Missing Minimum Monopsony Moods

Monopsony in labor markets is seems to be becoming a hotter topic year after year.  I can’t help but cynically think that some of its popularity is due to the fact that it can be used as a rationale for policies that are widely popular with the public, but less so with economists, like minimum wages, worker safety legislation, unemployment insurance, and legal protection for unions.  However, the reasons why monopsonies are a problem for economists don’t seem to be the reason why more politicized commentators oppose them.

A quick review on the economics: a monopsony is a lesser-known cousin of the monopoly, so lesser known that any economist discussing monopsony with noneconomists will hiss the “s” in monopsssssssssssssssony to emphasize the difference.  The difference here is that, in a monopsony market, there is only one buyer of the good in question, which, in this case, is labor.  The existence of a single employer means that the number of people who that specific employer can hire increases with the wage that he/she offers, which looks like this graphically:


Since the single employer can’t price discriminate (pay workers in the same position different wages), the employer maximizes their profits by paying workers Pm, a wage below their marginal productivity Pc.  This reduction in wages transfers some of the worker’s producer surplus to the employer, pictured in green.  However, as a nasty side effect, there are some workers who would be willing to work for the employer at a wage lower than their productivity, but who are unwilling to work for the employer at the lower wage Pm.  This creates a deadweight loss, pictured in black.  This subset of employees are worse off because they don’t get jobs, and the employer is worse off because he/she cannot profit from these workers.  A minimum wage, therefore, could increase employment in this monopsonistic labor market by forcing the employer to pay wages high enough to get these marginal workers out of bed, creating net societal gains.  Sure, the minimum wage would also give the green surplus back to workers who were already willing to work at a lower wage, but, from a economic standpoint, it’s the increased employment of those marginal workers that really matter.

The puzzle then, is why those who advocate for a minimum wage on monopsony grounds seem to focus almost entirely on the transfer effects of monopsony rather than the deadweight loss.  The pattern is for minimum wage advocates to write describing the monopsony model, with optional jabs at economists who the “ECON 101” method of modeling the labor market, and conclude that a minimum wage would be efficiency enhancing, while simultaineously waxing about how workers need better pay now, completely ignoring the efficiency-enhancing properties of the minimum wage that they just spent so much time proving.  Almost no where will you read an advocate for minimum wages write something like:

Around the nation, there are thousands of potentially hard-working employees who could be a real asset to our nation’s employers, but are unfortunately unwilling to go out and work for $7.25.  Employers would love to hire them at higher wages, but market structure prevents them from doing so without hurting their business.  A higher minimum wage is the push these workers need play a productive role in America’s labor market.

If you take the monopsony model seriously, this is the primary and intended benefit of minimum wages, while any change in the income distribution is just a side effect.


Guest Blogger: Bernie Sanders on Austrian Business Cycle Theory


I really appreciate Tucker giving me the opportunity to give my thoughts on a topic that I recently discovered: Austrian business cycle theory.

When it comes to macroeconomics and business cycles, I find mainstream discussion lacking.  I, of course, largely concur with my friends on the political left when it comes to what the government should do about recessions.  As the United States experienced following the great depression, the only way to fix a recession is to invest in working people by asking the rich, who prevent recovery by hoarding their income, to pay a little more.

However, when it comes to what causes recessions, I find their hand-waving about “animal spirits” and “aggregate demand”.  I’ve long pined for a theory of the business cycle that knows where the real blame for recessions lies: coperate greed, and these theories of aggregate demand shortfall seem to imply that the problem is that firms aren’t being greedy enough.  Thanks to some very generous constituants, I’ve finally found a theory that confirms what I’ve long known to be true: the Austrian theory.

The pitch goes like this: it all starts when the federal reserve lowers interest rates, hoping to lower unemployment.  However, greedy investors see an opportunity for easy profit where there really isn’t one: snatch up this easy money and start investing in lavish projects.  If big business was really as future-minded as Republicans would have you believe, they would know that the real resources for these projects don’t really exist.  As you and I know though, the gambler mentality of the billionaire class wins out, and the recklessness begins.  After all, they have savings they can fall back on if their projects go belly-up, while their workers aren’t so lucky.

Some proponents of ABCT like to place the blame on the central bank, but if anything they are an accessory to the greed and recklessness of the oligarchs who put a small chance of short-run profits over the well-being of their employees, their company, and the entire economy as a whole.  Higher taxes on the wealthy and more oversight of investment could help prevent these downturns from happening by limiting the potential returns from what Austrians call “malinvestment”.

Now, I certainly don’t know as much about ABCT as most of you; I even think that Tucker might understand it better than me.  However, the fact that so many people can hear the ABCT story and not see it as a cautionary tale against corporate greed strikes me as a missing mood.

Is Sleep a Preference or a Constraint?

Despite being more than a decade old, Bryan Caplan’s The Economic of Szaz is still sparking debate.  The gist of the article:

Building on psychiatrist Thomas Szasz’s philosophy of mind, this article argues that most mental illnesses are best modeled as extreme preferences, not constraining diseases. This perspective sheds light not only on relatively easy cases like personality disorders, but also on the more extreme cases of delusions and hallucinations.

The economic nature of mental illnesses, addictions, and compulsions is a topic worth exploring.  The problem is that mental illnesses aren’t experienced by everyone, and those who do experience it might experience it in radically different ways.  It’s very tempting for someone struggling with one of these conditions to say that Caplan and Szasz just don’t understand what it’s like to be depressed or addicted.  Other analogues have been floated that might straddle the preference/constraint line, like sexuality or cilantro aversion, but I think there’s another option that both more closely resembles the debate over mental illness and is more universal: sleep.

Is the amount of sleep you get the result of preferences or constraints?  I’m not entirely sure, just like I’m not entirely sure about mental illness.

In favor of sleep as a preference:

  • Sleep is responsive to changing incentives.  For example, people sleep more on weekends when the opportunity cost of sleep is lower, or sleep less before a big test when the opportunity cost of sleep is higher.

In favor of sleep as a constraint:

  • Some decisions about sleep aren’t responsive to incentives.  An insomniac probably wouldn’t fall asleep even with a gun to her head.  Similarly, someone under a sedative would not be able to wake up regardless of the benefits of doing so.
  • People consume caffeine or set alarm clocks to make themselves want to sleep less, or diphenhydramine to make themselves want to sleep more.  This makes sleep seem more like an exogenous constraint than an endogenous preference.

The good thing about this question is that everyone on earth sleeps, so what do you think?  Do you see the amount of sleep you get as a preference or a constraint?

Betting Markets as Insurance Against Psychic Losses

The existence of insurance, which dates back to at least 1000 BC, is one of the great arguments for the enriching power of markets.  Insurance allows people to transfer wealth from their hypothetical future self who is not in a car accident to their alternative hypothetical future self who is in a car accident, which is mind boggling to think about.  Still, as I think about the topic more, the puzzle isn’t why insurance exists in the first place, but why there isn’t insurance for more types of losses.

Take this 2017 The Atlantic article discussing some people’s reactions to Donald Trump’s presidential victory:

…as state after state flipped to red, her friends left quietly, one by one. Caffrey yelled at her husband for being too confident Hillary Clinton would win. She blamed herself for not volunteering more. Then she cried herself to sleep, “thinking about all the people who would die and suffer and become fearful and hated and hateful unnecessarily under a Trump presidency…”

…Tony Doran, a married gay man from New Jersey, was on a tropical vacation when he heard the news. “I left the luau and went back to our room and cried,” he said. “I was physically sick with terror…”

…It’s easy to mock reactions like these, but for liberals and conservatives alike, losing at the polls can produce an all-encompassing sense of despair. Conservatives experienced something similar after the election of Obama, whose socially liberal platform was anathema, for example, to many religious people.

These extreme psychic loss events aren’t just limited to politics.  One paper shows a drop in France’s suicide rate following a World Cup win, an effect that would presumably work in reverse too.  Another finds higher risk of stress-related deaths in cities whose team just lost a Super Bowl, and vice versa.

If the stakes of events like elections or sporting competitions are so high for bystanders, why not insure against these psychic losses in the event that your team or candidate loses?  We don’t even need to imagine what such a market might look like, as betting markets already exist.

The solution is simple- if you know you will be devastated if Trump wins in 2020, place bets on Trump to win until the payout you receive upon his victory is enough to offset your psychic loss.  The less likely that Trump is to win, the less you have to bet to get a given payout.

The question for another day: why aren’t people already doing this in droves?


When a great thinker like Scott Alexander designates the strongest argument for a policy you oppose, you should certainly pay attention:

This strikes me as the strongest argument for the minimum wage and other job-killing labor regulations: that they are turning otherwise-miserably-employed people into unemployed welfare recipients.

It’s rare, and refreshing, to read arguments for the minimum wage that acknowledge even the potential for disemployment effects, let alone acknowledge them as a feature rather than a bug.  The problem with Alexander’s analysis is that, ceterus parabus, we would expect the minimum wage to disemploy workers in less miserable jobs.

Imagine an economy with two jobs for low skilled workers: elevator attendant and fry cook.  Let’s assume for simplicity that demand for fry cooks and elevator attendants is roughly equal.  However, since fry cooks have to slave over a hot grill rather than in an air-conditioned elevator, the supply of fry cooks is much lower than the supply of elevator attendants.  If employers are to induce a worker to become a fry cook, they’ll have to pay her a higher wage, or a compensating differential.  Graphically, both markets clear, with some low skilled workers choosing to have a fun job at lower wages and other choosing to have a more difficult job at higher wages:


Now, let’s say that the government institutes a minimum wage somewhere between the wages of both jobs, which creates a surplus of workers in the market for elevator attendants but not in the market for fry cooks:


Using Alexander’s terminology, the only jobs that weren’t destroyed were the miserable ones!  It’s then possible that some unemployed elevator attendants, rather than waiting for a fun job to free up, will decide to enter the fry cook market…


…which lowers the premium for being a fry cook in the first place!

This is obviously a very simple world we’ve built, but, even in this model, it’s possible for a wage subsidy or even just a cash transfer to boost low skilled workers’ earnings without creating involuntary unemployment or forcing workers to consume a different bundle of wages and job enjoyability.

Moral Blameworthiness and Elasticity

Bryan Caplan’s next big project is entitled Poverty: Who to Blame, a cornerstone topic of which will be “moral blameworthiness“.  Caplan discusses the concept in short in a recent exchange with Robin Hanson:

Unlike Robin, I should add, I’m a big believer in moral blameworthiness.  Whether we’re discussing poverty or involuntary celibacy, I think we should always start by investigating whether the sufferer is culpable for his own woes.  And empirically, I think the sufferer usually is highly culpable…At the same time, though, I freely admit that a sizable minority of people suffer blamelessly. A severe congenital handicap could easily lead to both severe poverty and isolation despite exemplary behavior. Should government do anything about this? I don’t know…

This analysis, and the forthcoming book, relies on deontological moral philosophy that is pretty alien to economic analysis, which almost presoposes consequentialism.  While Caplan would reply that even the most deterministic and consequentialist person would still likely be angry at a close friend for committing a morally blameworthy action, I think that moral blameworthiness actually squares quite nicely with a topic covered in every ECON-101 class: elasticity.

To review: elasticity can be thought of as a measure of a person’s responsiveness to incentives for a certain action.  An activity with low elasticity- like buying food- is not very responsive to changes in incentives.  If the price of food were to double, for example, I wouldn’t reduce my consumption of food by nearly that much, so we say that food is an “inelastic” good.  On the flip side, activities with high elasticity- such as buying tickets to a concert- are very responsive to changing incentives.  If the price concert tickets doubled, I would reduce my consumption of concerts by more than 50%, so we call them “elastic” goods.

What’s the connection to moral blameworthiness?  It seems to me that actions considered morally blameworthy are also actions with very high elasticities, and vice versa.   We can see this apply to some of Caplan’s examples of morally blameworthy and non blameworthy behavior:

If your girlfriend misses your birthday, “My car and phone both broke down” is a better reason than “I forgot.”

Maintenance of your car and phone is much less elastic than maintenance of your memory.  The promise of a hundred dollar payout would likely be enough for you to remember even an incredibly arbitrary fact for a year, but an equivalent promise for a year without car or phone trouble probably wouldn’t change your behavior at all.

If a co-worker goes home early and asks you to cover for him, “I have the flu” is a better reason than “I want to play Skyrim.”

Again, if I were to pay a group of volunteers a large sum to go a month without videogames, most, if not all, would likely follow through.  A similar group paid to go a month without getting sick would probably see only a small reduction in the rate of flu cases compared to the general population.

In a future post we’ll cover why the similarity between moral blameworthiness and elasticity strengthens the case for means-testing- a case typically made using moral philosophy rather than economics.

“Choice Agony”

Previously, I discussed behavioral challenges to the cornerstone economic assumption that adding more choices can’t make people worse off, and posed the idea that the real driver behind the “paradox of choice” was the mathematical tendency for more choices to reduce consumer surplus.  Today, I’d like to examine a related, but I think more empirically relevant concept that, surprisingly, I first heard about in this talk on role-playing game design of all things.

The basic idea behind what is termed “choice agony” is that, in certain situations, the calculus of choice-making can itself induce disutility.

Sophie’s Choice might be the best example of this phenomenon, where (spoilers for a 39 year old book), the titular character’s trauma stems from her experience in Auschwitz, where a guard forced her to choose one of her children to die in order to save the other.  If the title isn’t evidence enough, it’s clear that Sophie’s misery isn’t primarily, or even mostly, about the loss of her child, it’s about the horror of the choice-making process itself.

Unlike the paradox of choice, however, surplus or opportunity costs don’t seem to be a driver here.  Indeed, we would expect Sophie to have higher surplus after making her choice as opposed to if the guard were to pick one of her children randomly.  After all, her choice, essentially, saved one of her children, while losing a child at random has no such negative opportunity cost (opportunity benefit?).  But readers of the book are supposed to be more horrified at the guard’s behavior for inducing choice agony.

Healthcare might be one area where choice agony is relevant, as described in this article, or this quote from the great Robin Hanson:

Here’s one possibility: Maybe we are just terrified of thinking about death. Would imply that we just don’t want to think about not trusting our doctor. Want to push decision off on someone else. By buying health insurance, the vision is that someone else is responsible. This explains some of the puzzles, but not all. Explains apparent excess medical amount. If you were terrified of not getting enough food, you’d go only to all-you-can-eat restaurants. Wasteful-appearing care is not wasteful because it consoles you.

Hanson actually rejects this explanation in favor of a signalling-based approach,

If I were to attempt to steel man proposals for universal health care, I would probably begin from these foundations.