Bryan Caplan  

Problems with Age-Testing

Macroeconometrics and Science... A Strange Preference on Wage S...
Let me clarify my question for David.  I support both means-testing and age-testing, but they're more similar than they seem.  It's true, of course, that you can change your means, but not your age.  But in both cases, there are perverse incentives for people near the cut-offs.

For means-testing, the problem is clear: If you're near the cut-off, there's little incentive to help yourself.  But age-testing has the same basic problem.  If Medicare kicks in at 65, for example, a 64-year-old could easily be better off doing without insurance, and - if necessary - delay treatment for a year.  For Social Security, similarly, raising the age cut-off increases the wedge between taxes paid and benefits received - especially for groups (e.g. men, the poor, and blacks) with below-average life expectancy.

When you put it that way, of course, it's also easy to see why some people might see age-testing as "unfair."   Low life expectancy is bad enough; age-testing adds insult to injury.  Right now, for example, black males have a life expectancy of about 70.  If we raised the age cut-off for Medicare and Social Security from 65 to 70, roughly half of black males would never collect.

Since I want to abolish the welfare state entirely, I'm always in favor of making it harder to collect benefits.  But appearances notwithstanding, age-testing and means-testing really do have a lot in common.

Comments and Sharing

COMMENTS (10 to date)
Colin K writes:

IMHO there is no single group of people in this country more screwed-over by SocSec than black men. They live shorter lives and make less money, so they only get to collect a return on their taxes for a couple years, and when they die, there's nothing left for their heirs. As Chris Rock once said, there's plenty of rich blacks, but not enough wealthy ones.

Steve writes:

"Right now, for example, black males have a life expectancy of about 70. If we raised the age cut-off for Medicare and Social Security from 65 to 70, roughly half of black males would never collect."

This estimate is incorrect to the extent that the distribution of expected black male lives is asymmetric. It doesn't seem reasonable at all to say that a person's life is symmetrically distributed about a mean of 70. I would imagine that a lot more than 50% of black males make it past age 70 due to the disproportionate effect on the mean of people dying young versus people living extra long. Individual lives could be short of the mean by up to 70, but very few will exceed the mean by any more than 30.

Perhaps this could be your response to those who use the argument that you articulate in the quoted section.

Philo writes:

So your only point is about "unfairness." Well, I have little insight into what people will judge to be "unfair"--a lot of it, I suspect, has to do with "framing." But I see a big difference between Henderson's obvious point about the unfairness of means-testing--that the lazy or improvident will get the benefit and the industrious and prudent will not--and your point, which somehow involves the ("cosmic"?) unfairness of differential life-expectancy. I presume that setting Medicare and Social Security eligibility at age 65 is already supposed to have been unfair to the short-lived. If the age-limit were kicked up to 67 (or 68 or 70), would that be *more* or *less* unfair? Note that (practically) nothing is being proposed to address differential life expectancy itself. The proposal is to eliminate a benefit that presently goes to the fairly old (65-67), reserving it just for the very old (67+). Why think that *increases unfairness*?

Loof writes:

Put aside means-testing, age-testing and consider health-testing for a bottom-line norm. If someone smokes, gets cancer, why should society as a whole pay the bill? Stop smoking, get health insurance; stay smoking, no insurance. Simple test. Simple choice.

Nick writes:


Ostensibly that is the point of cigarette and alcohol taxes

tom writes:

Bryan, you've identified some unfairness of age-testing. But what about effiency arguments? Means-testing creates what I think would be huge disincentives to accumulate capital. Accumulation of capital and incentives to strive are kind of a big deal in a capitalist country.

Age-testing may encourage some people to delay services until coverage kicks in, and that may be an unwise risk. But I'd guess that's small potatoes compared to the other positive economic effect of age-testing for the country as a whole: people would have to working longer and keeping a job with insurance. Isn't that much better than the incentives of means-testing? (I'd acknowledge that for some moderately wealthy people means-testing would probably encourage them to work harder to reach economic escape velocity. But it would do that at the same time it encouraged a large group to give up and go low on the High-Low game. Age-testing doesn't create such a big disincentive to work in any group.)

And if you were to try to hang your pro-means-testing argument on the 'unfairness' of means-testing, you'd still have to have an answer for the huge unfairness of telling tens of millions of middle-age people that the SSA has been funning with them when it sends out those estimated benefits forms every year or so and that we're going to change the whole thing

So I'd say both solutions have unfairness problems while means-testing also give horrible incentives to a big group of people and age-testing doesn't.

Justin Dailey writes:

A basic income grant, such as Charles Murray suggested in 'In Our Hands', is looking better and better all the time, if we're going to have transfer payments.

Fixed commitment to the government, and spending on the grant is virtually guaranteed to shrink as a proportion of GDP over time.

Lump sum amount with no means testing, so other than the work/not-work decision there isn't a huge impact on incentives (and I'd imagine anyone who has ever made more than $20,000/yr isn't going to be willing to switch to the $10,000/yr lifestyle).

The existence of the grant should make it easier abandon various interferences into the market, such as the minimum wage. It should also make it easier to get a flat tax, since (for example) a 25% tax rate on $40,000 of earned income is completely offset by a $10,000 basic income. We might even imagine a regressive income tax, so that somewhat lower incomes are no longer being supported on-net by the government.

Loof writes:

Nick writes: Loof, ostensibly that is the point of cigarette and alcohol taxes

Yes, purportedly, but not actually when generating so much revenue for government with proof in the soup of general revenue and putting pittance back into the health care system for making improvements to health.

What’s proposed as a base-line norm is not a tax that gives people a choice; it is a progressive base that gives people choices.

The government wouldn’t benefit with revenue; it benefits with less costs. The people benefit by having a choice to be bad or not; benefit by choosing health over disease; and the icing on the cake is the benefit of universal access to a basic health care system. Appears like a good bridge between social liberals and anti-social libertarians. What’s the problem?

And, it’s a BASIC universal health care system the government would insure and only there would the base-line norm apply. This doesn’t preclude private insurers, who’d insure beyond basics and might want to risk insuring bad apples. Being their business they'd certainly know the odds and undoubtedly profit from it. Perhaps someone more knowledgeable about Canada’s basic system can correct me, but L recall it covers only about a third of total health care costs. When L was there, also had private insurance for other costs beyond what was considered “basic”, which included dental, ambulance service, out of hospital drug costs, etc.

So, it seems what’s generally proposed is basic health insurance similar to Canada: run federally; applied provincially, which also creates economic efficiencies with competing provinces. The main difference proposed for America would be a base-line norm in the form of a health-care test.

Is ‘A base-line norm for basic health care reform’ a good slogan and possibly poetic justice for Obama?

HispanicPundit writes:


I wish you would explain why you want to abolish the welfare state. I dont mean to explain this as you would to a fellow libertarian but instead to an intelligent liberal.

I know how many others would make the argument - but how would you?

Philo writes:

I think David's main concern was the effect on incentives (not really "fairness"). This is substantial with means-testing, almost negligible with raising age qualification.

Comments for this entry have been closed
Return to top