I'm surprised that the phrase moral maze doesn't appear at all in the post. Rules that punish big corporations but don't punish smaller ones tend to push the world in a direction with fewer maze levels.
Upvoted, but disagree. The addition of complexity and arbitrary (and movable!) cut-off levels adds fear and weird incentives to those near the margin, including those below it. Also, modern mazes infect across the procurement/services/contracting web, so even if ONLY the biggest companies are directly impacted, they'll work their contracts, acquisitions, warrants, and operations to minimize their liability while adding Molochian pressure to everyone involved (which is everyone).
There was no mechanism that seemed like it would have reliably stopped these provisions if they had been an order of magnitude or two worse, and indeed the original BBB bill seemed to have a number of things in that category.
The general mechanism for stopping provisions that are an order of magnitude or two worse is lobbying and in this case, that's likely exactly what happened. The original BBB bill had a bunch of those things and then lobbyists came and fought the bill.
Lobbyist power is not absolute and there will be policies that damage business interests that lobbyists can't prevent. On the other hand, at present Washington does to me not look like a place where lobbyists have too little power.
Is it at all possible that the bill could promote greater honesty towards investors? After all, if you're taxed on book income you're that much less likely to make it appear to investors that your company is more profitable than it really is. Not saying thus will necessarily happen, but certainly worth watching and keeping an open mind.
I'm not sure how much different taxing book income is from reversing a key provision in the TCJA from 2017 - namely, the TCJA raised the income threshold where companies are required to use an accrual basis to report income.
The ‘Inflation Reduction Act’ includes a corporate minimum tax provision. The top 150 American corporations have to pay a minimum percentage of book profits as tax.
The whole thing has strange implications, with Marginal Revolution offering several claims that this could end up doing quite a lot of damage by setting up bad incentives. It is an interesting fact about the world that this is the source of the most interesting criticisms known so far.
There seems little question the thing is not first best policy and is quite the tangled mess. But that is normal. I also agree that taxing investment rather than consumption is quite bad – and if anything it seems like in that post Tyler ignores the best argument for not taxing investment, which is that in general private investors capture only a small fraction of societal returns to investment, so they are effectively quite highly taxed to begin with, and this is a lot more basic than issues of uncertainty.
(I do not agree that this implies that Matt Yglesias should therefore be a conservative, although that is the direction of the appropriate update.)
The question as always is the magnitude of the tangled mess and resulting consequences of the strange incentives on various forms of investment.
Book Minimum Tax
The latest claim is this:
I see no reason to doubt the factual claim, but doesn’t this undercut the other ways this is bad?
As I understand it, this is saying that if you pay the corporate AMT then this carries forward as a credit against future regular tax liability.
The quote also says this is mostly zero sum over longer periods, which implies that these corporations will eventually use those credits because they would otherwise owe normal corporate taxes.
Rather than a true minimum tax, then, this is a minimum down payment against future taxes. Its value is determined by the interest rate and how long before the full bill comes due.
This is clearly a stupid move. The government pays a lower interest rate than any corporation does, so the cost of this early payment to the corporation is higher than the value of early payment to the government. The new rule only could make sense to the extent that a given corporation never owes regular corporate taxes.
That’s still a relatively small stupid move, in terms of the size of the economic losses.
The complexity cost of all this can be divided into the cost of figuring out the tax owed and the cost of trying to avoid owing tax. The first one seems very small, as large corporations already compute these numbers. So the real question is the cost of trying to avoid owing tax. But if this is mostly about postponing tax then the amount of work worth doing to avoid this goes way down.
By the same logic, damage to accounting standards should be relatively small versus if the credit was not offered. Tyler Cowen asks how much stress the book profits concept can handle, and suspects this is too much. In which case, good news?
It also makes all other effects relatively small. Consider the original concern:
Should we be providing bigger or smaller tax incentives for new investments of various sorts by large corporations? I don’t know. I can imagine the right answer being exactly zero such incentives, or it being substantially more than currently available. I certainly don’t think we should presume Congress ‘got it right’ here.
My assumption is that if you look at the subset of the generally good corporate tax incentives that we could well be doing too little of them and so cutting down their impact is bad, but also that there are a lot of generally not good corporate tax incentives that are mostly rent seeking and regulatory capture, and that are also similarly reduced, making the trade worthwhile.
In terms of this punishing our biggest corporations perhaps being a bad idea, I too worry about the mood affiliation here, but also a lot of things government does greatly favors large corporations over small ones. The pattern as I understand it is that there is by default increasing regulatory capture and rent seeking and general insider behavior by large corporations, which is then only partially balanced out by explicit favoritism in some places for small business. Once again we should consider the possibility this is bringing things closer into balance rather than the other way around. The exact worry is that this makes it doubly punishing to be a large corporation that doesn’t engage in regulatory capture and rent seeking.
It’s a rhetorical ‘gotcha’ and somewhat contradictory to be reducing those baseline incentives while now providing new other similar incentives in other places, but it is not obvious one should have a strong opinion which kind are more efficient. It is often better and important to be logically consistent – if you do X to encourage Z, and also Y to discourage Z, and both cost money (e.g. you subsidize corn syrup production then try to get people to stop eating it) then that seems worth pointing out. But here, it’s more like we do X (deduction) and Y (subsidy) to encourage Z (investment), and we’re cutting X while increasing Y, which could be correct. If those writing the bill thought it was wrong, they wouldn’t increase Y, they’d instead increase X. If anything, increasing Y while not increasing X implies we probably should decrease X somewhat, to improve our efficiency of spend and because the ideal adjustment being zero seems unlikely.
Either way, the effect on incentives seems greatly reduced versus the alternative scenario, both for better and for worse, by the carry-forward credit.
This also reminds me of the electric vehicle credit in the bill, which the usual suspects are pointing out applies (as currently written) to exactly zero vehicles that physically exist. Which, again, isn’t great if you’re into subsidizing electric vehicles, but is pretty awesome if you’re into saying you’re subsidizing electric vehicles because voters love hearing that but you realize that subsidizing sold out things with yearlong waiting lists is grade A stupid.
So yes, by all means dunk on things getting more complicated and making no sense and largely undoing themselves and being accounting gimmicks but also give them credit for undoing most of the actual damage by ensuring the thing doesn’t work.
Inflation Reduction Ho!
As for the name Inflation Reduction Act versus the CBO not predicting any inflation reduction, at this point I don’t entirely get why they don’t call everything the Cute Puppies With Rare Diseases Rescue Act of [Current Year] Part [# of Laws Passed].
I would also say that if my calculations are correct the Inflation Reduction Act really will reduce inflation. This is not me disagreeing with the CBO so much as having a different calculation framework and baseline.
My reasoning is:
The low-hanging fruit here are things like IRS funding and Medicare negotiating prescription drug prices. These things create deficit reduction, and thus are used as ‘pay-for’ methods to justify doing other things. Those ‘other things’ will generally increase inflation, because they involve spending money on things. Thus, if these moves probably happen eventually, implementing these moves now rather than later reduces expected inflation, whether or not they are good moves.
The IRA also was, in terms of CBO scoring, ‘deficit reducing’ rather than merely deficit neutral, as scored on the absurd ten-year metric that somehow determines what laws America is allowed to pass. Every incremental law that lacks 60 votes needs to score above zero. This scores well above zero. If I was writing this law, and Manchin would allow it, I would find a way to save that surplus so I could spend it later with 50 votes. Instead, this did not happen. Even granting that a lot of that reduction is accounting gimmicks, those accounting gimmicks have now been ‘used up’ in order to sound like you are reducing the deficitrather than in allowing you to increase the deficit. That is very good for inflation.
The IRA is undoubtedly various somethings, reducing the urgency of Doing Something. Compare it to the Somethings that were previously be considered, which would have been actively bad for inflation, and we can be happy that the urge to do such Somethings has been reduced.
Finally, the Federal Reserve’s ability to raise interest rates and otherwise set expectations for monetary policy is central in reducing inflation. Does the IRA passing increase or decrease the amount they will feel permission to do so? My guess is it will increase, and also the real amount of room they have will increase as well, leading to somewhat higher rates. I expect on net the provisions to be medium term good and to be seen as medium term good, even if some provisions have issues. If inflation and unemployment (or inflation and economic growth) are effectively semi-fungible due to the monetary authority then an act like the IRA ends up being good for all of it or bad for all of it after resulting adjustments, based on whether or not you’re moving towards or expanding the production possibilities frontier.
There’s even another possible reason. In exchange for other things in the bill, Manchin got a deal to allow badly needed permitting reform, which may not end up happening because some Democrats are against civilization and do not want us to have access to energy, and Republicans might like civilization but they’ll be damned if they will prioritize that enough to vote for a Democratic bill. So either this passes, which is good for inflation by reducing energy prices, or it does not pass, which is neutral for energy prices but also makes it much less likely that any further deals can be made with Manchin. Which is, presumably, good for inflation.
Prescription Drug ‘Negotiations’
The health care provisions issues raised seem more complicated. The magnitudes of the various effects listed seem non-obvious even if all the dynamics listed are correct, and also my initial reaction to the scenario involved is that it is getting the dynamical response wrong.
As I understand it, the provision means that Medicare gets mandatory percentage discounts on drug prices for the top drugs by spend in Medicare Parts B and D, starting at 25% and going up to 60%. Which does not sound like ‘negotiation’ so much as dictating a discount. And the timing schedule makes this happen faster for small molecule drugs than biologics, on the margin moving development towards biologics.
What I’ve never been able to understand is how drug prices end up being set. The whole incident with Martin Skrelli seemed to indicate that the price of a drug that’s proven effective is largely ‘whatever we damn well feel like charging’ because what is the system going to do, not buy it? So there’s some amount of shame and some amount of potential bad press or other retaliation and if you go too far the system is human and will come after you, but companies charge outrageous prices all the time. So what happens when you are told that (arbitrary numbers here to keep it simple) half of the American medical system is going to get a lifetime 50% average discount on whatever price you charge here?
I do not think the answer is ‘well then you get 25% less expected lifetime revenue and thus invest a lot less in R&D.’ I think the answer is instead ‘Well if we raise the price by 33% then we end up with the same total amount of revenue.’ If you think that this discount will in turn lead to additional demands for more discounts elsewhere, you would raise your initial price even more.
Thus, it’s more like the generalization of this point:
The ever-higher launch prices seem like they should dominate.
They dominate even more because the bill also introduces limits on price increases. That’s even more reason to start out with an absurdly high price.
In this scenario, insurance premiums go up while Medicare costs go down an equal amount, and the new ‘negotiation’ requirement is a tax on private purchasing (and insurance to cover the purchasing of) prescription drugs. Which will result in a small decrease in demand, but likely only a very small one because demand here is mostly inelastic except in places where people already couldn’t pay, in which cases companies were already giving profit-maximizing (and/or PR-maximizing) discounts that wouldn’t change.
That makes this more like a small increase in income taxes combined with a modest taking from owners of existing drugs. Which isn’t something I would have supported but isn’t changing things that much.
There is also a bit of a taking from existing already-developed drugs, which raises expectations for future takings, but also lowers the expectation of this particular taking because it was clearly coming at some point, so it’s not obvious to me which direction this one goes in terms of future willingness to fund development.
The exemption through 2028 for orphan drugs doesn’t seem like it will have much effect. That’s only 6 years, so presumably we are talking here about existing drugs not being considered as aggressively for additional indications during that interim period, where the drug in question is already big enough to be on verge of eligibility, which sounds like a rather small window to find.
Discouraging single-drug companies from being bought out by Big Pharma does seem like a downside given that this damages the future value of single-drug companies, and thus discourages development. Again, magnitude and ability to adapt with different arrangements are difficult to measure.
There’s also the two-sided dipping temptation again. ‘This encourages gaming of the system by providing a generic competitor’ is certainly better for everyone else than not doing that, so if someone avoids the law by doing something good, then that’s not a disaster that seems worth worrying about. The fact that it games the system is mildly annoying but I don’t see a non-gaming-the-system norm we are damaging here.
The core problem is still that drug development is both highly expensive and highly worthwhile from a welfare perspective, and we want there to be more of it. At the same time, the marginal production cost of prescription drug production is very low, but is also what is paying for all that drug development, but it’s all retrospective funding which is harder to get right and to justify to people.
The best solutions seem clearly to be some combination of more efficient retrospective funding (e.g. buying patents, prizes and so on, and also getting other countries to pitch in more), and also reducing the development costs because a lot of them are due to government imposing gigantic costs for little or no benefit.
The bill also contains a $2,000 prescription drug out-of-pocket cap for seniors. Many seniors will therefore (1) notice that they will definitely hit the cap and therefore (2) stop caring at all about out-of-pocket costs, so for those who need a lot of drugs all the costs are effectively set to zero. This means those with the highest costs will become completely price insensitive. That does seem like rather horrible mechanism design, and people might be sleeping on how bad the impact might be.
Good News, Bad News
It is generally a worthwhile exercise to notice when things are tangled messes of non-ideal solutions and accounting gimmicks and mood affectation, to stop and point this out and to understand exactly how things are likely to be messed up. There was no mechanism that seemed like it would have reliably stopped these provisions if they had been an order of magnitude or two worse, and indeed the original BBB bill seemed to have a number of things in that category.
That does not mean the overall package is bad news on the margin, nor were these criticisms claims that it was, they were more like claims that if it had been sufficiently bad no one else would have noticed or stopped it. Which seems right.
Given the source of these criticisms, we cannot be confident that there aren’t other larger bad effects we have not noticed, which should be worrisome. Given the mechanism designs I’ve seen are universally terrible, one should assume similarly terrible other mechanism designs.
Yet as far as politics goes, the package still seems well above expectations.