Context: Sadly, FTX
FTX defrauded users in a way that is normal for cryptocurrency. But the FTX fraud is a function of the normal system working normally. Like ordinary financialized firms, FTX grew by making leveraged promises. Spotty regulatory attention to cryptocurrency gave it sufficient legal cover to make it easy for people to speculate on it, while effectively allowing participants puff up a speculative bubble by engaging in more aggressive leverage than is tolerated in other areas, often shading into overt fraud.
If you were to randomly audit the books of institutions run by people who look from the outside like Bankman-Fried did prior to the FTX blowup, the level of shenanigans he engaged in would not look like an outlier; his ability to do unusual things with a disproportionate amount of capital was approximately titrated to his willingness to take on liability, i.e. borrow more than he could pay.
I do not have a strong opinion on whether South Africa's Truth and Reconciliation Commission was too merciful, but I do not think anyone can legitimately think that it was not merciful enough; amnesty extended to those who have not yet confessed, and continue to occupy positions of power that can choke off their critics' access to resources and attention, is not part of a reconciliation, but license to continue to offend. If the investigation of the FTX fraud goes no farther than the individual at its nominal head, then it is extending such a license to those who created and endorsed the system in which Bankman-Fried was trying to do the right thing.
Journalist Kelsey Piper recently published a text message exchange with Bankman-Fried in which the latter, despite giving up on any pretense at a likeable or sympathetic PR persona, was unable to express a coherent perspective that could help a third party understand why it seemed like a good idea to him to do what he did. This kind of inability to articulate one's own perspective is inconsistent with the type of analytic and logistical skill needed to rise to his position, but entirely consistent with the behavioral patterns of institutionally inflicted trauma that rob power players of the ability to speak (see The Trauma Coup, Moral Mazes as Transformative Treatment).
The institutional structures that designated Derek Parfit as the prior generation's foremost ethicist promoted Will MacAskill - co-founder of Giving What We Can, 80,000 Hours, and the Centre for Effective Altruism, now a professor of philosophy at Oxford - as his legitimate intellectual successor. MacAskill's organizations have publicly endorsed earning to give via employment at the sorts of institutions (e.g. elite management consulting and investment banking firms) that inflict exactly the sort of institutional trauma Bankman-Fried demonstrates, inducting people into habits of correlated, systematic fraud, normalizing each new step as just the way things are done. Surely, thinks the new entry level hire at an investment bank or elite consulting firm, what I'm being asked to do will make sense to me later, right now I'm here to learn skills and acquire career capital.
If we look at the procedures being recommended, rather than the conclusions endorsed, by Effective Altruism, it seems to be calling for behavior like Bankman-Fried's. I shared an early draft of this with Simon DeDeo, and he responded:
Consider the following thought experiment:
"A talented and well-connected young man gets involved in the financial industry, ending up in a highly-placed position as a result of his intelligence and network. He gradually learns how to funnel money into secret accounts, without getting caught. These funds he channels to effective charities that produce a significant effect on net-QALYs, both through direct intervention, and through funding EA research.
The other consequence of this funneling of money is to lower the amount of money investors have in their accounts. People in the First world have less money than they would otherwise. Some people, for example, are forced to work at Starbucks after their investment portfolios go bust. He is careful to make sure that, in the event he is caught, the resulting fallout will involve only a relatively niche part of the larger system, will be seen by most as a recherche scandal caused by a lone wolf, and will not undermine confidence in capitalism as a whole.
The young man takes significant risks in doing this—risks of bankrupcy, and even jail time. However, he believes very strongly that the harm done to people by reducing their portfolios is far outweighted by the benefits provided to others. He knows the risks and takes them anyway."
My guess is that, if this scenario were better tuned for in-group language, the majority of people in Effective Altruism would endorse this behavior strongly.
It now seems likely that the majority of money moved by people inspired by Effective Altruism to "earn-to-give" is in practice fraud-to-give, even before counting the fake cost-per-life-saved numbers produced by EA institutions. While these organizations have backed away from some of their early public endorsements of earning to give, disown anyone who's been caught and publicly disgraced, and reject in an unprincipled way unsavory-seeming consequences of their ideas, I am unaware of any serious attempt to investigate the systemic problems with such institutions, any serious public engagement with my (Oppression and production are competing explanations for wealth inequality) or others' call to distinguish productive from destructive economic activity in a principled way, or any public apology to people who were lured into a life of crime by taking EA career advice seriously.
A serious attempt at applying philosophy to help people at scale would consider the procedures and principles needed to achieve these ends not as an afterthought, but as a central concern - perhaps the central concern.
Visavis the bit framed in ingroup language, it is quite important that SBF has caused huge reputational fallout for EA and that his bets now look ex-ante stupid. For the portion of EAs who would accept your story, they would likely not if clear numbers were given. Would they take $500m in funding for those outcomes? No. So your analogy doesn't hold.
Nor do I buy that the median earn to giver causes nearly these issues.
Thanks for the article.
Update the story to include the following language: "the talented young man also makes careful efforts to avoid being associated with EA in any way."
Do you think the average EA would now endorse the story? We seem to be getting quite close!
> If you were to randomly audit the books of institutions run by people who look from the outside like Bankman-Fried did prior to the FTX blowup, the level of shenanigans he engaged in would not look like an outlier; his ability to do unusual things with a disproportionate amount of capital was approximately titrated to his willingness to take on liability, i.e. borrow more than he could pay.
can you name an example of an institution that you think looks the same from the outside?
3 Arrows Capital, Binance are some opening shots
We have spoken to the former staff at Alameda and we know the whole org was just mind-bogglingly incompetent. The view of SBF as trying to do good things or as formerly competent is difficult to understand.
Dogman, in my experience (and analytically) it is a LOT (definitely orders of magnitude) more likely that someone competent who hates regulators and who thinks ends justify means (in a crude way) and who thinks money has roughly linear utility will be called incompetent *after* people catch on (despite passing an audit a year earlier) than that someone incompetent will become, age adjusted, the wealthiest self-made person ever, even briefly.
Your inside view is what it is, but LOL.
Happy to chat privately if you open your DMs or whatever. Not as inclined to post the contents of private conversations with stupid people who have unfortunately been defrauded for all of their money in a comments section.
I can also of course send you my many prior postings of similar opinions about the competence of FTX over the past two years, if you like.
Pretty sure my DMs are always open and that my email, my name at gmail, is fairly public knowledge. Ping me.
@dogman, I would be interested to hear more because I want to learn more about why the information you had about Alameda's competence didn't propagate.
If you want to talk privately you can DM me on Twitter ( https://twitter.com/orellanin ) or propose a different channel.