Bryan Caplan  

Tim Kane and Means-Testing

Hummel on Fed's Billion-Dollar... Garett Jones on the Supply of ...
Whenever an economist or libertarian opposes means-testing of Social Security and Medicare, I immediately ask: "So should we extend all currently means-tested programs to the entire population?"  Listeners often admit that it's a persuasive challenge.  At our last lunch, however, Tim Kane, one of my favorite libertarian economists, bit the bullet.  In his view, any program that we make available to the poor should be available to everyone.

If I understand Tim correctly, his main argument is that means-testing is a high implicit tax rate, with all the standard effects.  As a behavioral economist, I suspect this a serious overstatement.  Framing and transparency matter. 

Even if Tim is right, though, we face a choice of evils.  With means-testing, we'll have a high implicit marginal tax rate on everyone who might plausibly collect benefits.  Without means-testing, however, we'll have a high implicit marginal tax rate on everyone.  Why the difference?  Because means-testing saves a ton of money, allowing lower average tax rates.

Or so it seems to me.  My challenge for Tim and anyone who agrees with him: Offer a plausible estimate for the total cost of expanding all means-tested programs to the entire population - year in, year out.  Start with Medicaid, then add unemployment insurance, TANF, and everything else.  Even ignoring disincentive effects, can you really come up with a figure under 10% of GDP?

Next step: Ignoring disincentive effects, calculate the average tax increase you'd need to pay to end means-testing.

Penultimate step: Adjust the final equilibrium to account for (a) the better incentives at the bottom of the income distribution, and (b) the worse incentives for the remainder of the income distribution.

Final step: Tell me if you've changed your mind.

Please show your work.

COMMENTS (19 to date)
Tracy W writes:

I remember once, when working on tax forecasting for the NZ government, we got to discussing a universal basic income (which was being proposed a lot at the time) and for our own curiousity did a simulation of how much it would cost.
Assuming that the universal basic income was $10,000 a year and it replaced all benefits and pensions, this would have implied an average tax tax (not marginal) of 50% across everyone. (If you had lower marginal rates on the poor then it would imply proportionately higher rates than 50% on everyone else). This was assuming no behavioural effects.

And $10,000 a year would have been a 1/3 cut on the-then basic pension rate.

Tracy W writes:

We estimated this when I was working for the tax forecasting team for the NZ government, out of idle curiousity (not a full research programme) arising out of discussion over lunch of policy proposals for a universal basic income that were floating around at the time by people like the Green Party.

We estimated, ignoring behavioural effects, that a $10,000 a year basic universal income would require a 50% average tax take (average, not marginal), assuming that it would displace all benefits and pensions spending. And $10,000 a year was about 1/3 lower than the pension for a single person at the time.

Brandon Berg writes:

It's worth considering that the cost of having a high implicit marginal tax rate at the low end of the economic spectrum is limited by the low marginal productivity of the workers in question. If means testing keeps 10% of the population on welfare and out of the labor force, that might result in only a 1% reduction in total output, since those 10% are drawn heavily from the ranks of the low-marginal-productivity workers. Means testing has only a limited disincentive effect on high-marginal-productivity workers, because they can earn incomes far in excess of the means-testing threshold.

liberty writes:

Tracy W: Repeating it three times won't make it true. It does not take a 50% rate to pay for a $10k income -- the math just doesn't add up.

If anyone is interested in the Basic Income Guarantee, check out where you can find studies, papers, estimates, policy outcomes where it has been implemented, etc. Also see Palgrave-MacMillan's series of essay collections, the next one to come out of which I am editing - on Austrian economics and the BIG.

jeppen writes:

I think the question is improperly framed. If there could be no means-tested programs, many programs would be scrapped - not extended to everyone - and that's a good thing. Others would be changed or merged to achieve more reasonable outcomes and incentives.

Bostonian writes:

At least with income taxes, people know what marginal rates they face. If in 2030 Congress decides to means test Social Security for people turning 65 that year, based on the lifetime income reported on their tax returns, this would be retroactive marginal tax increase. It would not have disincentive effects for those people (they already worked), but I think it would be unethical.

If you announce well ahead of time how means testing will be done, the disincentives to work and save kick in.

Daublin writes:

It helps if you make the safety-net programs not that good. In such a case, people will means test themselves, so you don't need to formally means test them. It's a way of letting people vote with their feet.

As a simple example, there is no need to means test for use of public defenders. Public defenders are so awful that you would only use one if you can't personally afford a lawyer.

Bostonian writes:

Tim "Kaine", not "Kane".

David R. Henderson writes:

No, it's Tim Kane. Bryan spelled it right.

Brad Strang writes:

Means test every program, take the savings and place it into a forced savings program for all program recipients. Those with the lowest needs get the highest share of the savings. The saving disincentive has been partially remedied.

John B writes:

In 1948 the Socialist Labour Government in the UK launched the so-called Cradle to the Grave Welfare State.

There was to be no means testing on ideological grounds and in any case net contributors were significantly greater than net takers.

However it was not long before another Socialist Government started to introduce means testing becasue otherwise the Welfare State would be unaffordable. Other highly socialised European Countries have the same experience.

Whatever argument you are having seems somewhat redundant if Socailists decades ago determined, contra-ideological, that means testing was required and nevertheless all the welfare systems in Europe are bankrupt - hence all the debt and crisis.

The problem with highly developed welfare and means testing is, because of the complexity of benefits, it is impossible to provide taper relief above the threshold, so unemployed or those on low incomes receiving support find welfare is a disincentive to work or better paid employment, as they would lose more than they would gain, at least in the short term and you still need to eat and pay the rent in the short term.

They are trapped.

The bullet that needs to be bitten, is not to fuss over means testing, but to stop welfare except as a safety net for the destitute, and remove income taxation at the low end of the pay scale.

By the by, in the UK the mothers are entitled to weekly Child Allowance irspective of means. A well off friend of mine uses her Child Allowance for her two boys both at private school, and who want for nothing, to get her hair done or to have a manicure.

What sense does that make?

Tracy W writes:

Sorry about multiple postings, I thought it wasn't getting through as I got time out errors and when I reloaded the page nothing was there.

Liberty: we assumed that all other areas of government spending (education, health, defence, police, etc) continued at existing levels. If you planned for the universal basic income to replace all NZ government spending then I agree that $10,000 a year would not require a 50% average tax rate. But if you are not only proposing to cut pensioner's state funding by 1/3, but also require them to pay out of pocket for all their own healthcare, police, etc, then the proposal gets even more politically implausible.

I've had a look at the website and I can't find any articles on policy outcomes where a universal basic income has been implemented on a country-wide basis funded by that country. The only articles I could find were where a universal basic income (of some sort, eg child rights) had been given to everyone in some villages in a very poor country at very low levels (eg €12 per month). These studies, whatever their virtues, don't address the cost issues of a universal basic income in a rich country which already has a welfare system. Of course this may be a sign of my poor search skills.

[N.B. The multiple postings have been removed.--Econlb Ed.]

Dark Space writes:

I have no problem with this, but the programs must be scaled back. I have no expectations of ever receiving SS or needing to rely on medicare, but if some unforeseen tragedy or even just poor decision making on my part leave me in a state where I need a safety net, I like the idea of one being there, for everyone who needs it.

That being said, 15.3% of every dollar I earn up to 109k and then 12.4% on every dollar above that amount, gets paid into Social Security and Medicare funds. You may say that yours is not as high because you work for someone else, but your employer is effectively paying half to the government that otherwise could be going to you. This is insane. We're investing enough for a safety-yacht, and we're not even getting a decent safety-net in return.

David S writes:

Some thoughts:

John B: "stop welfare except as a safety net for the destitute"

Under this operational plan, you could still use a basic income guarantee - it would just be a very small guarantee. I personally have lived on about $6,000/yr (in 2001 dollars), and I didn't consider myself destitute. I did need a roommate to make ends meet, though, even though the apartment cost very little.

It seems that the cost is pretty obvious - income time the population of the US. Let's say $2,500/yr is the base rate - so a family of three gets $7,500/yr, which is probably (as in, I state without evidence ;) destitute level. So the cost is $750 Billion.

This is approximately equal to the current individual income taxes received by the US government - which is about half of all Federal receipts apparently. So, all things being equal, taxes would need to go up by 50%.

Of course, that would likely cause armed conflict as 1% of the population shouldered the entire burden. According to Wikipedia, the total income of the United States population is $10T. So providing bare minimum sustenance level support in this manner would take about a %10 income tax on everyone.

If you made less than $25,000, it is a net benefit. If you make more than $25,000, it is a net drain.

Another way to conceptualize this:

tax rate = worker rate x [ benefit rate / average salary ]

In a country where the worker rate is high like the US, low benefits are fairly easy to give. Much harder in failing countries like Greece, of course.

Milo Minderbinder writes:

$200/wk for every adult citizen. ($10,400 a year)

That's $2.7 trillion.

Kill all other transfer payments and subsidies. No SS, Medicare, Medicaid, Section 8, Import-Export Bank, farm subsides, Pell Grants, EITC.

That's close to 2 trillion.

Flat income tax rate of 30% kicking in starting at $30,000.

To drive down the 2.7 trillion expense: No one who takes the benefit can vote. (You need to be benefit free for 2 years and then you get the franchise back)

You can also lose the benefit by: 1) Leaving the country, 2) Going to jail 3) Having or adopting a kid

I am also massive defense cuts to defense and most other federal programs.

David S writes:

One other comment: "means-testing saves a ton of money, allowing lower average tax rates."

This is probably not true. This is a transfer payment, so transferring less money will not "save money". What "saves money" would be decreasing the transfer and enforcement costs.

I would suggest that transfer costs are likely minimal for transfer payments - the simple solutions I have heard in the past are things like "everyone gets a bank account, and money is transferred in each week." So I posit that the true cost is in enforcement - auditing and testing recipients to make sure that the governments objectives are being met.

If the plan was simplified such that everyone received identical benefits, the enforcement costs would drop dramatically, while the transfer costs would rise. So as long as the current enforcement costs are larger than the transfer costs, a "basic income guarantee" could save money. In addition, a lot of that cost is placed on the recipient under the current system (they have to spend a lot of time waiting in government offices, and many deserving recipients do not get benefits due to not understanding the bureaucracy).

See - much better! At least less value is destroyed, even if more is moved around.


Joe Teicher writes:

@Dark Space
you have it somewhat backwards. The 12.4% for soc. security only goes up to $109K. Its the medicare tax that is on all earned income. I believe that the medicare tax rate was 2.9% for 2012 and is 3.8% for 2013.

Andrew writes:

Eliminate all transfer programs and create a forced savings plan utilizing the EITC.

I have tried to explain it this way. A 10% flat-tax on all income. Half goes in your 'savings' account, half distributed on a per-capita basis to everyone else's 'savings' account.

Savings can be transferred by choice or death.

MingoV writes:

Social Security currently is means tested on the back end. A person who earned an average of $75,000 per year (inflation adjusted dollars) does not get Social Security payments that are three times more than someone whose earnings averaged $25,000 per year. The same is true of other payouts such as disability or family survivors.

The problem with means testing Social Security is that it penalizes those who saved the most. Social Security will become a retirement plan for people who pissed away money throughout their earning years.

I'd prefer to have Social Security phased out instead of turning it into another way of screwing high earners (who constantly are demonized despite paying a huge proportion of federal taxes).

Comments for this entry have been closed
Return to top