KnotGodel
6mo ago
Yeah - along with the related:
- "evolution says I should care, so I will"
- "intelligence uber ales"
Alas, I'm still working through my own utilitarian/math unhealthiness, so I shouldn't cast too many stones.
I guess it depends what you mean by EA.
I haven't seen evidence that the core EA organizations (GiveWell, Evidence Action, 80000 hours, etc) suffer from "extreme overconfidence". As far as I can tell, all are large net-positives for the world that don't really engage in huge risk-taking behavior - if anything, I think the opposite is true (e.g. Evidence Action shutting down one of its programs, since it was less effective than others).
The median EA survey-taker donates to global health charities and not anything weird, which seems pretty much the opposite of "extreme overconfidence". You could argue the weirder causes (animal welfare, AI, etc) display "extreme overconfidence", but I think most of them would say they're donating, not because they're extremely convinced that, say, AI will destroy us all and only MIRI can stop them - I suspect most are donating, because they think there is at least a small chance (~5%) this is true and its worth taking seriously - it doesn't seem like "extreme overconfidence" is the mistake being made here. [Edit: or if there is a mistake at all; I remain agnostic]
So, I don't really see why SBF should be considered a central example of an effective altruist.
a whole mess of people falling into the exact failure mode that critics of their approach predicted
What do you mean?
Status is a person's placement in a social hierarchy
I guess, among the people I spend time with (friends, family, coworkers), it does not feel to me like status is very coherent concept. Like, at work, there is a formal hierarchy whereby a manager has more power than those beneath them, but plenty of non-managers will be deferred to over managers. Moreover, who is deferred to is subject-dependent and, even there, we talk about things says more about who acts most confident and is more argumentative. None of the above is really related to how much people talk to you at lunch or who gets invited to socialize after work. So, to summarize:
- the formal status hierarchy only correlates somewhat with who is listened to most
- "how much people listen to you" is itself not terribly consistent between topics
- how much people defer to you about work isn't correlated at all with how much they want to talk to you about non-work things or socialize with outside of work
And even on that last point, much more of the variance is related to whether you display an interest in talking about non-work things or an interest in hanging out outside work.
Moreover, there are clear cliques, and I don't really get a sense that some cliques are preferable to others. Cliques are, as far as I can tell, the opposite of a status hierarchy: a group of peers socializes frequently because of idiosyncratic socializing value they provide each other (e.g. shared interest, complementary personalities, etc)
Personally, I don't really have a desire to be "popular" - I have a desire for specific people to continue liking me or, occasionally, start liking me.
Among friends and family, I'd say a "status hierarchy" is even less coherent.
I will admit that, in high school and for dating, I do think there is some coherence in a hierarchy of sorts, but I have not experienced that dynamic much outside those contexts.
Re banging your wife, we can add your peer group to the dynamic - do you think their opinion of you would change at all if I banged your wife in front of them?
I think their opinion of you and me would both change pretty drastically, though I think their opinion of you would fall far more. But I don't think "status" is the best framing of the situation. You having sex with my wife in front of them would cause them to feel incredibly awkward and (I'd hope) sorry for me. You'd associate me with negative emotions, which would make it less desirable for them to socialize with me.
But, again, I don't think any of this is zero-sum. I don't think if I went to a separate group the "low-status" would follow me. I don't think I would be significantly listened to less at work. etc.
That is, to the extent to which I have "status" (outside high school and dating) it seems to very dependent not just on me, but the the judge and also the context. I don't think the uni-dimensional model has much value to offer.
If you walked in on that what would you think my opinion of you was?
Not great? The affair would reveal you're narcissistic and think little of me.
What about your wife - if you saw that would you immediately assume she loved you as much as she did on your wedding day?
I would think she loved me less. But, I don't see that as an affront of my "status" writ large - I see it as her (a specific person) loving me less.
Edit: in case it wasn't clear, I'm interpreting the "status" theory similar to the g-factor intelligence theory: as a relatively stable common factor that explains significant variance in various other variables, which implies there should be significant covariance between
- how much Bob likes/respects you and how much Dan likes/respects you (stability over judge - i.e. "objectivity")
- how much Bob likes/respects today versus tomorrow (stability over time)
- how much Bob respects you in one context versus another (stability over domain)
I conversely believe the above are (a) not very will correlated and (b) significantly under your control (i.e. you laugh, smile, ask people genuine questions above themselves, are generous, talk to strangers, ask people to go bowling with you, etc ==> people will want to spend more time with you)
KnotGodel
6mo ago
For one, it's perfectly calibrated in terms of probabilistic forecasting or predictions
Is there a link for that?
Is that a dodge, or are you actually saying that you wouldn't feel like you lost status if I banged your wife in front of you?
Could you maybe describe what "status" means to you?
I don't really walk around thinking "I should to X at work to gain status" or "I should make fun of Y to gain status" or "Person Z lowered my status in that meeting - I've got to be sure to get even with them." I don't think that, in order to have more/closer friends, it is important that I become more popular than someone else. I do occasionally feel embarrassed (e.g. I said something wrong in a meeting) or ashamed (e.g. I forgot about a friend's birthday).
I guess I just don't think any of these as "zero-sum".
Even in the "banged your wife" scenario - does that give you status? I don't think it would among my peer group... Would I become less popular? Would people at work think I was less competent? I don't think the effect would be very large...
Do people actually think like that? To me, it doesn't seem like a good way to approach life from either a personal-happiness perspective or a social-welfare perspective. I don't know, I find the amount of emphasis you're placing on its importance confusing. So, I see three options:
- There is some disconnect between what you and I mean by "status"
- I actually do care immensely about status - I'm just repressing it.
- Some people viscerally care a great deal about status. Others don't.
KnotGodel
6mo ago
You also don't have to feel like you are losing status if I fuck your wife in front of you, or force you to blow me, but I would suggest not doing so demonstrates a lack of self respect
IMO, the problem with both of those is not that I'm losing status.
Where is “p”?
Sorry. I ended up using "x" in the actual math. I've edited the post to be correct
I’m gonna need a template for these equations
You might find the Modern portfolio theory article more useful or even his original paper (scihub). If you want a quick explanation
- We can approximate any reasonable utility function using a second-order Taylor series
- We can approximate investments as normal distributions
- The expected value of a 2nd order polynomial normal random variable, X, equals (up to linear transformation) E[X] - e * Var[X], where e is a parameter determined by the 2nd order polynomial and the normal distribution
- Therefore, we can approximate the problem of "choose an optimal portfolio" as "choose the portfolio that optimize E[X] - e*Var[X]
- Once we've made that leap, we can use the properties of expected value and variance to convert (a) a vector of expected values and (b) a covariance matrix of returns into an optimal portfolio
Obviously, the usual caveat applies: all models are wrong, some are useful.
[ Edit: in case it wasn't clear, I'm saying a wealth tax is better than a capital gains tax in that it doesn't distort risk-taking while a capital gains tax does]
So a 99 % capital gains tax results in everyone investing in stocks?
Yeah. Note: people would probably save less, but what the people are saving would be invested in stocks rather than bonds in this model.
I call bullshit on that. A 100% tax on everything right now is more distortionary.
Sure. I mean that according to ivory-tower theory, even a 1% capital gains tax now is equivalent to a 100% tax on far-future consumption. A 1% tax on labor income or current-consumption doesn't have that pathology.
But we have welfare, and the income tax isn’t flat. You’re very theoretical today.
Right, my main point is that, contrary to textbooks, I don't think poll taxes are actually non-distortionary. I think a poll tax (and its opposite: welfare) is distortionary.
Ignoring the motivating effect of hunger, of course.
What do you mean?
KnotGodel
6mo ago
I think the answer is that "tolerable" is just a poor framing for such conversations.
There are specific policies/actions with specific benefits/costs and they should largely be evaluated separately.
Asking whether trigger warnings are good/bad is silly. Asking whether the MPA should include a "sexist humor" label on relevant films (e.g. like the "graphic violence" label) is a specific change that can be discussed. Asking whether college professors should be fired for not warning a class that a book contains a rape scene is a specific change that can be discussed. etc
People love arguing broad ideas, but insight usually comes from getting down to brass tacks.
KnotGodel
6mo ago
I'm saying psychologically health people don't see status as zero-sum.
I don't have to feel like I'm losing status if slaves are freed.
I don't have to feel like I'm losing status if I stop eating meat.
Any feeling that I'm losing status is a feature of my brain, not the world.
KnotGodel
6mo ago
some Christian charity dedicated to banning abortion is usually happy to switch method to boost efficiency
I'm not sure this is generally true. I think it's usually fairly difficult for nonprofits to admit a program is ineffective. Indeed, one of the reasons I like Evidence Action so much is that they turned down a program (busing farmers to the city during the off season so they could work) that turned out to be less effective than their other programs (deworming and chlorine in water).
But even acknowledging that non-EA nonprofits do sometimes turn down less-effective programs, my main point is that virtually zero Christians will ask themselves which non-profit will actually be most effective at banning abortion.
KnotGodel
6mo ago
What does that mean?... Please be specific.
This is exactly what I wanted him to do, but I was being snarky about it. I'd thank you for being kinder, but...
Have you heard of a guy called "Sam Bankman-Fried?" He was in the news a little bit lately.
A single guy in finance being over-confident is pretty minimal evidence that EAs as a group and as a constellation of organizations suffer from "extreme self confidence".
All else equal, all wealth should be taxed equally (say, flat 1%/y) , not income from wealth. Current tax laws encourage bubbles and poor investing. Just buy a garbage bond or shitcoin and uncle sam will barely touch it, but god helps you if you invest in a company actually making money. And don’t give me the hard-luck grandma story.
Hmm - I haven't heard this take before. Let me do some quick math, courtesy of the Markowitz model. Suppose you have two investment options:
- stocks that returns Norm(A, B)
- risk free interest rate: i
Let e be your risk aversion and let x be the the proportion of your portfolio you are investing in stocks
U = xA + (1-x)i - exxB
dU/dx = A - i - 2exB
x = (A - i) / (2eB)
Capital gains scales returns by k and variance by k^2 (where k = 1 - tax_rate):
x = (kA - ki) / (2ekkB) = (A - i) / (2ekB)
As taxes go up, k shrinks from 1 towards 0, which makes x increase. Therefore, capital gains taxes cause increased risk tolerance.
A wealth tax reduces returns by k:
x = ((A-k) - (i-k)) / (2ekkB) = (A - i) / (2ekB)
This is the same, so a wealth tax doesn't affect risk tolerance.
It’s like a poll tax on wealth, and like a poll tax, it’s very tax efficient.
I don't think this can be true. The chief academic argument against capital gains taxes is that they impose a 100% tax on consumption in the far future, which is maximally distortionary. The same is true of a wealth tax.
The problem with income tax is that it discourages economically beneficial behaviour, like working or good investing.... the state, counter-productively, eggs you on to be a bum and to stack your wealth under the mattress (ignoring inflation). Your lazy bum money should be taxed at least as much as superstar cancer-curing money.
I personally think that if society had no welfare, a flat income tax would be either not distortionary or push people to work more. Note: historically people worked much more and (e.g.) wages being 4x lower because your country is poor is equivalent to a 75% flat tax today.
I have an elegant mathematical model illustrating this result, but I think I've force-fed this forum with enough math already.
KnotGodel
6mo ago
I wonder if the well funded caravans of migrants we see in some areas of the world have to some extend to do with funding related to EA.
I wonder if your wondering is done in good faith 🤔
Then there is Open A.I. and Chat GPT and effective altruists have been influential in Open A.I. Chat GPT has liberal bias. https://www.foxnews.com/media/chatgpt-faces-mounting-accusations-woke-liberal-bias
I think extremely few people (maybe even no one) pursue making LLMs liberally biased for EA reasons.
Climate change and veganism are two issues that could well lead to hardcore authoritarian policies and restrictions.
Since when has a group representing 3% of the population (vegans) taken enough power to implement "hardcore authoritarian policies and restrictions"?
Like with all identity movements, to elevate one group such as animals you end up reducing the position of another group, such as humans
Only for unhealthy minds, I think? Whether freeing slaves "reduced" the position of non-slaves is a question without an objective answer - only psychological interpretations. For instance, many Indians never eat meat and would tell you they don't feel "reduced" by this.
It does seem that at least a few of the people involved with effective altruism think that it fell victim to its coastal college demographics
That post is just describing regression to the mean, which every informal group encounters. Nothing unique to EA here.
My other conclusion related to the open A.I. incident as well is that the idea of these people that they are those who will put humanity first will lead to them ousting others and attempt to grab more power in the future too. When they do so, will they ever abandon it?
The same could be asked about any group with any large goal: companies, nonprofits, religious organizations. Nothing unique to EA here.
That this action is dishonorable matters
How do we know it is dishonorable?
This means that Sam Altman won't be the first.
won't be the last?
It also means that we got a movement very susceptible to the same problems of authoritarian far left movements in general of extreme self confidence to their own vision and will to power.
Do you have evidence EAs suffer from "extreme self confidence"?
This... encourages the power hungry to be part of it as well.
Again, this isn't unique to EA. Any group with money/power attracts the power hungry. What's your point?
KnotGodel
6mo ago
If they believed their altruism was ineffective they wouldn’t do it
Everyone thinks their own altruism is effective. EAs deliberately try to choose the most effective, which the vast majority of people don't consider attempting and act confused if you suggest it.
KnotGodel
6mo ago
Matthew 17:24-27
Ironically, this seems to indicate the opposite: that the church shouldn't expect money from its members, but Jesus does simply to "not cause offense."
Mark 12:41-44
Hmm. I always interpreted this as Jesus condoning generosity as a virtue , but I can see why you'd interpret this as encouraging people to donate to churches specifically.
KnotGodel
6mo ago
From Wikipedia:
Men can potentially have many children with little effort; women only a few with great effort. One argued consequence of this is that males are more aggressive, and more violently aggressive, than females, since they face higher reproductive competition from their own sex than females. In particular, low-status males may be more likely to remain completely childless. Under such circumstances, it may have been evolutionarily useful to take very high risks and use violent aggression in order to try to increase status and reproductive success rather than become genetically extinct. This may explain why males have higher crime rates than females and why low status and being unmarried is associated with criminality. It may also explain why the degree of income inequality of a society is a better predictor than the absolute income level of the society for male-male homicides; income inequality creates social disparity, while differing average income levels may not do so. Furthermore, competition over females is argued to have been particularly intensive in late adolescence and young adulthood, which is theorized to explain why crime rates are particularly high during this period.
KnotGodel
6mo ago
[Warning: Bible nerding]
Giving Wealth
you are supposed to give it to the Jews instead
Well, or the government or the poor. [ Unrelated, but afaict, no one in the New Testament ever encourages donating to a church. ]
Celibacy
celibacy is strongly encouraged
This is debatable.
You point to Luke 20:34-36, but, lets look at the surrounding context. Here is Luke 20:28-36:
“Teacher,” they said, “Moses wrote for us that if a man’s brother dies and leaves a wife but no children, the man must marry the widow and raise up offspring for his brother. Now there were seven brothers. The first one married a woman and died childless. The second and then the third married her, and in the same way the seven died, leaving no children. Finally, the woman died too. Now then, at the resurrection whose wife will she be, since the seven were married to her?”
Jesus replied, “The people of this age marry and are given in marriage. But those who are considered worthy of taking part in the age to come and in the resurrection from the dead will neither marry nor be given in marriage, and they can no longer die; for they are like the angels. They are God’s children, since they are children of the resurrection.
So, Jesus is given an obscure thought experiment and his response can be interpreted to mean either
- that those why marry essentially won't be saved
- that marriage is not really an institution in heaven
It's not obvious, and (for example) Martin Luther denounced the policy of celibacy and, afaict, it is not really encouraged in most Protestant denominations. While Paul is a big fan, Jesus only directly spoke on the matter once that I know of (beyond your Luke citation): in Matthew 19:8-12:
Jesus replied, “Moses permitted you to divorce your wives because your hearts were hard. But it was not this way from the beginning. I tell you that anyone who divorces his wife, except for sexual immorality, and marries another woman commits adultery.”
The disciples said to him, “If this is the situation between a husband and wife, it is better not to marry.”
Jesus replied, “Not everyone can accept this word, but only those to whom it has been given. For there are eunuchs who were born that way, and there are eunuchs who have been made eunuchs by others—and there are those who choose to live like eunuchs for the sake of the kingdom of heaven. The one who can accept this should accept it.”
This does certainly sound like celibacy being encouraged, but note that this encouragement is not what I would call "strong". First, Jesus says only those who can accept this should. This is literally odd, since everyone literally has the ability to not have sex, so the reasonable interpretation is that this is qualified encouragement. Also contrast this to some other passages, where Jesus is actually strongly encouraging his followers:
And when you stand praying, if you hold anything against anyone, forgive them, so that your Father in heaven may forgive you your sins.”
Then Jesus said to his disciples, “Truly I tell you, it is hard for someone who is rich to enter the kingdom of heaven. Again I tell you, it is easier for a camel to go through the eye of a needle than for someone who is rich to enter the kingdom of God.”
Heck even Matthew 5:21-22
“You have heard that it was said to the people long ago, ‘You shall not murder,[a] and anyone who murders will be subject to judgment.’ But I tell you that anyone who is angry with a brother or sister[b][c] will be subject to judgment. Again, anyone who says to a brother or sister, ‘Raca,’[d] is answerable to the court. And anyone who says, ‘You fool!’ will be in danger of the fire of hell.
To my eyes, this is what Jesus looks like when saying something is crucially important. His denunciation of sex seems extremely tame by comparison.
Progress
There is simply no way you can square this with the idea of progress, unless progress simply means converting people to Christianity
Well, one might define moral progress as moral circle expansion, which is pretty inline with Christian morality of loving thy neighbor and even thy enemy.
which is supposed to be the only thing that matters
I don't think most Christians up to and including Pope Francis himself would agree that converting people to Christianity is the only thing that matters.
Accordingly, the traditional Christian view of history is that of decline, perhaps interspersed with divine interventions here and there
I'm genuinely curious: why do you think this? Is there some reading I can do on the topic?
- Can you please provide a link?
- I'm saying is that, for anyone who works with temporal datasets, what you are saying is not much of a red flag. Unless there is additional context (maybe a link?).
- Again, for anyone who works with temporal datasets, what you are saying is not much of a red flag. Unless, again, there is additional context (maybe a link?)
- "The Philly fed even put out a paper on this" - can you provide a link?
KnotGodel
7mo ago
You brought up your receipts contradicting the official data. They did not.
You’re interpretation of “the crux” is selective: claims about the official data being manipulated and useless were/are definitely central to both our discussion then and the surrounding discussion on this site then/now.
Cool post, but
It should not be a rocket science to isolate these factors, actually, it would amount to a very cool paper with plenty of citations. So where is this paper?
There are plenty of papers on the life expectancy gender gap in general. This one for instance looks at why the gap has narrowed in Sweden by looking at the contribution of different causes. They find "External causes of morbidity and mortality" (i.e. accidents, self-harm, assault) explained 15.1% of the gap in in 1997 and 21.2% of the gap in 2014 - conclusions that seem to broadly match yours.
However, I take your point about the lack of such papers. Given the political leanings here, I suppose commenters will point out that this hasn't stopped similar gender pay gap analysis. There is truth to that, so allow me to point out factors that don't boil down to "it isn't fashionable in this zeitgeist."
Such papers on the gender pay gap are almost always just glorified linear regressions. This makes the causal claims questionable, but also makes it clear why such papers are easier to write than a mortality gap paper would be. To do pay analysis, you can just survey 10k people, which will give you about 10k pay datapoints. You ask those people a battery of standardized questions. You plug-and-chug in a linear regression model. Done. If you tried that for mortality, you'd survey 10k people, which will give you ~150 death datapoints after a year. That's not enough to do any kind of definitive regression with lots of controls.
So, for mortality research, you will realistically need hundreds of thousands of people, which means you are realistically using either medical records or government records. If the latter, privacy and lack-of-standardized-questionnaires will make it a struggle. If the latter, privacy and lack of asking-relevant-questions will make it a struggle.
And, so, to do This Paper on the life expectancy gap, you are stuck integrating results from multiple other papers/fields. It's easy to subtract specific causes-of-death (as you did and the paper I linked to above did). But you want more, you want to control for smoking, drugs, alcohol, "meat consumption and overall life style". Here you run into serious correlation≠causation issues. Sure, the link you used claims that smokers live 10 years shorter, but there is no RCT - how much do we trust that as a causal claim? Ditto for the other things. Another issue is that the linear-regression-doer can include interaction terms, while the literature-review-using-various-sources paper cannot.
I'm not saying there is no impact on what-is-popular regarding what is published in academia, but there are serious differences between researching the gender mortality and pay gaps.
KnotGodel
7mo ago
Yes, I think the postulate that science, a fortiori physics, works is irrational. I also think it's true. On faith alone.
This is a pedantic philosophical piece of sophistry. You don't believe in Newton's laws of motion through "faith alone". There may be, in a pedantic sense, "faith" in that believing in induction requires faith, but there's a reason you believe "an object at rest tends to stay at rest" and not "an object at rest tends to morph into elephant ears". This is not "faith alone" - it is a mustard seed of faith abetted by enormous tons of evidence, reason, and (yes) utility.
If you do value methodology purely for its utility as you say, then you must agree with me that economics is untrustworthy based on it's poor prediction power or disagree that it makes poor predictions.
"Trustworthy" is not a binary and "economics" is a sufficiently diverse discipline that you shouldn't lump it all together. Can economists measure inflation well enough to be useful? Yes. Can economists predicts stock price movements? Not really (with some exceptions). But using economists' lack of ability to predict stocks to infer they can't measure inflation is pure hogwash.
As for the quibble about measure and prediction being different, it's simply wrong. To measure phenomena and to model phenomena is part of the same inextricable process and chain of implication and trust.
No? I can measure how tall you are without any kind of model regarding what makes people grow or why some people are taller than others. By the same token, this
CPI doesn't make sense without a theory of inflation and price
is wrong. So wrong. Measuring something doesn't require predicting it.
Yeah, my bad. It should read
x = ((A-k) - (i-k)) / (2eB) = (A - i) / (2eB)
So, my conclusion remains: the wealth tax doesn't affect the allocation into stocks, while the capital gains tax does. This also explains your second point: the two conclusions are no longer the same.
Do you believe venture capitalists and the startups they fund are net-bad for society? I think they're (a) net-positive and (b) the platonic ideal of high-risk-taking, so I don't think we want to discourage risk-taking in general. I do think you can argue that leverage should be discouraged more, but I don't think a capital-gains-versus-wealth tax is the appropriate tool to do that, even if you insist on using taxes rather than regulations. For instance, you can remove the interest tax deduction.
I don't see how MPT implies this. This sounds like you're "reasoning from a price change" - i.e. what matters isn't that risk is reduced - what matters is how it was reduced.
Hence my use of the term "ivory tower" :p
Just to make sure we're on the same page - you're saying welfare is a "progressive tax that goes into the negative". A poll tax is a fixed sum demanded regardless of income.
You can decompose any tax system as the union of
In this sense, welfare and poll taxes are opposites:
Well, if you want math - suppose my utility function is
U = ln(wage * labor) - labor
dU/dLabor = 1/labor - 1
∴ labor = 1
With a poll tax
U = ln(wage * labor - poll_tax) - labor
dU/dLabor = wage/(wage * labor - poll_tax) - labor
∴ labor = 1 + poll_tax / wage
So, in this model, a poll tax would cause me to work more.
If you want a story: consider someone who barely makes enough to survive - a poll tax would force them to work more hours to continue surviving.
More options
Context Copy link