Hi Everyone,

I'm new to the forums. Question for all of you. I am undecided about what way of spending my money would be best. Until I have more confidence (I'll never have enough to be very comfortable, but I want more than I have now), I'm not going to give away large sums of money. Until then, I plan to invest my money.

I have a little over $100k to invest. If I am risk-neutral (i.e., I am indifferent between a deal that gives me $50 for sure and a deal that gives me a 50% chance of $100) for all practical purposes, where should I invest my money until I decide to give away more of it?

If you think I should put it in an index fund, what kind of index fund (S&P 500, emerging markets, etc.)? Why not a fund of hedge funds, or something like that? I'm looking into it more, but right now, the best proposal I've heard is to put the money in a donor advised fund, with some high-risk high-return index, such as emerging markets or frontier markets.

I figured some of you might have ideas on this, since many of you are likely in a similar situation and have given large amounts of thought to the question of would be expectably best.

Welcome, utilitymonster! It's great to have you on the forums.

I can understand the sentiment behind investing instead of donating at the moment -- the rate of return of further information can indeed be very high. Out of curiosity, do you have any preliminary ideas on what sorts of causes you might consider? My own recommendation is to fund research into which causes are most cost-effective (cf. the recent discussion about the Gates Foundation). In particular, I think the Singularity Institute for Artificial Intelligence is about as close as you can come to such an organization at the moment, although with sufficient funds, it might indeed be possible to finance something slightly more optimal for utilitarians.

As far as your question, I think I would agree with the donor-advised-fund suggestion, both for tax benefits and to counter the possibility that you change your mind about donating later. (It can feel hard to imagine ever changing your mind about such things, but it's worth considering outside view statistics on the matter, since I would guess that most people who end up changing their minds never imagined that they would.) Of course, a DAF has the downside of requiring donations to 501(c)(3) public charities, which makes your giving less flexible. We had an earlier thread discussing the pros and cons of DAFs, private foundations, and just investing the money privately.

Where to invest? The academic literature suggests that it's probably not worth spending much time on this decision, although even theory does predict some differences in risk-vs.-expected-return ratios. For instance, the Capital Asset Pricing Model would recommend funds with the highest beta, which is easy to look up on any mutual-fund service. That's probably what I would do personally.

Since you have a large amount to invest, if you really don't care at all about risk, you could even pick two or three individual stocks whose returns you think are greatly correlated with market returns and buy lots of shares of those, in order to have lower fees than even index mutual funds. I would recommend doing a web search of "stocks vs. mutual funds" for some articles explaining more details on the trade-offs between those alternatives (including tax implications). Also, if you plan to invest a few thousand dollars regularly (e.g., surplus from salary every two weeks), there do exist some low-fee broker accounts that allow you to make purchases fairly often while still maintaining lower fees than would be implied by the expense ratios of index funds. On the other hand, investing in particular stocks like this may involve more hassle than it's worth, especially if you are trying to diversify, so consider the cost of your time as well.

Finally, if you haven't already, consider spending a few hours reading general investment advice online, especially regarding tax rules, which are often non-obvious. For example, it's important to avoid double-counting mutual-fund gains if you go that route, as well as to keep in mind short- and long-term capital-gains tax rates. You may already know all this, but I figured I'd mention it in case any readers haven't thought to read about these topics.

Anyway, do keep us posted on what you decide, both about investing and then eventually donating. And feel free to participate in other forums as much or as little as you like. If you only have time to read passively on occasion, that's cool too.

Thanks for the informative reply.

My inclination is to give to whoever GiveWell recommends, but recently I've been reading some Bostrom stuff and I fear that I'm irrationally discounting highly improbable outcomes. Still, alarm bells go off every time I hear transhumanists and their ilk talk about almost anything. If I become sold on the Bostrom line I'll have to rethink things, and I have no idea what I'd end up supporting.

Right now, my money is being managed by a personal financial advisor that charges very low fees and has basically mixed the money across a variety of index funds. This has done well for me so far (36% last year, beaten S&P 500 (even after fees) since I've been with him), but I did not make the decision to use this planner in a very thoughtful way; as it happened, many of my relatives use this person and it had worked well for them. I checked out a book on funds of hedge funds that I'll skim through. I'm going to call Vanguard soon and ask a bunch of questions. I haven't spent enough time thinking about this, but I'll probably just set some kind of artificial time limit and then move everything over to whatever is the best thing I've thought of at that time. I'll probably go with an emerging markets index fund.

Nick Bostrom's work is indeed awesome, and his papers represent a significant fraction of the links on my recommended reading page. I'd be curious to hear what arguments you find persuasive and what you find alarming. What would it take for you to decide one way or the other?

The arrangement with your personal adviser may indeed be optimal, assuming the fees with him don't exceed the costs (including time and research costs) of making a switch. Investing in a bunch of index funds is pretty close to optimal already, in my mind.

His paper on the infinitarian problem is very interesting, but hasn't changed my mind about anything in a practical way. The paper on astronomical waste has me staying up at night.

Intellectually, the main concern is an unspecific hunch that something has gone totally wrong, given how crazy the conclusions are. While I think this hunch has some methodological import at the moment, as I gather more information, it's value will rapidly diminish. Specifically, the major identifiable concern is that the probabilities for some of these events are sufficiently low that the EU calculation wouldn't work out in Bostrom's favor. But I suspect they will.

The real concern is, I think, somewhat irrational. Being a crazy utilitarian for helping the world's poorest people, while it may make me a bit of a crank, puts me at a sort of respectable high ground. I find that many people at least recognize action for the world's poor as among the most noble kinds of lives for a person to live. But wholesale devotion to sci-fi-sounding scenarios just makes you sound crazy. And, on some level, I don't want people to think that I'm totally crazy. So I would really like to believe that Bostrom and company are wrong. That, I'm afraid, is the unflattering psychological truth.

My state of indecision might be resolved by letting things sink in for a while, researching far-out Bostrom scenarios, or maybe by forcing myself to get emotionally invested.

I am one of at least three or four utilitarians I know whose outlook was radically transformed by "Astronomical Waste," and I think the number is probably far higher. So if you do ultimately become more comfortable with the argument, you'll be in good company.

There was a good deal of discussion on the old Felicifia about existential risks, much related to AI risks. As I mentioned before, I recommend donating to SIAI. In particular, I suggest that utilitarians fund their own research projects through SIAI, by finding a topic of mutual interest and then paying SIAI staff to write papers on it. Even if existential risk weren't a concern, this is still one of the best ways I know at the moment to fund utilitarian research.

A friend of mine added these comments:

I would mention, regarding CapM, that historical empirical data suggests very strongly that maximizing fund Beta among domestic US stocks will not actually produce high returns. The closest thing that might would be either emerging market funds or minimizing Beta among US stocks and then using leverage. Note that donating appreciated stocks enables you to deduct the appreciated value from taxes while never paying taxes on the gain and that $3K/year of stock losses can be deducted from income if you itemize. These considerations push moderately for holding some leveraged low-risk individual US stocks in addition to any developing markets fund. Finally, it's worth investigating whether the Utilitarian or SIAI crowds can help with career assistance which might increase income expectations.

Regarding the aversion to science-fiction sounding ideas, it's a big problem. The very need for Givewell, and the fact that it didn't exist until recently, argue very strongly for a world sufficiently insane and lacking in the inclination to effectively pursue even common and sincerely held goals that the posterior odds on initially incredible claims should be seen as low. Ultimately, this boils down to asking questions about the track record of the world in allocating resources to important issues more than a decade ahead of time. If you believe the world to have essentially never done this in the past, the fact that it is not doing so now provides very little evidence against SIAI or FHI compared to the strength of the object level arguments. Regarding expected hedonic dis-utility considerations, it's extremely easy to demonstrate that animal welfare efforts can be more efficient than human poverty reduction by a very large margin. Ultimately, with a philosophy as extreme as utilitarianism, anything that sounds conventionally sane is right out. If the goal is something like human flourishing, fairly objective calculations put asteroid defense ahead of conventional charities using most plausible assumptions. Asteroid defense, poverty, and animal suffering (not to mention factory farming chickens on heroin) are so different that it seems likely that more research into what sort of consequentialism is most promising is high value. It turns out that SIAI and the Future of Humanity Institute (FHI) are among the only organizations that do such research with actual applications in mind, and thus among the only organizations likely to be motivated by a genuine desire to figure things out, not just to get tenure via publication track record. This would seem to argue for one of them. In general, SIAI pays people less than FHI and avoids university overhead, but doesn't get to put Oxford's brand on such papers.

Another consideration is that SIAI staff currently believe that we are in a position to turn money into growth at an unusually high rate, substantially faster than likely investment returns, though one could afford to wait a few months for us to put together a stronger argument than we currently have before deciding between SIAI and FHI if the argument is not yet overdetermined.

I'll take that into consideration. I'll probably leave most of my money where it is for now, and put the remaining part into a Vangaurd emerging markets fund (VEIEX).

Re: the Bostrom stuff. Have you heard any good objections? I searched Astronomical Waste on google scholar but didn't find any dissenting citations...

The doomsday argument would be a good counterargument against astronomical waste. According to it, if you cap off the total population at 10^50 people, then the expected number is on the order of fifty times what we've had so far. Also, while we can go extinct any time while we're limited to this solar system, all reason suggests that once we escape, destruction will be virtually impossible until the end of the universe. This cuts out all orders of magnitude between this solar system and our galaxy, taking out around 11 orders of magnitude. (This doesn't continue for groups, clusters, etc. because how well they're colonized can make up the intervening population sizes).

Also, I've shown the doomsday argument to be overly optimistic. All it proves is that you don't need to cap it off for the expected numbe of people to be finite, though. It doesn't say what it is.

Related to the anthropics point, the Simulation Argument is an alternative possibility that also downplays the urgency of preventing existential risk, as Carl Shulman noted.

I would add that researching these types of questions (as Bostrom and Shulman are doing at FHI and SIAI) is precisely what's called for in this situation: We know these issues are important, but we need to work out exactly what they imply about our actions. So the uncertainty here is in some ways an argument for SIAI (especially as opposed to, say, the Lifeboat Foundation), unless you think the expected value of further research is small.

Sim hyp VS AW: Interesting, I hadn't thought of that before. It provides one good reason for why it's imortant to do more research on the simulation hyp.

What does AW stand for again?

Astronomical Waste. See http://www.nickbostrom.com/astronomical/waste.html.