You Can’t Eat Lipstick

A Case Against Settling for Social “Good”

The other day, I met with a startup that wants to re-envision how we buy makeup. On one of the last slides in the deck titled “Social Impact”, the CEO informs me that they are not just another direct-to-consumer makeup company, rather, they will be the “TOMS Shoes of Makeup” — for every lipstick they sell, they will donate one to those in need.

I couldn’t help but head-scratch. Hey, I like makeup as much as the next girl, but this felt like a vanity-driven philanthropic cause — pun intended.

I couldn’t fully articulate why it bothered me at the time, but upon reflection, the makeup company fits a dangerous trend:

When considering social impact, we too often dole out gold stars for good intentions instead of calculating the real efficacy of the initiative.

We’ve just experienced an especially tumultuous political year, and in response, we’re witnessing a dramatic uptick in social enterprises in Silicon Valley. It’s cooler than ever to rebuke the greed-driven norm and create a company that will “do well by doing good”. I’m excited by the trend, but cautious. How do we determine if good is good enough?

Capitalist enterprises aim to maximize shareholder value so it would be reasonable for us to assume that Social enterprises seek to maximize social value. Unfortunately, it’s a heck of a lot harder to quantify impact when returns are measured in human lives and prosperity rather than dollars.

In 2006, Toms Shoes broke onto the scene and made the equation simple. For every pair of shoes you buy, they buy one for a child in need. Boom. Social impact quantified. Consumers rejoiced — not only are these wedges adorable, but they also empower previously shoe-less kids to go to school for the first time! Win-freaking-win.

However, in recent years, the Toms model has been criticized — from sources ranging from the Harvard Political Review to The Economist, who found that the one-for-one gifts “had no effect on the overall shoelessness, shoe ownership (older shoes were presumably thrown away), general health, foot health or self esteem of the recipients.”

Beyond effectiveness, we also do a poor job assessing the relative value of one social impact initiative versus another. Yes, shoes and makeup are useful, but as Bruce Wydick of the Economist laments, “You can’t eat shoes.” The same dollar that goes into producing and distributing a pair of shoes or a tube of lipstick could likely be both stretched further and put to better use doing something else that is more critically needed — but what might that something be?

At its core, social good is a math problem and we’re doing the calculations all wrong.

William MacAskill offers a compelling solution in his book Doing Good Better (highly, highly recommend) — he creates a framework to assess the effectiveness of altruism through five key questions below.

Framework to Assess Effective Altruism

1. How many people benefit, and by how much?
2. Is this the most effective thing you can do?
3. Is this area neglected?
4. What would have happened otherwise?
5. What are the chances of success, and how good would success be?

Let’s break them out:

1. How many people benefit, and by how much? We have to collectively prioritize and make hard trade-offs between 1 person dying and saving 10 others from losing limbs. In healthcare, economists use a metric to do this called the quality-adjusted life year (QALY) to assess cost effectiveness of different treatments and determine which provide the largest aggregate benefit for a given limited spend.

Example: In the chart above, MacAskill compares two medical treatments. The first provides 20% (self-reported) improvement in the quality of the patient’s life each year for the entirety of their 60 year lifespan, equalling 12 QALYs. In the second, a patient, who is currently at 70% health’s, has her life extended by 10 years. This equals only 7 QALYs. An impossibly tough comparison becomes easier through math — the first treatment is more net beneficial.

2. Is this the most effective thing you can do? It’s not enough to just ask whether a program is a good use of money, we need to ask, is it the best possible use of this money?

Example: You have $100 to donate. You can either donate to Charity A that will provide 1 QALY worth of care for $100, or you can donate it to Charity B that will provide 3 QALYs. If you’re optimizing for QALYs, B wins.

3. Is this area neglected? It is more effective to spend money on neglected causes because there is an asymmetrically high impact potential.

Example: In the chart below that circulated during the 2014 ALS Ice Bucket Challenge, you can see the money raised by certain high profile charitable events vs. the number of US deaths from each disease.

Using these donations as a barometer for public sentiment, both Chronic Obstructive Pulmonary Disease (blue circles)and Diabetes (green circles) may be “neglected” causes based on the high death tolls vs. low money raised.

Source: Vox. Quick note: Be wary of a few things on this chart: a) it’s a bit misleading because it doesn’t show total funding for each disease, simply a high profile social initiative, and b) we don’t know how much funding would be needed to eradicate each disease — some may cost more than others to solve.

4. What would have happened otherwise? If you don’t solve this, who will? Are you the best candidate to solve it?

Example: I do not know how to perform the Heimlich maneuver. I can still run over and help perform it on someone in a restaurant to get treated like a hero, but there is a strong possibility that another person in that restaurant — a doctor, a trained staff member — would have performed the function better and more safely without my intervention.

5. What are the chances of success, and how good would success be? This is an expected value problem — if you do it, how likely are you to win, and what does winning do?

Example: You’re a physicist working on nuclear fusion — a potentially limitless, clean energy source, the same one that powers our sun. The chance of you being the one to solve fusion may be infinitesimally small, but the impact would be monumental, making such a moonshot worth your while.

This framework makes it easier to size the true impact of the paint company’s social initiative. By running through a mental list of the 5 questions, it becomes painfully apparent to me that providing free beauty supplies to those in need is simply not a good enough problem to focus on.

Who is leading the way?

Overall, reading MacAskill’s Effective Altruism framework has altered the way that I assess philanthropic endeavors altogether. I’m convinced that we need to move beyond social “good” and strive for something greater. Let’s commit to refocusing attention on the big needs — sustainable energy, poverty alleviation, clean water, economic empowerment. Then, let’s hold all organizations — nonprofits, for-profits and investors alike — accountable for systematic efficacy within those areas.

I took a look at some prominent social impact investors within the startup-sphere to see what frameworks they use and how they compare to MacAskill’s. Here are the ones I’ll be watching:

• Bill & Melinda Gates Foundation: The foundation is the gold-standard in social impact. As an outspoken advocate for “outcome investing”, they employ scientific research methods to continuously assess & improve the efficacy of all of their initiatives. Check out this video and their “Guide to Actionable Measurement” for more.
• Chan Zuckerberg Initiative: The power duo was heavily critiqued for the mismanagement of their first big investment in 2010 — a$100M donation to the NJ public school system (you can read about it in this book), but it seems that they’ve learned a lot since then. The CZ initiative, formed in 2016, is now the largest private philanthropy, and it leverages cutting edge engineering technology for social change across education, justice/opportunity, and science. I couldn’t find a lot of information on their frameworks and/or assessment methods, but it’s a relatively new organization, so I expect that we will see a lot more Gates-like data from them this year.
• Y Combinator: Since 2013, YC has funded 25 nonprofits, including the ACLU and Noora Health, with $100k (non-equity) grants. YC’s nonprofit Program Lead Robby Walker wrote a blog post that explains what YC looks for in nonprofits that seemed eerily similar to MacAskill’s framework…until I realized that MacAskill himself is a YC alum with his own nonprofit 80,000 Hours/the Centre for Effective Altruism. Likely not a coincidence that the philosophies are well-aligned!
• Obvious Ventures: I’m particularly interested in for-profit social investors like Obvious because I’m a firm believer that the two aren’t at odds. Plus, for-profit companies have a lot of inherent advantages. OV is a traditional VC model focused on 9 investment areas across 3 themes: Sustainable Systems (Energy, Agriculture, Resources), People Power (Work, Money, Community) and Healthy Living (Food, Wellness, Health IT). As a traditional fund, $$ returns will be one good indication of success, but I’d love to see more information on their specific framework.
• Acumen Fund: Similarly, Acumen Fund makes only for-profit investments with a specific focus on wiping out global poverty through agriculture, clean energy, education and healthcare. They advocate “patient capital” — a way to bridge the gap between market-based investing and pure philanthropy. You can read the full approach here.
• Omidyar Network: Started by eBay Founder Pierre Omidyar, the philanthropic venture fund employs a mixed model. They call themselves a“flexible capital” source that includes impact investments alongside traditional grants. Here is their investment philosophy and approach.

Passing the mic — Who am I missing from this list? What organizations (both for-and non-profit) are also leading the way in creating and abiding by strong social impact frameworks? Eager for names and critiques here or to @nataliefratto on twitter