. The TMT sector model can improve the portfolio construction process for technology managers through enhanced security screening based on metrics relevant to their coverage universe and the ability to target factors that capture the unique dynamics of the sector. The sector model also allows managers to better quantify the technology-focused factors that are driving risk and performance in their portfolio, which can otherwise be washed out when using a broad market risk model.

Stay tuned as we continue to dive in to the sector models and highlight areas where these bespoke models provide added transparency for portfolios that are susceptible to heavy sector-specific effects.

US & Global Market Summary

US Market: 02/08/21 - 02/12/21

Screen Shot 2021-02-13 at 11.33.29 AM.png
US Stock Market Cumulative Return: 2/8/2021 - 2/12/2021
  • The rally that started in early February continued, with the major indices up 1-2% this week and again reaching all-time highs. The positive trajectory has been fueled by optimism around improved vaccine distribution and fiscal stimulus, as well as a strong earnings season thus far.
  • According to Factset, 74% of S&P 500 companies have reported, with 80+% of them beating earnings estimates (the highest such ratio since FactSet began tracking it in 2008), and 78% reporting a positive revenue surprise.
  • Fed Chair Powell helped with some dovish words on Wednesday, advocating a “patiently accommodative monetary policy” as the labor market remains “a long way” from where it needs to be.
  • President Biden’s $1.9T stimulus bill progressed this week, with House Democrats releasing a detailed draft including “non-negotiable” measures for $1,400 individual stimulus checks. Democratic members of Congress target getting the bill passed by mid-March.Market liquidity has remained elevated, with an average of 15.8B shares being traded each day across U.S. exchanges over the last 20 days, according to Bloomberg. This is close to the levels seen at the end of March.

Normalized Factor Returns: Axioma US Equity Risk Model (AXUS4-MH)

Screen Shot 2021-02-13 at 1.56.38 PM.png
Methodology for normalized factor returns
  • Earnings Yield and Volatility were tied for biggest winner in a week with muted factor movement, both up +0.13 standard deviations.
  • Market Sensitivity saw a mild upwards move, and appears to be close to exiting Oversold territory, while Sizedid in fact inch its way out of Oversold space, now sitting at -0.98 SD below the mean.
  • Momentum saw a slight drift down as it remains an Overbought factor.
  • Growth was the biggest loser on the week after hitting a recent peak of +1.49 SD above the mean on Feb 5, and is still an Overbought factor at +1.3 SD above the mean.
  • US Total Risk (using the Russell 3000 as proxy) decreased by 28bps.

Normalized Factor Returns: Axioma Worldwide Equity Risk Model (AXWW4-MH)

Screen Shot 2021-02-13 at 2.01.50 PM.png
Methodology for normalized factor returns

  • Earnings Yield shed its Oversold label after a +0.24 standard deviation move, now sitting at -0.83 SD below the mean.
  • Volatility continued its positive MTD trend, moving up another +0.19 standard deviations.
  • Size and Market Sensitivity saw incremental positive normalized gains, with both exiting Oversold space.
  • Exchange Rate Sensitivity drifted downwards by -0.11 standard deviations, and looks poised to cross into Oversold territory.
  • Profitability was the biggest loser of the week, falling -0.3 SD and landing exactly at the historical mean.
  • Global Risk (using the ACWI as proxy) decreased by 13bps.


Related Insights
See All Insights

What Forces Are Impacting Your Performance? Find Out Now...

Schedule a Call