As someone who likes to flirt with libertarian thoughts on occasion, I have pondered this one for a while. My desire for people who are less well-off to get more money is certainly there, I'll concede - but I also like the idea of getting rid of stupid regulations and letting the marketplace regulate itself (ideas about limiting CEO pay or bonuses or other stuff like that seems weird to me, for example).
I think I've found a way to think about the issue that actually makes sense to me:
Not raising - or even eliminating the minimum wage would make perfect sense in a perfectly functional free-market economy for labor. I will argue that #1) We don't have one now, and #2) we aren't going to get one.
A Free-Market Economy Example
Let's talk about what a perfectly functional free-market economy looks like, in a place I am very familiar with: the PC hardware market. You've got many corporations (HP, Acer, Toshiba, Dell, Lenovo, Fujitsu, etc.), competing against each other tooth-and-nail in order to win marketshare and earn profits. While they're fighting with each other, kicking and screaming, the consumers win - at least in terms of price. And yet the profit margins for these companies - for these product lines - are razor thin. Doesn't that seem strange? Here we have a microcosm of an "ideal" market economy, and yet we have very tiny profits.
We can explain this strange attribute pretty simply however - let's imagine we have two companies - "A" and "B". They both sell similar products. Company "A" however is taking a very healthy profit on their sales. Whereas Company "B" does not. If their costs are similar, it means that Company "A" is charging more for their product - and thus Company "B" will be cheaper, and will sell more. If "A" doesn't eventually drop their prices and reduce their profit, they will be driven out of business by Company "B". There's a real life example for this - Amazon.com. Jeff Bezos had famously said something like, "Your margin is my opportunity." Meaning, if you make too much of a profit margin, Amazon will step in and make little to none, and drive you out of business.
(Note for completeness's sake: Apple is an example of a company "A" that is not dropping prices but is doing quite well. They're a rare exception and don't have exactly the same product as the "B" companies, so the analogy doesn't work for them, but I should at least mention it, especially as an Apple Guy.)
So if we can agree that that is what a competitive marketplace ought to look like - why are Corporate profits the highest that they've ever been?
The Minimum Wage Effects on Employment
Studies have shown that moderate increases in the minimum wage do not affect employment significantly (caveat: they're from the left. The ones from the right say the opposite). Does that mean you can triple it all of a sudden? No. Moderate increases are what have been studied.
The math here is pretty simple - corporate profits are at an all-time high, and minimum-wage changes don't seem to affect employment - it seems like we might not have a fully-functioning free-market for labor. What's wrong with our assumptions?
The non-free, inelastic-demand, non-rational market of incomplete information for labor
In a completely fluid, well-functioning marketplace, a worker will always move to where the jobs are. And if they find themselves working in an industry that doesn't have a lot of jobs, and notice that another industry does, they will make sure to retrain themselves for the new industry so they can make more money. And furthermore, a worker will always have acccess to all of this information so they can make well-informed choices. Applying any level of scrutiny to these assumptions will show they are not always valid. A worker can't always pick up and move to the slightly higher-paying job across town, they've got to find an apartment, they have to drive further, the day-care center where little Bobby or Sue goes will be too far away, family lives nearby, or any number of hundreds of other reasons from trivial to the serious. A worker can't always re-train and enter into another industry that's higher-paying than the one they're in - this might entail them spending time they don't have, or spending money they can't afford. And a worker does not have perfect information on what jobs do/do not exist and where, or how to go about them. Perfect information like that doesn't even exist - and if it did, not everyone has broadband, a computer, access to the Internet, and the ability to use it all.
The current unemployment rate is 6.6%. Now, that doesn't mean the number of people who want to be employed and are not is 6.6% - that number doesn't include people who have given up and dropped out of the workforce. The best proxy we can come up with for that is the employment-population rate (Epop). That number is 58.8% right now. 58.8% of our population is currently working. The high was 63.4% in Dec. 2006 (source: Bureau of Labor Statistics). Meaning that not only is there the 6.6% of population which is unemployed and looking for work, there is another 4-ish percent who've given up looking for a job actively (presumably because they couldn't find one). So if you think that you can ask your minimum-wage employer for a raise, or you can ask not to work Saturday to spend time with your kids - what do you think is going to happen? That's right, you're fired, and replaced with someone else. Marx's idea of "the reserve army of the unemployed" is perfectly real. (I also think his solutions to this problem, and others, are the worst solutions to any problem, ever - but that's not the point of this piece).
In supply and demand terms, instead of imagining that 'labor' is the thing that is supplied-and-demanded, let's reverse it and look at 'jobs' as the thing that's in supply or demand. A high 'bid' for a job means "paying more hours for it" - accepting a lower wage, and a low 'bid' for a job means offering only a smaller number of hours for it - thus only accepting a high wage. What does this supply-and-demand curve look like? We can certainly argue that the demand for this commodity is somewhat inelastic - people need jobs and will even take jobs far beneath their skill level and training if they're desperate enough. And supply of this commodity, while it's not 'fixed', isn't highly dynamic. The end result is that people are going to 'overspend' - driving 'prices' up - in our reversed chart, this means that people will accept wages much lower than they deserve.
So with all of that groundwork laid out, let's get back to that video. I disagree with it, but I still like it and think it's well done. When you get to 3:24, I think their misconception is laid out bare - "The owner will either give Bob a raise, or a competitor will scoop him up to work in his restaurant." Does this really ever happen in fast-food restaurants? Managers of competitor restaurants in disguise sneaking in, taking notice of more productive workers and surreptitiously offering them a job with better pay? No. No it does not. Because it doesn't have to. The fast-food owner can fire Bob the moment he starts making noise about wanting better pay and snatch someone else up who's unemployed. That job won't go unfilled for long. And neither will the 'competitors' fast-food job go unfilled for long. Plenty of people want work.
And the next thing to note - their pretend math is all wrong. I'm fine with pretend math; but any restaurant owner who is running at such a tight margin that a minimum wage increase would make some of his workers unprofitable is not going to be in the restaurant business for long. Want some more realistic numbers? Sure - I'm making these up but they're at least closer to realistic: $8/hr labor, a fry cook can make around a burger every 5 minutes (12 burgers an hour) - they sell for maybe 5$ each. Your average fry cook is earning you $60/hr revenue and you're paying him $8. A bump to $10.00 is not going to break the bank. And your less-profitable cook is bringing in only $50/hr, but is still useful.
Corporations Must Maximize Profits for Shareholders
The real problem of depressed wages is at the bottom part of the worker pay-scale, but to show that the problem exists all throughout various different pay-grades - look at the anti-poaching agreements a whole bunch of silicon valley giants had. These are for engineering jobs at various tech companies, so I don't think a single salary we'd be talking about here would be less than $100,000 - probably more like $150,000 and more. These anti-poaching agreements which were deemed anticompetitive by the Department of Justice, by the way. They were entered into to prevent Silicon Valley companies from poaching each other's employees. Why? To keep salaries from escalating. And these are the same companies that constantly beg to increase the number of H-1B visas. Is that because they can't fill jobs? No, it's because they can't fill jobs as cheaply as they want.
Now there were people like Henry Ford who deliberately paid their employees more than they needed to - in order to improve retention, get higher-quality employees, and improve employee performance. But that's the rarity. I don't think we can bank on everyone being as smart as he was. (He was also a antisemitic asshole, but we're just looking at a small piece of the business side of him). Costco is another modern example of deliberately paying higher wages in exchange for heightened productivity - but it also seems to be the exception rather than the rule.
The rule will always be: A CEO must maximize the profits of a company for his or her shareholders. There can't really be any other way that I think makes any sense. You could even argue that a CEO who did not do these kinds of things is not acting in the best interest of shareholders, and should be fired. So they will pay the minimum they feel they can for as long as they can to maximize profit - and one might argue that's what they should do.
So I would argue we can assemble these bits of information together - we don't have a free market (when it comes to labor), and we probably can't get one. Corporations are reaping the highest profits they've ever had, as a fraction of labor. And they're going to pay as little as they can for wages. So what's the simplest, bare-minimum piece of legislation we can pass to help try to fix the problem? Crank up the minimum wage a little bit.
There are also some side-effects of extremely low-paying jobs - they end up costing the government money, in terms of food stamps or other assistance. It effectively means the government is subsidizing these extremely low-paying jobs - allowing corporations to pay less than they should for labor, and making up the difference out of its own pocket. Now of course any libertarian worth his/her salt will immediately say "there should be no such government assistance programs!" but they're there, and I don't think they're going away (And those programs aren't what this is about).
So the problem - people working hard and still being in poverty - has a lot of causes and reasons behind it, but the simplest workaround for our inefficient markets remains the same - just bump up the minimum wage a bit.
If the labor market was as competitive as the PC hardware market I described above - there would be no need for a minimum wage, and if there were one you certainly wouldn't need to raise it. This is the view of most libertarians that I have read. But the two pieces of evidence - highest corporate profits ever, plus minimum-wage-increase doesn't affect unemployment - really seem like a solid one-two punch. Why do they (we?) get it so wrong?
Because the libertarian viewpoint on supply-and-demand for labor is like a junior high school student's view of physics. Sure, junior high physics isn't really wrong - well, at speeds close to the speed of light I guess it is - it's just simplified. But every example in that physics class starts with "assuming a perfect sphere, and a vacuum, and no gravity..."
But real physics examples deal with nonuniform bodies, friction, air-resistance, gravitational pull, non-perfect springs - and so on. And real economics problems need to deal with "friction" of their own. They don't deal with friction, only with 'perfect markets' - which are probably more the exception than the rule.
And really at the end, what does a minimum wage mean? I would argue it's the minimum price of human dignity. A true, pure, traditional libertarian would say that if someone wanted to work for a nickel an hour, he or she should be able to. But a place where that would actually be happening doesn't sound to me like America. This is one of those rare occasions where I actually want the government to step in and say, "No, you can't do that. Pay them a reasonable amount that they can live on, or figure out another business model. What you're doing is bullshit." If your company can't operate without labor that costs less than the minimum wage, then your company shouldn't exist. Most libertarians don't believe that a person has a right to sell him or herself into slavery (only "most") - so we are already setting a lower bound on wages, in a sense. I'm just arguing that bound can, and should, go up.