Bryan Caplan  

The Means-Testing Club

Good Students Rule... The Political Economy of Gover...
We advocates of means-testing need a name for our club.  Singapore and Tyler Cowen (somewhat surprisingly) could be charter members.  Tyler, from Singapore:

This is oversimplifying of course, but you can think of the Singaporean system as "2/3 private money, 4/5 public provision," with private hospitals on the side.

You can think of the UK system as "public money, public provision."  Again with some private supply on the side.

The US system is "lots of public money, lots of private money, mostly private provision."

Many other systems are "public money, private provision."  In all cases there are various complexities piled on top.

Singapore now is making some changes, outlined in brief here.  For the most part, Singapore is adding on some public money, but in targeted fashion (one of the changes is for people over 90 years old, another is for people over 60).

Here's from The Straits Times (gated, I write from the paper copy) from Saturday:

The first [priority] is to keep government subsidies targeted at those who most need them, rather than commit to benefits for all.  Universal benefits are "wasteful and inequitable", and hard to take away once given, he [the Finance Minister] said.

That's exactly the liberaltarian line and sometimes the conservative line as well.  It is a principle I strongly agree with.

So what should the Means-Testing Club be called?

Comments and Sharing

COMMENTS (18 to date)
Alex Godofsky writes:

Why not the "high implicit MTR club"?

Thomas Boyle writes:

It's not at all clear to me that means testing is a libertarian approach.

Milton Friedman, for example, likely would not be a fan (he advocated the basic minimum income, for everyone, without means testing).

The problem is that you create a new source of marginal taxation as the subsidy goes away.

Much of the point of libertarianism is to let people make their own decisions without the government distorting those decisions. Higher marginal taxes - or phase-out of subsidies - have that effect.

Means testing of subsidies is not as simple as it seems: TANSTAAFL, and all.

Eric W writes:

The Reduced Breakfast Club?

David R. Henderson writes:

@Thomas Boyle,
Milton Friedman, for example, likely would not be a fan (he advocated the basic minimum income, for everyone, without means testing).
Actually, Thomas, he didn't. He was always careful to distinguish between a guaranteed income and what he advocated: a negative income tax. A negative income tax implies means testing, although the means are income, not wealth.

Handle writes:

Not names, but two alternatives to simple means testing.

1. Combine the implicit-MTR of Means Testing and regular income tax into a 'total-effect Flat Tax'. It doesn't get rid of the labor-discouragement incentive, but hey, you gotta get the money from somewhere.

2. Guaranteed universal income (in cash or in kind, and which might require some immigration rules, alas), but optional, and distributed in exchange for low-interest debts to the government of no particular term, but which can be collected from tax-rebates and estates.

Daniel Kuehn writes:

Libertarians and conservatives like Medicaid and CHIP? This is news to me.

Daniel Kuehn writes:

Hayek, of course, also thought means-testing was extremely problematic.

Salem writes:

The one-brick-in-the-wall club?

I think both the pro and anti means testing positions make good points. But to me, the crucial issue is that means testing benefits makes them less visible, and hence, over time, can undermine public support for them. Therefore we should be in favour of means-testing for undesirable programmes that shouldn't really exist, as a gradual way of undermining them leading to eventual abolition. See e.g. Osborne's means-testing of child benefit.

David R. Henderson writes:

@Daniel Kuehn,
Libertarians and conservatives like Medicaid and CHIP? This is news to me.
That's what I thought Tyler wrote too. But look at it more carefully. The word is "liberaltarian."

Thomas Boyle writes:

David R. Henderson,

I confess I may have misunderstood Friedman's proposal, but perhaps you can educate me further.

As I understood it, his idea effectively amounted to setting an income level at which taxes would effectively be zero, then causing income tax to vary from there. For example, if the marginal rate were 35%, the taxpayer might pay no net income tax at a gross income of $7,145 or so; would pay $1,000 in net income tax at a gross income of $10,000, and would receive $1,000 of negative income tax at a gross income of about $4,276.

This is arithmetically equivalent to giving everyone a guaranteed income of $2,500, then applying a 35% marginal tax rate to all income.

My numbers will not have been his, and I understand he was trying to work a proposal through the existing complexities of the tax code (making use of unused deductions and allowances), but was this not his "target" concept?

Seth writes:

@Thomas Boyle: Means-testing is a guarantee for everyone. That's the point, I believe. I think you may be confounding a guarantee with a subsidy. Anybody making over $7,145 in your example, is not actually receiving a subsidy from other taxpayers, though arithmetically, some people think they are.

Re: a name for the club, I rather liked Tim Harford's use of the term 'keyhole economics' in The Undercover Economist. How about the Keyhole Club?

Thomas Boyle writes:


I don't understand your first 3 sentences - can you expand on them?

Another way to understand my example is this: in a world where the "tax rate" is zero, a subsidy of $2,500 that is means-tested such that it is reduced with income and disappears for anyone with an income of $7,145 or more, is an implicit 35% marginal tax rate on those with an income below that threshold. If the "tax rate" is higher than zero, the overall marginal tax rate would be correspondingly higher.

It is easy to forget, when means-testing subsidies, that the transition imposed by the means test is a marginal tax superimposed on the existing tax system, and can quickly result in prohibitive marginal tax rates.

Therefore, better to set the marginal tax rate, and then adjust the zero-net-tax point, instead of creating individual transfer programs with separate means tests.

Daniel Kuehn writes:

David - Aha! Sneaky Tyler!

I still think this is somewhat wishful thinking. Maybe I'm crossing my terms but when I hear "liberaltarian" I think the BHL crowd, and I still don't think of them as liking means-tested welfare programs. If anything that sort of libertarian is more likely to outline Hayek's point that special treatment is illiberal (hence the advocacy of a basic guaranteed income). When you move out of that crowd I feel like you're even LESS likely to get that perspective. Certainly I wouldn't have expected before reading this that Bryan would sign on. Tyler is hard to classify so it doesn't surprise me that he likes means-tested programs. But it's not a view I generally ascribe to libertarians or liberaltarians.

BLM4L writes:

[Comment removed for crude language. Email the to request restoring your comment privileges.--Econlib Ed.]

David R. Henderson writes:

@Thomas Boyle,
If I understand your example correctly, then yes, it is right. That also means that there is means testing.

Alex Tabarrok writes:

The Meanies.

Alternatively, one should reframe:

No Welfare for the Rich.

The Justice League

Defunding the 1%.

Floccina writes:

So what do you call those supporters of the Current SS structure who supporter it giving more to those who need it less? Answer: Politicians? So means testing supporters are the non-politicians.

How about the phaseouters or minimalburdeners or the Miniredistributors or just redistributors.

Phil writes:

The Bleeding Hearts Club.

Only callous people would advocate spending government money on the middle class when there are poor people out there who need assistance so much more.

Comments for this entry have been closed
Return to top