I recently switched from earning money to donate to doing something directly useful. I think this work is very important (or I wouldnât have switched!), and if I were still earning to give itâs the kind of work Iâd be excited to fund. I also think this about Juliaâs work; what should this mean for our donations?
Weâve been donating 50%, and say we decide to continue doing that in future years. Letâs say our employers are willing to pay us a combined $300k, so we donate $150k and have $150k left over. Then we pay $51k in taxes and our employers pay another $22k, for a total of $73k in income taxes. This leaves us with $99k for ourselves:
| Donate 50% | |
|---|---|
| Income | $300k |
| Donations | $150k |
| Employee-paid taxes | $51k |
| Employer-paid taxes | $22k |
| Post-tax post-donation pay | $99k |
Since Iâm equally happy for us to donate $1 as for either of our employers to keep $1, however, another option would be for us to ask for lower pay instead of donating. Letâs say we want to keep the same $99k in post-tax post-donations pay; what does that look like?
| Donate 50% | Reduce Pay | |
|---|---|---|
| Income | $300k | $130k |
| Donations | $150k | $0 |
| Employee-paid taxes | $51k | $31k |
| Employer-paid taxes | $22k | $9k |
| Post-tax post-donation pay | $99k | $99k |
Our family is equally well off in the two cases, but in the donation case $33k goes to taxes that would otherwise go to something much more valuable. Thatâs 11% of what our employers would be willing to pay us, and 1â3 of our post-tax post-donation pay. This happens because while donations are âtax-deductibleâ, thatâs only for Federal Income Tax (10%-37% in brackets). Social Security (12.4%), Medicare (2.9%), and State (5% for MA) taxes are all calculated on pre-donation income.
While this is more efficient, there are some reasons why this might not be a good idea:
This only works if there isnât some other organization you think would do more good with your money. People may overvalue their particular organization when they would be more objective in choosing among organizations at large. People have floated the idea that EA should have a norm against donating to your employer.
Your employer might say they are willing to pay the larger amount, but if it actually came down to it they wouldnât follow through. Perhaps they think they can get someone else to do the job for closer to what they had been paying you, or they just donât have the funding. One way this could happen is that they would have been fine with the larger amount initially, but over time that no longer makes sense for the organization, and no one notices because money that doesnât need to be spent feels less like real money. The more pay you are waiving, the larger an issue this is.
If you know youâre costing your employer less, you might hold yourself to a lower standard in your efforts. This would be ignoring major costs like you taking a spot that could go to someone more productive, and the way many top organizations are primarily constrained by management capacity, but I could still see someone falling into this.
If youâve pledged with Giving What We Can, it was initially not clear to me whether a voluntary and easily-reversed decision to waive pay would count. I wrote to
community@givingwhatwecan.org, however, and they said it would.-
Itâs less legible. When you donate to a 501(c)(3) you get a tax receipt (âno goods or services were provided in exchange for your contributionâ etc) and have something clear to list publicly. But if you instead waive pay the best you can do is write something up explaining what youâre doing and get a document from the organization confirming it. Itâs a lot less clear! Some situations where this could be a problem:
In trying to get a mortgage or other loan: âI understand sir, but for the purposes of this loan application what matters is what theyâve actually been paying you.â
In media: âThese so-called âeffective altruistsâ claim to donate 50% of their income, but weâve reviewed their finances and they arenât actually donating anything.â
Social Security tax, unlike the other taxes, is nominally to fund your retirement. Paying less here means lower Social Security benefits later: how worried should we be about this? Even if we assume that Social Security is still paying out in full when youâre ready to retire, however, itâs far less efficient than investing in index funds: it only takes ~$3k/ây in 401k contributions to give a higher retirement income than $19k/ây in Social Security taxes [1].
Money going to the government as tax pays for a lot of valuable things. But itâs nowhere near as valuable as funding for, say GiveWellâs top charities, or other organizations that are at least as beneficial.
As with taxation, if you have other means-tested things in your life such as college or affordable housing these two cases are very different.
If your opinion of your organization decreases, requesting to resume being paid your full salary could appear as disloyal or not being a team player. I would hope that this would not be a concern at the kind of organization Iâm talking about here, but an organization can be excellent in terms of its impact on the world while still having harmful aspects to its internal culture.
It makes discussions around raises confusing. Letâs say your organization would be willing to pay you $100k, but you actually only take $75k. After some time, in which you gain experience and become more valuable to the organization, you think they should be willing to pay you $120k, of which you would take $90k (still 75%). Asking for a raise when you essentially have the power to give yourself an equivalent one is very strange!
Itâs much harder to pressure people into making donations than it is into accepting lower pay. For example, imagine funders start caring about employees waiving pay, treating it as a signal of how confident employees are in the organizationâs performance. It would be really easy for this to filter down into pressure on individual employees and build a harmful culture. (To be clear, I donât think funders should do this. Not only does it have harmful cultural effects, but it isnât even a strong signal: since this is uncommon behavior, it tells you much more about whether the organization has a culture of pay waiving than it does about employee views.)
Iâve laid this out as two options that leave you equally well off. You could instead set it up as two options that have equal altruistic value but where one leaves you with more money. While I agree one could do this, I donât think we should.
While donât have the full funding details for either my work or Juliaâs, my impression is that they are the kind of project that multiple major EA funders (ex) would be happy to support. This means that the effect of donating to them is pretty tricky. Whichever funder would otherwise be covering this part of their funding gap would instead have more money available for their other grants. Since I generally like the way these other EA funders allocate their giving, I think this is fine. Note that this complexity around funding replaceability isnât particular to donating through forgone pay; itâs just as much of an issue if you donate money.
There is some amount of undercounting here: if someone is considering giving up a $400k job earning to give to do direct work at $200k theyâre essentially already contemplating donating 50% in foregone compensation. If they then feel like they would need to waive additional pay they may not be willing to make this switch, even if it would be much more impactful than continuing to earn to give. This isnât an issue with waiving pay, however, but instead is about whether we should have a culture where EAs who are taking pay cuts to do directly valuable things also donate.
I think these are all real concerns, but an option to save your organization 1â3 of your take-home pay is very significant. I currently think that waiving pay would make sense for most people who would otherwise be donating to their employer or another organization whose funding trades off against their employerâs. Iâm less sure about our family in particular, since weâve generally prioritized legibility more highly due to being given as examples in various situations. Iâd be very happy to read additional objections, or reasons why the objections above go farther than Iâd thought.
[1] Hereâs a simple model, all in 2022 dollars. If you earn $150k,
you (and your employer) pay $19k/ây in Social Security tax, and you get
$36k/ây in from age 67. If you put $2.8k/ây into index funds starting
at age 22, earning 6% real returns, you can withdraw $36k/ây at 67 for
your whole retirement.
My personal view is that targeted small-dollar political donations (which large donors cannot simply fill, due to campaign finance laws) are likely to be vastly higher value on the margin than corresponding-sized (equivalent size plus tax savings) non-political donations to organizations that large donors can fill, insofar as such targeted political opportunities arise. So if I was in the situation youâre describing, Iâd accept the higher salary with the intention of donating to such political opportunities when they arose. Of course, this logic is specific to a particular kind of donation opportunity, and wonât generalize to most areas that EAs currently donate to.
Another possibility could be asking your employer to set aside some of your pay in a fund you can direct. (I intend to make a post on this soon.)
I suspect there are some tax issues there? In the US employer donation matching is common, and my previous employer gave all of its employees a sum they could choose where to donate, so that is presumably also fine. Iâm less sure you can do what youâre proposing, however, without it counting as income.
Can you do 100x donation matching? Then the deadweight loss is minimal in comparison.
The bit that I would expect to be tricky is trying to let employees choose how much of their pay they receive through this kind of âchoose where we donate moneyâ. If youâre just trying to let every employee direct a certain amount of funding, thatâs already fine.
Yes that was a concern. Will talk this point over with people before I post this
If the employer set aside the funds in a donation pot which you had no/âminimal control over thatâd probably be more likey to be legal. (Eg if staff voted where the funds should go)
For what itâs worth: I stopped doing salary sacrifice at a well-funded EA org mainly due to worries about fungibility.
I switched to (mostly) trying to seed fund and encourage individuals that other EA funders might miss, or would be less able to credibly and helpfully encourage.
This has gradually come to seem less useful as EA Funds has been growing and the FTX Future Fund started the regranter programme.
Nonetheless, my guess is that at least 1â3 of the value on this approach comes from encouragement and raising aspirations. So Iâm keeping going with this for now.
Another consideration in the legibility space is how permanent you expect your time working for a nonprofit to be. If you are returning to industry and negotiating pay, they will not care about how much you theoretically could have been making, they will care about what you were actually paid.
That doesnât sound right to me? Transitioning to a non-profit means a significant pay cut even if they donât formally pass up pay. If I were to return to tech, my current salary is low enough that I wouldnât expect it to affect our negotiations at all? And if instead I was being paid 2â3 of what Iâm currently paid, itâs still wouldnât have an effect?
If youâre trying to maximize your income in standard jobs in the private sector you want to get multiple offers at the same time and have them bid against each other. In which case it doesnât matter much what your previous role paid?
(And this is ignoring that in 22â50 states [EDIT: 15â50] itâs now illegal to require salary history)
I think thatâs right for someone who has an established private-sector job history. If someone is starting off in the non-profit sector but thinking it may not be permanent, salary could be seen as a proxy for level of responsibility in the non-profit role that is difficult to manipulate /â embellish.
I think the things you bring up a reasons why the effect that I mention might not be as strong as one might naively expect, but thatâs not a reason to expect that it wouldnât have an effect at all.
Even if there is a penalty either way, passing up pay could increase that penalty, in expectation.
Even for those with multiple offers, initial offers will be based in part on (perceptions of) current pay. The negotiating technique you mention can be useful, but also has a lot of downsides, and a candidateâs current job is often considered the main BAFTA.
Even if itâs illegal to ask for salary history, one might want to volunteer it. Also in most of the states you mention, the ban only applies to the state itself, not to private employers, or only to specific cities.
Thanks for pointing that out! I count 15 states (AL, CA, CO, CT, DE, HI, ME, MD, MA, NV, NY, OR, RI, VT, WA) where the ban includes private employers and is state-wide.
Some may also want to consider how passing up pay will affect benefit eligibility. If you can lower your income to 200% of the poverty line, under current Covid-related policy, you can get the maximum SNAP benefit for your household size in many states, including Massachusetts. That in turn entitles you to other benefits like broadband and school meal subsidies.
All together, a family of four in MA can gain about $12,000 per year by getting their income just below 200% of the poverty line ($55k). Thatâs based on my nonprofitâs free open-source app, PolicyEngine US; hereâs a graph. The application process and purchase restrictions can be annoying, but many online grocers now take SNAP and itâs tax free.
Iâd add:
TL;DR: Who is vetting this org? Are you able to pick a good org to work for with lower pay or would your rather have a funder vet orgs and you just go and work for one of them?
Specifically, I sometimes see people working for lower pay (or for free) for an org that brands itself as âEAâ without doing what Iâd consider reasonable vetting.
I do think thereâs added value in letting funders do centralized vetting, and if you work for less/âfree then youâre giving up on some or all of that vetting.
To be clear, if youâre think youâre able (and more important: willing!) to do your own vetting then I wonât argue with youâIâd only encourage you to do it on purpose rather than without noticing. (I specifically assume the author can do their own vetting, no argument there)
Maybe? On the other hand, as someone working within the organization (depending on your role, level, and involvement in strategy) you may have a lot more information about whether it is executing well and achieving its goals than a funder.
Yes, I definitely agree.
My bar is much lower: âif you considered for 3 seconds whether a funder is way more informed than you [and potentially decided ânoâ]â, thatâs enough for me
The situation Iâm imagining is more like a developer going to work for free on a project branded as âEAâ, trusting the CEO to know what theyâre doing, and having the project seem like an elegant solution to something.
This was great and really helpful, thanks for writing it!
Extending this thought, thereâs a consideration that attracts meâa kind of personal worldview diversification. To take the example of Nucleic Acid Observatory, I might not want to go all in on pandemic prevention by both working on it and directing my counterfactual donations to it through waiving pay. I might instead want to hedge and put some of my donations into, say, animal welfare (or neartermist global development, or what have you).
Iâm not sure if this kind of personal hedging has strong objections though. E.g. maybe itâs a strategy that only makes sense if youâre maximising âthe chance that I do some goodâ, but is in tension with maximising the overall good.
Since in effect this frees up funding for the kind of funder that funds the NAO, and those funders tend to be diversified, there is still some amount of diversification? See the complicated bullet on funding replaceability.
Oh true, I didnât think that bit through.
I think thereâs some weaker version that still applies, for two reasons. First because the money that gets freed up may only be freed up within a specific cause area of the funder. Second because even if the freed money goes back to the funderâs general pot, itâd be distributed according to that funderâs cause-area split. But if you think youâve already put a lot into one cause area, maybe youâd want to push your donations more heavily in the direction of another cause area, rather than just deferring to the funderâs split.
My usual reply to âconvert taxes into donationsâ type arguments is that evading taxes is bad. I donât think it applies here as such, because the variability in pay makes it a bit unclear what the ârightâ amount of taxes is that you should have been paying.
However, I think the underlying reasoning still applies somewhat:
Taxes are not your money. They are compensation for the state, for the various services and infrastructure it provides that enable you to do your job.
Therefore I donât really think it makes sense to compare the impact of paying X as tax money to paying X as a donation. Itâs more akin to worldview diversification: thereâs some amountâyour donationsâwhich is allocated according to your flavor of expected impact; and another oneâtaxesâthatâs allocated by democratically chosen representatives in your country. Both are important, and we wouldnât like any one of them to stop existing.
I still think both options you presented are acceptable. But this would change how you take into account the different uses of that amount of money when deciding between them. Ultimately, it looks to me like the disadvantages you listed now outweigh the advantages. Or at least like some point in-between would be better than going all the way.
Tax evasion is a crime. Tax avoidanceâlegally reducing your taxable incomeâis good and more people should learn about it. You shouldnât unnecessarily âcompensate the stateâ for things you donât want it to do.
This is right, and you should just minimize the amount you allocate to the state, whose effectiveness is very low.
I strongly disagree.
The relative value of taxes vs donations underlies a lot of EA thinking and doesnât get discussed much, so Iâm glad you brought this up. I think itâs important how one defines âevading taxesâ. If we grant the argument that âtaxes are not your moneyâ (which is plausible and appeals to me aesthetically), itâs pretty critical to identify the âcorrect amountâ of taxes which one owes. I might say the correct amount is whatever the tax authorities say I need to pay, which basically amounts to âwhatever I can get away withâ. Or you might say a bunch of the normal loopholes arenât morally legitimate, and that the correct amount is âwhatever your tax bracket saysâ. Or if youâre a tax protester, you might say one or taxes are not morally legitimate, and so the correct amount of taxes you owe is in fact less than the tax authorities say it is.
My point is, establishing how much money I owe in taxes (and therefore how much of my income belongs to the state) is as much a political question as it is a legal or administrative question.
In my opinion (and it seems you agree) Jeffâs proposal is sufficient far away from what most people consider âtax evasionâ that it doesnât really run into the problem youâre identifying. But I occasionally see other EA proposals that look closer to âsteal money to buy bed netsâ.
Yeah, I basically agree.
I think there are two orthogonal axes hereâone of property rights and âstealingâ, and one of the comparison between your groupâs concerns and your own. And the view that my own concerns are more important on the second axis often comes together with a view about the first one, that says taxes are mine as long as I manage to hold on to them.
But here since (the vast majority of people would agree) Jeffâs suggestion is ok on the first axis, I wanted to highlight that this doesnât cancel the need to ponder on the second one.
Sometimes even when you do steal, it may end up being fine to prioritise your own concernsâe.g. if youâre a poor person stealing from a store to feed your family. On the other hand, we wouldnât want everyone to do this. In the same way that we canât sustainably help the world if we go live in a tent and donate 95% of our money, we also canât strive to stop paying taxes and redirect all that money to bednetsâwhich would be analogous, but on the level of society rather than the individual. So we need to somehow still drill it into ourselves and our philosophy that itâs important to pay taxes.
This problem is even more relevant in other EA contexts right nowâwhere a lot of the movementâs money comes from the profits of companies in tax havens (the Bahamas). And this (rightfully) doesnât look good to an outsider. [Edit: this was claimed as false about FTX, and I may very plausibly indeed be wrong]