How strong is the evidence that the EITC boosts labor supply?

New work from Henrik Kleven:

This paper reconsiders the impact of the Earned Income Tax Credit (EITC) on labor supply at the extensive margin. I investigate every EITC reform at the state and federal level since the inception of the policy in 1975. Based on event studies comparing single women with and without children, or comparing single mothers with different numbers of children, I show that the only EITC reform associated with clear employment increases is the expansion enacted in 1993. The employment increases in the mid-late nineties are very large, but they are influenced by the confounding effects of welfare reform and a booming macroeconomy. Based on different approaches that exploit variation in these confounders across household type, space and time, I show that the employment effects align closely with exposure to welfare reform and the business cycle. Single mothers who were unaffected by welfare reform (but eligible for the EITC) did not respond. Overall and contrary to consensus, the case for sizable extensive margin effects of the EITC is fragile. I highlight the presence of informational frictions, widely documented in the literature, as a natural explanation for the absence of extensive margin responses.

Via A. Dube.  The effectiveness of EITC used to be a consensus view, so if this result holds up, it would require some substantial revisions in how we think about both welfare and job incentives.

Comments

A contribution showing that elasticities are zero.

so if this result holds up, it would require some substantial revisions in how we think about both welfare and job incentives
---

My priors are sacred, so please change reality.

Respond

Add Comment

Respond

Add Comment

I can tell you how the EITC works in the real world here in California. My housekeeper who cleans my house for 4 hours a week (she brings her cousin to help) will only take cash. She does at least 8 other houses in our high end neighborhood. She is a great worker. Based on what I pay her she makes around $35 to $40,000 per year.

I asked her if she files an income tax return. She said she does. I asked her how much income does she report. She says the preparer tells her the amount of day labor income she should report. I said and how much is that and she said you know the amount that gets the maximum (EITC) credit. She tells me that the returns are all filled out when she goes in to the preparer. They just pick the the right one (the one with the right number of dependents) and they put the names in it.

This is the main argument for UBI over EITC, negative income tax, or any means-tested welfare/subsidy, which forces the recipient to choose between work, receiving the subsidy, or playing both sides when the optimal outcome is that they continue to work even if that means receiving a nominal subsidy.

"This is the main argument for UBI over EITC,..."

You can just expand the EITC to all workers.

Respond

Add Comment

Respond

Add Comment

"I highlight the presence of informational frictions, widely documented in the literature, as a natural explanation for the absence of extensive margin responses."

I guess the above is the kind of "widely documented in the literature", the literature being instructions for tax preparers targeting zero audit EITC filers, covering how to find them, how to use the software, how to bill them, how to sell high profit add-ons like loans for "instant refunds", etc. Must also be tied to contractors required to issue W-2s with withholding payment history, eg, government cleaning contracts. Maximum profit comes from spreading W-2 work thinly over the maximum workers.

EITC complexity is a real profit opportunity, with profits funding lots of lobbying to keep getting IRS compliance softare and audit budgets cut ever more.

Respond

Add Comment

In some cases, there is a rush to file in order to claim the dependents before someone else in the “family” does.

It’s interesting, as the California story notes, how well known certain aspects of navigating tax law become.

Respond

Add Comment

Respond

Add Comment

"I show that the employment effects align closely with exposure to welfare reform and the business cycle."

So welfare reform turned welfare queens into worker bees. Tell that to the Democrats attacking Bill Clinton's "neoliberal" policies.

Only in America can someone start with nothing and work their way to top. Only in America. And maybe China.

Respond

Add Comment

Respond

Add Comment

The EITC is considered by politicians and economists as the preferable alternative to an increase in the minimum wage, so any discussion of one has to be considered in that context. Politicians prefer the EITC because it allows for social engineering (the amount of the credit is a function of the number of children of the claimant), while economists prefer the EITC because it encourages work (the credit phases out slower than it phases in and it doesn't discourage employment). This study casts doubt on the reason many economists prefer the EITC. Increasing the minimum wage is certainly a lot simpler and cheaper, and doesn't involve the type of exploitation associated with the EITC (such as RALs and third-party debt collection). According to some economists, UBI is better than either the EITC or the minimum wage, but UBI may put social security and Medicare at risk.

Putting people on the dole (UBI) as a lifetime status has all sorts of deleterious effects on the individual and the family, as well as the public fisc. Doing it with de facto open borders is suicidal - but perhaps that’s the intent.

What are the deleterious effects of free money on Alaskans or Trump and his progeny?

The deleterious effects of a UBI would be much reduced if A) they were as tiny as Alaska's permanent fund dividend, and especially B) you had to live in Alaska full-time to receive them. You, yourself, could enjoy that sweet Alaskan UBI (of $133/month or $4.50/day) right now if you were willing to make the move -- what are you waiting for?

You'll use your UBI to drink seasonal depression away at the Hard Rock Cafe in Anchorage along with your only friend and neighbor Rudolph the Reindeer.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

The conceptual framework in this paper seems crazy. There was a transition from traditional Welfare to the EITC.

The elimination of traditional Welfare mattered more?

They are not separable. This is one change, a new design of state welfare (broadly speaking) for low marginal product workers.

That was my reaction too. I thought the argument for the EITC was that it would not reduce employment, whereas traditional welfare would reduce employment?

I see where they are going with this. However...

EITC without children is $519 per year.

EITC with 1 child is $3,461; 3+ children is $6,43.

The incentive is to claim a child. It's not going to have much of an effect on anyone without a dependent to claim.

Furthermore, the EITC starts phasing out at $19K in income per year. As a commentor above pointed out, there's a strong incentive to not report income above that threshold.

https://www.taxpolicycenter.org/briefing-book/what-earned-income-tax-credit

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

"Overall and contrary to consensus, the case for sizable extensive margin effects of the EITC is fragile. I highlight the presence of informational frictions, widely documented in the literature, as a natural explanation for the absence of extensive margin responses."

This conclusion strengthens the case for wage subsidies because of their salience and immediacy vs. the EITC. Edmund Phelps makes this argument in his book Rewarding Work.

+1 on this. If we are going to choose a wage subsidy over Welfare or a higher minimum wage, it should show up on the first paycheck.

But you don't want to subsidize wages, you want to subsidize the income of low-income households. So no wage subsidy for middle-class teens or for a lower-earning spouse in a middle-class family.

I'm fine with shaping the benefit, but I'm on board with Harry's comment on immediacy.

From everything we know about behavior, from Skinner to Ariely, immediate reward is important.

That's fine, and I understand the rationale. But the 'instant subsidy' is going to have to be based an estimate of annual household income, and it had better be on the low, conservative side because lower-income households seem to love to get those tax refunds and would absolutely hate (and probably not be able to handle) having to write a check at tax time.

Respond

Add Comment

You do want to subsidize wages, as opposed to incomes. The purpose of the wage subsidy is to incentivize socially beneficial transactions that would not happen privately (same as with food stamps and rent vouchers, for example). Phelps' approach has nothing to do with a no-strings income-support program like welfare or UBI. It is a mechansim to engage people at the lower-end of the income and employment spectrum to get a start in the workforce and be paid for productivity, albeit with subsidies.

A key belief is that productive work is highly rewarding and highly socially beneficial in and of itself, as opposed to non-work and income support.

Also, and very importantly, the economic incidence of the subsidy depends on the relative elasticities of labor supply and labor demand, not who gets the subsidy. This distinction between economic incidence and legal incidenc is the same for the EITC and the minimum wage. This point is missed by the pubic, the media, and almost all politicians. It is crucial to understanding who the subsidy subsidizes but very few people understand it. It's so important that Alex has an MRU video on wage subsidies covering exactly this point. Alex also mentions Phelps several times. Highly recommended.

Respond

Add Comment

Respond

Add Comment

"But you don't want to subsidize wages, you want to subsidize the income of low-income households. So no wage subsidy for middle-class teens or for a lower-earning spouse in a middle-class family."

Why not? Just have a broad funding mechanism, FICA or VAT or some such and make the wage subsidy across the board. Then you get away from the stigma and downsides of only catering to a select population. Social security enjoys broad support because everyone benefits, not just a select group.

Given the high levels of support, there must be a fair number of people who both support Social Security and work in the cash economy to avoid it. If you could bring them into the formal economy by lowering the barriers like payroll taxes, they could earn SS credits instead.

Funny, people are surprisingly willing to fund transfers to the wealthiest age group as long as everyone has to work for it.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Not so sure. If EITC doesn't boost labor supply, it's a more effective anti-poverty program - the incidence of the credit actually falls on workers rather than employers.

That's true if you're looking for supply effects by which employers capture some of the benefit. But if instead the benefit comes in the form of lower turnover costs, you don't need a supply elasticity for employers to benefit. When business organizations talk about why they like EITC (eg Chamber of Commerce), it's the reduction in turnover costs that they tend to highlight.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Disappointing.

Perhaps instead of EITC, a universal limited payroll tax exemption? Basically, you wouldn't pay SS/Medi for the first X$ of income. Employers would stop collecting it, budget would record this as an expense paid to SS.

Could also solve another big barrier to low-hour, low-skill hiring by eliminating the onerous payroll tax requirements for the lowest-income workers -- remember, those are a fixed cost to the employer, so they are a bigger penalty to employees making less money. Let people make a certain amount w/out the government getting involved more than once a year, and suddenly jobs up to that amount will be everywhere.

How are they "fixed" costs? They scale with the size of the worker's pay check.

Right, not the tax itself, but the compliance cost to the employer -- that labor market distortion results in every Cabinet nominee now having a tax audit first b/c the law is violated so often.

If you're a skilled employee for an employer with thousands of employees, your additional overhead is relatively trivial. If you want to do part-time childcare work in a moderately well-off household the various compliance costs (such as filing payroll tax) are much larger compared to your pay than is the case for the big-company skilled worker. You don't pay those costs directly, but they decrease your usefulness to your employer, which depresses your wages and your job prospects.

Even with 2.7% UE and $60K PPP GDP per capita, there's a lot of ruin in a nation.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

I recall reading years ago that the original impetus, or perhaps excuse, for EITC, was exactly that - to offset payroll taxes. And that was considered in me EITC amount.

Respond

Add Comment

Respond

Add Comment

Looking at the charts on Page 77, it sure looks as if there was a large difference between labor force participation rates and that difference rapidly closed up after the 1993 welfare changes. By 2003, there's essentially no difference in the labor force participation numbers between the two groups.

That's for weekly participation rates, the annual participation rates are even better. There was a 14% difference in 1968, by 1997 there was no difference, and after that point, Employment participation is higher for Single women with children.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment