[glenra]

Doesn't really answer my point. An average increase in costs of 5% or even 3% is still enough to kill restaurants on the margin.

For instance, suppose you are a McD franchise owner whose restaurant has 50 "crew members" (source on that number: https://answers.yahoo.com/question/index?qid=20070917131148A... ) and your restaurant recently became profitable to the point that it currently makes an annual profit of $250k. A quarter million dollars in annual profit isn't a lot but it's respectable - that is a sustainable business! That's something you can afford to keep operating indefinitely (and try to grow over time). You have some room to invest, some room to make mistakes.

Now suppose the state passes a law requiring you to give all 50 of your crew members an extra $5/hour. How much does that cost? 50 people * 2000 hours/year * $5 = $500,000. So your restaurant just went from gaining a quarter million to losing a quarter million a year.

Now it is not a sustainable business. Your pockets are not deep enough to keep losing that year after year; you should probably close the business or scale it way back.

Are you going to claim that every restaurant in the chain has enough buffer to take a half-million-per-year hit to the bottom line and stay standing? Of course not. There are always some marginal businesses; those are the ones that are likely to shrink, cut benefits, or close.

[bakhy]

it would be foolish to expect that such a measure can be passed without negatively affecting some businesses. a lot of people, including the electorate of Seattle, and the authors of the here oft cited study, obviously consider that the gains outweigh the losses.

surely you don't suggest we should allow majority of workers to be exploited because some weakly profitable restaurants might fail? (and 50 employees with 250k annual profits sounds pretty weak.) all these workers are paid the same miniscule wage, regardless of whether their restaurant makes more or less profit, so most of them are exploited, simply because they can easily be replaced.

[glenra]

> surely you don't suggest we should allow majority of workers to be exploited because some weakly profitable restaurants might fail?

Who is exploiting whom here? Most restaurants fail. Nearly all restaurants lose money in the first three years of operation. During that time, when the firm is losing money but still paying salaries, isn't the firm owner the one being exploited? And aren't you exploiting the owner even more by forcing him or her to lose even more money on the way to a slim chance of profitability?

People who consider that "the gains outweigh the losses" probably don't realize that even those who get a raise due to a minimum wage law might be on net worse off than before that law was passed. Certainly those who lose their job or fail to get a new one are worse off, but even some who keep their job are also worse off, because they lose other benefits in lieu of salary that they presumably valued more.

There is no plausible theory that says low-skill workers are made better off by making it illegal to hire them for less than a specified price. Legally fixing one term in a contract in general makes both parties to that sort of contract worse off. There are many reasons why it should be hard to measure the damage the minimum wage does so it's not surprising that a few empirical studies return the "man-bites-dog" result, but we shouldn't fool ourselves into thinking it doesn't do any harm at all just because that harm is occasionally hard to measure.

The minimum wage is callous and cruel; it harms those least able to afford being harmed. It is at heart indefensible.

Fortunately, we live in a rich enough society that being legislatively forced into a suboptimal job situation or involuntary unemployment is usually recoverable. It's not a death sentence. Yes, Seattle is effectively kicking poor people in the face with this law, but in the grand scheme of things it's probably not the worst legislative indignity we inflict on them.

[bakhy]

"when the firm is losing money but still paying salaries, isn't the firm owner the one being exploited?" - no. please consult a dictionary. low-educated workers are abundant, and are thus forced, if they want to make an honest living, to take whatever they can. anyone who has been in that situation or has been close to people in that situation should easily be able to see that.

"they lose other benefits in lieu of salary that they presumably valued more." - key word being "presumably", which is, given the election results in Seattle, apparently not true. and, there is of course no reason for all of them, or even a majority, to loose more in benefits than they gain in higher pay. makes no sense.

"we shouldn't fool ourselves into thinking it doesn't do any harm at all just because that harm is occasionally hard to measure." - ah, but we should :) measurement is king. if any losses will occur, they should be measurable. otherwise, i guess we should throw the entire history of science out the window.

"There is no plausible theory that says low-skill workers are made better off by making it illegal to hire them for less than a specified price." -- see my very simple explanation here: https://news.ycombinator.com/item?id=7840655 - generally, libertarians, a behavior typical of followers of any ideology, simply ignore all unfavorable aspects and just assume that the market automagically solves everything, without considering parameters under which a certain market operates.

"The minimum wage is callous and cruel; it harms those least able to afford being harmed." -- your concern for the poor is very moving. i personally, OTOH, would use the same words to describe the libertarian approach to economy and to the poor. but, in the end, those are just emotional soundbites.

[glenra]

> no reason for all of them, or even a majority, to loose more in benefits than they gain in higher pay. makes no sense.

I'm not sure you understand the argument here: A job is a bundle of benefits. There's a money component but there's also all sorts of tangible and intangible other factors that are of value to the worker and cost money to the employer. Other factors might include things like: Free food. Free training. Company-provided uniforms. A well-lit work environment. Flexible hours. Sick leave. Vacation time (paid or unpaid). A well-lit and clean and safe work environment. Job security. A 401k plan. Health coverage. Convenient employee parking. Educational assistance. Prescription drug discounts.

Suppose the value of your work to the company is $16/hour, of which you get paid $10 in salary, $5 in benefits, and the company keeps $1 as profit. If the minimum wage is increased to $15/hour, the only way the company can afford to keep you on is to get rid of ALL the benefits; they would then pay you $15 in salary and $0 in benefits.

The problem is that if you had WANTED that deal you could have BARGAINED for it. You CHOSE a 10/5 salary/benefit split because that's what you preferred, so when the government forces you to take a 15/0 split instead you are being forced to accept a deal you like LESS than your prior option. You're worse off.

Yes, you're making $15/hour, but now you have to work HARDER and the work conditions are worse and the timing is less flexible and the schedule is less relaxed and the job security is nonexistent and you have to pay for your own food and uniform and training and so on where before those were all provided. It is now a miserable job even at the higher wage, where before it was an okay job with a merely not-great wage.

But wait, I hear you cry: "minimum wage workers have no negotiating power so they can't just demand exactly what they want!"

Ah, but they kind of can. They do so not by arguing with employers directly but merely by CHOOSING employers. When considering whether to work at McDonald's or Burger King, workers look at the whole package and pick the one they like MORE. (The company cares about the SUM, not how much is salary versus benefits - it just wants to get the best workers they can at a low OVERALL cost) Over time, the company that offers a better overall blend of benefits+salary will find it easier to hire the people they need so the packages offered will tend to evolve to best meet the needs of the average workers.

Then the government steps in and says screw that, you can't HAVE those benefits anymore, because the company can't afford to give them to you along with your new higher salary.

If the minimum wage went from $10 to $15, people whose value to the company was between those two simply lose their jobs; people whose value to the company was $16 or more keep their job but the job is now a really crappy one they (according to revealed preference) probably like less than what they had before. So they aren't actually made better off by the change, even though they got a raise.

Which means there's no net benefit to the higher wage, only costs.

> key word being "presumably", which is, given the election results in Seattle, apparently not true

Election results are mostly about signaling and group affiliation. They don't tell us anything about what minimum wage workers want, because minimum wage workers are such a small portion of the electorate.

> measurement is king. if any losses will occur, they should be measurable.

Bad study design could easily render them unmeasurable in any particular study. And most minimum wage increases are essentially designed to be nearly unmeasurable by the sort of study you're thinking of. Yes, the theory says that, all else being equal, businesses will tend to respond to higher labor costs by such measures as: cutting benefits, cutting shifts (so there are fewer hours but not necessarily fewer jobs), investing in automation, shrinking or failing to expand, not opening (as many) new businesses. But theory doesn't say exactly WHEN this will happen or even in which business sectors. An exceptionally stupid study design might compare just the number of jobs right before the law takes effect to one year later, and find no effect. But wait: the law was argued about in the legislature for two years and then passed but didn't take effect until the next calendar year, and took effect in multiple small stages. A smart, pro-active businessman who knew he needed to make changes might see the writing on the wall and start making those business changes before the law even passes much less takes effect, confounding that study design. On the other hand, a lazy, re-active businessman might just suck it up and change nothing, losing some money for a year or two after the change, either hoping to make it work or hoping the law gets repealed. Eventually he is driven to make those changes but does so after the study window, again confounding the results. In short, what sort of delayed or premature response we should expect for a particular sector and a particular sort of legal change is an empirical question - you might have to run one study to find the shape of the expected response and then run another to estimate its magnitude. Doing the sort of thing Card/Krueger did is almost guaranteed to miss most of the signal, even if it's really there.

[bakhy]

"The problem is that if you had WANTED that deal you could have BARGAINED for it." -- it's plain as day you have no idea what you're talking about :D you guessed correctly later on - "But wait, I hear you cry: "minimum wage workers have no negotiating power so they can't just demand exactly what they want!"" -- that's right. some people have little to no choice, it's amazing that you are unaware of that...

as for the whole reasoning, it goes under the assumption that all the gain will be taken out of benefits, which is unfounded. also, why would the worker not cover benefits him/herself? with cash in hand, s/he has a much wider choice then picking McD's or BurgerKing's bundle package or nothing... ultimately your argumentation also goes to show how important workplace safety regulations are, since many business owners think like you, and history has already taught us how that can end.

as for the reasoning about studies, this is so full of fallacies and arbitrary pseudo-reasoning it's hard to even start answering. these things are hard to measure, yes. maybe you will be surprised, but people did already realize that long before you wrote that comment. but, when a couple of unemployment figures suit your point, then, OTOH, i am supposed to immediately blindly agree with you? :)

this is getting boring.

P.S. I cannot resist - do you even realize that after so much market-efficiency-based reasoning, you completely took a shit on it in order to discredit the study? the effects you expect are products of a theory based on ideal markets and market agents, yet when we fail to measure the expected effects, that's because they are anything but ideal. genius :D

[glenra]

> ultimately your argumentation also goes to show how important workplace safety regulations are

I'm not sure if you realize that across multiple industries the rate at which worker fatality stats improved either stayed the same or even declined when OSHA was introduced. In short, it's not clear at all that federal worker safety regs actually improve worker safety. You can perhaps see that best in the graph (Figure 34.1) on the third page of this PDF:

http://object.cato.org/sites/cato.org/files/serials/files/ca...

Regarding the study, I'm not saying you can't learn anything from studies. I am saying that when we have hundreds of studies showing that the minimum wage generally causes unemployment - and we do - one or two that say "but it's hard to see the effect of this one small change in this one small area with this one specific way of looking at the data" aren't enough to rebut that general presumption. If you want to "even start answering", you should read at least ONE study on the other side, not just look at data sources that agree with you. For instance, you could read this:

http://www.econ.ucsb.edu/~pjkuhn/Ec250A/Readings/Neumark_eta...

The unemployment stats I pointed to were comparing having a minimum wage AT ALL to not having one - bigger changes are more likely to be highly visible in the output data. BTW, to answer your question (somewhere else in this thread) about Germany, the last time it had very high unemployment was around 2005 when it had a high minimum wage and lots of "worker protection" laws. After the German labor market was freed up a bit (including getting rid of that minimum) the unemployment rate there dropped quite dramatically. Here's a chart of what that looked like:

http://mediaserver.fxstreet.com/Reports/95e05c46-b989-46ee-a...

(UPDATE: Wait, that's showing values in millions of workers, not as percentages, which is a little misleading. So here's a chart that shows the rate while comparing Germany to France over the same period:

http://thinkingoftheworld.files.wordpress.com/2013/06/chart-... )