A progressive-ish case against the minimum wage


On the first statement, Bernie and I are in complete agreement. In fact, I’d go as far as to say that no one should be living in poverty. Aside from the word “must,” I agree with his second statement, too. On the left side of American political discourse, these two statements often seem near-synonymous, but are they? Is a higher minimum wage really the best way to reduce poverty and help low and middle income people?

If you forced me to decide whether the United States should implement a $15 national minimum wage, I’d say yes. The evidence is mixed, but on net I think it would have a positive impact on Americans’ quality of life.

But that’s not the point of this post. Instead, let me explain why the minimum wage is a stupid policy that gets way more attention than it deserves and why it should feature less prominently on the progressive economic agenda.

A thought experiment

Imagine you’d never heard of the minimum wage, or any other progressive economic policy for that matter, (mandated employer provided healthcare, UBI, progressive income tax, etc.) and were tasked with coming up with a way of increasing the wellbeing of economically disadvantaged Americans. That said, you’re well-versed in social science, from the most neoliberal-shill caricature of neoclassical economics to the most radical form of Modern Monetary Theory.

There is approximately a 0% chance that a minimum wage is the policy you’d come up with. Why? Politics and economics.

1. Politics

Direct vs. indirect policy

There are no free lunches, but some lunches look a hell of a lot cheaper than other ones. Certain welfare programs have the government provide something (food stamps, money, Medicare) and others have the government force somebody else to provide something (mandated employer-provided healthcare, minimum wage).

All else equal, policies will be more politically feasible when the government is seen as providing a good thing instead of forcing someone else to do the same. Of course, all else is never equal. Government provision might require the creation of a new agency like the Social Security Administration and will require more “direct” financing via taxation or deficit spending.

I am not a libertarian, but the genre of libertarian, strawmanny, anti-minimum wage memes like this one carries a grain of truth.

Minimum wage hurts the poor : libertarianmeme
Yes, I know that IRL the employers are often faceless megacorporations instead of mom and pop shops, and that there is a serious debate about when, if ever, a minimum wage causes unemployment.

Fundamentally, the government is using its monopoly on violence to forbid citizens from consensually agreeing to exchange labor for some amount of money. This isn’t an intrinsically terrible thing, and it certainly might be appropriate, but the burden of proof should at least lie on proponents to explain why such a prohibition is warranted.

Broad vs Targeted Costs

A corollary to the minimum wage’s “indirect” nature is that it creates a set of people (employers and workers, to a lesser extent) whom the government is explicitly preventing from doing something. Once again, not intrinsically awful and plausibly appropriate, but they’re likely to form a powerful constituency against the minimum wage.

In political terms, this is just one more barrier to implementation, and one likely to slow implementation and moderate the final policy outcome. A stylized fact of politics is that a small group of people directly, significantly impacted by a policy will form a much stronger lobby than a larger, more lightly impacted group, even if the two total effects are the same. From a purely political perspective, then, it will be easier to pass “diffuse” policies that don’t require the government to forbid specific people from doing specific things.

If the goal is to get money into the hands of people who need it, having the government forcibly prevent people from making consensual agreements, thereby creating a concentrated lobby of affluent business owners (and workers, perhaps) in staunch opposition seems like a particularly hard way to do it.

Sweeping generalizations: Where the left goes wrong

I hate false equivalency just as much as the next guy, but sometimes “both sides” really are wrong about something. I think this is partially true with respect to economics. Now, I’m going to make sweeping, oversimplified, unqualified generalizations about “the left” and “the right.”

My sense is that the right has at least rhetorically aligned itself with “economic thinking.” Indeed, the stereotypical Econ 101 supply and demand Neoclassical thinking kinda does comport well with certain conservative values: small government, the social utility of capitalism and of pursuing one’s self interest, the notion that even well-intentioned intervention tends to screw things up.

Clearly, conservatives tend to ignore some inconvenient aspects of even this school of thought. Negative externalities, according to Econ 101, must be taxed in order to reach a “social optimum” level of production. And that means a large carbon tax, to take just one example of something the conservative movement isn’t exactly enthused about.

The American left has its own complicated relationship with economics. One of the genuine contributions of “leftists,” as opposed to their more mainstream liberal and even progressive counterparts, is that there exist real conflicts in society between groups with irreconcilable interests (say, workers and business owners) and sometimes we have to pick a side.

Their contribution comes from the fact that this is partially true. The problem comes from the fact that it can make complex empirical disagreements look like straightforward value disagreements.

For example, leftists are going be awfully skeptical of an argument that the minimum wage might empirically be bad for some workers, since the leftist worldview would tend to understand the fundamental question as “should wages be higher?”that is, a conflict between workers and employers—rather than “what are the empirical effects of a minimum wage on wages, hours, benefits, and employment?”

The left needs to take more seriously the possibility that the answer to the first question is “yes, wages should be higher even at the expense of employers” but the answer to the second might be something like “well, it elevates the hourly wages of 60% of low-wage workers but modestly reduces their hours worked, and decreases employment offers for the least-educated cohort of workers.” I just made that up—it’s probably not what the evidence says, but it could be.

2. Economics

After psychoanalyzing voters and ideological movements, I’'ll move on to the simple economic absurdity of a minimum wage. This is where the word “arbitrary” really shines through. The fundamental problem with a minimum wage is that it is attempting to empower workers during contract negotiations without doing anything substantive to increase workers’ bargaining power.

Why are software engineers better paid than home health aides, even though the latter job is generally less pleasant? Because software engineers provide a lot of economic value to their employers and are relatively scarce in the labor market. The first part means that demand for their labor is high, and the second means that supply is low. As you learned in Econ 101, this implies a high “equilibrium price” for software engineer labor.

Home health aides, on the other hand, are relatively abundant in the labor market. Don’t want to help old folks take their medications and move around the house for $10 an hour? Too bad, the next guy in line will.

Now, there are two broad classes of solution here

1. Increase workers’ bargaining power.

This can take the form of strengthening collective bargaining through unionization, or to increase the economic productivity of “low-skilled” (which really just means “doesn’t have a scarce, highly-demanded skill”) workers. Then, they can credibly threaten to move to a higher-paying industry.1

2. Just give them money (or something like it).

This can take many forms: universal basic income, negative income tax, unemployment insurance (UI), refundable tax credits, free healthcare, or restrictive food stamps if you want to be an asshole about it (WIC food stamp restrictions, from Matt Yglesias’s The case against coupon government below).

Obviously, these categories are not distinct. Giving people money, UI, or Medicaid makes it much easier to walk away from the bargaining table, which in turn increases wages at both individual and systemic levels.

The Minimum Wage

So where does a minimum wage fit into this framework? Well, it definitely isn’t “giving people money,” but it’s kinda increasing worker bargaining power. From one perspective, the policy is organizing workers to say “none of us will work for less than $X an hour,” and forbidding defection. It also doesn’t actively increase worker productivity; low wage workers can’t credibly threaten to find a better-paying job because of the minimum wage.

To be fair, the MW does an ok job of organizing workers. After all, any individual worker can refuse to work for less than $15 an hour, but this doesn’t do much unless everyone else is doing the same. And, as I said up top, I’d rather we increase the minimum wage than do nothing if those were our only two options. So what’s the problem?

Minimum wage as pseudo-union

If the minimum wage is a pseudo-union, it’s the shittiest union on earth. It doesn’t listen to its members, doesn’t negotiate with management, and never changes its demands in response to circumstances or local conditions.

Sure, we can layer complexity on top by allowing states to lower the minimum down to $12 or something, but this only goes so far. The pseudo-union doesn’t have anything to say about working conditions, scheduling, due process before firing, vacation days, or anything else.

Yes, we can pass federal laws to address these things as well, but how many worker-friendly bills do you expect to get through budget reconciliation while Democrats control both houses and the presidency and don’t lose Joe Manchin’s vote? Fair enough, you say, but unions don’t always act in the best interest of all workers—are they really any better?

I don’t know enough about unionization to have a strong opinion here, but I do know something that is in the interest of all workers: money! Don’t want to strengthen worker bargaining power directly? Fine with me, let’s give everyone $10,000 a year, or couple a steep negative income tax with unemployment insurance, or whatever floats your boat.

Joe the mechanic can use his cash to work fewer weeks annually, Amy the line chef can use hers to get a Bachelor’s degree, and Tony the receptionist can use his to pay for the childcare his employer doesn’t provide.

Oh yeah, and we don’t have to worry whether giving people cash is going to accidentally harm them, perhaps by pricing them out of the labor market. Which gets me to my next point…

Yes, at some point the minimum wage becomes bad.

You’ve probably heard some politician or online commenter or talk radio host say something along the lines of “$15 an hour?! Last week, it was $12! Why not make it $20, or $50, or $100 an hour then?” In general, this critique seems to come in bad faith, but just like the libertarian meme up top, it carries a grain of truth.

Indeed, I believe that a $15/hour minimum wage would good but that a $50/hour minimum wage would be bad. Somewhere in between, the sign on the policy’s net effect flips. This isn’t an indictment of the minimum wage—it’s just unambiguously true that the minimum wage has different effects at different levels.

As one very recent, reasonable, relatively pro-minimum wage paper concludes,

Of course, there is some level of the minimum wage at which employment will decline significantly. The empirical literature on the minimum wage should re-orient itself towards investigating the determinants of that point. The observed range of minimum wages apparently does not include the turning-point, but recent initiatives suggest we may be about to observe the impact of much higher minimum wages in the near future in the US and elsewhere (as I finish writing this, the residents of Geneva have just voted for a $25 minimum wage). Together with an increased use of high-quality data, we may in this way be entering a new chapter of minimum wage research.

In other words, current, “starvation”-level minimums don’t seem to have much effect on employment, but nobody really knows how high we can go before that starts to kick in or how large employment effects will be above that point.

To be clear, this is true of virtually anything. As the Laffer curve shows, the revenue-maximizing income tax rate isn’t 100% since virtually no one is willing to work solely for Uncle Sam.

Unemployment insurance is good, but the disincentive effect isn’t entirely a bad-faith conservative rhetorical invention. If we gave unemployed folks $10,000 per month, I believe the net effect on society would be bad—that is, until we have a fully-automated, hyper-productive economy thanks to superintelligent AI.

Other safety net policies are the same way, but the difference between the minimum wage and other policies is that going ‘too far’ on the former directly harms those it is supposed to help, whereas going too far on the latter does not.

Raising UI too high won’t directly harm people who choose not to work because of it. It might decrease overall output or drive up inflation, but at the end of the day unemployed people are still getting money with which they can buy stuff. Raising the minimum wage too high, on the other hand, would decrease output and drive up inflation as well, yes, but it would do this while preventing people from finding a job.

Back to Politics

This isn’t just bad in and of itself—it is also rhetorical fodder for minimum wage opponents, whether cynical Chamber of Commerce types or those genuinely concerned about its effects on unemployment for the poor. By comparison, no one can say that Biden’s recent COVID relief checks are secretly bad for the people getting them, since they’re directly getting the money!

Ok, so why does the minimum wage get so much attention?

I don’t really know, but my best guess is a combination of intuitive appeal, state capacity, and path dependence.

1. Intuitive appeal

The notion that full-time workers should not be living in poverty and therefore that employers should not be permitted to pay meager wages is intuitively compelling. As I said up top, I agree that workers should not live in poverty, but wage mandates aren’t necessarily the best way of going about this.

The obvious, immediate cause of low wage workers’ plight is that paychecks are too small, and forcing the employer to pay better just seems like the thing to do. I could imagine a world in which this wasn’t the case and cash assistance was the intuitive, salient solution, but (in America at least) that’s not the world we live in.

2. State capacity

The minimum wage doesn’t really require the government to do anything. The weakest, most incompetent government on earth could “have” a minimum wage, even though it would struggle to enforce it.

The same isn’t true for many labor organization policies, worker training programs, or direct assistance programs, which require active, deliberate action by a competent bureaucracy. This can range from having the government mail everyone an annual check to directly hiring people to clean up national parks or something.

This is directly related to…

3. Path dependence

Early progressives really started advocating for a minimum wage around 1890, and the first state laws were passed in the early 20th century. Did the pre-WWI, very pre-New Deal U.S. have the state capacity to undertake more active welfare policies? I don’t know. Maybe we could have done UI, but wouldn’t have been able to do Medicaid. State governments probably had less capacity than the federal government, too.

Anyway, the activists weren’t stupid. If you’re going to agitate for a new policy, you better make sure that the government can physically make that policy happen. Surely, these early activists knew that ‘general law enforcement’ was something in the government’s wheelhouse of competence.

Once the minimum wage became an established part of the U.S. safety net, it probably settled in as one of the questions in American politics that everyone agrees to have an opinion on. This, to some extent, crowded out other types of pro-worker/safety net programs. In other words, raising the minimum wage has become something of a Schelling point for liberals and progressives: the default, consensus policy and rallying cry.


In his recent interview with Ezra Klein, Chuck Schumer said that the Democratic Party’s most fundamental strength is that “we believe government is there to help people, help them economically…”

I agree. Thing is, there are a lot of ways we could do this: free college, free healthcare, child tax credits, UBI, minimum wage, food stamps, negative income tax, jobs guarantee, the list goes on (quite literally forever, as there are an infinite number of permutations and parameters to adjust).

I am not arguing that the minimum wage is bad. I am arguing that it has received an undue amount of legislative attention. In 2020, 80 states and localities raised their minimum wage.

On its own terms, I think this is a good thing! But progressive activists only have so much time and money, politicians only have so much political capital, and voters only have so much attention. How many UBI, UI, or negative income tax ballot measures and bills has the minimum wage crowded out? We can’t know, but I’d guess a lot. I’m happy to see Andrew Yang’s proposal for a modest UBI in New York City, but I can imagine a world in which this kind of thing wasn’t such a novelty.

So here is my call to action for those on my side of the political spectrum: let’s take policy prioritization seriously. I don’t mean demanding more when doing so isn’t politically feasible, like calling for a national ban on firearm ownership. No, I mean making sure that we’re asking for the best policy among many given a particular level of political feasibility.

If both options are politically comparable, should we prioritize a carbon tax or negative emissions technology research? Background checks or a bump stock ban? Public housing or zoning deregulation (you can guess where I stand on that one)? A negative income tax, or a higher minimum wage?

Political capital is a scarce resource. To make the most of what we have, we need to choose priorities wisely. And, I believe, de-emphasizing the minimum wage in favor of more effective economic welfare policies is a good place to start.

Leave a comment


Even if a home health aide getting paid $10 an hour secretly knows that she won’t change jobs, having the mere option of moving to a higher paying job increases her bargaining power with her current employer.