« Dean Baker Speak | Main | Fairness and Equality »

October 25, 2006

The Minimum Wage

Every once in awhile, I like to reengage the minimum wage debate. It's such an article of faith on the right that minimum wage increases lead to widespread unemployment, and such an intuitive argument, that society would have to be a pretty bizarre place not to abandon the wrongheaded policy altogether. Except for the fact that, intuitive as the argument may be, and faithful as its rightwing advocates may prove, there's just not much evidence that minimum wage increases have a measurable effect on unemployment.

The foundational study in this area -- which various economists have sought to confirm or reject, all with varying, and often contradictory, success -- is the Card/Krueger survey of New Jersey restaurants that found a slight positive impact on employment. Lots of controversy on that result, but now there's new data from Arindrajit Dube, Suresh Naidu, and Michael Reich checking out San Francisco's restaurants after they instituted an $8.50 minimum wage. As Kash Mansori points out, the results are best explained in this graph, which also shows neighboring Alameda, which didn't have a minimum wage bill passed:

As Kash explains, "The minimum wage goes up in one place, but doesn't change right next door. Employment in restaurants goes up in both places - if anything, by more in the place where the minimum wage went up...Putting all of these different types of papers together, my conclusion is that the best evidence that labor economists can gather from US data seems to indicate that we need not fear major employment losses if we were to increase the minimum wage. The effects may be slightly negative for teenagers, but overall the effect on jobs may be zero to even slightly positive."

Meanwhile, the invaluable folks at EPI have a briefing paper on the likely impacts of raising the national minimum wage to $7.25. For all the talk of teenagers and so forth, 80% of the 15 million affected workers would be older than 20, $7.3 million children would see their parent's income rise (46% of families with a minimum wage worker rely mainly on that income), and so forth. Well worth a read.

Update: Relatedly, here's a recent letter signed by over 650 economists, including five nobel prize winners and six past presidents of the American Economic Association calling for an increase in the minimum wage. That's not to say opponents of an increase couldn't furnish a similar letter -- the point here is that there is not, as the right would like you to think, an economic consensus that raising the wage would be a bad thing.

At Tapped too.

October 25, 2006 in Economics | Permalink

Comments

Cue Fred Jones with the "Why not raise it to $20???" in five ... four ... three ....

Posted by: Thlayli | Oct 25, 2006 4:33:57 PM

Ruben Bolling's cartoon on the minimum wage packed in so much in such a small space.
http://www.salon.com/comics/boll/2006/09/28/boll/index1.html

Posted by: Carl | Oct 25, 2006 4:45:01 PM

It's Alameda, not Alameida. And for the life of me I can't think of a decent restaurant in Alameda

Posted by: Brendan | Oct 25, 2006 5:11:36 PM

Also, is that the city of Alameda, or Alameda County, which is Oakland + some inner ring suburbs?

Posted by: paperwight | Oct 25, 2006 5:14:19 PM

According to the link Alameda County is used in the comparison.

Posted by: talboito | Oct 25, 2006 6:38:02 PM

You can't outsource a restaurant.

Posted by: slickdpdx | Oct 25, 2006 6:46:35 PM

article of faith on the right that minimum wage increases lead to widespread unemployment

Let's be honest here: the GOP and businesses don't hold this as an article of faith, they hold this as the rabbit-out-of-the-hat excuse on why they want more of the profits going to owners and stockholders, and less or the same bottom-line profit share going to workers.

Does anyone really think they have are truly concerned about preventing higher unemployment? Higher unemployment is GOOD for business: they can hire workers for less money with more competing for jobs.

Capitalism/Corporatism has no moral conscience except in very rare exceptions (and wages are not one of those exceptions), and they have little concern for the social consequences of workers being paid less than the minimum costs of sustaining a decent living.

Accepting seeming facts based on data in the case of the wages/employment relationships is just as likely as Bu$hCo accepting the data findings of the global warming issue. They don't want facts, they want ideology and profits.

Posted by: JimPortlandOR | Oct 25, 2006 8:00:38 PM

How come this site works fine witrh FireFox while Tapped is funky ... I never get to preview my comment over there!

Anyway, to sort through the logic, you have to go to the premises.

The simple economics says that wages are both an input cost and a source of income, and that while rising costs are bad for business, rising incomes in the market is good for business.

Of course, the impacts won't be distributed evenly across industries ... the negative wage-cost effects will be focused in low productivity jobs, and the positive income effects will be spread more widely, and so on average will go toward high productivity jobs. That is, while the net employment effect tends to be around zero, there is some swapping out of bad jobs for better ones.

Pursuing a declining real wage economic policy is pursuing a declining real income economic policy, or in a phrase, economic underdevelopment. Pursuing a rising real wage economic policy is pursuing a rising real income economic policy, or in a phrase, economic development.

Note, however, that the closer the minimum wage is pushed toward the median wage, the more likely we are to see a negative net employment effect. $7.50 required to get above the 1 adult and 1 dependent poverty line on 40 hours a week, 50 weeks a year. In highest the minimum wage has been in terms of present CPI purchasing power has been about $9, so that is well within the range of experience we use when we say that there is no evidence of significent net negative employment effects. Push much beyond that, and we are in uncharted terrain.

Cue Fred Jones with the "Why not raise it to $20???" in five ... four ... three .... {holds up two fingers} ...

Posted by: BruceMcF | Oct 25, 2006 8:18:18 PM

I think the only relevant comment along this thread is that the right makes shit up. To that, I add once y ou realize they are just making shit up, it makes it incredibly easy to understand how to handle their arguments with all the seriousness that they are due.

Posted by: akaison | Oct 25, 2006 8:29:40 PM

Until Bruce McF did his thing, I was going to go along with slick and say that minimum wage jobs, which I think are something like 2% of the labor force, are low elasticity. (I got some jargon, I just use it badly). Like Walmart or McDonalds, the minimum-wage industries are probably very close to maximum efficiency and very low margins already, and cannot cut payrolls without losing sales. So the response would be to shut down stores, or increase prices. Or seek efficency by other means, perhaps in consolidation within industries.

Posted by: bob mcmanus | Oct 25, 2006 8:46:24 PM

Excellent point Ezra: the parents of $7.3 million children are often an overlooked demographic.

Posted by: foolishmortal | Oct 25, 2006 9:13:28 PM

Posted by: bob mcmanus | Oct 25, 2006 5:46:24 PM ... Like Walmart or McDonalds, the minimum-wage industries are probably very close to maximum efficiency and very low margins already, and cannot cut payrolls without losing sales. So the response would be to shut down stores, or increase prices. Or seek efficency by other means, perhaps in consolidation within industries.

This is projecting. However, it is important to point out that the Card and Kruger study referred to above compared fast food restaurant employment on both sides of a state border before and after one of those states raised the state minimum wage.

When you work out an abtract prediction, you get out what you put in. In the abstract prediction above, part of what is included is the assumption that wages affect costs, but wages do not affect income.

There is also a tacit assumption that the economy is at full employment, because if the economy is below full employment, there will be a gap in many individual labor markets between the maximum amount that a business can pay its workforce, based on the commercial benefit of their work, and the minimum amount that has to be paid to get them to work. If employers hold the stronger negotiating position, wages will tend to be toward the bottom end of that gap.

In a relative sense, Wal-Mart is better off in a low-wage, low-income economy, while higher margin competitors are better off in a high-wage, high-income economy. But there's no reason to think that retail employment as a whole will drop because of an increase in the minimum wage up to the poverty line. In the neighborhood of the poverty line, retail and fast food ought to have income elasticities greater than 1 ... on average they should see a percentage increase in sales to low-income individuals that is greater than the percentage increase in the incomes themselves.

Posted by: BruceMcF | Oct 25, 2006 11:22:03 PM

The Australian Fair Pay Commission has announced its Federal Minimum Wage decision for October 2006.
The Commission announced an increase of $27.36 per week for minimum wage rates up to $700 and $22.04 per week for minimum wage rates $700 and above. This brings the Australian minimum wage to around $13.47 an hour which adjusted by purchasing power parities is around USD 9.76 an hour. (At the moment the Australian unemployment rate is 4.8%)

The decision and associated documents can be found at http://www.fairpay.gov.au/fairpay/MinimumWageDecision/Minimumwagedecisionno1-2006.htm

There is also a range of supporting research at http://www.fairpay.gov.au/fairpay/ResearchAnalysis/Researchandanalysis.htm.

Posted by: Far Away | Oct 26, 2006 7:11:18 AM

Way to go Ezra:
'It's such an article of faith on the right that minimum wage increases lead to widespread unemployment,'

Start off with a straw man argument.

No, the article of faith is not that it will lead to widespread unemployment. It is (at least amongst the economically rational righties) that there will be, for modest rises, modest effects on total employment, effects negative but sufficiently modest that we won't be able to see them in the overall churning of jobs in the economy.

There are a number of extremely good arguments against using the minimum wage as a poverty reduction method: I've written about them at length elsewhere. But perhaps one that might be of interest here, advanced as it is by both Card and Kreuger and Paul Krugman?

It's a grossly expensive way of trying to raise the incomes of the working poor: because of the interaction of the tax and benefits system the rise in incomes pre tax and benefit provided by a rise in the min wage becomes a very small rise in post tax post benefit incomes after the EITC, rent subsidies and so on are clawed back.

In effect, as well, it becomes a substitution of aid from all of us (the taxpayers) to the working poor to a tax paid only by the employers of minimum wage labour.

If you want to raise the incomes of the working poor then the best way is by expanding the EITC (yes, something I support until we actually get to a rational system of a Citizen's Basic Income).

Posted by: Tim Worstall | Oct 26, 2006 7:59:34 AM

And -- wouldja believe it -- I support doing both!

Also -- if you don't believe widespread unemployment from minimum wage increases is an article of faith among some of yoyr compatriots, you should;ve been around last time we had this debate...

Posted by: Ezra | Oct 26, 2006 9:19:05 AM

Widespread underemployment, yes, but not unemployment. There's good empirical evidence from the UK (actually from the Low Pay Commission, the entity that decides what should be recommended as the minimum wage) that a rise in the minimum wage causes small rises in unemployment (about equal to a few days worth of the normal job churn) and a more substantial drop in hours on offer to min wage workers.

Posted by: Tim Worstall | Oct 26, 2006 10:23:26 AM

Ezra,

You have provided me with the best example of letting localities set their own minimum wage standards and eliminating the federal minimum wage!

According to your information, it just makes good sense for local communities to be allowed to set minimum wages that reflect the needs of the community on a community-by-community basis.

Since a national minimum wage can not possibly take into account every local condition then we can eliminate that meddling in local affairs and get back to the enumirated powers of the federal government.

Posted by: Guy Montag | Oct 26, 2006 10:35:51 AM

I can't believe this is even being debated. If the evidence isn't crystal clear by now, raising the minimum wage doesn't effect unemployment.

If the effects were that large, states that hadn't raised the minimum wage over the last decade would have shown spectacular relative growth.

They haven't.

If the effects were that large, states that hadn't raised the minimum wage over the last decade would have shown would have shown much lower unemployment rates compared to states with higher minimum wages.

They haven't.

This isn't really debateable at this point. At most, a few thousand workers nationally might lose their jobs, while rasing the wages of millions.

Its a little difficult to square the concerns of the libertarians on this. Remember, EVERY policy issue has effects like this, where some people are favored over others. I don't understand how the great concern they are showing for a few thousand workers at a maximum can be reconciled with the overall lack of concern for the millions of people whos lives will be materially improved. Its the old marginal utility of dollars, my friends.

Posted by: mickslam | Oct 26, 2006 12:06:16 PM

Posted by: Tim Worstall | Oct 26, 2006 4:59:34 AM It's a grossly expensive way of trying to raise the incomes of the working poor

The working poor ought to be restricted to those who cannot obtain full time work. The full time working poor ought to be a category that does not exist.

Posted by: Tim Worstall | Oct 26, 2006 7:23:26 AM There's good empirical evidence from the UK (actually from the Low Pay Commission, the entity that decides what should be recommended as the minimum wage) that a rise in the minimum wage causes small rises in unemployment (about equal to a few days worth of the normal job churn) and a more substantial drop in hours on offer to min wage workers.

The first point that may be made off the bat is that the UK minimum wage was to 60% of median wage, not 53% of the median wage represented by a poverty-line minimum wage of $7.50 in the US, so the UK study prediction of about 1 hour less work per week in the UK would likely be lower for a US$7.50 minimum wage.

However, more critically, note that this empirical evidence is based on an assumption of absolutely no income affect on employment. Some people may note that survey evidence regarding whether restaurants feel that they employ people for fewer hours under a minimum wage is not only subject to an obvious bias if the respondent feels they will have the same employment but less profit income ... but even if the responses are 100% accurate, the question of how they would respond to an increase in customer demand is not asked. It is simply assumed that there is no income affect.

And the actual research paper finding that Tim is referring to here does not do a before and after comparison across a boundary, with a rise on one side of the boundary and no rise on the other, it does a hypothetical comparison that depends on the assumption of no income affect for its conclusions.

The hypothetical comparison uses workers who are just over the target minimum wage mark. A regression analysis is performed to model what happened to workers directly affected by the minimum wage and workers just over the minimum wage mark, and finds that those affected by the minimum wage were employed about one hour less per week than they would have been had their employment acted like the employment of those just over the minimum wage mark.

Of course, with no income affect, this is one hour lost compared to those who were unaffected by the minimum wage. With an income effect, this is a net difference of one hour, with no telling how it was distributed between general income affect gains and direct cost of employment losses.

Posted by: BruceMcF | Oct 26, 2006 1:27:13 PM

I would expect restaurants to be a poor choice to get an answer to this question. Restaurants cannot relocate for cheaper labor, they have to be in the same locale as their customers and similarly the customers cannot outsource their big macs.

Most restaurants, at least from my expiriences as a teen-ager, already make considerable efforts to minimize labor expenses, so one wouldn't expect a marked reduction their either.

So the only possible negative effect of raising a minimum wage requirement on restaurants would be if the price increased enough that consumer behavior (cooking at home rather than going to restaurants) lowered the demand for Restaurants overall. I wouldn't expect a modest increase to have that effect.

Restaurants are not the only minimum wage industry though. Another example would be a call center. Call centers can, as their name implies, be anywhere and they are quite sensistive to the price of labor. Even then though, I wouldn't expect the effect to manifest immediately, as a call center does have signifigant capaital investment involved, and the most likely effect, new call centers going to alternate places is very hard to measure.

I do think Ezra's characterization of 'wide spread' unemployment is inaccurate of the conservative posiition on this. More than would otherwise exist is more accurate. An additional negative effect would be increased inflation. Most minimum wage jobs have low margins and high labor costs so any increase in labor costs will have a direct effect on consumer prices.

If their are studies that show this is not the case, I am not aware of them. I would expect that the Restaurants in the studies above have had to raise their prices in response to this law.

Posted by: Dave Justus | Oct 26, 2006 2:00:11 PM

Okay, I have to keep reminding people that, regardless of whether or not there is some marginal reduction in employment (either from increased unemployment or fewer hours worked), raising the minimum wage will increase the incomes of lower wage workers, so long as wage elasticity is below 1. And you can find a real consensus among economists that it is substantially less than 1; in fact, I'm not sure you can find any economist anywhere who would argue that it's greater than 1.

Which means that lower wage workers make more money and work less when the minimum wage is raised. This is a problem for those who believe that the only thing keeping the rabble from mischief is constant, grinding labor, but most people would be fine with it.

So the debate on whether or not raising the minimum wage causes unemployment is a side show (albeit an interesting one). The idea that it hurts workers is simply pernicious. It does represent a transfer of wealth from employers to workers, but then the current social security / capital gains tax structure works to transfer wealth from workers to investors and no one is screaming about that.

Posted by: James Killus | Oct 26, 2006 3:05:46 PM

Wait!

Are these restaurants even paying the "minimum wage" to begin with? When I worked in a few of them as a teen we had a different "minimum wage" even if we did not get tips or get tipped out.

Also, everyplace I am familiar with in VA, DC and TN, the restaurant workers get their entire hourly pay check taken by the feds assuming that they actually got some percentage of tips for what they served.

In short, no restaurant worker that I know of is impacted by minimum wage at all, other than it being more money for the government to take from them before they ever see it.

Ezra, you can easily check this out when you are hoisting a pint with the editors at their favorite watering hole. Just ask the staff, they are pretty free with that info.

Posted by: Guy Montag | Oct 26, 2006 3:47:11 PM

Posted by: Guy Montag | Oct 26, 2006 12:47:11 PM Are these restaurants even paying the "minimum wage" to begin with? When I worked in a few of them as a teen we had a different "minimum wage" even if we did not get tips or get tipped out.

California (I don't know how many other states and locales) does not allow tips to be included as part of the wage, so there is not "tip credit". Indeed, that was part of what killed the poison pill minimum wage and Paris Hilton tax cut bill ... the provision that would have driven San Francisco wages in locations like that from around $9 to less than $3.

Posted by: BruceMcF | Oct 26, 2006 9:26:02 PM

BruceMcF,

So, the feds just get a bigger paycheck and the worker sees no difference?

You see, CA may not include tips in the wage, but the feds sure do.

Posted by: Guy Montag | Oct 27, 2006 7:38:58 AM

Posted by: Guy Montag | Oct 27, 2006 4:38:58 AM BruceMcF,

So, the feds just get a bigger paycheck and the worker sees no difference?

You see, CA may not include tips in the wage, but the feds sure do.

That seems to depend on whether the Democrats or Republicans control the House of Representatives when the law is passed. Of course, in all cases there is a formal requirement to ensure that the employee earns the minimum wage, but just like the formal requirement to not employ people without work rights in the US, many employers seem willing to break the law so long as it is not effectively enforced.

1996 was the first time that the minimum wage after subtracting the tip credit was left at the previous amount ... in all previous minimum wage increases, that amount was given as a fixed percentage of the minimum wage.

So I guess we'll see in a week and a half.

Posted by: BruceMcF | Oct 27, 2006 1:39:39 PM

The comments to this entry are closed.