A Rebuttal to Caplan on UBI
Back of the Envelope
For Caplan (and many others), the unprecedented costliness of a UBI is a cheval de bataille. From this view, we can eschew most of the tricky, philosophical hot buttons by resorting to some elementary budgeting: why bother with the difficulty of moral theory when the math is so definitive? UBI proponents, they claim, are mistakenly brushing past this fatal error: any difference-making UBI would be absurdly expensive, and therefore, UBI is straightforwardly defeated by rough calculation.
Final point: While we have many moral and empirical disagreements, the biggest comes down to cost. Forget every other issue. Simple cost estimates for a UBI are simply astronomical. $10,000 a person times 330M Americans is $3.3 trillion. That’s more than double what the U.S. will spend on Social Security in 2023.
This analysis is a little pessimistic. Firstly, it’s unclear to me what percentage of the 330M number (if any) are non-citizens and whether they’d receive UBI. Secondly, although people have recently been suggesting more financial benefits for parents, merely restricting UBI to people over eighteen cuts into that figure significantly. This variety of argument also downplays the fact that most people recommend UBI as a (near-)complete replacement for federal entitlements, which totaled well over $12K per person in 2022 anyway (using $4.1 trillion in mandatory spending divided by 330M) and which desperately need revamping.1 And while our fiscal situation is already pretty hairy, it’s useful to remember that the US tax-to-GDP ratio is well below the OECD average.
More importantly, however, these accounting criticisms don’t fully appreciate the special logic required for analyzing a universal program like this. Caplan’s comparison to Social Security, for example, belies the crucial detail that only a fraction of the population receives Social Security—the same goes for Medicare, food stamps, etc. With UBI, however, the affluent people footing the bill will also receive a check; this inflates the total numbers and makes comparisons with targeted programs misleading in regards to overall cost.
Example: Suppose a two-person welfare state; person A makes several hundred thousand dollars per year, and person B makes very little. Now suppose a UBI program collects $30K in tax revenues from A and distributes $15K in UBI handouts to each A and B. The total size of the program is $30K, but it only really costs A $15K after the handout. With a targeted program, the government merely collects $15K from A and awards it to B, so that program looks like it’s half the size of the universal alternative, but really it costs person A the same amount in either scenario.
Frankly, the persuasiveness of these charges of innumeracy against UBI proponents relies more on uncharitable framing than rigorous thinking. What’s remarkable to me, actually, isn’t how exorbitant UBI proposals would be but that they’re even within the ballpark of feasibility at all. Rather than scare me off, cost estimates like this intrigue me, especially after some proper adjustments. Most crucially, even if championing UBI is currently imprudent, decent GDP growth (US real GDP per capita is up more than $10K from just ten years ago) puts a modest half-life on that objection. It reminds me of this Robert Solow quote:
Suppose someone sits down where you are sitting right now and announces to me that he is Napoleon Bonaparte. The last thing I want to do with him is to get involved in a technical discussion of cavalry tactics at the battle of Austerlitz. If I do that, I’m getting tacitly drawn into the game that he is Napoleon.
Ultimately, contra Caplan, I view these bean-counting arguments as ephemeral—and thus the preoccupations with scale as wrongheaded. In short, these are arguments about when UBI is a good idea masquerading as arguments about whether UBI is a good idea. Moreover, this pragmatism steers us away from more interesting philosophical conversations.
Narrow Tailoring
Caplan also objects to the universality aspect of UBI: he plainly views it as a bug rather than a feature. These counterarguments about universal programs aren’t always as perspicuous and exacting as I’d like, but they should broadly fall into three categories of gripes about over-inclusivity: (1) dislike for giving any money to the rich whatsoever, (2) moral outrage at reallocating resources to undeserving layabouts, and (3) the economic improvidence of reallocating resources toward squanderers.
Complaints about shifting resources toward rich people forget that rich people are the ones funding these kinds of programs; they’re not ultimately gaining money, and if they are, then that’s a problem with the tax scheme rather than the universality of the handouts. Separately, one might object that collecting some money and returning it to the same person seems pointless or wasteful, but overall it’s much more efficient than trying to both adjudicate and police who ought to receive the benefits.
Regarding the moral outrage of reallocating resources to the undeserving, I’ll have to redirect you to my upcoming review of Robert Sapolsky’s book about free will for some reason to doubt instincts here about desert, but also, in a certain way, giving those people money makes the most sense. UBI might generate some neighborhood effects (i.e., positive externalities) like those discussed in my post about school choice. And the most profound neighborhood effects will probably arise from awarding UBI to very undeserving people: for example, maybe kicking felony drug offenders off of SNAP benefits in the 1990s was a dumb idea. Think about it: employers probably view these people similarly to how the anti-UBI crowd views them; they don’t want to hire the people you dislike giving benefits to. Moreover, alleviating desperation or inequality could reduce property crimes, and mitigating the financial stress of imminent bankruptcy might likewise mitigate some concomitant negative societal impacts from that stress.
Regarding the related concern of disincentivizing work, I’m somewhat doubtful about the supposed hegemony of pecuniary incentives, especially towards the income extremes. UBI wouldn’t be sufficient to satisfy many people over the long term, and to the extent that it does, they must greatly despise the work available to them and/or so thoroughly enjoy leisure that it’s unclear whether their defecting from the workforce is such a moral travesty—and, were they likely very industrious or productive anyway, to be tempted by such paltry revenues?
UBI could also enhance the labor market in key ways. People often reference how UBI could alleviate structural unemployment through facilitating training and education, but what about labor market mismatches (due to insufficient search time) or geographic underemployment? Greater play in the joints of the labor market could allow workers to locate better-fitting jobs. Professor Caplan champions lower housing costs and more immigration on the basis that merely relocating people to maximize their productivity would substantially improve the world; doesn’t UBI grease these wheels? Or, think about JK Rowling relying on government benefits to write Harry Potter; even if such an outcome is rare, you wouldn’t need many tech startups or superstars to emerge from UBI for that counterbalance to be nontrivial. Thus, the labor market implications of UBI are unclear.
Importantly, these objections about UBI being overinclusive aren’t really about UBI in particular anyway, but apply to universal benefits more broadly, and both the rich and the indolent regularly benefit from government services. Should we exclude layabouts from using roads or postal services? Should we wave down affluent families and tell them to vacation elsewhere when they arrive at national parks? Furthermore, public goods like defense are non-excludable and therefore necessarily universal, but few people even view that as problematic. In sum, the opposition to universality isn’t universal; maybe UBI is logically distinguishable from these other scenarios, but the exceptions seem ad hoc to me.
Unlike the objections proffered above, however, the spendthrift argument (that we shouldn’t reallocate assets to squanderers) is particular to UBI, as it empowers recipients to make spending decisions. I’ll tackle this spendthrift issue in another post, but the alacrity of otherwise libertarian-minded people to suddenly convert to paternalism on this front is notable. I’ll also highlight that UBI opponents can’t have it both ways here: if people truly wasted their UBI immediately, then they’d still need to work the same as before—a meager sum can’t simultaneously be set aflame and bankroll a cushy life of unemployment.
I’m still pretty uncertain about the wisdom of UBI, and I’m even less certain about the timing, but these counterarguments are comparatively weak. The checkbook-balancing routine is too glib, and it’s more like a stalling tactic than the definitive leg-sweep that Professor Caplan imagines. Arguments against universal handouts are more attractive, but it seems a bit twisted to prefer paying salaries to bureaucrats to police handouts and ensure they don’t accidentally aid someone who arguably didn’t deserve (but still needed) the help, rather than simply help needy people regardless of their backstory. Plus, my affinity for simplifying the government and its labyrinthine, piecemeal benefits schema (along with much else) prevents me from finding such arguments genuinely persuasive. Other complaints about UBI provide more enduring worries (and some that I share), but we can improve these conversations by burying the criticisms mentioned here.
And, of course, these figures don’t include some oft-mentioned-but-perhaps-not-humongous cost-savings from UBI like bureaucracy shrinkage or countervailing economic gains from avoiding benefits cliffs, etc.
Every local experiment in guaranteed income for lower-income households (that I know about) has shown overwhelmingly positive outcomes--the squandering argument doesn’t hold up. In terms of land taxes, it seems to me there’s way more available economic rent money available in financial investments than real estate.
You hit on the core reason I prefer a UBI or NIT-like structure. In my experience, the government has shown near zero ability to delineate who “deserves” aid from who doesn’t. These categories are completely arbitrary.
Most importantly through, a UBI would be best funded via land taxes that target economic rents and thus are not pulling money (earned labor) from one person and putting it into another person’s pocket.