Contra-Scott Alexander on Money is Not Fungible

that wasn't actually the title, but money is fungible and I will say it

I don’t really want to defend a process that results in getting close to getting rid of good and cost efficient life-saving programs, but someone needs to do it, because the usual people who would do this are all completely bamboozled because they have good effective altruist instincts that are short-circuiting them. At least that’s my theory.

Scott Alexander made a post called Money Saved By Canceling Programs Does Not Immediately Flow To The Best Possible Alternative, and I think we disagree on at least four things. I would summarize them most succinctly by saying I think he could retitle the post, Money Saved by Canceling Programs is Not Fungible. Which makes my disagreements more clear.

Or I could summarize them more fully; I would say:

1. A good (the optimal?) budget cutting process likely puts “good programs” on the chopping block.

2. Money and charitable giving are fungible.

3. Government spending can, in fact, be cut.

4. Just because spending is 0.XX% of GDP, does not mean it shouldn’t be cut if wasteful.

Now, to be clear, I think funding PEPFAR is good and the US should fund it with tax dollars for all sorts of reasons, though I can’t say I know anything near enough about foreign aid to lavish all the relative praise it has gotten overnight from gray tribe circles.

If you cancelled PEPFAR - the single best foreign aid program, which saves millions of foreign lives - the money wouldn’t automatically redirect itself to the single best domestic aid program which saves millions of American lives.

-Scott Alexander

[Emphasis added]

I think the credit for the high status PEPFAR got in no time at all mostly goes to Kelsey Piper, who has very well earned cred with the technocratic elite gray tribe.

The point I’m going to make here is subtle, but it is primarily that everyone is doing their jobs here, except the people arguing money is not fungible, who are missing the forest for the trees.

Let us back all the way up, and perhaps use generalities rather than specific names, because if there is one lesson of the last 10 years that both parties might be counseled to take to heart, it is not to forge a sword you don’t want your enemy to wield.

If you task some group with cutting wasteful spending, and want them to do it efficiently, which is something you might want to do from time to time, you absolutely cannot expect them to be able to make perfect decisions. They will need to cut first and ask questions later, and they might be especially likely to decide that since the usual mode of things is to add spending and not cut, it is much better to err on the side of cutting more and trusting that important things will be added back later. As I wrote last week, at least one person thinks:

“Delete any part or process you can. You may have to add them back later. In fact, if you do not end up adding back at least 10% of them, then you didn’t delete enough.”

If you want this sort of cutting to be possible in practice, you need to accept that things like PEPFAR will end up on the chopping block. This is the cutter doing their job well.

However, others are also doing their jobs well. Kelsey Piper should certainly be pointing out what a bad mistake this would be, especially if there is time between going to the chopping block and being chopped. And the people who think Kelsey Piper knows what she’s talking about should say so, and let this become an important issue. And then it should be taken off the chopping block. And everyone did a good job and we got a good result. And that is what happened!

But we start to go wrong in all sorts of ways (though I do agree with about half of Scott’s post).

For example, Scott makes the argument that government spending can never be cut, only grown or redirected:

But probably something would happen, deals would be made, Congress would think about its existence when deciding how much deficit spending to do, and eventually it would in some sense go back into the general pot of all other federal spending - taking the pot from its current $1,500 billion dollars all the way to . . . $1,506 billion dollars. From there would go to the same kinds of programs¹ the rest of the pot goes to - like the Broadband Equity And Deployment Program, a $42 billion effort to give rural Americans Internet which, after endless delays, has failed to connect a single rural American.

I’m quite sure that’s an ungenerous interpretation, but I also think it is the natural one, and what most readers will take away. And it’s wrong. Of course we can cut spending.

The other thing that Scott is missing, which I can only attribute to PEPFAR being catnip for the gray tribe, is that money is fungible. If there is a tremendously cost effective thing to do, by definition we don’t need a government to fund it to the tune of $6B per year, Americans alone gave over $500B to charities in 2023!

I would certainly bet that if PEPFAR is canceled in the future, or had been canceled here, money would have found its way into the void. The real value of the government having funded PEPFAR isn’t that it continues to fund it forever, it is the finding of it in the first place! And that’s already done!

But importantly, the fact that PEPFAR found itself on the chopping block is not necessarily an indictment on the cutting strategy, unless like Scott, you have already given up all hope that cutting the budget is even possible, in which case I don’t know what to tell you. That’s because with literally every spending item, you can expect someone to object with roughly the same words and emotion as Kelsey. Almost every spending item in the budget has someone who wants it to be spent, and no efficient cutter of spending can possibly know a priori who is credible and who is not. They sound alike.

Now, the DOGE hypothetical expense cutting experiment may well end poorly, though I hope not. But if you believe they should cut, and that cutting is possible, and that it is hard to tell which programs are needles in the haystack, and that it is okay to put 10% of the programs back when you learn which are most valuable, then you can come to the conclusion that in a case like we have seen with PEPFAR, everyone did what we expected, and what we wanted, and the result was quite good, actually.

I would also be willing to suggest that any group tasked with cutting spending that took the level of precautions necessary to avoid any false-positive errors like thinking PEPFAR should be cut, would in fact be too cautious to actually get anything material done.

I really do not like what I think of as the 0.XX% discourse. Just because the US has an enormous GDP, does not mean that cutting waste is pointless. If the amount of money is so small, then by the same token that spending it can’t matter that much, then neither can cutting that spending matter much. Especially because money is fungible and we just agreed it isn’t that much.

This discourse rhymes with the belief that spending can never be cut, in that thinking that cutting waste will never amount to anything material is a ridiculously defeatist attitude. I think everyone can agree that cutting $1T would make a real difference. A president’s term is 1461 days. Cutting $1B per day for just the weekdays of that term would get you over a trillion. A billion here and a billion there and pretty soon we’re talking about real money.

As a bonus observation: While it is absolutely true that balancing the federal budget, or anything close, would require addressing the larger spending items (medicare/medicaid, social security, and defense) or raising taxes, if someone cutting started with those, I could just as easily see a line of attack that says, “why cut these things that clearly give benefit to broad swaths of the citizenry when so much lower hanging fruit exists to cut?” And I would probably find myself defending the unpopular view again.