You Need More Money
Part 1 of the Inefficient Markets Sequence.
[Epistemic Disclaimer: I am not rich yet. I feel like I’m just barely starting to understand this stuff and perhaps that is the best time to teach it: while I still remember what wasn’t obvious, both to cultivate my comprehension, and to enrich the rationalists.
For educational purposes only! Double-check anything I say. Read the comments. You have been warned!
You are responsible for your own money. Do your due diligence and don’t rely too much on random internet bloggers :) I do not know your financial situation. I am not your financial advisor.]
Convergent Instrumental Rationality
Sufficiently advanced intelligent agents, almost no matter their ultimate ends, will tend to pursue instrumental means such as self-preservation and resource acquisition.
What of sufficiently advanced humans?
The Outside Perspective
Can you imagine a world you’d rather live in? Perhaps your life is comfortable, for the moment. But the world is truly awful right now. Children are dying as we speak, of hunger, or war, or disease. A cheap shot, but I’m sure you could come up with many other ills. The world was even worse, in many ways, in the recent past. It could become worse again, or better. Perhaps a lot better.
If you truly have something to protect, what are your means?
If, when pondering a question, you discover what something smarter than yourself would answer, perhaps you have found your answer also.
What would a superior intelligence do in your shoes? What would a more advanced culture think of ours?
You can’t know all the answers to these questions, but we do know this much: acquire resources.
“Need” Is a Relative Term.
Distinguishing a “need” from a “want” is one of those elementary-school tasks that we all think is easy, at the time. It’s a basic budgeting skill: buy what you need first, then buy what you want with what’s left over.
But upon deeper examination, it’s not so simple.
Do you need to see that doctor? Your ancestors only a few generations back lacked access to competent medicine. The physicians of the past often did more harm than good. Were their basic needs being met? And in the more recent past, many diseases which now have effective cures were fatal. Were their basic needs being met? Not by today’s standards.
From the perspective of a more advanced culture (if we knew more, thought faster, were more the people we wished we were, had grown up farther together), are your basic needs being met? Is effective cryonics a need or a want? (And are the current options effective enough to bother with? That’s a point of some contention. Wikipedia still calls it pseudoscience and quackery.)
If, when pondering a question, you discover what something smarter than yourself would answer, perhaps you have found your answer also.
I posit that if you are not clinically immortal, then your basic health needs are not being met!
This is not a want from the outside perspective of a competent culture. Just like how we know antibiotics are a basic medical need for someone with a life-threatening infection, clinical immortality is a basic need for a culture that has a cure for aging. And a humane technological culture will eventually achieve that. There’s nothing in the laws of physics that requires you to age. The Dragon Is Bad. So we know what the smarter culture thinks. Shouldn’t we think so too? That it even occurs to you that this might be a want is an anomaly caused by the peculiar time and place you find yourself in. You’re weird.
So you see, there are degrees of need. There are the immediate needs that you already think of as needs, and less immediate, yet vital needs that are far more than you can afford.
Self-preservation is an instrumental goal of sufficiently advanced humans.
Are you clinically immortal? That was rhetorical.
How do we get there? Resource acquisition is an instrumental goal of sufficiently advanced humans. You need more money! A lot more money! We all need more money!
Effective Altruism
If you’re rich and have a pressing medical need, perhaps there’s an option open to you that others in your situation couldn’t afford: fund medical research for your condition.
Money can buy you a lot of safety. (Self-preservation through resource acquisition!)
But besides donating to research that may save your life one day, there are a lot of other problems you could help with if you had more money.
Friendly AI is the most important. If we get that one right, it will solve all of our solvable problems. If we get it wrong, nothing else matters, because we will no longer be in control.
Much ink has been spilled on the topic of EA and FAI. I’ll not say more here.
They Say Money Can’t Buy Happiness. Jealous much?
Poverty is misery. Money gets you at least halfway to happiness.
But are you even that far yet? This Princeton study suggests that you need an annual income of at least $75,000 for peace of mind. But that was in 2010. Based on the consumer price index, that would be about $90,000 in 2020 dollars. Does it need to be more if you live in an expensive area (like the Bay Area)?
Is your income at least that high? No? Then you need more money!
Maybe you can do better than halfway if you spend it wisely.
Money Is Time
Perhaps the most valuable thing you can buy is time. It’s a shame that we still have to work 40 hour weeks. It used to be worse. Civilization should be past that by now. Free time gives you the opportunity to work on everything else. You can become that thousand-year old vampire, or at least eight hours closer to it per work day than your peers. The right skills probably have compounding returns.
More time for family. More time for saving the world.
To get $90,000 per year without working, on a fixed income of 4%, you need $2.25 million.
Do you have at least that much? No? You are not rich yet! You need more money! Yes? Maybe you are (barely) rich, but you are still not immortal yet! You need more money!
Time Is Not Money
How long will it take you to earn that $2.25 million? 40 years? Too long! We’re trying to buy free time here, not waste it on a 40 hour work week for 40 years.
You need more money. Rationalists Should Win. So why aintcha rich? Because we are not winning hard enough yet. We are doing it wrong. You can’t wait 40 years. You’re supposed to be a rationalist. Get creative. Solve this problem.
Move to Costa Rica where the costs of living are lower? Maybe that’s a backup plan. But you’re still not immortal!
Live in a trailer while doing remote work for a software company? Remote work seems like more of an option now than ever before. The pandemic has changed things. How about your parents’ basement?
Getting better, but no, the answer is clear: You have to decouple time from income. As much as you can.
Having the $2.25 million will do it, but it takes money to make money, doesn’t it?
Rationalists Are Already Halfway to Quant
We’ve got a chicken-egg problem. The solution to chicken-egg problems is bootstrapping. You need to develop sources of income that don’t take too much time.
Sell cheap Alibaba junk on Amazon for five times the price? Sell easy Facebook ad development services to local small business that barely know how to use a computer? Flip houses? Do the Bay Area startup thing until you hit a home run?
I don’t particularly care how you do it (I mean, don’t be a criminal), as long as you’re decoupling. Whatever business you think you can handle, more power to you!
What talents do we have, as rationalists? Reasoning and acting under uncertainty. Computer literacy: web searches, spreadsheets. Data science: code, statistics. The discipline to act on the numbers. What else?
It sounds to me that the typical rationalist is already halfway to quant.
Early Retirement
Retirement should not take you 40 years to earn. If you can save money, and I mean a significant fraction of your income —preferably half or more, and manage your investments well, you can retire a lot earlier than your 60s, or at least have a shorter workweek doing trading instead of whatever it is you’re doing now. (Assuming, of course, that you aren’t already too close to that old.)
If you earn 4% per year, then you need the aforementioned $2.25 million for the $90,000 half-happiness income. If you earn 10% per year, you only need $900,000. If you earn 15% per year, you only need $600,000. At 18% you need $500,000; at 24% you need $375,000. And of course, you can acquire that nest egg a lot faster if you’re earning a good return on your smaller investments.
And if you really want your free time sooner (and I do), perhaps you can settle for a quarter-happiness income of $45,000 plus lots more free time to start? You could keep saving money and grow your income over time. At 18% you only need $250,000.
Now maybe that still seems like a lot to you, but it’s much less than $2.25 million!
I’m oversimplifying a bit here. While I do think 24% returns (or more!) are achievable, they would be volatile. You still have to pay taxes and health insurance. You’d need a cushion for drawdowns, which could last for years if you are unlucky. Maybe it’s just an extra $90,000. Maybe you could arrange to move back in with your family for a short period. Maybe your previous company would take you on for a while. Perhaps you have a spouse who could work part time (or full time, that helps with insurance too).
Do those returns seem unrealistic? I hope to convince you otherwise.
Wealth Is a State of Mind
Not literally true, but hear me out.
Your talents are not just your skills. They’re also who you know and what you own. Develop them.
If you found yourself broke in a poor neighborhood, how long would it take you to get out?
Borrow a phone and contact a wealthier friend or family member, and they’ll probably have you out the same day.
Suppose that way was closed to you, but you had resources comparable to the people living there. Poor neighborhoods are a “poverty trap”. How long? Really think about it. For at least five minutes by the clock. How would you do it?
Maybe your education is an advantage. Suppose you were unable to prove your credentials. Your university lost your records in a fire. You don’t have your diploma. But you’d still have your skills, what you can remember of them.
How long?
Now suppose an upper-class person found themselves in your neighborhood with your resources? I don’t mean a lazy trust-fund kid with no skills. Someone who culturally understands money, who earned it themselves, or had parents who taught them well. How long would it take them to get out and back up to their standard of living? How would they do it? Really think about it, for at least five minutes.
Maybe their credit with financial institutions is an advantage. In his period of financial ruin, Donald Trump once famously remarked, “See that bum? He has a billion dollars more than me.” Say what you will about Trump, but recently, he had the finances to fund a presidential campaign, and win. The bum is probably still homeless. What’s the difference? What advantage was enough to overcome that?
If you only had the requisite talents, a mere lack of savings would be no object. Saving for retirement is middle-class thinking.
Suppose that way was also closed, due to whatever reason they’re stuck in your neighborhood. The banks won’t go for it. No loans bigger than you could get yourself. How long? How would they do it?
You have the same resources!
If, when pondering a question, you discover what something smarter than yourself would answer, perhaps you have found your answer also.
How long would it take you?
I first heard about this thought experiment from this 2015 talk by Douglas Kruger. He has more to say about the topic. I think it’s worth a listen.
Systematized Resource Acquisition
I hope you now have a sense that more is possible. Why aren’t the “rationalists” surrounded by a visible aura of formidability? Because we have not developed the Art sufficiently. We haven’t really gotten together and systematized our skills.
Resource acquisition is a core rationality skill.
This sequence is the beginning of developing that skill into a system.
Perhaps you don’t have the talents to woo investors and run a business well, or to even recognize and negotiate with people competent in these areas.
But perhaps if you made the right kind of friends, or at least knew the right kind of people, you could learn some of these things and share them with us. I’m certainly not there yet. This sequence is only a start.
So let’s work on saving for early retirement for now. If only to buy us the free time needed to develop the skills to become truly wealthy.
Up next, the foundational skill of trading: how to not lose money.
Part 2: Repeat Until Broke
Part 3: How to Lose a Fair Game
Part 4: The Wrong Side of Risk
Followed by an introduction to alpha.
Part 5: Market Misconceptions
Part 6: Charting Is Mostly Superstition
- How to Lose a Fair Game by (14 Aug 2020 18:41 UTC; 36 points)
- Charting Is Mostly Superstition by (23 Aug 2020 20:26 UTC; 33 points)
- Market Misconceptions by (20 Aug 2020 4:46 UTC; 27 points)
- The Wrong Side of Risk by (16 Aug 2020 3:50 UTC; 22 points)
- Investment idea: basket of tech stocks weighted towards AI by (12 Aug 2020 21:30 UTC; 18 points)
- Repeat Until Broke by (13 Aug 2020 7:19 UTC; 14 points)
- May I ask for some help getting started with crypto? by (1 Nov 2021 21:52 UTC; 5 points)
This is an interesting post — you’re covering a lot of ground in a wide-ranging fashion. I think it’s a virtual certainty that you’ll come with some interesting and very useful points, but a quick word of caution — I think this is an area where “mostly correct” theory can be a little dangerous.
Specifically:
>If you earn 4% per year, then you need the aforementioned $2.25 million for the $90,000 half-happiness income. If you earn 10% per year, you only need $900,000. If you earn 15% per year, you only need $600,000. At 18% you need $500,000; at 24% you need $375,000. And of course, you can acquire that nest egg a lot faster if you’re earning a good return on your smaller investments. [...] I’m oversimplifying a bit here. While I do think 24% returns (or more!) are achievable, they would be volatile.
You’re half correct here, but you might be making a subtle mistake — specifically, you might be using ensemble probability in a non-ergodic space.
Recommended reading (all of these can be Googled): safe withdrawal rate, expected value, variance, ergodicity, ensemble probability, Kelly criterion.
Specifically, naive expected value (EV) in investing tends to implicitly assume ergodicity; financial returns are non-ergodic; it’s very possible to wind up broke with near certainty even with high returns if your amount of capital deployed is too low for the strategy you’re operating.
Yes, there’s valid counter-counterarguments here but you didn’t make any of them! The words/phrases safety, margin of safety, bankroll, ergodicity, etc etc didn’t show up.
The best counterargument is probably low-capital-required arbitrage such as what Zvi described here; indeed, I followed his line of thinking and personally recently got pure arbitrage on this question — just for the hell of it, on nominal money. It’s, like, a hobby thing. [Edit: btw, thanks Zvi.] This is more-or-less only possible because some odd rules they’ve adopted for regulatory reasons and for UI/UX simplicity that result in some odd behavior.
Anyway, I digress; I like the general area of exploration you’re embarking on a lot, but “almost correct” in finance is super dangerous and I wanted to flag one instance of that. Consistent high returns on a small amount of capital does not seem like a good strategy to me; further, if you can get 24%+ a year on any substantial volume, you should probably just stack up some millions for a few years and then you could rely on passive returns after that without the intense amount of discipline needed to keep getting those returns (even setting aside ergodicity/bankroll issues).
Lynch’s One Up on Wall Street is an excellent take by someone who actually managed to make those type of returns for multiple decades; it’s not exactly something you do casually...
(Disclaimer: certainly not an expert, potentially some mistakes here, not comprehensive, etc etc etc.)
Part 3, How to Lose a Fair Game, is up now, which addresses some of these concerns.
I like this comment. This is the kind of discussion I was hoping for. Yes, I’m oversimplifying, but we have to start somewhere and it takes too long to say everything at once. Some of the things you’re bringing up are planned for later posts in the sequence [EDIT: see part 3]. Anything left over means I will have learned something, which I consider a win.
Part of the reason why money buys happiness is that money buys *absence of financial stress.* That means that strategies that involve increasing stress in order to get more money are not necessarily going to increase your happiness. This consideration weighs in favor of:
1) More traditional, low-risk, low-effort investment strategies, like your bog-standard automatic investment into index funds. They’re more likely to get 7% than your theoretical 18-24%, but they require very little effort. Strategies like flipping houses, building a real estate empire, or startups *might* get you a higher return, but they also take significant time and effort and come with a high risk of loss or bankruptcy. (I personally have met a number of people who tried real estate and wound up bankrupt.) Maybe risk-loving people like Trump can sail through bankruptcy with no stress, but I sure couldn’t!
2) Considering financial stress also weighs in favor of reducing expenses, especially fixed costs that are hard to alter, such as debt payments, mortgages, and car loans. The person with a $5000/month income and $1000 rent is safer and feels richer than the person with a $7000/month income and $3000 rent.
I also think you leap a little too fast from “$75k income associated with happiness” to “therefore, you need $75k income.” The person with 75k income isn’t spending 75k, they’re likely paying ~20% in tax and saving some amount too.
I will have more to say about this in the next post.
Definitely preferable to not investing at all. You should at least be doing that much for now. But I think you can do a lot better than that, for less effort than you think. (And better still with more effort, if you’re in a hurry.) Which I will be explaining how to do later in the sequence.
This is part of the poor-person mindset you need to break away from. Reducing expenses is not necessarily wrong per se. But you need to focus on acquiring resources instead. Think 20% conservation, and 80% acquisition, not the other way around. It’s one of the things covered in that Douglas Kruger talk I linked to. Particularly that story about the buffalo.
It was $90k in today’s dollars, but I did suggest settling for $45k to start, and this is an oversimplification, because the exact numbers are not my point. These numbers will vary with time and place. But I want to emphasize that I am not suggesting you settle for $90k, or $45k, but that that an income is the bare minimum to be doing OK, and we will need a lot more than that to hope to afford some critically important goals that a hypothetical more enlightened culture would consider the very basic of needs: you are not immortal!
I wasn’t asking for advice, and I would appreciate it if you would avoid the lecturing tone. Assume that people you are talking to know about as much as you do about this subject.
Tone policing is a violation of my commenting guidelines.
This is a public forum with multiple readers. Don’t assume that everything I say is directed to you personally, even if it’s in a reply to your comment, unless I PM you. It’s not about you.
If I believed everybody knew as much as I do about the subject, I would not be writing the sequence at all.
It’s not about me, but you should know that many people will be less willing to listen if you take a lecturing, dictatorial tone. You can teach without doing so, and you will be a more effective teacher if you teach with emotional intelligence.
Potentially interesting and important topics, but I can’t quite tell what those topics are—it’s not exactly concrete enough to be advice (or its more useful sibling, personal retrospective), not really objective and specific enough to be data, and not deep enough to be about agent and utility modeling. Writing is too far toward the persuasive style to fit well in LW.
Starting to talk about values and state of the world in terms of goals, then shifting to personal comfort (retirement) is confusing. I think you’re over-abstracting “resources”.
This assumes that you believe that individual personal control of resources is the best way to reach your goals. You don’t explore other options like “increase resources to other aligned agents” or “destroy resources being used against you”. You do talk about alliances and friendships with other humans, but only in the context of increasing your personal comfort or position, not in terms of actually furthering goals.
Money is not resources—it’s a somewhat abstract representation of motivation for other humans to provide resources in the future. I fully agree that I can use more, and that most people can, but only if not everyone has more. In order for my money to be useful, there needs to be someone who wants some of it more than I do, and will exchange actual goods or services for the money.
This is only the introductory post to a new sequence. It will become more concrete and actionable later in the sequence, promise. But maybe not all that concrete by the next post, because inferential gaps. Knowledge acquired in the wrong order can be dangerous. I don’t want to explain the practicals of how to trade before explaining how to do it safely.
This is something I worried about while writing this post. This is the hook post of the sequence to persuade you that reading and acting on the rest is worth the bother. The rest of the sequence should feel more informative, rather than persuasive as compared to this post.
That said, persuasion is not itself an evil. EY certainly had some in his sequences. Having been on the interwebs for a while, I very much understand the desire to resist politics as a failure mode. It can get ugly. Nevertheless, I want to see more instrumental rationality on LessWrong. Among other things, this means coordinating cooperation between people. I don’t know how to do that without using a little persuasion. May I gently suggest that we could use a little more cooperation on instrumental goals on LessWrong?
Please don’t take the rest of this the wrong way. I welcome constructive criticism, so think of this as accepting your words and correcting myself in the comments, rather than defending my honor at the expense of rationality, due to the norms of not editing the substance of posts too much after publishing them.
I do not mean to emphasize “comfort” (although why not, on utilitarian grounds?) The main benefit I see to “retirement” is freeing up your valuable resource of free time to do more important things with it. I feel that 40-hour work weeks are not a good deal, considering the alternatives available.
That said, I can’t possibly hope to predict all the ways my words may be misinterpreted before it happens. However, I noticed (too late for this post) that it is possible to share drafts with individual LessWrong users before publishing them. But even had I known, I wouldn’t know who to ask. Since you seem willing to criticize now, would you also be willing to review drafts of future posts in this sequence before I publish them? It would allow me to make advance corrections without violating the post-publishing edit norms.
Oh, but I did mention that I’m doing this to “enrich the rationalists” and had a section on EA. If I can make a lot of rationalists individually richer by sharing my knowledge of trading, then that is increasing the resources of other aligned agents. If they also donate to EA causes we can agree on, then we are again increasing the resources of other aligned agents. But this sequence isn’t about ends, besides this first post pointing out that you have them and they are important enough to take even more seriously than you are, it’s about means.
Destroying resources being used against us sounds interesting, but isn’t really the topic of this sequence. Did you have something in particular in mind?
The amount of money is not fixed for all time. The economy is not zero-sum, although one category of trading I will be teaching is (but not the other). And by “we” I mean us in the rationalist movement. Money by itself won’t help those who can’t use it wisely. I think we can find better uses for it than those we are taking it from. Money is to some degree a positional good, but if you resist buying bragging rights, due to current social inequality, you’ll need a lot more money before that starts to matter.
I don’t mean to say that money is the only resource that matters. Knowledge, credit, friends, and, especially time were other resources I mentioned. In our current society you really can’t avoid needing money, and if you have excess, you can trade it pretty directly for other resources.
Using a style that’s seen as too much focused on persuasion will lead to a lot of rationalists taking your posts less seriously.
Not having read the following parts, nor watched the video, my immediate reaction is...
1) Ok, you got me motivated. 2)
Pick up a phone and call their upper-class family or friends, would be my first guess.
A sufficiently smart and diligent working-class guy could copy my strategy in a decade, because it would take him a lot of time to learn to code during weekends, and then he would still be missing university education and job experience. And this is still because the profession of software developers is extremely open and supportive; trying to become a surgeon the same way wouldn’t be that easy.
And I suppose the upper class is way more protective of its assets. I could start at the bottom and spend a decade or two fighting my way up the pyramid, competing against people who have the same goal and who all know that only one in a hundred can achieve it.
A formerly upper-class person who was magically stripped of all their assets (how exactly do you remove the contacts and references? ok, suppose they moved to a different country and had to change their name) would at least still remember how the game is played, what are the winning moves, and what are the fatal mistakes. Also, the fact that the person already was there, suggests they have the necessary character traits. (Just like a programmer who had to leave their country, change their name, and start using a different programming language, would still have intelligence, curiosity, systematic thinking, etc.)
For a person who wasn’t born in an upper-class family, probably a good move would be to befriend someone who did, and try to learn from them. The obvious problem is that upper-class people are often already surrounded by tons of people who try to befriend them for various reasons, and they have the necessary defense mechanisms.
I guess what I am trying to say is that attitude is not enough, and attitude together with high general intelligence is… better than attitude alone, but it still only gives you a ticket in the lottery with many participants and few winners (who are then disproportionately visible).
Love this; I like seeing rationalists rise above the meek ineffectual I’m-right-but-no-one-cares quokka stereotype. Rationalists should win.
OK, I listened to the talk, and...
I completely agree that wealthy people take more risks, and that wealthy people often think about others as resources to be used instead of authorities. But I object connotationally—if you are not wealthy, trying to copy this strategy is probably not a good idea. Wealthy people take more risk precisely because they can afford it. And they see other people as resources because they can hire them. And they don’t worry about authorities too much, because they can hire a good lawyer that will get them out of trouble.
The thing that separates you from your goals is not the time but the actions you need to take? Well, it’s both. Doing non-trivial things takes time. Getting skills to do those things well takes time, too. Doing all of this only on weekends, because you need a daily job, that makes the time even longer. And by the way, “trying ten big crazy ideas” also takes a lot of time.
You should stick out of the crowd, because celebrities earn more, regardless of whether they actually do a better job. Probably the best advice in the video. But self-promotion also costs you money and time, so the average person now has to split their time between… improving their skills, working on their project, doing the self-promotion, doing the daily job, and dealing with whatever life randomly throws at them. This is doable, but it’s like having two or three jobs, for a few years. But it’s doable.
One issue is that you seem to be equivocating between increasing your personal income to more-than-a-base-level-of-$90000 and increasing your personal income to help solve immortality. I think we can safely say that achieving clinical-immortality is something that will take significant amounts of resources—trillions of dollars worth at the bare minimum. (Cancer research seems to be about $50B/yr right now and that’s only one small part of achieving clinical immortality.) Suppose I stress myself for the rest of my non-immortal life (say, for 50 years) to earn $190,000/yr, in the hopes of achieving immortality before I die of old age. This is a stretch in a lot of cases. Taking out the base level of income for happiness, this gives a total lifetime extra income of $5M.
Assuming that all of that is donated towards the cause of developing immortality, what effect does that extra income have towards achieving personal immortality? This is impossible to calculate, but I’d assume a very small effect, since it is only about one one-millionth of the funding needed. As a rough estimate, I’ll assume that donating all of that toward immortality research gets you an extra one one-millionth chance of becoming clinically immortal, which seems possibly too generous since if the research finishes the year after you die, you get nothing (modulo cryonics). If you assume that an immortal person will live for a million years on average, then this means that dedicating your life to immortality research donation gets you one extra expected year of lifespan. Of course, all of these numbers of very arbitrary and include assumptions, but I think it gives the right magnitude of expected value.
Contributing toward solving immortality is a wonderful and noble thing, but the personal benefit to an extra donation seems to be low—significantly lower than the personal cost of giving that donation.
That’s not what “equivocation” means. Equivocation is the fallacy of changing the definition of a term mid-argument. What you’re describing is me using multiple arguments in support the thesis (that “you need more money”), which is perfectly valid. For a broad range of worthy goals, including some you might not realize are “needs”, more resources (such as money) will improve your chances of success. Personal income and clinical immortality were just two examples. There could be others.
Aubrey de Gray is not nearly so pessimistic. In this 80,000 Hours interview he suggested that $50 million per year might be sufficient for the SENS program. That was in 2012 though.
That’s still more than I can individually afford, even if I were (barely) rich, but if more of the rationalist/EA community were a lot wealthier and agreed it was a priority, this seems more achievable.
The idea isn’t to cure aging overnight, but to achieve escape velocity: for rejuvenation to buy enough lifetime for the next intervention to be developed.