Current Events > Supply and demand no longer explains the job market

(+) Yeah! (+)

https://www.bloomberg.com/view/articles/2018-04-05/supply-and-demand-does-a-poor-job-of-explaining-depressed-wages

The battle over the effects of minimum wages has been one of the most protracted and bitter fights in the history of empirical economics. Some researchers, such as David Neumark of the University of California-Irvine, continue to insist that pay floors kill jobs, and a few studies find negative effects. But a series of very careful, large-scale studies is finding that the minimum wage is as benign as its advocates have suggested.

One of these is a study by economists Doruk Cengiz, Arindrajit Dube, Attila Lindner and Ben Zipperer, which looked at state-level evidence and found no negative effect of mandated pay increases on employment. They found that minimum wage hikes tend to decrease the number of jobs just below the new cutoff, but increase the number above the line -- implying that the wage hike isnt killing jobs, but simply giving people raises.

Now, Kevin Rinz and John Voorheis, a pair of researchers from the U.S. Census Bureau, have an even more comprehensive study with even more detailed evidence. Looking at data on individual earners from 1991 through 2013 -- a very long time period -- the authors take careful account of factors like mobility and transitions into and out of the labor force. They find that minimum-wage increases tend to raise incomes for people at the bottom of the distribution, and that the effect doesnt fade with time. Meanwhile, they find that the probability of people losing their income entirely -- i.e., unemployment or dropping out of the labor force -- isnt significantly affected by minimum-wage increases.

This evidence is about as good as were likely to get, and it is likely to be highly persuasive. On his blog, my Bloomberg View colleague Tyler Cowen wrote:

On the pro-minimum wage side, you should consider that those immediately affected by the wage hike do seem better off, and their higher income in the meantime may itself bring some efficiency-enhancing gains.

But the importance of this research goes beyond its implication for minimum-wage policy. Along with other research, its forcing us to rethink our basic understanding of how labor markets work.

The standard framework that economists traditionally used to understand job markets is just supply and demand, the theory taught to every introductory econ student. But since the 1990s, a steady drumbeat of empirical results has led to questions about that simple models usefulness.

First, economists found that even very large, sudden waves of low-skilled immigration didnt hurt the wages of native-born Americans, as the theory suggested should happen when theres a positive shock to labor supply. Adjusting the model to match this reality wasnt too hard. Economists just had to assume that immigration produces a positive shock to labor demand as well as supply -- immigrants buy things locally, creating jobs for the native-born.

But the minimum-wage effect posed more of a problem to the theory -- no matter how you slice it, price controls should lower employment in a competitive market. The likeliest reason that this doesnt happen is that employers have market power -- that its so costly and difficult for workers to find new jobs that they simply accept lower wages than they would demand in a well-functioning market. If employers have market power, modest minimum-wage rise will tend not to increase unemployment, because they force companies to move back toward the wage levels that would prevail if competition were working the way it should. In that model, a small increase in minimum wage could even increase the number of jobs.

New evidence is showing that employers have more market power than economists had ever suspected...

---

Tipping culture ruined it.
---

fenderbender321 posted...

So basically we need to de-regulate.

Foppe posted...

Tipping culture ruined it.

These are not, in fact, the case being made here.
---

Of course employers are holding almost all the cards. No surprise there, and the reasons for the minimal negative effects are what I have been suspecting for a while.
---

Bootstraps

dey dook our jerbs!
---

antfair posted...

New evidence is showing that employers have more market power than economists had ever suspected...

Entry level Job
Masters Degree
5 years job experience
---

Squall28 posted...

antfair posted...

New evidence is showing that employers have more market power than economists had ever suspected...

Entry level Job
Masters Degree
5 years job experience

unpaid position
---

Wanted: Senior Level Experience for Entry Level Position with An Intern's Pay Rate
---

Tyranthraxus posted...

Wanted: Senior Level Experience for Entry Level Position with An Intern's Pay Rate

No benefits. Contract position
---

Musourenka posted...

Of course employers are holding almost all the cards.

It actually is, there should be a more equal playing field between employers and employees. This is showing that there is zero impact on raising minimum wage, which means we could likely make the wage minimum 15/hr and 60k salaried and have mostly positive impact, even for employers. Of course republicans don't give a shit about improving anything so until they all leave the country that won't happen.
---

fenderbender321 posted...

Oh I know. Those fucks down at Bloomberg ain't smart enough to see the solution.

Which is um...?
---

That's still supply and demand tho. If employers didn't need those jobs filled, they wouldn't be paying people to do them. This just (further) debunks the myth that employers can't pay their employees.
---

antfair posted...

They found that minimum wage hikes tend to decrease the number of jobs just below the new cutoff, but increase the number above the line -- implying that the wage hike isnt killing jobs, but simply giving people raises.

I'm having trouble seeing how they come to this conclusion. If the data shows that the lowest paid workers lose jobs and more above-line jobs are created, that isn't a "raise", that's just eliminating one position and creating more that those eliminated aren't qualified for in the first place which is why their original salaries weren't the same. If a company fired all their call center workers and created more developer positions... that isn't a raise, nor is it any use to the call center folks.
---

lol employers would pay people $1/hr if they could get away with it. The supply of employees is saturated. It's not like they already did that in the early 1900's or anything, but pesky worker's rights got in the way.

fenderbender321 posted...

Deregulate the shit out of the entire country. Get government's hand out of EVERYTHING. Government basically controls the money supply, banking industry, healthcare, everything.

The study shows this to be entirely false. Employers control money through pay.
---

s0nicfan posted...

I'm having trouble seeing how they come to this conclusion. If the data shows that the lowest paid workers lose jobs and more above-line jobs are created, that isn't a "raise", that's just eliminating one position and creating more that those eliminated aren't qualified for in the first place which is why their original salaries weren't the same. If a company fired all their call center workers and created more developer positions... that isn't a raise, nor is it any use to the call center folks.

they looked at the individual income earners, not the jobs themselves

edit: oops, that is a different study! nevermind

though they did look at overall employment levels, and even disaggregated it by demographics, and found no substantial decrease in any
---

Rexdragon125 posted...

lol employers would pay people $1/hr if they could get away with it. The supply of employees is saturated. It's not like they already did that in the early 1900's or anything, but pesky worker's rights got in the way.

But they CAN'T get away with that, which is the point. Nobody's going to work for a dollar an hour and you'd be a fool to hire somebody at such a low wage because they're going to hate you and they're going to steal from you. If we simply abolished the minimum wage completely, the market for unskilled teenage labor would settle at around $5-$6, we'd create more unskilled jobs, and we'd end up with more useful employees down the line.
---

r4X0r posted...

But they CAN'T get away with that, which is the point. Nobody's going to work for a dollar an hour and you'd be a fool to hire somebody at such a low wage because they're going to hate you and they're going to steal from you. If we simply abolished the minimum wage completely, the market for unskilled teenage labor would settle at around $5-$6, we'd create more unskilled jobs, and we'd end up with more useful employees down the line.

none of these outcomes are substantiated, and the findings in this article cast them into serious doubt
---

ctrl-F monopsony

no results found

cmon

https://obamawhitehouse.archives.gov/sites/default/files/page/files/20161025_monopsony_labor_mrkt_cea.pdf

Balrog0 posted...

none of these outcomes are substantiated, and the findings in this article cast them into serious doubt

well yeah, they're r4X0r's knee-jerk assumptions
---

r4X0r posted...

Rexdragon125 posted...

lol employers would pay people $1/hr if they could get away with it. The supply of employees is saturated. It's not like they already did that in the early 1900's or anything, but pesky worker's rights got in the way.

But they CAN'T get away with that, which is the point. Nobody's going to work for a dollar an hour and you'd be a fool to hire somebody at such a low wage because they're going to hate you and they're going to steal from you. If we simply abolished the minimum wage completely, the market for unskilled teenage labor would settle at around $5-$6, we'd create more unskilled jobs, and we'd end up with more useful employees down the line.

yeah whats immigration, hey lets hire this guy from noweheresrepublic who normally make 3 chicken bones and a glass of milk for his day's wage and offer him a 100$ monlthly salary and fire our american workers!

they talk about monopsony, they just aren't using the word, for whatever reason

maybe because monopsony is a supply and demand phenomenon and that would kill their clickbait title

edit -- they actually link to 'oligopsony'
---

Balrog0 posted...

they talk about monopsony, they just aren't using the word, for whatever reason

maybe because monopsony is a supply and demand phenomenon and that would kill their clickbait title

seems like a legit explanation to me

Balrog0 posted...

edit -- they actually link to 'oligopsony'

damn you title editors! damn you to hell!
---

I just wish I was smart enough to actually interpret these studies, tbh, they are really fascinating to me but I do not know enough (math or theory) to really judge them
---

everyone who does works for a consulting firm supporting the big boys
---

monopsony, oligopsony, etc

sony always wins baby
---

I'd be interested to know how this breaks down by region, such as rich states vs poor and if there is much difference for large companies.
---

r4X0r posted...

Nobody's going to work for a dollar an hour and you'd be a fool to hire somebody at such a low wage because they're going to hate you and they're going to steal from you.

There are people working for free if it's billed as "for exposure" or "an internship."
---

Let's not forget university programs where you're placed in a full time nonpaid internship that you're paying 12+ credits a semester for. Some wish they were working for free/$1hr.
---

Here's a tweet storm about this article (which DOES use the word monopsony): https://twitter.com/Noahpinion/status/981891041425965057
---

So can we finally say the free market doesn't exist?
---

P4wn4g3 posted...

So can we finally say the free market doesn't exist?

Absolutely.

The problem is when so-called defenders of the free market want to remove regulations and government protections but keep the same economy (for example, defend Wal-Mart on the grounds that you don't have to shop there, etc.).
---

Anteaterking posted...

Here's a tweet storm about this article (which DOES use the word monopsony): https://twitter.com/Noahpinion/status/981891041425965057

not from that thread but

"The phenomenon of monopsony needs a catchier name ... otherwise it'll never be something that is generally known about"

"Agree, let's try to think of a cool name. Something that's fun to denounce in a Facebook group.

How about "capital dominance"?"

"The formal economics word for the difference between the wage & the Marginal Product of Labor in a monopsonized labor market is exploitation. "
---

Pretty much all classical economic models fail after empirical scrutiny.
---

COVxy posted...

Pretty much all classical economic models fail after empirical scrutiny.

supply and demand is pretty well-established in a myriad of markets, though

the basic model just doesn't work for labor markets

which doesn't falsify the model

but that's what the issue is; that specifying the relevant variables is hard, or maybe literally impossible, which would make the theories unfalsifiable in any case
---

Balrog0 posted...

supply and demand is pretty well-established in a myriad of markets, though

You have more expertise than I have on the subject, so I'll trust you on this. The issue is the latter half of your post, either it's a general phenomenon and it is falsfied, or it's unfalsifiable because the boundary conditions are left completely open.

I guess the biggest issue I have is that the evidence for these relationships, if it exists, is often way too weak and variable to justify all the precise computation in the field. Like, the quantitative side of the field gives people far more confidence in its theory then they should have, given the, at best, weak fit of empirical data.
---

COVxy posted...

You have more expertise than I have on the subject, so I'll trust you on this. The issue is the latter half of your post, either it's a general phenomenon and it is falsfied, or it's unfalsifiable because the boundary conditions are left completely open.

I guess the biggest issue I have is that the evidence for these relationships, if it exists, is often way too weak and variable to justify all the precise computation in the field. Like, the quantitative side of the field gives people far more confidence in its theory then they should have, given the, at best, weak fit of empirical data.

yeah, I agree, though I suppose it also applies to these studies which portend to find no relationship as well as social sciences more generally really. it is really a subset of the replicability crisis imo
---

That's a pretty shitty article, but I'm not convinced that anyone will actually read my post without bias if I were to type it up, so I'm just going to enjoy my day off trying out Crusader Kings II since it's free on Steam.
---

FLUFFYGERM posted...

That's a pretty shitty article

why?

FLUFFYGERM posted...

but I'm not convinced that anyone will actually read my post without bias if I were to type it up, so I'm just going to enjoy my day off trying out Crusader Kings II since it's free on Steam.

smh dog on covxy for not explaining his field and then do the same thing in another topic smhsmh
---

well, the tl;dr is that it's shitty because it uses all kinds of weasel words and basically tells you what to think, not why to think it. and it ignores the bigger picture of what really happens when there's a huge demand for a specific skill set and a huge supply of labor for skills that are not quite in demand. looking through that author's history, it's clear that he has an agenda which is why he employs these tactics.

i bet if i read the actual studies mentioned they aren't even going to make strong claims the way he does.
---

FLUFFYGERM posted...

i bet if i read the actual studies mentioned they aren't even going to make strong claims the way he does.

Why don't you just read them and decide?

This is essentially just a repeat of the time that you trashed on that simulation that tried to measure luck vs skill and didn't defend any of your points after I gave you the mathematical reasons why your complaints weren't valid.
---

Anteaterking posted...

FLUFFYGERM posted...

i bet if i read the actual studies mentioned they aren't even going to make strong claims the way he does.

Why don't you just read them and decide?

This is essentially just a repeat of the time that you trashed on that simulation that tried to measure luck vs skill and didn't defend any of your points after I gave you the mathematical reasons why your complaints weren't valid.

I don't remember what your complaints were exactly, but I do remember that no one refuted the fact that the simulation didn't say what Antifar said it said, and absolutely no one refuted the facts regarding how it's too simplistic a simulation to be worth anything. And how another assumption or two would've had the simulation proving the opposite result, IE that it's not mostly about luck.
---

Tag
---

I'll quote it:

FLUFFYGERM posted...

For example, they assume that everyone starts out with the same amount of capital and that the agents operate over a period of 40 years, enduring or enjoying "chance" events. They don't precisely define what chance events are, and they assume that a chance event has roughly the same impact on someone - either doubling or halving their capital.

They don't assume that everyone has the same amount of capital, they're trying to measure the effect of this model by fixing the amount of capital that everyone starts with. In addition, if you're doing a random walk in a space determined by a power law, you WOULD assume that you're moving exponentially (e.g. always halving or doubling).

FLUFFYGERM posted...

In fact, this type of concession is not beyond the writers. After all, in a section later on they run an exercise where they assign a value to the amount of education individuals (and the collective) have. In an effort to show that more education can increase the probability of succeeding if you have high talent. In other words, more education can mitigate the adverse effects of bad luck.

So if they're able to bake in that type of assumption into one of their runs, they could have easily ran a session where each agent acted responsibly by investing some portion of their proceeds over a 40 year period and ended up very wealthy. But they didn't, because they care about their narrative more than they care about reality.

They modeled education by changing the distribution of talent (increasing the mean in one case and the standard deviation in the other). That's a fairly easy thing to change in the model. I don't know how you would suggest changing to include "good investment". You could have a group of people who less than double their capital, but in exchange they less than half their capital on bad luck. But you'd still get the same overall effect among those people (since what's mathematically causing most of this is that your random sample of talent has a reduced effect overall compared to how many lucky events you encounter).

FLUFFYGERM posted...

What if unlucky events don't actually occur with the frequency they chose? (Six months).

Unlucky events in the model don't occur every six months, it's that every six months you move within the world and can encounter an unlucky event.

I'm not going to go into too much of the rest of it, but most of your arguments are based on making the model more complicated. The merits of the simplified model is that there were very few decisions made by the authors; ultimately, they chose the parameters of the distribution on T and the ratio of lucky events to unlucky events, both of which they explored the effects of changing. For example, your suggestion that talent should impact how often you encounter lucky events is based on the assumption that the study is trying to explore. If your model assumes "Higher talent gives you more opportunities to double your capital" (and in this case succeeding at this opportunity is only dependent on your talent), you'll of course come to the conclusion that talent has a bigger effect on your capital. Because you set it up that way!

There are different ways of building the model that would come to different results, but calling it trash because they didn't match yours is short-sighted, especially since your gripes seem to have more to do with the conclusion (luck is more impactful than skill) than the math behind it, so you're hunting for "flaws" in an attempt to reverse the conclusion.

---

Someone would have to be an idiot or clueless of economics to think it did.
---

I'm not going to pull up the topic, re-read it for context, and then respond to your post months later. Especially given what I just said about how no one refuted what I said about Antifar being wrong, how another assumption or two would've resulted in the model proving the opposite point, etc.

I mean if you're really that interested in the topic you can make another thread and I'd gladly participate. You're wrong in saying that this is in any way a repeat of that though. That was me reading the study and pointing out that Antifar's topic was wrong and that the study didn't even say what he said. This is me pointing out the flaws in the article without spending additional time to read the study itself to see if the article is even right in its summaries of the study. Big difference.
---