15 Comments

Without having read Arnold's essay, I can't say whether this is a compelling response. But on its own terms it feels weaker to me than your usual content.

The "Philanthropizing Democracy" section is an interesting policy idea, but isn't really responsive to the question of whether philanthropy can be undemocratic.

The "Strings" and "Fairness" sections convince me that "procedural unfairness" as you've described it is not the right concept to be using: we should rather want more good things and less bad things. I would have been interested in more depth on this point: the side of the argument you're on would seem to lean toward "it's worth it overall to let rich people have more influence," so I would have liked to see you directly accept or reject that. But perhaps that wasn't necessary in the context of Arnold's essay.

Lastly, this sentence seems fallacious (although maybe i'm misreading it?): "Given that strict equality of influence is a non-starter, I don’t really see much force to the complaint that it’s unfair that rich people ... have more influence than ordinary people." Replace "influence" with "wealth" and it seems to fall apart: one can sensibly complain about a concerning degree of wealth inequality without advocating for strict equality of wealth. You argue before this that fairness just isn't the right concept in this case, but as far as I can tell this sentence isn't advancing that argument -- it's making a new, invalid point.

Expand full comment

On the latter point, the thought is: influence has many sources. Most of them don't seem particularly objectionable. (We're not the slightest bit tempted to try to get more people to listen to homeless folks than to NY Times columnists.) So it doesn't seem that inequality of influence is objectionable *in general*. So that raises the question of why inequality of influence *that stems from the particular source of wealth* is any more objectionable.

I agree this doesn't decisively establish that it isn't objectionable. Maybe there's some argument there that could be given. But I'm just highlighting the need for such an argument, and explaining why I don't personally find the conclusion to be immediately obvious.

Expand full comment

In some CEE countries, a small proportion of individual's tax burden - 1 to 2% - is discretionarily allocated to "socially beneficient causes" eg Romania, Poland, Lithuania, Hungary. Not universal (you have to be a tax payer) but close. In Italy, just under 1% is allocated either to specific religious denominations or government welfare programmes based on people's declarations. So there's already a tiny precedent for something akin to "philanthropy vouchers" albeit for taxpayers only. How deep this could be extended realistically until we'd end up with even more desperately underfunded prisons and even more money for the Worthy Elderly Who Fought for Our Freedom (despite majority of them being now boomers born after the war) is hard to tell....

As to money and influence and "fairness". I feel this is not so much about unequal influence or sources of wealth but because I suspect people, even people who are ok with wealth inequalities, don't feel that greater wealth should AUTOMATICALLY lead to greater influence. It feels like cheating because people feel that influence should be allocated not so much equally, but on a different basis.

Expand full comment

I like the out of the box thinking but I have concerns.

First, it creates a lot of collective action problems that can't be meaningfully solved at that scale. For instance, you want both A and B funded and others really want B and C funded but you don't care about C. You will want to somehow commit to not funding B so you can effectively move money from C to A. In small legislatures with repeated interactions this can be overcome in ways that don't make sense at scale.

Second, a big part of what people want out of elections -- and charitable contributions -- is to express their values and affiliate themselves with those values. Unfortunately, often what expresses your values is very different from what achieves those values.

For an extreme example, imagine people had to fund our nuclear deterent by choosing to designate their vouchers to building nukes rather than helping orphans with cancer using blind rescue dogs. Even though many people do believe that it America's nuclear deterent does really make the world better (otherwise Russia/China could bully everyone) few people want to expressively associate themselves with weapons and war. Tons of people at software companies I worked at demanded their company not work for the military even though they certainly didn't believe the us shouldn't have a military or that military shouldn't but new weapons systems.

Ultimately, I think this is why it's so important we have indirect governance. What's really going on is that we all pay an emotional cost for connecting ourselves to things that feel more distasteful even if we ultimately understand they are net good. When you distribute the choices to individuals their small influence will struggle to overcome that cost. When they elect people who then vote for them that person both pays less of a cost (they can vote to fund many things because it makes sense to devote the time) and the large impact of their vote can overcome that cost. Indeed, there are psychology studies showing that people actually report that they would vote for different policies in situations where they are one vote out of a tiny number versus one out of many.

--

I'd rather go the other way and let people transfer their electoral votes to organizations which can use judgement and make deals to pick the best canidate on their behalf.

Expand full comment

Yeah, there's certainly a strong general case for indirect over direct democracy on those sorts of grounds. I've just grown skeptical that our representatives are actually any better. (I agree that stronger, less "democratic" party structures would also be an improvement, but popular ideology is so deeply opposed to that that I don't see much of a pathway to implementation. It's easier to imagine legislators passing a "charity vouchers" bill, by contrast.)

Part of my thought was that people vote expressively *because* it is so "cheap" for them to do so (cf. Caplan on rational irrationality). They can easily ignore tradeoffs, express their support for everything bagels, and generally be stupid and obnoxious, without any noticeable (to them) downside. Charity vouchers would force people to confront those trade-offs. Security hawks could make their case for why people need to choose to invest more in the military (or whatever). I imagine their pitch would be appealing to many conservative men, at the very least. But yeah, in general people would need to become more comfortable with instrumental reasoning for this to work at scale.

Given all the uncertainties, I wouldn't recommend immediately decentralizing 100% of the federal budget in this way. As I note at the end of the post, the experiment could start small, and gradually build up from there if it seemed to be working out OK.

Expand full comment

That's a very interesting approach -- raise the price of expressive voting for people -- that I'd once have been a fan of but I've come to the conclusion that people -- indeed society -- actually get something of value from expressive voting.

It's just not feasible for even highly educated, intelligent and very motivated individuals to reach an informed deciscion on even a relatively small fraction of the problems/issues we need government to handle (w/o considering coordination) so we really should want some way that we can communicate what our values are so agents/politicians can do that for us. Besides, I suspect we are literally evolved to both want to express our values and to defer to the society when our values are rejected with sufficient clarity. So other things being equal it's better to help people express them via some vote.

And I don't think indirection is as hard to sell people on as you expect. The trick is just to frame it as increasing our control/options not limiting them. I think the natural way to do that is via transferable voting interests. Ideally started in the primary system.

Since lots of people don't vote in primaries telling people they can assign their voting interests to some institution/respected politician/activist group (ACLU, Greenpeace) who will assign their vote feels like an expansion of democracy. But the effect is to both allow people to more clearly express the values they care about (especially if you can divide up your vote into pieces) while at the same time moving the actual work of optimizing for those values with a smaller group of full timers.

I don't think people mind if they assign their vote and their trusted institution tells them "we know it's not our ideal canidate but an unelected canidate gets nothing done" as long as they personally feel they expressed their values via their choice of vote assignment.

Expand full comment

What's the advantage of transferable votes over transferable funds? An advantage of the latter, as I stress in the post, is that you avoid artificial thresholds around the 50% mark. I think trusted intermediaries could still bargain / co-ordinate to address the problems you raised in your first comment.

Expand full comment

It's that transferable votes still appoint someone/some group to do the hard task of hammering out a complete budget (not to mention non-budgetary concerns) that trades off competing needs appropriately.

Transferable funds make that virtually impossible.

The transfered votes effectively create a mini-legislature whose members are incentivized to compromise (if you refuse to do that they'll make the deal with someone else to pass a bill/choose a canidate). You have common knowledge there that failing to compromise means no one gets what they want.

Transferable funds creates the opposite incentives. Instead of the vote holders who are less willing to compromise ending up outside the majority coalition the fund holders who are less willing to compromise keep all their funds for their unique goals while the others fund the less sexy/widely seen as important.

And voters themselves just can't possibly devote enough time to actually work out the optimal funding ratios for space exploration versus NIH versus helping the poor even to order of magnitude (time and too big a coordination problem).

--

Besides why would that look any different that the utterly disastrous distribution of charity money given in cash now? A bit to givewell but alot to dogs with sad music and paying people to uselessly raise awareness or pay themselves giant salaries.

Expand full comment

Interesting! I was thinking that there would be some incentive for the holdouts on either extreme to coordinate on "moral trades" that better advance their collective goals (it's hardly a one-shot prisoner's dilemma, after all), but it would be interesting to hear a game theorist map it out.

(I agree that voters themselves would generally be pretty hopeless, but that's why many would naturally want to transfer their budgetary power to a trusted intermediary, like I donate to EA funds rather than trying to work things out myself.)

Why would it look different to conventional charity? Partly because people currently rely on the govt to fund core public services (including altruistic obligations like foreign aid), freeing up charity to be more "personal" and frivolous. I think there would be more of a sense of civic duty to use one's "civic voucher" (as we could call it) wisely, much as many people feel a sense of civic duty towards voting responsibly.

Still, your counterargument that the centralization of voting is helpful for *forcing* coordination is an interesting suggestion that I hadn't come across before. Thanks for pointing it out!

Expand full comment

Philanthropy is by its very nature undemocratic.

A government is essentially a system for distributing resources. A complicated one, no doubt, but that is its fundamental role. A democracy is a type of government that basis its rule making on people and the franchise, meaning that people either directly or through delegation are the agents for the resource (really surplus) distribution.

If a subset of people can independently distribute resources, then that is an inherent subversion.

This is the mortal flaw of E/A. It purports to be a separate (superior) system for determining the correct ends of resources, but if there were a superior, deterministic approach to solving society's ills, then presumably we would eliminate our current method (government and voting), because that is what the government is supposed to do!

The reason we don't do mathematical resource management isn't because folks haven't thought of it or tried it before (E/A is just fancypants Utilitarianism), but because there is no formula for determining the best, most efficient, or most just distribution of resources (largely because we can't define best, efficient, or just!). So we use government as a least-worst algorithm for assessing the current system, predicting outcomes, and deploying resources to agreed-upon ends.

The dumbest statement that is constantly repeated is that "The US is the most generous population on earth" because we have massive "philanthropy." The US is not generous because the money that we "donate" is really a mishmosh of tax dodges, political action committees, donor money laundering, and vanity projects that wastes phenomenal resources.

If you want an interesting angle on the idea of philanthropy, read the description of the history of the 7th tradition of Alcoholics Anonymous. I have not read anything as to-the-point on the subject.

Expand full comment

What an odd argument. Do you think that personal consumption is also "an inherent subversion", or do you only object when people spend their money on someone other than themselves?

Expand full comment

And seriously, you should read this. Especially the section starting at the last paragraph of 163.

https://aa-netherlands.org/wp-content/uploads/2020/03/en_tradition7.pdf

Expand full comment

There is certainly an argument that personal consumption is a subversion - that underlies certain left positions, but it is not one I am on board with.

It may be odd, but it is coherent.

It gets to the question of surplus, and who has a right to that surplus. If you presume that there is a natural right to obtain, then my argument will not make sense. I believe that society/government exists in proportion to surplus, and the disposition of surplus is why we have government. The very concept of "personal" implies ownership and property, both of which are definitional constructs of society. They are useful constructs, but artificial.

If we separate the form of government (who is it by, for and of) from its function, we would want to ask two questions:

1. What is the best disposition of surplus/spoils?

2. What form of organization/system of government provides the greatest likelihood of achieving that disposition?

Because we are nominally decent people, we would very much like to allocate the surplus in ways that achieve some good, usually divided between forms of investment to maintain/increase the surplus and forms of spending to achieve direct ends (health, happiness, etc.). As a species, we have been immensely creative at deriving different rule sets to deal with these two questions, including ones with property and without, with "personal consumption" and without. But the underlying objectives are the same: how do we determine the "best" allocation of surplus.

After about 10,000 years of civilization, our current model believes that the objective is personal satisfaction (freedom writ largely as freedom to consume) and that some combination of markets, capital, labor, property, and representative democracy is the best algorithm. Of course, this has only held true for about 200 years, and is probably almost entirely dependent on fossil fuels.

And the current model, while fabulous at extraction and consumption, is unbelievably bad at solving non-market problems (transactional conditions like education, healthcare, and shelter where the "consumer" cannot choose to "not buy"). So we need a non-market arbiter, and we have chosen taxes and public spending steered by democratically-elected representatives.

And the algorithm will change over time as it always does, shifting both the means of allocation and the system of determining the optimum outcome.

The issue with contemporary philanthropy, especially EA, is that it monopolizes substantial portions of the surplus and subverts the roughly agreed-upon algorithm in favor of what is essentially just neo-oligarchy or neo-monarchy. When you advocate for aggregating capital under an investor or a set of trustees that are unaccountable to the people, you are essentially saying that democracy is not, in fact, good at either picking the right outcome or achieving it. Given that there is ample evidence that philanthropic organizations are largely quite bad at deploying resources, we should ask why do we still believe they should be the allocators?

The Jedi mind-trick of EA has been to suggest that the Oxford Philosophy department has deduced an algorithmic, deterministic way of divining the correct outcomes and allocating resources with precision to attain them. This is obvious hogwash. EA is just a new set of Saudi Princes picking winners without accountability. And this is essentially undemocratic.

I think you are aware of this, because you suggest a thought-experiment of distributing the surplus evenly to all citizens to deploy philanthropically. Logically, this is no different than direct democracy except the administrative apparatus would be non-profits that presumably would operate more efficiently and effectively than governments. Having been on the boards of Habitat for Humanity and a high-quality charter school, I will tell you that a large philanthropy is often less productive than an equivalent bureaucracy, and always much more expensive to administer.

Honestly, I think we as a society are so drunk on the right to personal consumption that we are shockingly bad at using surplus to alleviate suffering. I suspect that the most effective person-level altruism is mutual aid - people helping families and neighbors - though I do believe that our government has a critical role to play (FEMA is a good example). But large-scale philanthropy, especially in its new incarnation, is laughable. And undemocratic.

Expand full comment

Also, how exactly does this differ from giving people cash? Is there some limitation that my donation can't go to the people with SSN ... charity? If so I fear that the rules about what qualifies as a valid cause really are doing all the work and that's no more democratic than it was before.

Expand full comment

The idea is to restrict the funds so that they can't be used for purely self-interested purposes. E.g. restrict to registered charities from which one receives no direct personal benefit (I think there are already similar laws in place governing the eligibility of charitable tax deductions). There are certainly big, important questions surrounding the precise specification of those rules/limitations.

One advantage of charity vouchers over basic income is that (if designed successfully) plausibly more of it would end up going to global causes rather than domestic consumption.

Expand full comment