Investing in Women-led Fortune 1000 Companies | Hacker News Reader

Investing in Women-led Fortune 1000 Companies

quantopian.com

45 pointskarenrubin37 comments

One possible explanation (pure conjecture) is that this is a consequence of the still prevalent sexist/old-boys-club mentality. Only the women who are so exceptional that there can be no question about their capabilities are the ones who end up on top. And the women who are equally qualified as the typical male CEO simply don't make it through the gauntlet.

This is an effect that minorities face in general. They have to be the absolute best in order to break through the glass ceiling. There are plenty of male CEOs who are utterly mediocre, and mediocre women never get a shot at the CEO spot.

A much more likely explanation is selection bias. There would never have been a post like this if the result would have the other way around. Statistical variation makes it easy to find a starting point to show what you want.

Or, female CEO's are a more risky prospect than male CEO's and companies have different risk/reward motivations based on their situations. If a company's facing a tough grind for a few years they may be risk averse and stick with "safe bets" wherever they can.

Or, years of fighting "the patriarchy" has left females uniquely qualified to lead companies in the fight against changing market conditions.

Or, prospective female executives are probably very aware of the social and political climate we live in and that may make them more risk averse in regards to accepting positions, they may place a higher value on taking positions at companies on the upswing.

We can do this all day but I doubt the data's going to give a clear answer, we'll all see what we want to.

I like your theory. Whenever there's talk of some sort of affirmative action, some people say "Imagine getting your job just because you're a woman". Well with this theory, there are men getting CEO job partially because they are men, even if they aren't very good.

There's no discussion here of survivorship bias [1]. Without a discussion of how it is addressed, these results are highly suspect. Specifically, the benchmark takes into account those companies that lost enough value to fall out of the "bottom" of the index and are therefore not in your current list of companies. However its far from clear that your subgroup includes similar companies.

A naive approach to this analysis is as follows. Take all the companies currently in the Fortune 1000, and look to see which of them has had a female CEO in the period under study. Backtest with buys and sells on the dates of hire/termination.

If this is indeed how it was done, the benchmark will almost assuredly underpeform simply because of the underperformance of components that are no longer in the index today.

[1] http://en.wikipedia.org/wiki/Survivorship_bias

Looking at your graph, it's clear that you need to adjust for volatility. Your "Women CEO" portfolio is much more volatile during that period. Compare Sharpe ratios, not total performance.

After that, you may have to worry about dividends. I'm not sure if the platform takes them into account when you send buy sell order, but if it does, make sure you look at a portfolio that purchases SPY over that period, not merely the price, as the appreciation won't reflect the dividends paid.

Also, short-term (and even long-term) capital gains from having such a high turnover in the women CEO portfolio...

If you're looking for systemic mispricing in the market, that shouldn't matter so much as there are enough tax free vehicles (such as IRAs) to smooth out these effects.

The Quantopian backtester pays dividends out in cash, and since I am rebalancing the portfolio each time I buy or sell stocks, I think that is covered.

Looking at the volatility is a good suggestion and definitely something to consider. I noticed the trend as well, but haven't dug into it yet.

Ok, but you're extracting the price of SPY, so you're not counting SPY's dividend. Use a portfolio that simply buys SPY at the beginning of the period and holds it.

The volatility thing it the biggest effect though.

You also may want to try betting on the spread between male-only companies and ones with at least one woman on the board. If the investment thesis is sound (board gender diversity is beneficial), this would be a good way to test it. It would also bring you into market-neutral territory.

I wanted originally to go for board diversity, but haven't found that data set yet. I believe Morningstar has some of it, and am hoping to get my hands on it sooner rather than later.

I'd like to analyse the data as well. Can you open source your methods and sourced data? I'm very in favour of spreading this analysis, but I'm afraid that there might be some sort of bias going on (for example, I really want to adjust for industry, because I suspect that tech and media companies are leading this trend, while materials are not).

Edit: Just saw that this is an interactive notebook! Awesome! Going through it now.

Edit2: It looks like you've already done what I planned to do! Awesome work!

@murbard2

I finally got the annualized sharpe calculation figured out. You can see what I did here: http://imgur.com/HlcjNYz but it shows my sharpe as 0.66 and the SPY as 0.3....

Honestly, this seems implausible to me...so I will keep digging. :-)

The conclusion I'd draw that female CEOs are a indicator of a company culture likely to be more accepting / more meritocratic / less nepotism internally which likely makes for a for a more agile company.

Regardless, stats like this hopefully help erode long standing views of women and leadership.

Expressed gender is not a good measure of a person's ability to create (or maintain) a good working environment.

The person above didn't claim otherwise.

They said the opposite, that the culture of the company has to be a certain way to have a female CEO in a world of mostly male CEOs. They didn't say that a female CEO creates that culture, and that would have the chicken-egg problem (i.e. how does a CEO create a culture that within itself helped cause a CEO get into the role? They could continue it however).

PS - I should clarify that I don't agree with the OP's point about culture. I actually subscribe to the notion that becoming a female CEO is so difficult the only ones who manage it are just "that damn good" that their companies do better.

Many reject the Efficient Markets Hypothesis. Few do so while lying atop a gigantic pile of money.

So if the markets are all efficient then how did they get the pile of money?

There will always be outliers.`

No one has a pile of money. That's the point. Many people come up with theories on how to beat the market by using some novel theory or algorithm. These people generally have not successfully tested these theories (and made the aforementioned piles of money).

(This is why I follow patio11. Reading his tweets and HN comments is not unlike unpacking a particularly obtuse if-else-if statement block. Difficult, but rewarding!)

If you plot based on blood types of CEO's rather than gender, you may get other type of results. So should we take blood type as another type of parameter like gender?

Point I want to make is, people are different starting from blood type, height, weight to "n" number of parameters and we can get different graphs for different parameters but that won't conclude anything.

Each person is different and just because some one belongs to a category does not make that person any extra attractive for any job.

In my personal opinion, this type of validation rather than competence,track record,experience is not beneficial to organization or society and completely shortsighted. So please stop these generalizations like "women are so and so", "men are so and so" ...etc. There are good and bad elements in every category.

I think you're looking at the trees, and you should step back and look at the forest.

The fraction of CEOs that are women is dramatically smaller than the fraction of the population that are women. There is no qualitative explanation as to why that should be true. So long as that remains true, it's worth looking into why it is true. The relative performance of the group is fair game for investigation.

It's obvious, women are better than men at certain things, such as running companies. That's what we're all trying to say here isn't it? Oh, wait, we want EQUALITY for both sexes. Well, then in that case why differentiate one sex as potentially better than the other? Why create that competition?

There is no qualitative explanation as to why that should be true.

I think there are plenty of qualitative explanations especially if look back a few decades (since most CEOs are in their 50s).

Women didn't obtain the name number of university degrees as men until recently. Women often step out of the workforce in order to care for family, etc, etc.

All of these would explain where there would be a smaller pool of qualified women to take a CEO role.

This reminds me of a study I did on CEOs who came to speak to my university. I found that investing in a company if their CEO was visiting college campuses averaged an annual return of ~20%. The logic was that CEOs of poor performing companies didn't have time to visit schools.

For a disclaimer, the portfolio was also more volatile than most benchmark indexes, so looking at the Sharpe ratio in both instances (mine and OP's) is important.

In the first cut of your analysis (without rebalancing) the portfolio actually goes negative (I'm looking at the first graph in the notebook). An equity portfolio cannot go negative unless you are using leverage. Then from late 2012 to 2015 the portfolio rises by 400%. Looking at the companies in the portfolio, I don't see how they could have generated that kind of return. Is this algorithm using leverage? If not, I don't see how you can get those results. If the algorithm is using leverage, then comparing to a simple investment in the S&P is not valid.

You are absolutely right that the first version of my algo is using tons of leverage. I didn't realize what leverage was when I did this the first time.

The second version of the algorithm factors in leverage and there it hovers around 1.

How much is this skewed by Alibaba and yahoo?

Also, what would Apple look like for openly gay CEOs?

"Ethical" investment funds also did well for a while. Even if the women in this study are better CEOs (and that share price reflects this) any attempt to capitalize on this will change the population. ie assume equal talent pools, top 30 from each beat top 60 for either.

This is really cool, but I'd love to see this narrowed down to something super scientific. Cause I can already hear the haters, screaming about control groups, sample sizes, and all that stuff. Which is actually important.

Right now I take it for what it is a quick mock up of some shallow data. But it seems compelling enough to take a deeper Nate Silver ( http://fivethirtyeight.com/ ) style deep dive.

I’d love suggestions about how to make this more scientific. I think the biggest potential pitfall is the variability related to the stock size. My approach is pretty simple right now. I just took all of the companies from the Fortune 1000 that had female CEOs since 2002, and bought their stock when they were women-led, and sold it when they no longer were.

From an investment perspective, I think the best thing to do next would be to make sure I have the right benchmark. Ideally they Fortune 1000 would be the best one to use, but I need the historical Fortune 1000 companies for the last 12 years….that will take some manual work to pull together.

At some point, though, the control group is an academic exercise. If the strategy makes money - invest.

In particular, your purchases aren't cap-weighted. You simply divide your portfolio evenly among women-lead companies. So aside from biases that might exist for women-led companies (they have to work harder, so they're better performers; or, they tend to be more highly represented in consumer cyclical, etc), you're also "over-weighting" small-cap equities, which have done quite well relative to the S&P500 (all large cap) over the period[0].

I would be interested in a breakdown of other factors, but I expect very little variance actually derives solely from gender of ownership once you account for other conventional biasing factors of the chosen index.

[0]: http://performance.morningstar.com/funds/etf/total-returns.a... (Compare to S&P500 TR)

Calculating four factor model alphas (http://en.wikipedia.org/wiki/Carhart_four-factor_model) would be a good starting point.

It's not a sample, it's a census.

sample_size == population_size

This comment was into 0 territory. It's actually a very important point that people often miss when discussing things... you don't do "population statistics" when you're getting data from every member of the population. So in this case, "sample size" is irrelevant.

Many other elements of the analysis can and are being challenged, which is fine to discuss, but there's no need to discuss "sampling" issues, as there is no sampling going on.

Reply on news.ycombinator.com