Show me the harm

It is often claimed that philanthropists do more harm earning money than good making donations. We saw this idea raised many times during the recent press coverage of Earning to Give. Our response is that although the objection may be true for typical examples of philanthropy, when donors are giving effectively it’s difficult for the expected harm to outweigh the good done by the donations.

In this post, I make some very rough estimates of how harmful finance would have to be in order for it to outweigh the good done by the donations of someone Earning to Give to effective charities.

The Direct Harm

Introduction

Left leaning advocates sometimes claim that Bill Gates did more harm through the Windows monopoly than good through his charity work. On the other hand, some on the right claim that he did more good with Windows than with charity. Both views are implausible.

I’ll focus on the first claim, that the Windows monopoly did more harm than the Gates Foundation managed to do good, since this is the more common type of objection we find raised about Earning to Give. The vaccination programs of the Gates Foundation alone are estimated to have saved around 6 million lives to date, and this number is expected to rise significantly in the future. Windows has had 800 million users to date. This means that in order for Gates to have caused more harm through Windows, using Windows would have to be worse than making a person in poverty play a 1-in-100 Russian roulette. Microsoft might be annoying, but it’s not that bad.

Zero-sum games

What about other cases? It seems likely that many industries involve zero-sum elements, which use up human effort with creating value for society. For instance, it has been claimed that certain types of financial speculation are just gambling. One side wins and the other side loses. No value is created. Similarly with corporate law, escalating amounts of resources are spent aiming to win cases, when mutual disarmament might deliver the same results without enormous legal costs. In both cases, it’s tough to show that these industries actually create no value. Some level of speculation leads to more efficient markets. Some level of competition is probably required in order to have a functioning legal system. Even if we assume that they indeed do not create any value, that is not a robust enough argument to show that pursuing a legal career and doing significant effective philanthropy is net harmful; in order to make this case, one would need to show that the industry is causing harm on a large scale. Here are some calculations to show roughly how large the harm would have to be for such a claim to be substantiated.

Harm outweighing the good in finance?

Estimate 1

Finance is often taken to be the legal high earning career that’s most harmful to society. The average Goldman Sachs employee earns around $500,000 per year. If someone joined Goldman and donated half of his earnings to Against Malaria Foundation, that would be about enough to save 100 lives per year (or more accurately, saving 4000 QALYs), plus likely have substantial positive flow-through effects. For Earning to Give at Goldman to be net harmful, the marginal employee would need to be causing the death of a hundred people each year. This would mean that Goldman Sachs employees are several orders of magnitude more deadly than American service people in Iraq.

Goldman has 32,000 employees. An upper bound for the harm caused by the marginal employee is thus the total harm caused divided by 32,000. For the harm to outweigh the good, Goldman would therefore have to be killing at least 3.2 million young people each year, or doing something else that is similarly harmful. That would mean that Goldman Sachs would need to be responsible for around 5% of all deaths in the world. Bear in mind that Goldman Sachs only makes up 22% of American investment banking, and 3% of the American financial industry - if the rest of finance is similarly bad, then it would imply that finance is doing something as bad as causing all the deaths in the world.

This is an upper bound, if we assume the Givewell estimate of cost per life saved won’t go down, because (i) Goldman’s activities will likely have some diminishing marginal returns (ii) there will be some level of replacement of Goldman employees. By (i), I mean that the 32,000th banker will have less impact than the rest of them. That’s because the marginal bankers will have to take worse profit opportunities than the others, because the bankers will take the best opportunities first. If Goldman’s harm is roughly proportional to its level of profit, then the marginal bankers will be much less harmful than the average banker. By (ii), I mean that if someone turns down a job at Goldman, Goldman will take some actions to compensate for that loss, for instance they’ll hire other people. This will probably result in a significant proportion of the harmful impact occurring anyway.

Estimate 2

Let’s consider the American financial industry in general. Upcoming Giving What We Can research estimates that it would take $200 billion a year to move everyone in the world above the $1.25 poverty line. That figure will only be $74 billion in 2030. The employees of the financial sector could do this if they transferred (e.g. via GiveDirectly) 30-75% of their salaries to those in extreme global poverty (depending on what date you want to achieve the goal by). In other words, if everyone in finance were Earning to Give, it would be possible to end extreme global poverty within the next twenty years. Harm would only dominate if the financial sector is doing something roughly as bad as single handedly causing all global poverty.

For an individual doing Earning to Give at the margin, the situation is even better, and the levels of harm would have to be even worse to get outweighing. They’ll be doing less harm due to replaceability, and can do more good since they can pick better donation opportunities than would be available if that huge amount of money were being spent on aid.

I find it implausible that finance is causing this much harm. The most strident critics of finance claim that the financial crisis will result in hundreds of thousands of deaths and around 100 million extra in poverty. The crisis is often regarded as the most harmful effect of the financial industry in recent times. But we’ve shown that if everyone in the financial sector were doing Earning to Give, it could save millions of lives and move everyoneout of extreme poverty - far greater positive impact than the harm done by the crisis. And (i) it’s difficult to claim that the people in the financial sector were wholly responsible for the crisis (ii) we’ve totally ignored any of the positive effects of finance. This post points out that the financial sector would only need to increase annual GDP growth by 0.1% in order to offset a financial crisis every twenty years that costs 15% of GDP.

Doing good in finance

It may even be that many people pursuing Earning to Give in unethical industries have a net positive impact in their work. That’s because bringing more ethically motivated people into finance seems likely to make it more socially beneficial.

All you need to do is to create less harm than your less ethically motivated replacement. For that, it just needs to be possible to turn down some opportunities to do harm. For instance, much of the harm caused by finance in the last crisis seems to have come from fraudulent activities in packaging mortage debt. If you were working in this sector, but didn’t engage in (as much) fraud you could have reduced how bad it was. Of course, there would have been pressures on you to earn more, which may have made it difficult to avoid and keep your job, but it seems unlikely there was nothing you could have done. Alternatively, there seem to be ways to reduce the harms of finance. By advocating for these reforms from within, you might have some scope to improve finance from the inside.

The whistleblower escape hatch

If you find yourself involved in something that’s really harmful, then there’s always the option to quit, and perhaps even to do some good on the way by becoming a whistleblower.

There have been many whistleblowers who have played a vital role in uncovering massive scandals. A recent financial example is Sherron Watkins, who played a major role in bringing down Enron.

In August 2001, Sherron Watkins, then Enron’s vice president for corporate development, sent an explosive, seven-page email to the energy company’s CEO, Kenneth Lay. She detailed what she called an ”elaborate accounting hoax,” which included inflating income and hiding epic losses. Although Lay claimed that he’d launch an investigation, Watkins said that she was immediately punished; her computer’s hard drive was confiscated and her desk relocated to the nether regions of Enron. Four months later, Enron could no longer sustain the fraud, and it filed for bankruptcy.

(quoting from here)

With whistleblowing, the more severe the harm being done in the industry, the more impact you have! Whistleblowers are rare and often have an outsized impact.

Other objections

It’s all signalling

An alternative source of harm from doing Earning to Give is signalling your support for a harmful or suboptimal system. This could have an effect on the actions of others, which leads to negative effects, in particular, it makes it harder to fundamentally change the system.

I think this effect is likely to be very small in most cases. First, I doubt many people will be strongly influenced if someone they know enters finance or law. It’s a widely accepted path that lots of people already take, and they generally take it in order to get the large salary rather than because they were influenced by their friends.

Second, it seems possible to mitigate the negative signalling (if you think this is an issue) by telling people “I don’t support the system, but I think that engaging with the system is currently the best way for me to make the world better.”

Third, doing Earning to Give has positive signalling effects too. In particular, you’re promoting the ideas of (i) spending lots of your income to good causes rather than spending on it luxury goods and (ii) effective giving. These signals seem much stronger than the negative ones. We’ve seen concrete examples of one person’s donations leading to others.

I expand on these effects more from section 5.3. of this document.

Collective action

A related worry is that considering Earning to Give might lead to a failure of collective action, in which lots of people end up propagating a system that is sub-optimal when a much better option was on the table. We address this objection here. The quick answer is that only happens if effective altruists coordinate badly. As effective altruists become more numerous, it’s seem far more likely that coordination will be good enough that they’ll be able to solve these big problems, rather than get stuck in a situation where all the investment banks are filled with people doing Earning to Give but the system is unchanged.

Could you do more good elsewhere?

An objection to Earning to Give that seems much more plausible is that taking a high earning job in order to donate more involves giving up an even better opportunity to do good. It’s implausible that the harm done within high earning jobs is greater than the good done by effective donations, but it might be even better to work in a more socially beneficial job but donate less. For this to be true, the extra good you do in socially beneficial jobs (compared to non-beneficial high earning jobs) would have to be significantly larger than the good done by the extra donations you could make.

We’re highly uncertain about for which jobs this is the case. In the past, we have advised effective altruists with a particularly strong comparative advantage for earning money to go into Earning to Give, while we’ve advised others to go into jobs that directly do good in effective cause areas.

There may be other situations when it is better not to do Earning to Give. For instance, if you think boosting economic productivity is highly important, then it might be a mistake to swap entrepreneurship or engineering (which presumably increase society’s wealth significantly) for finance or law (which presumably don’t help nearly as much) in order to earn slightly more. We’re planning to write more about this soon.

All told, the question of who can make the most difference through Earning to Give seems very unsettled. See this post for some of the arguments against pursuing Earning to Give.

Thank you to Carl Shulman for providing comments on a draft and several of the examples.