14

Carl_Shulman comments on Setting our salary based on the world’s average GDP per capita - Effective Altruism Forum

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (23)

You are viewing a single comment's thread.

Comment author: Carl_Shulman 28 August 2017 05:54:31PM *  13 points [-]

As with Eric, I'd like to express praise for your altruism, respect for your choice, but raise some cautions about the idea of a global human mean income as global norm.

I think it makes sense to think about this in terms of market compensation (including wages and nonpecuniary benefits) and explicit and implicit donation thereof. Depending on people's opportunity costs that salary could represent a large boost in income relative to their outside prospects, a 10% donation rate, 50%, or 99%+. I'd also think about to what extent the change in donation affects your impact (positively and negatively). The degree of sacrifice and magnitude (and sign) of impacts will be enormously different across cases.

This approximate world average has a very strong intuitive appeal to us, because it’s what somebody would get paid if there was complete equality.

Some thoughts on this:

  • If you include nonhuman animals then mean income is orders of magnitude less; even adjusting for cost of living, that would imply low subsistence wages, i.e. absolute poverty and <$200 for humans (which would be clearly highly counterproductive)
  • Equal allocations of income would leave no extra for those expensive needs, e.g. health conditions that require hundreds of thousands of dollars per to survive, young vs old
  • If equality meant bringing up productivity and conditions for poor people, then total and per capita output could rise several fold; if it meant everyone allocating their efforts altruistically optimally then per capita wealth could explode (or collapse alongside skyrocketing total wealth)
  • Conversely, if equality were achieved by taxation at 100% rates and transfers then incomes would collapse, ceteris paribus
  • Median income is dramatically lower than mean, but has a plausible better claim re living high while others die
Comment author: Robert_Wiblin 29 August 2017 03:18:39AM *  9 points [-]

I'll put Carl's third sentence in stronger terms.

Such very low salaries are fine and may be the most moral lifestyle for people whose direct altruistic efforts aren't that time-effective.

If applied to the key figures of the EA movement - take for example someone like Holden - who is generating the equivalent of millions of dollars (perhaps tens of millions) worth of additional EA-style donations each year via their direct work, it would be highly destructive. Every hour of Holden's time is worth ~thousands of dollars in marginal donations, so virtually any time he uses to save money isn't worth it. I would feel fine if Holden were spending a million a year to eke out every moment of his time and happiness and never give a second thought to money.

Because I'm optimistic about the value of the work Joey is doing leading an innovative anti-poverty organisation, I suspect he would do more good for the world if he took a higher salary and used it to buy more time and attention, even at expensive hourly rates. But I'm only weakly confident about that as he surely knows his personal circumstances better than me.

All of the rational analysis aside, I find the intensity of commitment to personal sacrifice in the name of a moral ideal demonstrated here highly admirable and wish more people shared it.

Comment author: Peter_Hurford  (EA Profile) 29 August 2017 01:46:14PM 4 points [-]

Every hour of Holden's time is worth ~thousands of dollars in marginal donations, so virtually any time he uses to save money isn't worth it.

I'm skeptical of this reasoning.

First, Holden is an extreme example, and many EAs (even some "key figures of the EA movement") are thinking overly highly of themselves if they think their time is consistently worth "~thousands of dollars in marginal donations".

Second, the time you spend thinking about frugality may not trade directly off of productive time. I know it doesn't for me. (Though I'm also not all that frugal.)

Third, it may be possible to save thousands of dollars with an hour of thinking about frugality (especially if you are already spending a million a year).

To be clear, I'm of the opinion that frugality in EA is supererogatory -- I'd rather see most people think about improving other aspects of their effectiveness first. However, I definitely call BS on 95%+ of arguments of the "I can't be frugal/vegetarian/etc. because it would harm my productivity" type. I think we should just own up to having room to improve on the frugality angle and just not wanting to make that sacrifice. That's how I feel, at least.

-

I would feel fine if Holden were spending a million a year to eke out every moment of his time and happiness and never give a second thought to money.

I would agree that I still would think positively about Holden and still judge him to be very high performing from an EA perspective even if he spent $1M, but I am skeptical that there really are $1M in such improvements.

-

Because I'm optimistic about the work Joey is doing leading an innovation anti-poverty organisation, I suspect he would do more good for the world if he took a higher salary and used it to buy more time and attention, even at expensive hourly rates.

I can confirm that Joey does spend money to buy time and I can't think of any other things he should be doing in this area.

Comment author: Robert_Wiblin 30 August 2017 03:57:44AM *  3 points [-]

I agree with all of this in principle so we're now down to an empirical question.

Holden is the most extreme single case I can think of, for the purposes of demonstration. There are a handful of other managers who generate bottlenecks who I think might generate an expected $1,000+ worth of donations with an hour of extra work. But yes they're not the typical case.

I doubt Holden could find ways to spend $1m usefully to save time. But $200k? Probably.

Options include:

  • Live next to your office in the centre of town (not an issue for Joey as he works at home).
  • Get an executive assistant to manage your house and personal life so you never have to think about it or buy anything yourself.
  • Always fly direct in a seat you can work in. Always catch UberX wherever you're going.
  • Update you phone and laptop, etc to the best equipment each year.

In SF this stuff could easily run in to the hundreds of thousands, and I'd be psyched to see Holden spending that kind of money so he can work every hour God gives without any other stressors in his life.

"However, I definitely call BS on 95%+ of arguments of the "I can't be frugal/vegetarian/etc. because it would harm my productivity" type. I think we should just own up to having room to improve on the frugality angle and just not wanting to make that sacrifice. That's how I feel, at least."

I suspect that many EAs doing direct work are committing the reverse moral error of undervaluing their time in order to seem i) humble, ii) morally dedicated by being conspicuously frugal. They would be more moral people if they spent more on themselves. I didn't used to think this, but seeing how time-effective direct work can be I have changed my mind.

"Second, the time you spend thinking about frugality may not trade directly off of productive time"

In my experience during the middle of the workday I think it translates close to 1:1. Hence I get UberX if I'm moving between work tasks. During random evenings, or weekends it's more like 3:1 (i.e. a third of the time I save goes towards extra work). So I often use Uber Pool if I'm just relaxing. But YMMV.

I don't want to tell Joey he has opportunities he thinks he doesn't. His circumstances are a bit unusual - my impression is that people working in professional positions in the centre of expensive cities (e.g. high up positions in government) can easily spend up to $100k on sensible things to free up their time, and even more importantly, their attention. And I admire them for figuring out how to do it so they can fully apply themselves to their comparative advantage.

Comment author: Joey 30 August 2017 05:37:23AM 3 points [-]

I like getting down to the empirics :) Some thoughts/empirical data on some of the suggestions. Specific to me as it’s hard for me to speak for Holden, but I would be somewhat surprised if my situation/work loads were way different than other folks working in EA orgs.

i) Indeed at CSH our Canada office is small enough to fit in a single home which we do to save money and travel time. Our India location does have an office, but I have found its location to have near 0 effect on staff work hours (this is tracked so we have decent data on it). This might be due to the fact that most staff seem to work a pretty set 35-40 hours in all cases.

ii) We have in fact hired a PA to manage our life operations, but I found the benefits to be pretty minimal (also via hour tracking). Sadly the between the coordination time between us managing the PA and the work they were able to accomplish without our physical presence, our total time saved was pretty limited. This was in large part due to bureaucracy type issues like banks needing to talk to the account holder that I would expect to be true for most PA’s. We do use a bunch of technological systems which mimic a lot of their work (calendly, online groceries, etc)

iii) My job (even though we are based out of India) is pretty minimal in terms of flights, so maybe I am just lucky that I always have enough work-reading to cover the time in them. Maybe this is an issue for folks that fly a lot more or have less reading intensive work tasks? I also am lucky small enough (6 feet) to work on a laptop in even an economy class seat.

iiii) I do spend a decent amount on business class laptops that are updated before they break, but not sure that technological progress is fast enough to see huge differences in this in a single year unless someone is doing pretty computer intensive tasks like video editing.

Regarding any of these or other ideas I would be really interested in seeing any empirical data on them improving work hours as I have found this historically pretty hard to find and hard to create in my personal data. Even data on overall time trades off 1:1 vs 1:3 etc or data on really huge amounts of well time tracked hours worked by people using systems like these would be super handy.

Comment author: MichaelPlant 30 August 2017 09:49:46AM 3 points [-]

Joey, am I right in thinking your don't count work expenses in your estimate?

Your circumstances seem quite unusual for a couple of reason.

It seems you work in your home, and so don't commute to work. Most people don't get to move their office to suit their preferences. In general, if your office is in a city, you either pay to commute to it, or you'd pay more rent if you move closer, both of which are more expensive than your set up. (You could cycle, but you'd expect to you pay more to live within cycling distance).

I'd also be curious to know how you spend as part of your work and whether you count those trips in your budget or class them as business expenses. Just to push the point, if I spent 52 weeks of the year on business travel I claimed from my organisation, my personal expenditure would be tiny. I think there's also something in that travelling abroad for work will be a partial replacement for holidays (at the least, it's a change of scenery), which I don't see in your budget either.

Maybe the other thing is you're living with a partner. This isn't something one can guarantee, and if you doubled your rent, utilities and internet numbers (leaving aside, for the moment, the normal costs of dating!), because I'm assuming you split those, that adds $3300 dollars-ish, around an extra third, to your total.

I think what you're doing is admirable, but my concern is that because you run your own organisation and live with a partner, which not all EAs can or will do, you're able to reduce your own expenditure in ways that are hard to recreate. Hence I'm not sure how practical a standard this is, even leaving aside all the concerns you might be able to spend more to save time.

Comment author: Joey 30 August 2017 07:05:03PM 4 points [-]

It depends on the work expense. I would guess I generally err on the side of covering it under personal expense (e.g. using our home as an office space we do not get compensation for, or the free food we provide in Van comes out of our personal budget.). But we do put some things under it (e.g. my next flight to India will indeed be under work costs). I think in general our work expenses budget wise follow a similar pattern of lower cost than comparable organizations, so I do not feel my personal budget is offset by it any more so than the average EA org/earning to give job, probably a little less.

It's true I have no specific budget for travel (although this would go under “other spending”). I generally find there are less money and time consuming ways to maximize novelty and life satisfaction. I wish I could count my trips to India as holiday travel and I guess they are novel. Overall though the locations we are going to (low income cities in north India) would not count as a vacation for most people. Certainly that is the way I feel, although I can imagine other people enjoying travel as a whole more than I do.

I definitely think you're right it's hard to cross-apply any specific rule, and there are things that could pull in both directions (I do think living with a partner makes this possible where it would otherwise not be). Of course there are other things that would pull in the opposite direction (there are cities cheaper than Vancouver for example). As mentioned in first comment, this was the number we felt we could both be comfortable and optimally effective at. If it was different circumstances we would have picked a different number.

We do not think we are so atypical in terms of skills and life circumstances in the EA movement that some EAs wouldn’t benefit from this post. Of course we believe that many people will have different life circumstances that prevent it, but there are also many who could do something more like this. Many people cannot donate 10%, but I still think it's very worthwhile to talk about and I expect many people that hear about folks donating 10% increase their net donations. I think the same applies to stronger commitments (e.g. 50% or more).

Comment author: Peter_Hurford  (EA Profile) 30 August 2017 04:22:21AM *  1 point [-]

Thanks for the thoughtful response.

-

Rob:

I agree with all of this in principle so we're now down to an empirical question.

-

Me:

"However, I definitely call BS on 95%+ of arguments of the "I can't be frugal/vegetarian/etc. because it would harm my productivity" type. I think we should just own up to having room to improve on the frugality angle and just not wanting to make that sacrifice. That's how I feel, at least."

-

Rob:

I suspect that many EAs doing direct work are committing the reverse moral error of undervaluing their time in order to seem i) humble, ii) hardcore by being conspicuously frugal. They would be more moral people if they spent more on themselves. I didn't used to think this, but seeing how time-effective direct work can be I have changed my mind.

-

I suspect you're right here that this would be empirically answerable, but the exact methods I'd want to employ elude me right now. Potentially worth thinking about, as it would definitely affect how I think and message around EA. There are certain people I'd want to be more hardcore and certain people I think who worry too much and should relax. It's hard communicating that just right, and I don't really meet up with enough EAs, in person, regularly enough, to get a good sense of what the balance is.

Comment author: Joey 29 August 2017 05:13:52AM 3 points [-]

One thing I wish I made more clear in the post is that I do already spend pretty generously on time to money trade offs. I stay fairly up to date with the various posts about how to spend money to save time. I get food and products delivered, have a cleaner that cleans our house monthly. I do value my time at about the cost I would expect it would take to hire someone into the same role, not the literal number I am paid. I do not spend a huge amount of time thinking about money and how to save an extra $0.50 on bread (although I did spend non-trivial time up front for large recurring costs like housing). I do not see the current level of spending competing against my free time. In fact, it was picked after we had already optimized spending more to save more time.

Maybe there is a bunch of interventions I am missing that directly save time that I have not seen written about?

Comment author: Joey 29 August 2017 05:16:55AM 3 points [-]

I definitely do not see frugality as my main way of doing impact. I see that as productive hours put into Charity Science Health work, and I prioritize my productivity there over saving money.

Overall I am not really concerned about this becoming a norm, but rather letting people know of a wider range of what is possible than what is normally talked about, encouraging a few by example. I do generally think the community could be more frugal while maintaining effectiveness in other domains.

Comment author: zdgroff 29 August 2017 02:55:54AM 0 points [-]

Good point about nonhuman animals. Going with the norm for humans certainly does have weird implications.

I suppose part of the answer might be that with humans, poverty is the problem we're trying to fix, but with animals, it's a lot more than poverty. (Though at least for wild animals, poverty/starvation is a large part of the problem.)

Comment author: Joey 29 August 2017 05:14:30AM 1 point [-]

At the end of the day the number is based off what I think would not impair my productivity. It’s nice having it anchored to something concrete, but I am not as sold on it being anchored as I am on the good done from spending less. I do think if I based it off the animal average (aside from it being way harder to calculate) it would not be enough to live off of without major time sacrifices.

Comment author: Carl_Shulman 29 August 2017 06:45:13AM *  1 point [-]

At the end of the day the number is based off what I think would not impair my productivity.

If this comes first (which makes sense to me), it might be better to start by framing the description in terms of that, giving yourself as a data point/example of that being low in your judgment of your case, rather than in terms of the global mean (which will often not track your primary criterion)?

Julia Wise's posts on her and Jeff Kaufman's frugal budgets take this form (data point).