« On understanding and spinning your own New Keynesian model | Main | Neo-Wicksellian indeterminacy in pictures »


Feed You can follow this conversation by subscribing to the comment feed for this post.

The positive effect would come from shifting the Overton window for acceptable wages. There would be obvious disincentives (like bad publicity) to paying "unfair" or "substandard" wages and incentives (like tendency to follow social norms) to not doing so. If you view labor share as at least somewhat sociologically determined, you could get a positive effect where labor costs are a relatively small percentage of total costs or where bad publicity conflicts with company marketing and/or combating it has costs. Maybe some clever graduate student could test this somehow.

Also usually window shifts move the center of the window, as well, which would have an even larger effect (but not necessarily a good one).

I recommend this paper by Bruce Kaufman... INSTITUTIONAL ECONOMICS AND THE MINIMUM WAGE:
There is a pdf available on the web... easily searched.

Thank you Edward. The abstract is below if anyone is interested:

Bruce E. Kaufman Institutional Economics and the Minimum Wage: Broadening the Theoretical and Policy Debate

Debate among labor economists on the pros and cons of a minimum wage law has come to focus on whether labor markets are competitive or monopsonistic. The purpose of this paper is, first, to suggest that this perspective is too narrow and, second, to use principles and concepts of institutional economics to demonstrate why. In particular, I use institutional theory to develop four theoretical rationales for minimum wage legislation: imperfect competition and inequality of bargaining power; involuntary unemployment and destructive competition; efficiency and growth; and labor market externalities and social costs of labor. One revisionist implication is that a minimum wage under plausible conditions may increase economic efficiency even in a purely competitive labor market.

Concerning the MACROECONOMIC problem aspects, two things come to mind:

1. The effect of increasing the minimum wage rate on aggregate wage income simply depends on the elasticity of labour demand. If labour demand is elastic, as in the benchmark case of a Cobb-Douglas production function, increasing the minimum wage rate will REDUCE aggregate wage income. This is the most likely outcome.

2. But assume, for the sake of the argument, that labour demand were completely inelastic. Since money can only be spent once, profits would decrease correspondingly so that aggregate income (wages and profits) remained unchanged. So, even under this extreme assumption, the expectation of positive macroeconomic effects of a minimum wage increase appears as voodoo economics.


since you seem to know about switzerland, any comments on their minimum wage ?

The Kaufman article is great - a must read.

@Herbert's point #2 - If, for some reason, demand for labour is completely inelastic, and the increase in minimum wages is purely paid from reduced profits, then within the scope of "Ontario," it's hardly voodoo to assume that there would be a net positive effect, since many profits flow to out-of-province shareholders.

My personal thoughts on the subject are that you really can't answer the question through theory...too many confounding variables. You really need some empirical studies. Just how elastic is the demand for bottom end labour? For many minimum wage employers the option is either shrink their business, or pay the higher rates. You can't really run a Mcdonald's on fewer employees than it currently uses, so maybe you can cut some marginal hours, but for the most part, shrinking the business just hits your bottom line through reduced volume instead of higher costs.

Furthermore, how much of the increase can get passed on to customers? Depends on the elasticity of demand for their product. In many industries, the bulk of the increase will pass on to consumers, and thus be a transfer from the higher wage earners doing the buying to the lower wage earners behind the counter.

Efficient, certainly not, but only a good empirical study is going to tell us whether there's a net stimulus or not.

Watching Ziggy run over a homeless man and then devoting her life to animal rights; it's the least ON could do.

> Moreover, as a policy measure to stimulate the economy, would minimum wage increases be superior to other stimulus measures such as tax decreases or increased infrastructure spending?

I think that the important policy measure is that a minimum wage increase is "free," from the perspective of government finances. Tax cuts, infrastructure spending, or other fiscal measures all come at a political cost, and of course Ontario cannot unilaterally use monetary policy.

From Kaufman:

"The reason is that the minimum wage reduces or eliminates the externality-like gap between the private and social cost of labor and thus improves economic efficiency. The effect is analogous to placing a tax on a paper mill that dumps pollutants into a river. The higher cost causes the firm to reduce production and cut employment, but economic welfare is improved—not hurt—because the tax corrects a market failure (a missing property right) that allows the firm to use a valuable social resource (the river) without paying the cost. A minimum wage is also, in effect, a tax on firms, but these firms—like the paper mill—are using a resource to make profit without paying the full social cost. The minimum wage, therefore, has exactly the desired effect: it ends (or reduces) the subsidy on low-wage labor and causes firms to cut back on production and employment to the efficient level that would prevail if the labor market were truly at a competitive equilibrium..."

@genauer: There is no general minimum wage in Switzerland. As a rule, those European countries which have a low unemployment rate do not have a minimum wage as yet. These are Switzerland, Austria, Germany, Denmark, Sweden, and Norway.

@Neil: Yes, profits may be distributed as dividends to foreign shareholders. But when you consider the circular flow, the money is not gone but also part of aggregate demand.

A bit off topic….

Minimum wage rules are a blunt tool. A smarter option would be to allow employers to pay sub-minimum wage rates, while state employment agencies helped those concerned look for more productive work, and limited the time those concerned stayed with any given employer.

There is nothing wrong with a $1/hr job if it’s all that’s available and assuming the state makes sure that take home pay is up to some socially acceptable minimum via in work benefits. What’s wrong is someone STAYING IN such a job when there IS SOMETHING better available. And the latter system under which no one can stay in a $1/hr job too long, and in which the state helps those concerned look for something better might achieve the latter ideal: allowing $1/hr jobs without the “low productivity encouraging” effect that such jobs can entail.


I seem to recall some recent literature which suggested alternative responses for employers, other than paying the minimum wage or cutting hours.

First, they can cut back on other non-legislated "perks". So employees get higher wages, but have to pay for their uniforms (or pay more), lose their employee discounts, get reduced training or see other benefits reduced. This mitigates the cost of the minimum wage for employers and the benefit for employees.

Second, they can work their employees harder (which, I suppose, might be linked to reduced employment). So they cut back on washroom breaks, increase pace, etc. In this scenario, workers may earn more money, but are potentially worse off to the extent that the increased effort is costly (in welfare terms).

But I suppose this goes to your general point that it's really an empirical question.

Why stop at raising just the lowest wages? Why not raise all wages? And all non-wage income too? A simple thought-experiment: suppose you passed a law that doubled all wages and prices, so everyone's income doubled. Would that raise aggregate demand, and increase employment?

A second thought-experiment:

"If we passed a law raising the minimum price of apples, that would increase the incomes of those selling apples, which would cause them to buy more goods, and increase Aggregate Demand."

Now substitute "labour" for "apples". What's the difference?

Gauti Eggertson has a paper that claims that, at the zero lower bound, higher minimum wages and more monopoly power is a way of implementing economic stimulus. He cites the New Deal experience as evidence. See http://www.econstor.eu/bitstream/10419/60658/1/522100546.pdf


On the macro-economic effects, the other possibility is that increased costs would be passed on to domestic consumers. Given the prevalence of minimum wage jobs in sectors that are likely to be immune from international competition (retail, food and personal services, etc) that seems like a plausible outcome (and more plausible than the suggestion that those costs would be born by international capital holders).

Angelo: the mechanism in the Gauti Eggertson paper is economically equivalent to passing a law requiring everyone to work with one hand tied behind their back. It reduces aggregate supply, which increase inflation, which reduces the real interest rate, which lifts the economy off the ZLB. Saturation bombing to destroy output would work too, in his model.

There was a useful Vox piece on claiming some basic stimulus gains that should be added to the mix of our discussions.

"Would that raise aggregate demand, and increase employment?"

This isn't on point. Simply raising aggregate demand isn't the purpose/mechanism of raising the minimum wage. Read the Kaufman paper. You have to refute his arguments. Employees aren't apples.

@Nick Rowe and Peter N: If think Nick's substitution of apples for labour makes the macroeconomic point under consideration quite clear. There is no effect on "aggregate demand", it's all only broken window fallacy.

Kaufman's paper mentions the main point in passing: Minimum wages are preferred by a majority of voters, hence politicians implement them. But I think minimum wages are inhumane as they effectively prevent less productive people from getting regular jobs.

I would think this was clear enough for anybody:

"Reducing poverty was expected to be a benefit of the FLSA, but an indirectbenefit achieved by accomplishment of other direct goals. What were these direct goals? A reasonably close synthesis of the language of the section leads to these four:

• eliminate labor standards that are so low they harm the ongoing efficiency, health, and well-being of workers.

• prevent unrestrained competition in labor markets from further lowering labor standards in affected industries, or spreading low standards to other industries.

• prevent low labor standards from interfering with attainment of full employment and sustainable economic growth.

• eliminate low labor standards because they lead to labor disputes and divisive relations between employers and employees, thus further harming economic activity.

None of these goals applies to apples. Kaufman proceeds to discuss the evidence for these.

Peter N: I was responding to Livio's post, not Kaufman's paper.

If you want a response to K's paper (or the bit you quoted from it), here it is: if there's a negative externality, you should impose a Pigou tax. Does K want to impose a special tax on low-wage workers?

Some people make their living selling apples. Apple sellers are people too.

Peter, what's the externality associated with hiring labor that is analogous to pollution? And if I understand the argument correctly, he's not saying that higher minimum wages will increase employment, just that lower employment is... a good thing?

This discussion and Nick's ever so droll response reminded me of an issue that arises each term with a handful of students in the business school 4th year capstone course.

In the student's analysis of a publicly traded company, each is required to, inter alia, compute the market cap for the last 3 years for the firm analyzed and compare to the market cap of each of the direct competitors and then normalize each market cap using the price to sales ratio. The logic of comparative market caps is that market capitalization is a not bad proxy as one measure of value creation - the purpose of the firm - see Schumpeter and Coase.

Each term, some students tell me in their analysis that the market cap of firm X increased BECAUSE the firm issued a large number of new shares.

Last week, in exasperation, I noted that I am the sole owner (shareholder) of my very practical Honda Element worth perhaps $15K. IF in a fit of madness (I am not a co-owner of her car), I added my partner on the ownership certificate - so that now there are 2 shareholders instead of 1 shareholder - does my Honda double in value to $30K?

IF so, as Nick noted re doubling all incomes, I want to acquire multiple partners and keep adding them to the title of my Honda (or my house). I am going to become fabulously wealthy overnight - as I drive up the value i.e. market cap of my Honda (or my house) - not to mention all those new partners.

I am quite giddy contemplating my micro solution to becoming rich by creating more shareholders - and the macro solution to all of us becoming more wealthy - by simply increasing wages for everyone by legislative fiat.

Ian Lee:
I guess what I found especially amusing about the argument the students were using to support a minimum wage increase was why they stopped at only 14 dollars? Based on their arithmetic, if raising the minimum wage to $14 gives the economy a 5.1 billion dollar boost, why stop there? Based on their argument, continually raising the minimum wage could provide an ever increasing economic impact.


I raised a somewhat German specific aspect about minimum wages recently on a somewhat more conservative blog


which was met with deafening silence : - )

Beyond that, I actually have here a personal example, where somebody can do work (house cleaning) at a value of 8 Euro = 11 Dollar, but not higher, because she is sloppy, working slowly (needing 3 – 3.5 hrs for what the more commercial competition does in 1.75 – 2 hrs), actually did some damage of a few 100 on the first occasion. But she lives close, needs the money, I have some sympathy.

Why should anybody else have the right to interfere with a mutually agreed contract?

But since she went through my (very modest) DVD collection, and “borrowed” some (“I hope you don’t mind”) not giving them back for a month now, the specific issue is moot.

But the principle case stays, who decides about contracts on wages which are only worth below a certain threshold?

"If you want a response to K's paper (or the bit you quoted from it), here it is: if there's a negative externality, you should impose a Pigou tax. Does K want to impose a special tax on low-wage workers?"

The minimum wage is, in effect, such a tax on employers, where the results of the tax are paid to the employees. Compensation of non-employees is indirect deriving from the changes resulting from the tax.

That's why the argument "if $7.50 is good, why isn't $12.00 better?" doesn't apply. A Pigou tax has an optimal level.

You can reinterpret lots of things as Pigou taxes, just as you can analyze many complicated financial arrangement as combinations of purchases and options. Overtime rules are taxes where the tax rate is the present value of product of the penalty and its probability.

The supposed advantage of making it a penalty is the belief that a regulation and penalty impose stronger social norms. The penalty should, ideally cover the cost of its collection and also a deterrent equivalent to the cost of what is deterred.

Establishing social norms is a perfectly valid way of preventing externalities, and often, though not always, the best and cheapest way. The danger, as with all propaganda, is its outliving its usefulness or exceeding its desired scope.

It doesn't seem unreasonable that in the short term there is an economic stimulus due to transfer from those with a lower propensity to spend to those with a higher one, in the medium term some loss of jobs and reduced job creation due to higher costs and reduced demand, in the long term creation of some higher paying jobs creating and installing technology with higher productivity. Not unlike inflation but beginning with labor rather than commodities, more useful at times like this to increase demand and in the long term productivity.

I agree schemes like a negative income tax and GAI where at the very least, is not used for addictions, and possibly should be used for some socially beneficial utility, are superior to a larger minimum wage. But a minimum wage increase is more politically viable. Basically, alot of why economics is inefficient is ecoonmists don't have time to study things. Multidisciplinary studies are impossible if you are marking tests and making sure the references on your Paper are factually correct and don't annoy the cited author.
It is better not to waste time with the university degree if you cannot select only improtant courses, if you are able to gaze what is important and what is vestigual or parasitic. Truly valuable people have an oscillating income, at least until or if they live off their capital gains. So you don't want to keep them poor forever if you can help them by ignoring AB/Texas/Italy-commies and taxing AB/Texas/Italy-commies. Probably the first few hundred hours worked/yr at a higher minimum wage is most efficient. The GAI should come with some forced Walden-Two-ish labour at the very least. People do need a Protestant work ethic, though you don't want to make them waste their lives in this manner. Nick, you don't just want double everyone's wage, because once you net about $15/hr you can save for your own retraining or education or whatever...perhaps some financial managment could be offered for the middle class.
The Eggerston paper asumes deflation of 15%/yr. Maybe in a pandemic or massive AGW drought it woyuld be relavent assuming gvmt actors that can't maitain civil order and/or enact stimulus.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad