I'm concerned that the control group doesn't seem like a representative comparison. It seems like Giving Pledge signatories are older, richer, more famous, and more successful than the sample of YC recipients.
A better comparison group might be the world's 100 richest people, would it be possible to redo this analysis for them?
I was mostly interested in establishing the base rate; I only included the comparison because it was easy. So I am unlikely to look more into comparisons, but I published my data above and others are welcome to do more analysis!
The base rate seems tricky to interpret without a real comparison. 3 out of 5 points in the summary say that the rate is high, but it doesn't seem like the evidence exists to support that claim. The implied reference class is "non EA super-rich people", but that class isn't actually examined in the data.
The rates do seem subjectively very high. Way fewer than 10% of people I know have been convicted of financial crimes! But I wonder if founders and CEOs are being blamed for financial crimes that their companies commit, and approximately all successful companies commit financial crimes, defined broadly.
Overall great data though! It raises real concerns.
I'm curious, do you think the Giving Pledge should have higher standards? My hot take is that we don't want to help launder reputations, but that I would personally be very happy if e.g. the Sacklers donated all their money to a real charity.
Strong downvote on the OP because the counting is done in a HIGHLY misleading way. If I were the moderators I would take the post down or demand SUBSTANTIAL corrections.
For Claim 1. (25% of signatories have been accused of financial misconduct, and 10% convicted[1]), the info provided in the footnote renders this claim outright false. Beyond the fact that many civil violations of the law are not even crimes, civil judgments are not convictions, require a substantially lower standard of proof, and from a legal strategy standpoint, there is often no reason not to settle a case that requires no admission of wrongdoing and where the legal consequences of a criminal conviction are not triggered and it's just not worth the opportunity cost of spending time and money litigating it.
Using the experience of non-US billionaires is also extremely misleading without evidence about base rates of crime in those countries. My guess would be that India, for example, has historically had such high rates of corruption that virtually every Indian over ~50, and certainly those in business, have done things that would result in jail time if done in the US. But that's not morally blameworthy; it's just the system they have to operate in.
Beyond that issue, some more mundane ones:
No evidence is presented for the idea that white collar criminals are less likely to be punished than other criminals. It is probably the case that rich people are less likely to be punished for the same crime than poor people. But the nature of financial crime in publicly traded companies is that the evidence is just way more accessible to the government than the evidence of stealing toothpaste from Target. So financial crimes in publicly traded companies are likely prosecuted at much higher rates than other types of crimes.
Overall the rates listed appear to generally be lower than the general population, which is extremely impressive considering that most billionaires are men, who tend to have much higher rates of criminality. So I would expect billionaires to have almost double the population base rate of criminality just by virtue of that! I don't know why Y-combinator funding is a remotely appropriate reference class.
Let's review the individual claims now:
25% of signatories have been accused of financial misconduct, and 10% convicted[1]
4% of signatories have spent at least one day in prison
Overall, 41% of signatories have had at least one allegation of substantial misconduct (financial, sexual, or otherwise
Claim 1. is blatantly false, as noted at the outset. Beyond that, we have only an explanation, hidden in the footnote, of what constitutes a "conviction". No definition of the meaning of allegation. Based on the stated methodology I assume at least a criminal charge or the filing of a lawsuit to count. No indication on whether e.g. a public allegation with no suit counts, and no indication that a lawsuit needs to be credible to count.
Were it to be true, this would be lower than the overall % of Americans with criminal charges, which is about 1/3 of adults. This doesn't include those who've been defendants in a civil lawsuit, which I can't find good statistics for. I can't find overall statistics for criminal convictions but 8% of American adults have felony convictions. Based on my experience as a lawyer, it's surprising what kinds of minor bs count as a felony but I'd still be surprised if misdemeanor convictions don't outnumber felonies (with substantial overlap), so I'd guess at least 16% have some kind of conviction. Miscount civil judgments as "convictions" and tons more people get thrown in.
Claim 2.--the population base rate is about 5% over a lifetime.
Claim 3. In addition to the problems noted with 1., it's unclear what constitutes "substantial misconduct". Crimes at least have a clear definition. Civil violations of the law have a clear-ish definition.
"Blatantly false" feels a bit hyperbolic but sure, if you have an alternative word for "convicted" that covers both criminal and civil cases I would be much obliged. I asked some lawyers and looked around online and couldn't find anything, but I agree it would be better if this information was in the main text instead of a footnote, all else equal.
I don't know but I would feel pretty conflicted about working on a grant from the Sackler family. I am even a bit unsettled about a lot of EA work being funded by Meta who is facing lawsuits e.g. about child addiction to their platform. I even feel uneasy writing this, but when I do, I feel an even stronger urge to say it publicly. I think if fairly criticizing our donors get us in trouble I kind of want to be in trouble.
Thanks for writing this -- I think everyone needs to at least be asking those questions to themselves, no matter what their funding source is! Some questions that might help clarify the conflictedness:
How does the conflictedness compare to the conflictedness (if any) you would feel if you were a business performing services for Meta? Would it matter if those services were product-focused (e.g., as a programmer) vs. background in nature (e.g., landscaping)? This is intended to help dissect whether you're describing a general conflictedness versus something more specific to accepting donations from an organization.
Does the exact nature of the funding affect your sense of conflictedness? On a continuum, you might have funding by Meta itself, funding from a controlling shareholder, funding from someone whose wealth significantly flows from Meta stock that they currently hold, someone whose wealth was significantly made in Meta stock but sold it (+/- significant management control in the past), and so on. The answer might help identify how much of the conflictedness arises from concerns about whitewashing reputation and similar factors that are more prominent on one side of the continuum.
How does the conflictedness compare to the conflictedness (if any) you would feel if you were a business performing services for Meta?
To me, selling services to a bad actor feel significantly more immoral than receiving their donation, since selling a service to them is much more directly helpful to them.
(This is not a comment on how bad Meta is. I do not have an informed opinion on this.)
Yeah those are good questions. I have already helped build wind farms to power the data centers of various big tech companies so my conscience is by no means "clean"! Moreover, I own stock, especially via index funds in Meta and other organizations. I also use and even rely on Meta, Alphabet, etc. for a lot of good in my life and make other purchases that are far less ethical. In the end, I think my analysis would show that no matter what, I am embedded in this super connected and complex society where a lot of different interests are pulling in different directions and I am part of that intricate web trying to nudge, in a very miniscule way, our shared future towards something a bit more hopeful and a bit more secure. Still, I would be worried if I get in trouble from being honest about this, and publicly honest about my uncertainty about whether my sources of funding might make my philanthropic endeavors net positive or not. I think for now it is net positive, but I am quite uncertain about this, especially if e.g. I would receive substantial funding from Meta due to an increase in its share price due to its push for open source, risky AGI. It would be fantastic if there was some rating of philanthropic sources that indicated the likelihood that using such funding would be ok or not. But I think that's a big ask and with the amount of risk and suffering around us perhaps it is better to just try to do something good.
Thanks for the feedback! The point I'm trying to get across is something like "if you take donations from an ultra high net worth donor, you probably shouldn't be that surprised if they end up having committed misconduct." I would like to mostly remain agnostic as to the question of whether you should be more or less surprised about UHNW donors committing misconduct than some other reference class (with the exception of the YCombinator thing that I included).
Is there a better way to phrase this? I understand the assumption that "high" is relative to some other reference class rather than just being high in absolute, but I'm not sure what other word to use.
That makes sense! After SBF, I've wondered about the causality around charitable giving and criminality. If GP signatories are more criminal than other billionaires, that would suggest that individuals may be accepting criminality that they can justify with altruism. But if they're similarly criminal to other rich people, and rich people are more criminal than other groups, then that suggests a broader societal problem that we should adjust our legal system to properly disincentivize criminality.
1. Often, things become crimes that get prosecuted when they are done by the wealthy vs. normal people. To be clear, the reason for this is that governments/prosecutors want money and there is a lot of money in going after Kjell Inge Røkke for an illegal boating license but there isn't for a father letting his 15-year old child drive in a parking lot. There's a lot of money going after a billionaire for tax evasion but not in someone having a side hustle where they make money under the table selling $50k worth of widgets per year. This is going to lead to billionaires' actions being surveilled more and thus gone after for crimes more often than the average person. The reward makes it worth it. Billionaires will have far more legal/monetary resources and thus you should naively expect more settlements, particularly without an admission of wrongdoing.
I suppose I recommend people think something like "ok, how bad was this really" when they look at billionaire crimes. Perhaps a litmus test of "would I remain friends with my average non-EA friend if I found out that they had done this"
2. The Giving Pledge has no restrictions on the charitable causes that signatories are allowed to support. As such, unfortunately, as you would likely expect, the vast majority of donations basically go to vanity projects to improve self-image that have little to do with trying to improve the world, let alone try to do so effectively. The median philanthropic gift in terms of impact is something like a generic climate organization with many philanthropic gifts going to building a building at a college to get their name on it or arts museums and such. A non-negligible amount is also tax evasion.
I suppose I recommend people think something like "ok, how bad was this really" when they look at billionaire crimes.
And also billionaire acquittals! Convictions just don't correlate that well with commonsense morality, in both directions.
This project has honestly made me kind of nihilistic regarding interpreting white-collar crime; it seems like you often need to spend a very substantial amount of time digging into the details to actually understand how you should update on someone's behavior. (I found this book helpful, for those interested in reading more.)
Point 1 strikes me as probably incorrect. While there may be more money to get in going after higher net worth individuals, it is also far more expensive because they can set themselves up with excellent legal representation. Unfortunately I don't have a reference for this but I have a vague recollection of reading in multiple places that e.g. gutting the IRS disproportionately benefits the wealthy because the IRS then does not have the funding to pursue more complex and better defended cases, even though there is more money to be had there.
Similarly for other crimes, including things that are obivously wrong and not just 'would I feel that bad if my friend did it' level, I would expect there to be less of this reaching prosecution among the very wealthy than among average people who committed similar crimes (such as sexual or generally violent assault, especially when committed against people who are not themselves very wealthy)
Data seems to indicate otherwise. https://www.gao.gov/products/gao-22-104960 https://www.usatoday.com/story/money/2023/11/09/irs-uses-funding-to-audit-wealthy/71486513007/
Sorry, but in your first link it literally says "But, audit rates have dropped for all income levels—with audit rates decreasing the most for taxpayers with incomes of $200,000 or more. IRS officials said audit rates declined due to staffing decreases and because it takes more staff time and expertise to handle complex higher-income audits." So I am confused by that reference, as it seems to be exactly what I said.
My general point was however not about taxes specifically, I just thought taxes might be one example. I am more generally surprised that people think for crimes overall that the wealthy are more persecuted. My impression is that most of human history indicates the opposite.
I misunderstood, perhaps. Audit rates are primarily a function of funding - so marginal funding goes directly to those audits, because they are the most important. But if the US government wasn't being insane, it would fund audits until the marginal societal cost of the audit was roughly equal to the income by the state. The reason I thought this disagreed with you point is because I thought you were disagreeing with the earlier claim that "This is going to lead to billionaires' actions being surveilled more and thus gone after for crimes more often than the average person. The reward makes it worth it. Billionaires will have far more legal/monetary resources and thus you should naively expect more settlements, particularly without an admission of wrongdoing." This seems to be borne out by the model it seems we're both agreeing with, that higher complexity audits are more worthwhile in terms of return.
And yes, I think that per person, the ultra-wealthy, rather than just the people making 200k+, are far more likely to be investigated and prosecuted for white collar crime, because they get more scrutiny, rather than based on whether or not they commit more crime, even though they are more likely to get off without a large punishment via lawyers. But the analysis looked at rates of guilt, not punishment sizes.
I agree that less money to the IRS will benefit the wealthy more but this is because the IRS already mainly audits wealthier people and thus less audits disproportionately reduces audits on the wealthy. This is similar to how tax cuts nearly always help the wealthy... because the wealthy pay the vast majority of taxes. It's nearly impossible to cut taxes without the wealthy paying far less dollars in taxes than a middle class/poor person.
While more expensive to pursue cases against the wealthy, it is more than worth it when it comes to the increase in reward.
I previously estimated that 1-2% of YCombinator-backed companies with valuations over $100M had serious allegations of fraud.
While not all Giving Pledge signatories are entrepreneurs, a large fraction are, which makes this a reasonable reference class. (An even better reference class would be “non-signatory billionaires”, of course.)
My guess is that YCombinator-backed founders tend to be young with shorter careers than pledgees, and in part because of this will likely have had fewer run-ins with the law. I think better reference class would be something like founders of Fortune 500/Fortune 100 companies.
The YCombinator search also appears to have focused on individual young and apparently successful companies rather than all their founders' activities. Branson's conviction pre-dated the Virgin businesses that actually made him rich, and his alleged sexual misconduct on his private island wouldn't show up in a company search either. I'm not sure the YC search was sufficiently in-depth to uncover founders guilty of yacht license violations, a day in jail for non-business-related offences or alleged sexual misconduct outside business hours.
A 10% conviction rate for financial misconduct still looks very high though.
Yep that all seems believable to me. I also noted the age thing as possible explanation (c):
Giving Pledge signatories are older, and have had more time to accumulate charges.
I wouldn't be that surprised to learn that giving pledge signatories have a lower rate of misconduct than some reference class, so I don't want to make a strong claim here; I'm mostly just noting that "altruistic people don't commit crimes" doesn't seem like a likely hypothesis.
I agree with your meta point (even though I half expected to read the opposite, in which case, I'd have had a point to make about the list being curated)
One thing I will say is that the list probably isn't at all representative of "altruistic people" or "people who join EA groups". Not only is it a set of extremely wealthy people, but there's no actual obligation of anyone on the list to follow through with their pledge or do anything genuinely altruistic at all. I'm sure many are sincere about it, probably including the likes of Branson with actual convictions to their names, but for others it may just be relatively straightforward form of reputation laundering without the expense of actually giving the money away!
Fair point about the company age, and yeah I agree that this list is not representative of "people who join EA groups" (for many reasons), but my intuition is that these people actually are relatively altruistic by commonsense morality standards. Anil Agarwal stuck out to me as arguably the most egregious person on this list, and he pledged 75% of his wealth to charity. Milken is maybe number two, and he's given > $1B. And of course SBF is probably high in the list of egregiousness, and seemingly was sincere in giving lots of money to charity. (Though note that SBF wasn't in my data set because the giving pledge kicked him out.)
I would be interested in someone filtering this list by "people who have actually given >$X" and seeing if that changes the results though.
From a charity/movement perspective, one might care mostly about the expected rate of moderate-to-major scandals[1] from previously non-scandalous[2] donors per donor-year ("per-donor yearly risk"). Many of the Giving Pledge folks have been very wealthy for several decades, so they add either a 0 or 1 to the numerator and may add 20, 30, or even 40 to the denominator. As a toy example only, if 10% of movement X's donors have experienced moderate-to-major scandals since becoming wealthy,[3] and they have been wealthy for an average of two decades each, the per-donor yearly risk can be estimated at 0.1/20 years = 0.5% per donor-year.
It strikes me that the per-donor yearly risk for the largest EA donors could be ~an order of magnitude above the risk faced by charitable communities and organizations as a whole. Many of them are fairly young for high-wealth people and thus have not been wealthy for many years. Thus, the denominator is pretty low. In particular, Ben Delo and SBF only had a few years of being very wealthy each before being convicted of felonies, adding one to the numerator but only a few to the denominator.
That brief sketch suggests that EA may have a much higher risk of unforeseen donor scandal than other charitable movements (e.g., American higher education). This probably stems from age of donors and involvement in crypto, but the sample size is too small to draw any real conclusions as to cause. If this possibility is correct, it strengthens the idea that EA orgs need to be well prepared for major donor scandals.
I am using the term "moderate-to-major scandal" rather than "conviction" as a means of encompassing settlements on unfavorable terms (think Prince Andrew, according to reports), bad things that couldn't be prosecuted due to legal issues (think most of Bill Cosby's crimes), and exceedingly slimy behavior that isn't actually illegal.
I am excluding donors who have previously been the subject of a scandal from the population here. For such a donor, the charity/movement should not be relying on base rates as much because the existence of a prior scandal should greatly alter its risk analysis.
Just focusing on the reputational effects, my quick guess is that the extent to which a donor/public figure is ~memetically connected to the movement/charity is really important, and I expect that most Giving Pledge signatories are significantly less connected to the Giving Pledge (or American higher education donors to American higher education) than the biggest ~5 EA donors are to EA. (Note that "EA donor is also a poorly defined category.) Without considering this factor, we might conclude that getting more donors adds significantly/unchangeably to ~scandal-based risk levels, or that the number of donors is the key thing to consider. I think this would be a false conclusion; more donors probably means that each is less central (which I think would also have the benefit of reducing the influence of any given donor) and because the public profile of the donor(s) probably matters a lot. (I do think there are separate ethical issues and considerations involved in taking funding from UHNW donors.)
Gina and I eventually decided that the data collection process was too time-consuming, and we stopped partway through.
Josh You and I wrote a python script that searches Google for a list of keywords, saves the text of the web pages in the search results, and shows them to GPT and asks it questions about them from a prompt. This would quickly automate the rest of your data collection if you have the pledge signers in a list already. Email me if you want a copy.
You can open the web console on the linked page https://givingpledge.org/pledgerlist and type $$(".pledger-list-item").map(e => e.textContent); (or you can just copy-paste the list directly)
[ "Bill Ackman and Neri Oxman", "Tegan and Brian Acton", "Margaret and Sylvan Adams", "Anil Agarwal", "Leonard H. Ainsworth", "Paul G. Allen (d. 2018)", "HRH Prince Alwaleed Bin Talal Bin Abdulaziz AlSaud", "Brian Armstrong", "Sue Ann Arnall", "Laura and John Arnold", "Marcel Arsenault and Cynda Collins Arsenault", "Lord Ashcroft KCMG PC", "Jon and Helaine Ayers", "Stewart and Sandy Bainum", "Sarah and Rich Barton", "Lynne and Marc Benioff", "Nicolas Berggruen", "Manoj Bhargava", "Aneel and Allison Bhusri", "Sheikh Dr. Mohammed Bin Musallam Bin Ham Al-Ameri", "Steve Bing (d. 2020)", "Sara Blakely", "Arthur M. Blank", "Nathan and Elizabeth Blecharczyk", "Michael R. Bloomberg", "David G. Booth", "Richard and Joan Branson", "Eli (d. 2021) and Edythe Broad", "Charles R. Bronfman", "Edgar M. Bronfman (d. 2013)", "Warren Buffett", "Charles Butt", "Garrett Camp", "Steve and Jean Case", "John Caudwell", "Brian Chesky", "Ron and Gayle Conway", "Scott Cook and Signe Ostby", "Lee and Toby Cooperman", "Joe and Kelly Craft", "Joyce and Bill Cummings", "Ravenel B. Curry III", "Benoit Dageville and Marie-Florence Dageville", "Ray and Barbara Dalio", "Jack and Laura Dangermond", "John Paul DeJoria", "Ben Delo", "Mohammed Dewji", "Barry Diller and Diane von Furstenberg", "Ann and John Doerr", "Dagmar Dolby", "DONG Fangjun", "Glenn and Eva Dubin", "Anne Grete Eidsvig and Kjell Inge Røkke", "Ric and Brenda Elias", "Larry Ellison", "Henry Engelhardt, CBE and Diane Briere de L'Isle-Engelhardt, OBE", "Candy and Charlie Ergen", "Judy Faulkner", "Charles F. Feeney (d. 2023)", "Andrew and Nicola Forrest", "Ted Forstmann (d. 2011)", "Melinda French Gates", "Phillip and Patricia Frost", "Mario and Regina Gabelli", "Mala Gaonkar", "Bill Gates", "Joe Gebbia", "Dan and Jennifer Gilbert", "Dame Ann Gloag", "Dave Goldberg (d. 2015) and Sheryl Sandberg", "Jeremy and Hanne Grantham", "David and Barbara Green", "Jeff T. Green", "Jeff and Mei Sze Greene", "Harold Grinspoon and Diane Troderman", "William Gross", "Gordon and Llura (d. 2020) Gund", "Sir Stelios Haji-Ioannou", "Harold Hamm", "Nick and Leslie Hanauer", "David and Claudia Harding", "Gordon V. Hartman", "Reed Hastings and Patty Quillin", "Lyda Hill", "Barron Hilton (d. 2019)", "Orion and Jackie Hindawi", "Reid Hoffman and Michelle Yee", "Chris and Kylie Hohn", "Elie and Susy Horn", "Sir Tom and Lady Marion Hunter", "Jon (d. 2018) and Karen Huntsman", "Yan Huo", "Dr Mo Ibrahim", "Carl Icahn", "Jared and Monica Isaacman", "Joan and Irwin Jacobs", "Badr Jafar and Razan Al Mubarak", "Sonia and Paul Tudor Jones", "John W. Jordan II \"Jay\"", "Ryan D. “Jume” Jumonville", "George B. Kaiser", "Nikhil Kamath", "Brad Keywell", "Vinod and Neeru Khosla", "Bongjin Kim and Bomi Sul", "Miseon Hyeong and Beom-su Kim", "Sidney Kimmel", "Rich and Nancy Kinder", "Robert E. “Bob” and Dorothy “Dottie” King", "Beth and Seth Klarman", "Robert and Arlene Kogod", "Michael Krasny", "Elaine and Ken Langone", "Erica and Jeff Lawson", "Liz and Eric Lefkofsky", "Francine A. LeFrak and Rick Friedberg", "Gerry (d. 2018) and Marguerite Lenfest", "Peter B. Lewis (d. 2013)", "Daoming Liu", "Iza and Samo Login", "Lorry I. Lokey (d. 2022)", "George Lucas and Mellody Hobson", "Richard and Melanie Lundquist", "Duncan and Nancy MacMillan", "Alfred E. Mann (d. 2016)", "Joe and Rika Mansueto", "Bernie and Billi Marcus", "Richard Edwin and Nancy Peery Marriott", "Strive and Tsitsi Masiyiwa", "Kiran Mazumdar-Shaw", "John and Marcy McCall MacBain", "Craig McCaw", "Red (d. 2023) and Charline (d. 2019) McCombs", "PNC and Sobha Menon", "Dean and Marianne Metropoulos", "Alya and Gary K. Michelson, M.D.", "Michael and Lori Milken", "Yuri and Julia Milner", "George P. Mitchell (d. 2013)", "Thomas S. Monaghan", "Gordon (d. 2023) and Betty (d. 2023) Moore ", "Tashia and John Morgridge", "Michael Moritz and Harriet Heyman", "Dustin Moskovitz and Cari Tuna", "Patrice and Precious Motsepe", "Elon Musk", "Jahm Najafi", "Arif Naqvi", "Jonathan M. Nelson", "José Ferreira Neves", "Rohini and Nandan Nilekani", "Gensheng Niu", "Pierre and Pam Omidyar", "Natalie and Paul Orfalea", "Bernard and Barbro Osher", "Bob and Renee Parsons", "Jim Pattison", "Ronald O. Perelman", "Jorge M. and Darlene Perez", "Melanie Perkins and Cliff Obrecht", "Peter G. Peterson (d. 2018)", "T. Boone Pickens (d. 2019)", "Victor Pinchuk", "Mark Pincus", "Hasso Plattner", "Vladimir Potanin", "Azim Premji", "Tom and Theresa Preston-Werner", "Ernest and Evelyn Rady", "Terry and Susan Ragon", "Emily and Mitchell Rales", "Chad Richison", "Julian H. Robertson, Jr. (d. 2022)", "David Rockefeller (d. 2017)", "Edward W. (d. 2016) and Deedie Potter Rose", "Stephen M. Ross", "Jeff and Marieke Rothschild", "David M. Rubenstein", "Chris and Crystal Sacca", "David Sainsbury", "John and Ginger Sall", "Henry and Susan Samueli", "Herb (d. 2019) and Marion (d. 2012) Sandler", "Vicki Sant (d. 2018)", "Jack and Renate Schuler", "Lynn Schusterman", "Steven Schuurman", "Stephen A. Schwarzman", "Paul and Jennifer Sciarra", "MacKenzie Scott", "Ruth and Bill Scott", "Walter Scott, Jr. (d. 2021)", "Tom and Cindy Secunda", "Dr. B.R. Shetty and Dr. C.R. Shetty", "Ben and Divya Silbermann", "Craig Silverstein and Mary Obelnicki", "Annette and Harold (d. 2013) Simmons", "Jim and Marilyn Simons", "Liz Simons and Mark Heising", "Nat Simons and Laura Baxter-Simons", "Paul E. Singer", "Jeff Skoll", "Robert F. Smith", "John A. and Susan Sobrato", "John and Timi Sobrato", "Michele B. Chan and Patrick Soon-Shiong", "Ted (d. 2016) and Vada (d. 2013) Stanley", "Mark and Mary Stevens", "Tom Steyer and Kat Taylor", "Harry H. Stine", "Jim (d. 2014) and Virginia Stowers (d. 2021)", "Dato' Sri DR Tahir", "Vincent Tan Chee Yioun", "Nicolai and Katja Tangen", "Tad and Dianne Taube", "Robert (d. 2023) and Jane Toll", "Claire (d. 2014) and Leonard Tow", "Dennis Troper and Susan Wojcicki", "Byron and Tina Trott", "Ted Turner", "Albert Lee Ueltschi (d. 2012)", "Hamdi Ulukaya", "Sunny and Sherly Varkey", "Shamsheer and Shabeena Vayalil", "David Vélez and Mariel Reyes", "Dr. Romesh and Kathleen Wadhwani", "Jian Wang", "David and Bonnie Weekley", "Sanford and Joan Weill", "Dr. Herbert and Nicole Wertheim", "Shelby White", "Simone and Urs Wietlisbach", "Andrew Wilkinson and Zoe Peterson", "Anne Wojcicki", "Sir Ian Wood", "Hansjörg Wyss", "Patti Bao and Tony Xu", "Samuel Yin", "YOU Zhonghui", "Charles Zegar and Merryl Snow Zegar", "Mark Zuckerberg and Priscilla Chan" ]
Feelings: I do not feel good reading this. I guess I feel defensive.
This is interesting. My general view is as follows:
I would recommend GiveWell take money from anyone who has finished earning it. GiveWell's work seems pretty inflexible and if the person is done with their activities then it's just making the world better.
I would recommend caution in regard to big donors. I think (though hard nosed friends disagree) that a major error in regard to EA and SBF was that many EAs happily let him use the brand without asking if they were trading something away. This is hard to think about. The counter is something like "people can call themselves what they want and people can take ideas that they want". But personally I think there was an assumption that we had vetted SBF (not least because I think we thought we had).
I would be more cautious in smaller orgs, more flexible orgs. If I were to be regularly taking money from a russian oligarch who had retired, it feels like this might cause me to me more relaxed towards his former harm. My reasons for being a vegan are roughly this - I find i think more clearly about animal welfare when I am not eating them. And I like thinking clearly. Just as honesty has big costs that may be worth it, I wouldn't be suprised if $100mns were not worth becoming sloppy about the damage of fraud.
I would not want any EA orgs to take money from someone who was still doing their crime. To me, that clearly looks like we are justifying or incentivising more of it.
We should be happy to call out their crimes. I would be wary of an EA that isn't capable of talking about billionaire crime for fear of upsetting billionaires. The rule of law is really important and I've heard Google, Amazon, Musk all doing pretty immoral and defecty things in order to make parts of their fortunes. Even if most are still positive on net, ill gotten gains should be fined off them, regardless of their giving.
So I think that perhaps GiveWell, GiveDirectly, etc should happily take money from these poeple regardless, but other orgs should be more careful. And if you find you can't say true things about someone for fear or not getting their money, I would be concerned about that.
That kind of spontaneous research guided by ethics is really one of my favourite aspects of EA. Thank you for undertaking this research and highlighting the very unhealthy dynamics of power exercised by those at the top. We know them, without really knowing about them: a reminder from times to times never hurt.
Personally I don't feel attacked at all: I am happy that these people give money for useful purposes, and that has nothing to do with calling them out for their behaviours or not. I don't believe in the 'don't bite the hand that feeds you' kind of thinking. It's our role, as EAs who benefit more or less distantly from this money, to be very aware of who gives the money and what price has to be paid for that.
When people criticize EA because they are choked that we accept being fed by blood money (this includes billionaires holding large shares in companies that exploit workers), I have no moral qualms to say that I prefer doing something good with this money rather than nothing; I am still working on structural changes with that money. As long as I haven't read any convincing. data-supported report on how being in EA and using these funds is more harmful than working only at a structural level to change these dynamics, I think that EA remains my best bet.
25% of signatories have been accused of financial misconduct, and 10% convicted
That's really high?? Oh - this is not the giving what we can pledge😅
I estimate that Giving Pledgers are not less likely, and possibly more likely, to commit financial crimes than YCombinator entrepreneurs.
At what stage of YC? I guess that will be answered later. EDIT:
I previously estimated that 1-2% of YCombinator-backed companies with valuations over $100M had serious allegations of fraud.
.
Gina and I eventually decided that the data collection process was too time-consuming, and we stopped partway through. The final dataset includes 115 of the 232 signatories.
Random, alphabetical, or date ordered? Not that it will really matter - although I guess I would expect the earlier pledgers to be more altruistic, maybe more risk taking though.
I found that the punishment of the criminals in my data set correlated extremely poorly with my intuition for how immorally they had behaved.
It would be funny if it weren’t sad that one of the longest prison sentences in my data set is from Kjell Inge Røkke, a Norwegian businessman who was convicted of having an illegal license for his yacht.
Ohhh ok 😂😅 Yeah that is funny and sad.
[Milken] was pardoned by Donald Trump in 2020.
😑
While not all Giving Pledge signatories are entrepreneurs, a large fraction are, which makes this a reasonable reference class. (An even better reference class would be “non-signatory billionaires”, of course.)
Agree
Despite this, I can find very little criticism referencing the fact that many of these signatories are criminals.
This is interesting, and is naturally raised by this post (v. interesting by the way). It makes me wonder about their screening practices. I'm guessing a random like me can't sign up (they check one's net wealth somehow?) but perhaps that's all? If any billionaire can sign up, then maybe it's not really the giving pledge that one should criticize?
Rates of Criminality Amongst Giving Pledge Signatories — EA Forum
This is a crosspost from the new Animal Welfare Alignment Newsletter by Anima International. You can subscribe on Substack if you are interested in following these efforts. Audio reading also available on Substack.
The goals of this post are to:
1. Raise a question I see as crucially important to the goal of aligning AI to animal welfare...
Hello! I'm Justin Portela. I got hired by GWWC to make YouTube videos after AI in Context did such a kickass job.
My channel is using that same cinematic, high-production value beauty to talk about everything in the EA universe that isn't AI.
...
“How long have you been v*g*n?”
This is one of the most common icebreakers at animal protection events. It’s a baseline assumption, and it mostly holds true: if you’re out advocating for animals not to be tortured or abused, realistically these days you are v**n, or close. And it makes for good conversation. It seems fairly safe to assume when you meet strangers.
But this assumption is hurting the movement in a way which we don’t always notice: someone new comes into the sp...
I investigate the rates of criminal misconduct amongst people who have taken The Giving Pledge (roughly: ~200 [non-EA] billionaires who have pledged to give most of their money to charity).
I find that rates are fairly high:
25% of signatories have been accused of financial misconduct, and 10% convicted[1]
4% of signatories have spent at least one day in prison
Overall, 41% of signatories have had at least one allegation of substantial misconduct (financial, sexual, or otherwise)
I estimate that Giving Pledgers are not less likely, and possibly more likely, to commit financial crimes than YCombinator entrepreneurs.
I am unable to find evidence of The Giving Pledge doing anything to limit the risk of criminal behavior amongst its members, though I have heard second-hand that they do some sort of screening.
I conclude that the rate of criminal behavior amongst major philanthropists is high, which means that we should not expect altruism to substantially lower the risks compared to that of the general population, and that negative impacts to EA’s public perception may occur independently of whether our donors actually commit crimes (e.g. because even noncriminal billionaires have a negative public image).
Methodology
I copied the list of signatories from their website.
Gina Stuessy and I searched the internet for “(name) lawsuit”, “(name) crime” and also looked at their Wikipedia page.
I categorized any results into “financial”, “sexual”, and “other”, and also marked if they had spent at least one day in jail.
Gina and I eventually decided that the data collection process was too time-consuming, and we stopped partway through. The final dataset includes 115 of the 232 signatories.[2][3]
How well do convictions correspond with immoral behavior?
It is a well-worn take that our[4] legal system overly protects white-collar criminals: If an employee steals $20 from the cash register, that’s a criminal offense that the police will prosecute, but if an employer under-pays their employees by $20 that’s a civil offense where the police don’t get involved.
I found that the punishment of the criminals in my data set correlated extremely poorly with my intuition for how immorally they had behaved.
It would be funny if it weren’t sad that one of the longest prison sentences in my data set is from Kjell Inge Røkke, a Norwegian businessman who was convicted of having an illegal license for his yacht.
One particular way in which white-collar offenses are weird is that they often allow the defendant to settle without admitting wrongdoing.[5] E.g. my guess is that Philip Frost is guilty, but his settlement with the SEC does not require him to admit wrongdoing.
I wasn’t able to find a single person who admitted guilt in a sexual misconduct case, despite ~7% of the signatories being accused, including in high-profile cases like people involved with Jeffrey Epstein.[6]
I was considering trying to add some classification like "Ben thinks this person is guilty" but decided that this would be too time-consuming and subjective.
Nonetheless, if you want my subjective opinion, my guess is that most of the people who were accused of financial misconduct are guilty of immoral behavior, under a commonsense morality definition of the term.
Less controversially, some of these cases are ongoing, and presumably at least some of them will result in convictions, which makes looking only at the current conviction rate misleading.
In any case though, I believe that this data set establishes that the base rate of both criminal and immoral behavior is fairly high among major philanthropists, no matter how you slice the data.
Some Example Cases
Here are a few cases to give some flavor of what Giving Pledge signatories have been accused of.
Mario Gabelli
Used sham businesses to fraudulently purchase portions of the US cell phone spectrum. Settled for $130 million. Sued for breach of fiduciary responsibility in an unrelated case, settled for $100 million. https://en.wikipedia.org/wiki/Mario_Gabelli#Legal_Issues
Michael Milken
Milken was indicted for racketeering and securities fraud in an insider trading investigation. Milken was sentenced to ten years in prison, fined $1.1 billion ($2.3 billion in 2023 dollars) and permanently barred from the securities industry. He was pardoned by Donald Trump in 2020.
“In 2004, a committee of the Indian Supreme Court charged that Vedanta had dumped thousands of tons of arsenic-bearing slag near its factory in the Indian state of Tamil Nadu, resulting in major poisoning of the environment and neighbouring population. In 2005, another committee of the Indian Supreme Court charged that Vedanta had forced over one hundred indigenous families from their homes in the Indian state of Odisha, where it sought to mine bauxite. According to the committee's report: "An atmosphere of fear was created through the hired goons", and residents were "beaten up by the employees of M/s Vedanta".”
Are Giving Pledge signatories less likely to commit financial crimes?
I previously estimated that 1-2% of YCombinator-backed companies with valuations over $100M had serious allegations of fraud.
While not all Giving Pledge signatories are entrepreneurs, a large fraction are, which makes this a reasonable reference class. (An even better reference class would be “non-signatory billionaires”, of course.)
We might naïvely think that Giving Pledge signatories (being charitably minded) are less likely to commit financial crimes—but if anything, it seems like the opposite is true.[7]
~10% of signatories being convicted of financial misconduct and ~25% accused versus ~1% of YCombinator companies is a stark difference.
I have a few possible explanations:
Giving Pledge signatories tend to run publicly-traded companies, whereas YCombinator companies are usually private, and it is easier to be charged with financial misconduct if your company is publicly traded.
Giving Pledge signatories are wealthier, and therefore more lucrative targets for prosecution.
Giving Pledge signatories are older, and have had more time to accumulate charges.
This book argues that white-collar criminals generally do not see their behavior as immoral, because humans’ moral intuitions don’t fit a white-collar world. E.g. someone who is unusually moral might be more willing to give money to a homeless person on the street, but they aren’t any less likely to aggressively depreciate assets in a quarterly earnings statement, because the latter does not pattern match to our intuitions of a (im)moral decision.
I have not looked into any of these explanations deeply, but from a pragmatic perspective it seems clear that we should not expect that altruistically-minded HNWIs are notably less likely to commit financial crimes than the average person with the opportunity to do so.
Giving Pledge’s Response to Criminal Behavior
As far as I can tell, the Giving Pledge is quite hands-off, and has not commented on the criminal behavior of some of its members or attempted to prevent criminal behavior. I can’t find a single press release from them saying anything; the profiles of people like Michael Milliken are sitting there without even a small banner saying “we don’t endorse crimes,” etc.[8]
The only thing I am aware of is two cases in which they have silently removed profiles: T. Denny Sanford was removed after accusations of child pornography, and Sam Bankman-Fried was removed after charges of defrauding investors.
I have heard secondhand that they screen signatories beforehand, but cannot find any public documentation of their screening (and, as evidenced by the statistics reported here, any such screening does not seem to be terribly effective).
I’m not surprised that they didn’t make any actually-consequential reforms, but I am surprised that they don’t e.g. issue press releases making some milquetoast statement about crime being bad.
PR Impacts
The Giving Pledge (and billionaire philanthropy more generally) receives a lot of criticism.
Despite this, I can find very little criticism referencing the fact that many of these signatories are criminals.[9]
My (perhaps uncharitable) interpretation is that there are some people who believe that billionaire philanthropy is immoral, and some who don’t, but very few who take the nuanced view that some but not all billionaire philanthropy is immoral.
Someone told me that they think this is a blind spot of EAs in general: The major risk of having wealthy/powerful people support your movement is that they may co-opt it toward their own goals, and that risk is largely independent of whether or not that person commits crimes, so the public is correctly not updating very much based on a certain movement’s donors being criminals.
This fits with information I've heard from university group organizers: University students sometimes distrust EA because of ties to "big tech" or "Silicon Valley billionaires", and the more specific worry that some of these billionaires may have committed crimes does not affect their trust as much.
Implications for EA
It seems fairly likely that any engagement with an ultrahigh net worth donor carries a substantial risk that the donor will turn out to have engaged in criminal behavior or will in the future.
Some EA organizations use services that screen potential donors, but I am skeptical that these services will meaningfully decrease the risk of criminal behavior among these individuals, or be able to accurately identify which people are most likely to commit white-collar crimes.
We should also be aware that engagement with ultrahigh net worth donors carries a risk of damaging EA’s public perception, and that this risk may be partially independent of whether the donors are actually criminal.
Thanks especially to Gina Stuessy for help with this article, but it also benefited from feedback from Jake McKinnon, Zach Robinson, Ben Rachbach, Will MacAskill, Jason, and Julia Wise.
Legal pedantry note: I am including here people who have had civil judgments rendered against them (e.g. lawsuits from the SEC), even though in civil judgments people are technically “adjudged liable”, not “convicted”.
I realized after doing this that the names are listed in alphabetical order by last name, so if there is some relationship between alphabetical position and criminality that bias would be leaking into the data set. This seems unlikely to me, but I guess you never know.
I populated the data for Michael Milken out of order because I found out about him through an earlier research project; he is therefore left out of aggregate statistics
I tried to use my best judgment when marking people as “convicted” or not, but I didn’t have an established set of best practices to rely on, and I think it might be valuable for someone to recategorize these using a more formal method, if any readers are looking for a small research project.
Please note that this is not a perfect comparison. A key difference is that the YCombinator article used a denominator of companies, whereas this looks at individuals. Nonetheless, I would be surprised if it turned out that giving pledgers are substantially less likely than YCombinator-backed companies to commit fraud.
I wouldn’t be surprised to learn that they have some screening process which secretly prevents sketchy people from joining, but I can’t find any reference to it online. And of course we can at least say that this screening process does not seem to be completely effective, given the rate of criminals amongst signatories.
This paper lists numerous criticisms of billionaire philanthropy: it diminishes the role of community-based giving practices, it requires grantees to “assimilate the dominant colonizer white culture”, it undermines democratic control, it increases income inequality, it overly pushes market-based solutions, it influences political decisions through funding think tanks. Nowhere is criminal behavior on behalf of signatories mentioned.
I'm concerned that the control group doesn't seem like a representative comparison. It seems like Giving Pledge signatories are older, richer, more famous, and more successful than the sample of YC recipients.
A better comparison group might be the world's 100 richest people, would it be possible to redo this analysis for them?
I was mostly interested in establishing the base rate; I only included the comparison because it was easy. So I am unlikely to look more into comparisons, but I published my data above and others are welcome to do more analysis!
The base rate seems tricky to interpret without a real comparison. 3 out of 5 points in the summary say that the rate is high, but it doesn't seem like the evidence exists to support that claim. The implied reference class is "non EA super-rich people", but that class isn't actually examined in the data.
The rates do seem subjectively very high. Way fewer than 10% of people I know have been convicted of financial crimes! But I wonder if founders and CEOs are being blamed for financial crimes that their companies commit, and approximately all successful companies commit financial crimes, defined broadly.
One interesting comparison is that the percent who have ever spent a night in jail is similar to the rate for the overall American population.
Overall great data though! It raises real concerns.
I'm curious, do you think the Giving Pledge should have higher standards? My hot take is that we don't want to help launder reputations, but that I would personally be very happy if e.g. the Sacklers donated all their money to a real charity.
Strong downvote on the OP because the counting is done in a HIGHLY misleading way. If I were the moderators I would take the post down or demand SUBSTANTIAL corrections.
For Claim 1. (25% of signatories have been accused of financial misconduct, and 10% convicted[1]), the info provided in the footnote renders this claim outright false. Beyond the fact that many civil violations of the law are not even crimes, civil judgments are not convictions, require a substantially lower standard of proof, and from a legal strategy standpoint, there is often no reason not to settle a case that requires no admission of wrongdoing and where the legal consequences of a criminal conviction are not triggered and it's just not worth the opportunity cost of spending time and money litigating it.
Using the experience of non-US billionaires is also extremely misleading without evidence about base rates of crime in those countries. My guess would be that India, for example, has historically had such high rates of corruption that virtually every Indian over ~50, and certainly those in business, have done things that would result in jail time if done in the US. But that's not morally blameworthy; it's just the system they have to operate in.
Beyond that issue, some more mundane ones:
No evidence is presented for the idea that white collar criminals are less likely to be punished than other criminals. It is probably the case that rich people are less likely to be punished for the same crime than poor people. But the nature of financial crime in publicly traded companies is that the evidence is just way more accessible to the government than the evidence of stealing toothpaste from Target. So financial crimes in publicly traded companies are likely prosecuted at much higher rates than other types of crimes.
Overall the rates listed appear to generally be lower than the general population, which is extremely impressive considering that most billionaires are men, who tend to have much higher rates of criminality. So I would expect billionaires to have almost double the population base rate of criminality just by virtue of that! I don't know why Y-combinator funding is a remotely appropriate reference class.
Let's review the individual claims now:
Claim 1. is blatantly false, as noted at the outset. Beyond that, we have only an explanation, hidden in the footnote, of what constitutes a "conviction". No definition of the meaning of allegation. Based on the stated methodology I assume at least a criminal charge or the filing of a lawsuit to count. No indication on whether e.g. a public allegation with no suit counts, and no indication that a lawsuit needs to be credible to count.
Were it to be true, this would be lower than the overall % of Americans with criminal charges, which is about 1/3 of adults. This doesn't include those who've been defendants in a civil lawsuit, which I can't find good statistics for. I can't find overall statistics for criminal convictions but 8% of American adults have felony convictions. Based on my experience as a lawyer, it's surprising what kinds of minor bs count as a felony but I'd still be surprised if misdemeanor convictions don't outnumber felonies (with substantial overlap), so I'd guess at least 16% have some kind of conviction. Miscount civil judgments as "convictions" and tons more people get thrown in.
Claim 2.--the population base rate is about 5% over a lifetime.
Claim 3. In addition to the problems noted with 1., it's unclear what constitutes "substantial misconduct". Crimes at least have a clear definition. Civil violations of the law have a clear-ish definition.
"Blatantly false" feels a bit hyperbolic but sure, if you have an alternative word for "convicted" that covers both criminal and civil cases I would be much obliged. I asked some lawyers and looked around online and couldn't find anything, but I agree it would be better if this information was in the main text instead of a footnote, all else equal.
I don't know but I would feel pretty conflicted about working on a grant from the Sackler family. I am even a bit unsettled about a lot of EA work being funded by Meta who is facing lawsuits e.g. about child addiction to their platform. I even feel uneasy writing this, but when I do, I feel an even stronger urge to say it publicly. I think if fairly criticizing our donors get us in trouble I kind of want to be in trouble.
Thanks for writing this -- I think everyone needs to at least be asking those questions to themselves, no matter what their funding source is! Some questions that might help clarify the conflictedness:
To me, selling services to a bad actor feel significantly more immoral than receiving their donation, since selling a service to them is much more directly helpful to them.
(This is not a comment on how bad Meta is. I do not have an informed opinion on this.)
Yeah those are good questions. I have already helped build wind farms to power the data centers of various big tech companies so my conscience is by no means "clean"! Moreover, I own stock, especially via index funds in Meta and other organizations. I also use and even rely on Meta, Alphabet, etc. for a lot of good in my life and make other purchases that are far less ethical. In the end, I think my analysis would show that no matter what, I am embedded in this super connected and complex society where a lot of different interests are pulling in different directions and I am part of that intricate web trying to nudge, in a very miniscule way, our shared future towards something a bit more hopeful and a bit more secure. Still, I would be worried if I get in trouble from being honest about this, and publicly honest about my uncertainty about whether my sources of funding might make my philanthropic endeavors net positive or not. I think for now it is net positive, but I am quite uncertain about this, especially if e.g. I would receive substantial funding from Meta due to an increase in its share price due to its push for open source, risky AGI. It would be fantastic if there was some rating of philanthropic sources that indicated the likelihood that using such funding would be ok or not. But I think that's a big ask and with the amount of risk and suffering around us perhaps it is better to just try to do something good.
Thanks for the feedback! The point I'm trying to get across is something like "if you take donations from an ultra high net worth donor, you probably shouldn't be that surprised if they end up having committed misconduct." I would like to mostly remain agnostic as to the question of whether you should be more or less surprised about UHNW donors committing misconduct than some other reference class (with the exception of the YCombinator thing that I included).
Is there a better way to phrase this? I understand the assumption that "high" is relative to some other reference class rather than just being high in absolute, but I'm not sure what other word to use.
That makes sense! After SBF, I've wondered about the causality around charitable giving and criminality. If GP signatories are more criminal than other billionaires, that would suggest that individuals may be accepting criminality that they can justify with altruism. But if they're similarly criminal to other rich people, and rich people are more criminal than other groups, then that suggests a broader societal problem that we should adjust our legal system to properly disincentivize criminality.