Scott Sumner  

Who benefits from a wage subsidy?

The Prohibition of Evacuation... Dan Ariely: Where's My Quarter...

One of the basic principles of public finance is that it makes no difference whether a tax is legally borne by the buyer or seller. The burden of the tax will depend on the relative elasticity of supply and demand, and the economic incidence of a tax doesn't depend at all on the legal incidence. Stephen Gordon did a post discussing this issue, and provided a nice set of graphs for comparison:
Screen Shot 2015-09-13 at 11.42.20 AM.png
In this case most of the burden of the tax falls on buyers, in the form of higher prices. But his post also shows the opposite case, which is more likely to apply to labor markets (where sellers of labor, i.e. workers, are believed to absorb most of the burden of a payroll tax, whatever its legal incidence (which is 50-50 in the US.))

The same is of course true of subsidies, which are merely negative taxes. If there is no minimum wage then it makes no difference whether you subsidize workers $4/hour by directly paying them that sum of money for each hour worked, or paying the company that hires them that sum of money for each hour worked. This is a point on which all well-informed economists agree, whether on the left or the right. (And of course the whole point of the proposal is to offer an alternative to the minimum wage.)

Perhaps this is all obvious, but I mention these facts because several commenters have told me that many "on the left" are opposed to subsidies on low wage employment, like the one proposed here. They believe it subsidizes firms who pay low wages. I've favored this policy since grad school, and still strongly believe it is the best way of reducing economic inequality. It is similar to the Earned Income Tax Credit, with two possible advantages---ease of preventing fraud, and better incentives in terms of the decision on how many hours to work. With an hourly wage subsidy, low wage workers would always be rewarded for working harder, something not true of the EITC.

Comments and Sharing

COMMENTS (24 to date)
Danny Kahn writes:

And there's a third advantage, which is that workers get the extra money with each paycheck instead of a lump sum with their tax return.

E. Harding writes:

"Perhaps this is all obvious"

-It's not.

Dikran Karagueuzian writes:

I agree with E. Harding: not obvious.

The unnamed commenters who say that many on the left are opposed to "subsidies on low wage employment" are correct. One example of many:

Even progressives who accept that the workers get most of the subsidy are rather negative about such programs:

Bill writes:

Steve Landsburg corrected Paul Krugman on a related matter:

Sam Hammond writes:

Moreover it basically all but solves demand side recessions. Germany traversed the 2008 crash with rising employment and productivity growth due to a lack of a minimum wage and wage subsidies to low skill workers. Thus they also avoided hysteresis.

Note their policy also greatly impoved matching efficiency in labour markets, which everything being equal could raise inequality, but that's a good kind of inequality, I hope all would agree.

Scott Sumner writes:

Everyone, Thanks for the comments.

Sam, Yes, I had heard that the Germans had done something similar, but wasn't familiar with the details. Was the subsidy based on hourly wages or annual income? (I prefer the hourly approach)

John Thacker writes:

Particularly, if you read the comments on the NYT article that you link (including the favorite comments in the sidebar), you'll see that the vast majority of commenters, who characterize themselves as being "on the left" make exactly this critique about it helping low-wage employers. To be fair, I've certainly seen this critique among those in the middle and on the right with less economic education as well.

It's the commonplace "common sense" view among the less economically educated. The equivalence really is counterintuitive (as Bryan Caplan can certainly tell you.)

John Thacker writes:


Note that some ("on the left," but also in the middle and right) who oppose the subsidies paid to firms also oppose the EITC, because they see the equivalence of the situations. However, their preference is for a high minimum wage (plus welfare for those who can't find jobs at that wage.) Disagreeing with that requires a different set of arguments, as you know.

Brian Donohue writes:

To me, it seem pretty obvious and intuitive. Party A has $X to pay party B for working an hour.

Party Gov says "I get 12.4% of that."

Why on Earth can everyone not see that it makes no difference if Party Gov extracts that 12.4% from Party A upfront or if instead it waits until Party A pays $X to Party B, then takes its 12.4%, or (as it does) takes 6.2% upfront from Party A and another 6.2% afterwards from Party B?

baconbacon writes:
The same is of course true of subsidies, which are merely negative taxes. If there is no minimum wage then it makes no difference whether you subsidize workers $4/hour by directly paying them that sum of money for each hour worked, or paying the company that hires them that sum of money for each hour worked.

This is "obvious" only in a hypothetical sense, in a world without frictions. Once you hit the real world and have to define "employment" things start to get messy. Assuming you define employment along these lines- "the employee has to show up and fulfill his job requirements and the employer has to pay the employee" you get some large asymmetries. If a worker gets the wage subsidy the he will be willing to work for $4 an hour less than he would normally take, but a company won't hire him for negative dollars so hiring stops when the employee has ZMP. If the Employer gets the subsidy he will happily pay the employee a portion of that wage to do the absolute minimum to qualify for employment. Employers will not care between the difference between a person that has -$1 an hour value and -$10 an hour value as they will be shuttered off to their own private room where they can get paid a minimal amount to do nothing.

At first it seems like a small difference, until you remember that huge potential chunks of the labor force face large opportunity costs for working. For example parents- going to work for them might mean having to pay most of your wages in childcare while also suffering the disutility of missing time with your kids. If employers got a $4 an hour subsidy you can bet your butt that there would be massive day care centers popping up all over the place where you bring your own kids and look after them all day and get paid $1-2 an hour to do it. If the employee gets the subsidy no such scheme is possible.

When you ignore frictions and real world heterogeneity of individuals you come up with completely wrong answers- which economists have been doing for centuries.

Scott Sumner writes:

Bacon, I specifically mentioned constrained solutions like the minimum wage, so I can't see how you can claim I ignored that issue.

ThomasH writes:

I've "heard" a lot more often from the right that they oppose anything like an EITC that makes the tax system more progressive than from the left that an EITC would subsidize employers.

A thought experiment:

Mitch McConnell stand up and says "I agree with my friends across the aisle that low and middle income workers are not earning enough, but I think his proposal for a higher minimum wage suffers the drawback of causing some increase in unemployment. We may differ in our estimates of how much, but why cause any? I propose instead that we increase the EITC [that he'd come up with an idea for a wage subsidy is beyond my powers of imagination] which for the anyone at the minimum wage today would represent an increase from $y/hr to $15/hr."

Will Harry Reid turn him down? [I'm not saying he'd be happy about being upstaged but would he object?]

John Hayes writes:


It sounds more like you're describing potential for fraud rather than a scenario from a normally operating business. Fraud is real and does exist with wage subsidy schemes; although it's cheaper to claim against a fictional employee rather than a real one who will have leverage to capture more of the money until you fire them. Fraud also exists with any welfare scheme so the issue should be relative, is this more or less fraud than existing more complex, more subjective programs like SSDI.

In normal businesses, management overhead of useless employees is large and harmful to productive employees. A useless employee will destroy more productivity than any subsidy could potentially compensate.

Hazel Meade writes:

If a worker gets the wage subsidy the he will be willing to work for $4 an hour less than he would normally take,

I don't think this is really true. I don't know anyone, especially workers at low wages, who would voluntarily forgo a higher wage because they are getting a subsidy from elsewhere, except in cases where the phase out of welfare penalizes working more. Or maybe in a subsistence situation where wages are driven down to the barest minimum, but we're not in that situation.

Maybe someone can explain the mechanism through which this is supposed to happen, because I don't get it.

baconbacon writes:


Bacon, I specifically mentioned constrained solutions like the minimum wage

You ONLY mention the minimum wage, and the manner in which you mention it implies that you are talking about a set $ amount for pay.

It is demonstrably false that a wage subsidy is the same as an employment subsidy. You apparently misunderstood this in my previous example so I will try a very straightforward one.

Imagine an otherwise free market that was to decide on a $4 an hour wage subsidy or a $4 an hour employment subsidy. In this totally free market there are no child labor laws, so I will find 10 children whose parents are paying for preschool/day care, and offer to give them free day care if they sign their kids up to be my employees. My company will produce nothing but macaroni pictures and skinned knees and I will receive $80,000 a year in subsidies to do so.

There is no eqiuvilant ability to do so under (normal) definitions of employment, and no returns to scale (though you can cop out and claim this is covered under "elasticities") for the reverse situation.

Lewis writes:

If you read Stiglitz' Learning Economy, you see there's a pretty stark contradiction in his support for the minimum wage. If subsidies make labor input cheaper relative to capital, then firms will develop technologies and techniques that use labor more intensively and capital less intensively---i.e., that raise labor productivity and bid up wages in the long run. This will alter the labor/capital share in favor of labor over time. Capital will become less valuable if firms do not develop expertise in using it.

An example might be: suppose Chipotle could hire workers at an after-subsidy cost of $4/hr. They might develop a 'burrito school' that trained their many cheap workers to be very efficient with their hands and to field customer complaints. They might learn how to set up restaurants with two counters. By contrast, here in Berkeley they are discussing a $19 minimum wage, and I am willing to bet the first burrito-making machine will be field-trialed on Telegraph avenue.

As an analogy, in the US we have developed a lot of techniques and technologies that have amplified the output of cars and gasoline. In Singapore, the UK and Denmark they have made cars and driving prohibitively expensive and over time developed patterns of life, infrastructure, firms with economies of scales, agglomerations of expert workers etc. that make alternatives to auto transport both cheaper and more convenient on a per-unit basis. The extreme in the other direction actually isn't the US but rather Saudi Arabia. See also water consumption in Israel and Australia.

Nathan W writes:

I strongly agree with Danny that it should be preferable to distribute the benefits each paycheque instead of at tax time.

People are not very good optimizers and they can plan more sensibly when the subsidy comes in regular intervals instead of in lump sums.

Anyways, I tend to favour minimum wages with a safety net, because I think it is better for long-run productivity - firms are forced to think of ways to employ people that are worth $10 per hour rather than $6 an hour, for example. Since the safety net presumably offers a very paltry standard of living, any unemployed workers are forced to seek ways to increase the value of their labour to the new and higher minimum wage. Compare this to a subsidy which accepts the current lower level of labour productivity.

I would think that the value of labour subsidies would not be a general case, but rather something that can be prodcutively employed during periods of sharp downturn, in order to keep people in the labour market (and specifically, a labour market which is strongly expected to turn around and resume growth) so they don't lose skills.

Why provide wage subsidies, which incentivize relatively unproductive uses of labour, instead of raising minimum wages, which incentivize both employers and employees to find more productive uses of labour?

J Mann writes:

Nathan, your theory is that firms are capable of using their existing workers at a use that would return $10/hour in t your example, but they just don't because they can hire workers at $6? Doesn't the oriental profit give them MORE of an incentive to discover that use (and hire some additional workers!) than they would have if they paid their workers $10?

baconbacon writes:
It sounds more like you're describing potential for fraud rather than a scenario from a normally operating business.

It only sounds like fraud because I'm intentionally describing it in a way that highlights the lengths that you could go to. If you put it slightly differently you can see.

Instead of starting a new daycare center, take someone already running a daycare center. On passage of this rule you offer a new position of "helper". This position pays poorly (1-2$ an hour) but as a perk you can bring your own kids. Your duties are "help keep the kids play area safe and clean".

Scott Sumner writes:

Thomas, You said:

"I've "heard" a lot more often from the right that they oppose anything like an EITC that makes the tax system more progressive than from the left that an EITC would subsidize employers."

That's my point. The left likes the EITC. So why not a wage subsidy?

Bacon, Any program is subject to abuse, including this one. Inevitably there will be regulations to deal with those cases. But even with regulations, I'd expect problems, as with the EITC.

Greg Jaxon writes:
Who benefits from a wage subsidy?
All the submarginal entrepreneurs whose lousy ideas for using labor are made profitable thanks to wage subsidies.

The subsidy allows such employers to purchase labor whose productivity is (Wage+Subsidy)/hour while paying out only Wage/hour for it. Like the subsidies to purchase electric vehicles that cause more than the economically optimal number of electric cars to be manufactured, this causes more than the economically optimal deployment of labor.

That may be "humane" or "progressive", but it's not going to make the market freer or less distorted.

ThomasH writes:


There I think there are two separate issues:

Why does (if it does) the left like EITC more than a wage subsidy?

Why does (if it does) the left like minimum wages more than subsidizing employers.

I guess that the reason for the first (if it's true) is that the EITC already exists and a "wage subsidy" sound new and unknown.

I guess that the reason for the second (if it's true) is that it sounds as if the main effect of the measure is to transfer income to employers with little increase in aggregate wages paid to workers.

But the point is that I guess it would be easier with good economic arguments to the get a "leftist" to support an increase in the EITC or a wage subsidy that to convince a "conservative" support it.

John Thacker writes:
I've "heard" a lot more often from the right that they oppose anything like an EITC that makes the tax system more progressive than from the left that an EITC would subsidize employers.

You do hear that from a large portion of the Right. However, it's also certainly the case that the EITC has been expanded in every major piece of tax legislation, and on a bipartisan basis. It seems to be as easy to expand the EITC as it is to increase the minimum wage, politically, even though the coalition of support is different. The EITC supporters are among the economically educated across the spectrum, whereas the minimum wage draws popular support from economically less informed across the spectrum.

Nathan W writes:

I agree with your reasoning.

Sure, they will hire a larger number of workers in the presence of a wage subsidy. For the arbitrary numbers picked, they may hire people for positions with labour productivity between $6-10/hr.

I wonder, though, whether it would be better for long run growth and economic potential if we were to have a larger reserve of labour, who could only be legally employed once the positions to be filled were at labour productivity of greater than $10/hr (for the case of the arbitrary numbers picked). If you are excessively concerned about those who are unemployed in the meantime, all you need is a moderate safety net and access to training which will enable them to partner with enterprise in the future once there are positions available where labour productivity is greater than $10/hr.


@ John

I suspect the differences in coalitions of support for the minimum wage are driven by people with a preference to earn their salary (at a higher minimum wage) as opposed to receiving money from the government (EITC). I think there are very few people who prefer handouts to earning the money for themselves, given half a chance to do so.

Comments for this entry have been closed
Return to top