Back to Community
Research: An Update to Investing in Women-led Companies

Last week, Pat Wechsler reached out to talk about my notebook on investing in women-led organizations of the Fortune 1000.

Through the course of our discussion, and her questions about the theory and data supporting my simulation, a couple of changes were made. This notebook shows the resulting algo, which her article is based on.

The changes include:

  • Updating the SPY benchmark to reinvest dividends. My previous algo just bought $100,000 at the beginning of the time period and held it until the end. In the example below, I import the benchmark from Quantopian's backtester which takes care of reinvesting dividends. This dramatically changed the results of the benchmark.
  • There were also 4 CEOs that Pat was able to identify that were missing from my list. These included, Stephanie Streeter of Banta, Paula Rosput Reynolds of Safeco, Dorrit J Bern of Charming Shoppes and Carleton Fiorina of HP.

I've removed a lot of the explanation that was in the original notebook, and focused on the changes. You can read the full explaination here in the first shared version: https://www.quantopian.com/posts/research-investing-in-women-led-fortune-1000-companies

Loading notebook preview...
Disclaimer

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.

9 responses

This is really great work, and an excellent template for "invest in companies when they are classified as X" research.

A little late to the party here, but we might want to adjust for the fact that most S&P 500 companies are matured, and are involved in markets/market segments that have reached the point of saturation - this is especially important to note considering we're looking at a significantly larger portfolio of companies from which women-led ones are selected, and therefore, many smaller companies with more room for growth are included. You can argue that they're supposed to have the innovation-based growth to overcome that, but the fact remains that their "cash cow" (high-revenue, low-to-no growth) divisions will likely account for far more stable revenue than any of their frontier/high innovation divisions with high growth, leading to significantly lower overall growth, and thus less equity value and dividend growth as well (and thus, of course, lower overall returns). Might want to consider that most of the new entrants into the Fortune 1000 might also be experiencing quite a bit more of what we call "bubble-driven growth" considering how frothy the markets were for certain periods.

From the graph, we can see higher levels of volatility, and more significant drawdown among the companies you selected from the Fortune 1000, which may be a reflection of their smaller size and lower stability compared to their S&P 500 peers. S&P 500 companies, being the usual "blue chips", are supposed to be stable, boring companies with good dividend payouts and decent (most likely, underperforming vs underdogs or newer companies that are doing well) share price growth, and that's pretty much their selling point.

From an anecdotal perspective, its like comparing Apple to say, Google at the time of its IPO. Going forward, obviously Google had more room to grow, across the same period. I'll admit it's pretty challenging to achieve statistical balance in situations such as these, where the sample size is relatively small, and where you pretty much have to "take what you can get" with regards to data, but it's important not to state conclusions too strongly.

That being said, I'm sure we're onto something with respect to gender balance in company leadership. It's great that someone around here took the time to establish rough correlation (but once again, far from causation) in the context of this issue. =)

Just saying. I was just a little bummed by how conclusive these results were made to sound by a certain article on Quartz. I agree with the above poster, by the way.

P.S. Feel free to call me out on any mistakes I've made, or if the algorithm is more comprehensive than it seems at first glance.

There is definitely still work to be done here, and it's an ongoing project.

I think you are right that the weak point is the benchmark. There are a couple things I am hoping to try to account for this.

  • A sector weighted benchmark
  • A benchmark of all other companies in the Fortune 1000, or a random selection of a smaller number of companies from the Fortune 1000. (I need the data to create this and it's expensive!)
    • Expand my universe of companies beyond the Fortune 1000 and invest in all companies with female CEOs (US Equities of course). I'm working on getting this data set.

You also mention the volatility. Interestingly, my algo is less volatile than the S&P during this timeframe. The sharpe of my algo is 0.6 and the benchmark is 0.2.

I am most certainly not trying to say anything with regards to causation here. I actually want to look at a gender neutral strategy that explores CEO change ups and if this has less to do with the gender and more to do with companies changing their CEOs. There are an awful lot of unknowns, but it's a backtest worth sharing regardless.

I think the Freakonomics guys would, ahem, freak! What else could explain the behavior you're calling out? Why not pick 10 other demographic filters and build the same comparisons? Industry, headquarters location, trans-global factor, employee count and/or productivity quotient, conservative/liberal markets, commodity dependent, interest rate sensitivity, do they allow dogs in their corporate offices, do they provide day care, paternity leave, are their logos red blue or green? Pick a cross section of filters and build new sets. Now that would be research that would be Freakonomic worthy

I think that there is a misunderstanding that research somehow follows smoothly from inspiration and a crisp question, through analysis, conclusions, and an answer to the original question. It's sometimes more of a sausage-making activity. I agree that one could pick just about any company factoid to see where it might lead. Quantifiable characteristics of CEOs seems like it might be a promising research area (e.g. age, level of education, number of years of experience as an executive, etc.) If the male/female factor turns out to be irrelevant, then move on to something else. Turn over another stone.

Is there a cloneable version of this yet ?

If not, perhaps consider re-posting :)

I'd love to have a copy to scan through !

James - check out https://www.quantopian.com/posts/female-ceos-a-sector-neutral-version-its-clone-able

Disclaimer

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.

Thanks Seong !

Great work.

Has anyone updated this algo with data up to 2019?