Deinonychus antirrhopus

The wage subsidies are phased out on a sliding scale, so that there's no cut-off effects.

What exactly does this mean? The subsidy declines over a course of X years of employment? So for the worker starting at $5/hr gets a $5/hr subsidy in year one, $4 in year two, etc.?

Not it would be more like the following

Assume the minimal wage is $10, so the subsidies would be as follows

$1 wage => $9 subsidy
$2 wage => $8 subsidy
$3 wage => $7 subsidy
$4 wage => $6 subsidy
$5 wage => $5 subsidy
$6 wage => $4 subsidy
$7 wage => $3 subsidy
$8 wage => $2 subsidy
$9 wage => $1 subsidy
$10 wage => $0 subsidy

The cost and scope would depend on the minimal wage one wanted.

Oh, my god.

I'm astonished that he could seriously post that. And I'm astonished, Steve, that you took it as seriously as you do.

If I was to pretend to take it seriously too, I would point out that there is no point to a sliding scale since no business in its right mind would pay any rate between the mandated minimum and the top of the subsidized wage.

Why stop at $10, though?

Shoot, why not make government subsidize a "living" wage, and make it set for, oh, NY or SanFran? That'd be at least $15-$20/hr, realistically.

Then, my next question is: Why would industry/business fork out the $5/hr? Why wouldn't they cut back to, say, $3/hr? Presumably, you actually retain the minimum wage, which means that whatever benefits are expected to accrue, are ameliorated, at best.

Then, more fundamentally, why is this government's job? If we believe that farm subsidies are a bad idea (and I think most folks agree that it distorts all sorts of prices), would a wage subsidy REALLY be a better thing? Indeed, if we wind up paying folks not to farm, at what point are we paying folks not to work?

Finally (not having read Poonwalla's original piece), I'm curious: What is it that this subsidy is supposed to fulfill? That is, even assuming it's somehow gov't's job to pay for better wages for people, exactly why are we paying these additional wages?

If I was to pretend to take it seriously too, I would point out that there is no point to a sliding scale since no business in its right mind would pay any rate between the mandated minimum and the top of the subsidized wage.

First, businesses do not have minds in that they are not living organisms. Second, you are wrong.

Okay, I see Dean is having the same problem.

Suppose we changed the subsidy from labor to say pork rinds. Everybody is given a coupon that they can use to purchase pork rinds. What would happen to the supply and demand of pork rinds? Would it look alot like the picutre I have drawn? Yes. Would you guys disagree with me? Probably not.

Labor works pretty much the same way. Its a commodity, it is bought and sold in a market. Solving the optimization problem for the firm shows that a decrease in the wage rate will increase its demand. This is what the subsidy will do.

Finally (not having read Poonwalla's original piece), I'm curious: What is it that this subsidy is supposed to fulfill? That is, even assuming it's somehow gov't's job to pay for better wages for people, exactly why are we paying these additional wages?

It is ideally supposed to replace the current welfare system of transfer payments. Instead people would work at something the firm is attempting to make a profit at.

Another counter argument is that the subsidy would allow firms to exist that otherwise would not. One could argue that in general society does not want these items badly enough to support such firms thus, they should not be made.

Now, I did make one simplifying assumption in drawing my graph, that the subsidy is constant at all wage levels. Thus, in reality the shift would not be a parallel shift, but one that is a rotation to the right at the point where the demand curve is equal to the minimal wage level.

Indeed, if we wind up paying folks not to farm, at what point are we paying folks not to work?

In a sense we already do that with the current welfare/transfer payment system. By paying these people who don't work, it makes it less costly to not work.

First, businesses do not have minds in that they are not living organisms. Second, you are wrong.

Bwaaahaaaaa. Got me. That was good.

Returning to pretending to take this seriously: Consider the problem with the increased cost/decreased demand for labor by those whose work is supposedly worth more than $10/hr once you include the enormous tax rates paying for this scheme.

Bwaaahaaaaa. Got me. That was good.

Hey, I can be a jerk too. :-p

Consider the problem with the increased cost/decreased demand for labor by those whose work is supposedly worth more than $10/hr once you include the enormous tax rates paying for this scheme.

Right, I pointed that out. Unless the wage subsidy is large many people wont support it as it'll just be an additional tax. Now if the current welfare state were to be dismantled at the same time (I know I'm a funny guy), then that could offset the increase costs. The wage subsidy could have a chance then...maybe. I agree that this is not likely to happen though.

A high tax rate could also end up reducing demand for labor to such an extent that the program isn't helpful. I haven't seen the specifics of Phelps plan, so I can't comment on them all that much.

I see this as sort of a half-way measure. Its better than the current welfare system, but it does come with some serous potential problems as well.

Huh.

Okay, here's where I see at least two problems.

Company A currently has folks on the payroll at $10/hr.

Question 1: Why wouldn't Company A simply reduce all of its $10/hr salaries to $5/hr (frex, call it minimum wage), and depend on the gov't to pick up the tab? It may well not even hire one add'l worker, but simply report this as an improvement on its bottom line?

Question 2: Why wouldn't a whole lotta folks at Company A start agitating for higher salaries (esp. since A's bottom line is improving), so that the real result is simply shifting the pay scale up the ladder? Wouldn't this prove inflationary for the economy as a whole?

There ya go, Steve, but isn't this a portrait of a recent visitor of yours?

Another problem:

All other things being equal (i.e. assuming that this doesn't merely replace existing welfare programs), doesn't this trigger a new round of inflation, as the bottom of the economy resets to $10/hr equivalent? At which point, you have a) diminished the value of work for those actually working without subsidy and b) reset the new 'living' wage from $10/hr to $10+x/hr where x equals the equivalent to the inflationary pressure. Also, isn't there considerable disincentive to work at higher rates/longer hours due to the tax burden added from the new expenditiures? (Note: I am deliberately ignoring any potential 'free-riding' behavior on the part of employers e.g. cutting wages to the bone to maximally exploit the subsidy.)

So we're interested in lump-sum transfers to lower income earners from high-income earners.

I want think differently about these topics; Steve's analysis works, but I want to frame the problem differently. First, in competetive labor markets, wages are determined by the intersection of supply and demand.

In my view (I'm really making this up as I go along), the demand curve for labor does not move, because the subsidy, from the firm's point of view, does not significantly change the marginal productivity of workers. It is the supply curve that shifts (to the red line in the figure below). If labor markets were perfectly competitive before the subsidy, firms paid $6 (let's say) to receive $6 worth of economic benefit. This does not change after the subsidy if markets are still competitive.

However, with the $4 government subsidy workers provide $6 worth of services at a cost of $10 to society. This inefficiency is represented by the yellow triangle.

Also, some workers (with higher opportunity costs) enter the workforce, facing the lower portion of the demand curve. The government subsidizes them too, and each person who is employed is a net loss to society, because although they get paid their marginal product by a firm, the government pays them the rest, leading to the pink are of inefficiency. (This chart ignores the minimum wage, which would cap the quantity at slightly higher than perfectly competitive (before), and less than shown for $10.)

This type of subsidy might be more efficient or less efficient than lump-sum transfers not tied to work. We might want people to work instead of doing nothing, but that has implications for the size and composition of the workforce. Effects on the capital to labor ratio are hard to figure out.

Walter,

I'm not sure what the impact on prices would be. We'd need a more detailed model to trace out the effects.

Kevin,

If labor markets were perfectly competitive before the subsidy, firms paid $6 (let's say) to receive $6 worth of economic benefit. This does not change after the subsidy if markets are still competitive.

This is exactly the opposite conclusion that Phelps arrives at. At first my thoughts echoed yours, but then I went through the math on it, and (assuming I did it right) the firm does indeed increase its demand for labor. While I agree there is no change in the marginal productivity of labor the cost of labor has changed.

Steve,

I agree that labor quantity demanded increases, and that the relative marginal price of labor decreases, but only because supply increases through the subsidy, increasing the quantity demanded, and lowering price. That's in my graph, although I didn't emphasize it.

It still appears to me that none of this is necessarily more or less efficient--in any total social welfare sense--than just giving people money and not demanding that they work in a paid job.

With a subsidy get more goods and services produced with subsidized labor, but fewer goods and services produced with unsubsidized labor.

In a strange twist, we subsidize students working in college bookstores who will make $100K later on, to the same extent we subsidize the poor who will never rise.

But I presume I should read through Phelps on this. I looked through his Structural Slumps, but it appears nontechnical. Which paper of his presents the formal model?

I dunno, I'd try his book which should have a reference to his technical writings....or e-mail him. I've found many professors are quite willing to share papers of their work.

Thanks.

I wrote Peter Gottschalk at Columbia U., requesting a copy of an unpublished paper in which he criticizes the Cox & Alm mobility figures.

I was pleasantly surprised by his very apologetic email 1 month later, to which he attached a PDF of his paper.

Damn, wish I'd checked in earlier.

Steve:

1. What incentive does a worker in Dean's Company A have increase his productivity? Employers can reward better performance with higher wages -- but in this case, any raise the employer decides to give is offset by the decrease in the subsidy (unless the raise takes the employee over $10/hr, and how many times have you gotten a 100% raise?)

2. What incentive does a manager at Company A have to give an employee a raise? Since the emloyee won't receive any benefit from increasing their wage rate, so why bother adding to the company's cost by giving a raise?

3. If I were a small business owner under this kind of program, since I couldn't offer employees monetary incentives, I'd think about using non-wage benefits (increased vacation, paying a bigger chunk of insurance costs, higher 401k contributions, etc.) to lure and keep employees. Maybe this is a good outcome, but it's not the intent of a wage subsidy.

Assuming that the increase in labor supplied comes out of the pool of net recipients of transfer payments (rather than, say, young mothers currently supported by their husbands who are lured back to work, or students or some such), then it should indeed increase social welfare. If we are currently paying these folks the equivalent of a $5 an hour wage to produce nothing, then paying them the equivalent of a $10 an hour wage to produce $5 worth of work is a net benefit, even before we factor in the increase in social capital as previously unemployed workers increase their productivity and skills.

Steve: I think your illustration of the sliding scale is off. I concur with Dean; as the chart is currently drawn, the long-term incentive would be for firms to pay all their workers whatever the minimum is and let the government pick up the slack between that and the subsidy level; ie, if the minimum is $10, and the system is designed to top all wages between $5 and $10 up to $10, firms will pay everyone worth less than $10 $5 and let the government bring their wages up.

On the other hand, if we give such workers a marginal phase-out, so that they are always better off, and set the phaseout to, say, 50% so that the worker is always better off if the employer increases their salary, the subsidy becomes somewhat higher, but should be more effective:

$5 - $10
$6 - $10.50
$7 - $11.00
$8 - $11.50
$9 - $12.00
$10 - $12.50
$11 - $13.00
$12 - $13.50
$13 - $14.00
$14 - $14.50
$15 - Parity

"If we are currently paying these folks the equivalent of a $5 an hour wage to produce nothing, then paying them the equivalent of a $10 an hour wage to produce $5 worth of work is a net benefit, even before we factor in the increase in social capital as previously unemployed workers increase their productivity and skills."

Emphasis added.

Jane wrote:

If we are currently paying these folks the equivalent of a $5 an hour wage to produce nothing, then paying them the equivalent of a $10 an hour wage to produce $5 worth of work is a net benefit, even before we factor in the increase in social capital as previously unemployed workers increase their productivity and skills.

I disagree, on the basis of the opportunity cost of working in a subsidized job. The current system is a pure transfer to people who value not-working more than working for their highest valued alternative. By enjoying their free time, those on the dole "produce" value for themselves. This value is real.

Some such people get, say, $8 per hour in benefit from not working, but could earn, say, only the minimum wage in a free labor market. That's why they don't work. [yes, huge simplification]

Now when we tell them they can get $10 an hour, they find it worthwhile to work for wages. Doing so, each produces the equivalent of $5.15 for the firm, and $4.85 is pure transfer.

Before they "produced" $8 per hour for themselves (because they value free time) and produce nothing for everybody else, and received $x in pure transfer. They got $8+x in total.

With a wage subsidy they enjoy $10 per hour for themselves, but produce only $5.15 for everybody else, receiving $4.85 in pure transfer.

Those formerly on the dole may be better off or worse off, depending on the size of the dole. If an individual got x>$2, then he's worse off with a wage subsidy. If x

But society is worse off, even ignoring the social transaction costs of transfer, because $8-$5.15=$2.85 less total value is produced.

I'm not trying to argue for the sake of arguing; I think the analysis is sound--even if it seems ridiculous. I do think there are other issues involved. But I wanted to point out some of the bizarre implications of a government policy that changes the relative price of paid vs. unpaid labor.

Kevin,

You're right in that there are other issues involved. I read somewhere that Phelps also discusses various positive and negative externalities associated with people being either on welfare or just not working (e.g., engaging in criminal activity to support themselves).

Now when we tell them they can get $10 an hour, they find it worthwhile to work for wages. Doing so, each produces the equivalent of $5.15 for the firm, and $4.85 is pure transfer.

Can you clarify a bit here. Are you talking about a $10 job under the subsidy or they get an actual job offer that pays $10 w/o the subsidy? My first response to what you have written is: Don't firms employ a factor up until the point where the marginal cost of that factor is equal to that factors marginal value? If you are talking about the subsidized wage, then no I agree with your point.

Jane and DaveL,

Good points. The way I have the subsidy scale set up there is no incentive to "get ahead". If you are making $6/hour and your boss says, "You're getting promoted you'll get $7.5/hour and you'll have to do a couple of extra tasks," the response would generally be, "Thanks, but no thanks."

Can you clarify a bit here. Are you talking about a $10 job under the subsidy or they get an actual job offer that pays $10 w/o the subsidy?

My first response to what you have written is: Don't firms employ a factor up until the point where the marginal cost of that factor is equal to that factors marginal value? If you are talking about the subsidized wage, then no I agree with your point.

I'm talking about the subsidized wage.

That should clarify things. If it doesn't, then perhaps this will. Let's talk about one guy formerly on the dole. With a wage subsidy, if he works he'll get $10 total--the market wage + the subsidy.

According to my graph, firms will pay the wage where supply equals demand. That is, the private marginal cost to the firm will equal private marginal benefit (which equals the marginal productivity of labor).

After the subsidy, firms will purchase a chunk of extra labor, with all these guys having lower productivity than the marginal productivity under the dole system. Firms will not pay more than the MPL --subsidy be damned.

I assumed that $5.15 would be the guy's personal productiviy. We could say it's $6 or $7 or $7.50. He still wouldn't have worked with a dole system, because he values free time at $8. And with a wage subsidy, "society"--everybody else & our guy--still gets less product than it used to.

If his productivity was over $8, then he would have been working.

If he is really low skilled, with a personal productivity of labor below any firm's margin, then he'll be unemployed under a wage subsidy system.

For all this nonsense I've put you through, you can thank none other than
Walter Williams, my micro I professor, for requiring an insane amount of reading of the works of
Armen Alchian.

I now leave this matter until I can hunt down Phelps' writing.

Happy New Year!

I now leave this matter until I can hunt down Phelps' writing.

If you find and e-copy send me one please. I'll try e-mailing Phelps to see if he'll send out pre-prints of some of his articles.

You're dead wrong, Steve.

This Brian flashback brought to you by the year 2004.

Thanks Slart, I was begining to go into withdrawl.

Thanks Slart, I was begining to go into withdrawl.

What, are you a Texan?