No Urban Wage Premium for Non-College Educated Workers

The Economist has a nice graph and article on the urban wage premium based on David Autor’s work. The graphs shown that in the past both college and non-college educated workers earned higher wages in more densely populated areas but today only college-educated workers experience an urban wage-premium.

Housing costs eat a large share of the college wage-premium so even college educated workers are not as better off in cities as the graphs make it appear. Autor’s point, however, is that wages for the non-college educated aren’t higher in cities so they might not move to cities even with lower housing costs. That could be true but I also suspect that the urban wage premium for the non-college educated is endogenous–firms employing these workers have moved out of the city but could move back in with lower housing and land costs.


'workers have moved out of the city but could move back in with lower housing and land costs'

And to think that some real estate people joke that land isn't being made anymore, and thus property will always go up in cost.

Land may be fixed, but housing need not be if zoning and building regulations are liberalized.

How do you rezone single family to alllow 200 single family homes with backyards per acre?

I grew up when conservatives understood the land was abundant for single family by simply building more roads and more schools.

Somehow, conservatives came to believe all roads were created by god and thus building roads to vacant land is to defy god.

Cities grow, politicians and businesses want them to grow. Eventually a city increases in size and/or population sufficiently that it can no longer be controlled. A big part of this problem is the politicians and their endless quest for more and more taxes and more and more regulatory control. The politicians are able to get elected and hold on to their power by giving away free stuff in return for votes which in turn requires more and more tax revenue. Eventually the city is too expensive to live in, too over regulated to move a business to (or stay) and it begins a downward spiral. It is at this point the politicians both want to raise taxes more and force businesses and individuals to pay for it while regulations limit their ability to make a profit. This tax, spend over regulate cycle must broken if you want large cities to remain livable.

Watch Singapore where land is in short supply and see how their leaders handle it. 80% of people get public housing and the rest can bid up more luxurious housing to their heart's content. The same conservatives that idolize Singapore as a pro business, low tax haven are also scared of its socialist redistribution of land. Pragmatics will see a useful mix of government and markets.

I would never sacrifice so much individual liberty in exchange for the Singapore system.

Yep. Still waiting for the "US is a socialist hellhole" crowd to explain to me which country on earth is less of one.

Forget about the urban rural spread. Look at college grad salaries.....Flat for fifty years. The peak stays near 25/hr while us gdp per cap doubles?

"No Urban Wage Premium for Non-College Educated Workers."
Maybe Brazilian Evangelicals are taking their money.

Interestingly, these graphs suggest that any decline in non-college fortunes has occurred entirely among urban non-college people, not rural ones as the conventional wisdom suggests.

It would also be good to include charts showing the percentage of the total population that is college-educated and non-college-educated, as the non-college share of population has fallen, particularly in urban areas. If (for example) 50% of urban workers were non-college in 1970 compared to 25% today, then a comparison of all non-college workers in 1970 to all non-college workers today would not be apples-to-apples. Arguably in that situation, you should instead compare the bottom half of non-college workers in 1970 to all non-college workers today, particularly if you believe that education is primarily about signaling.

"It would also be good to include charts showing the percentage of the total population that is college-educated and non-college-educated, as the non-college share of population has fallen,"

+1, the changing demographics are going to effect the output. It's noteworthy that in the 1970 graph high end non-college educated workers made more than low end college educated workers.

By 2015, even low end college educated workers were making more than high end non-college educated.

Granted, that's based upon averages. But still it's clear that the two groups are segregating out with a distinct difference in incomes.

+1 it is the quintiles that matter, measuring by education level introduces too many problems because people are staying in school longer.

Some people are staying longer. Others are staying shorter. Girls more, boys less. Etc.

And those who stay in schoool longer are more likely to move, so you can't easily sort out where those with more started out. Eg, few go to college where they graduated from high school, thus are less likely to work and live where they graduated from high school, primary school, etc.

We’ve also added 10-20 million illegal immigrants to the potential labor supply since about 1970, few of them college educated. I would expect the vast majority of these are urban.

In just a few.

Interestingly, in rural areas, they can compose a more noticeable segment of the workforce proportionally.

People respond to incentives and the marginal tax rate for low education workers has increased. Going from welfare to work often gives a low marginal return. You lose benefits, including access to housing support etc. If the quality of subsidized housing falls you have an incentive to leave. Part of the housing cost for wealthier income groups is a safety tax i.e. you pay a premium to live in the right communities. Part of the cost is the consumption of the amenities that go with some locations, including increased safety.

The gap might be smaller if you include support programs. Low income can often get more support services and community sponsored support and or amenities. They can also achieve some safety by being part of a support group (or even a gang).

Going to the city can be like running a gauntlet, you have a chance at very high compensation and access to amenities even if many will never achieve the higher levels of income or be able to afford some amenities.

Some people from minority groups, (gay, Jewish, some Asians etc) might be able to gather in groups that are not easily supported in small communities. Again you achieve an amenity that isn't reflected in wages.

People are rational, you might not be seeing what they are purchasing or how they when they decide to live in a community. A lower sticker price on housing might not have as big an impact as you suspect. The cost of various communities and the returns to residents are not always easily quantified

What data do you have for marginal tax rates are higher for low wage workers?

Simply paying more in taxes by consuming more does not make the marginal tax rate high.

First, sales, property taxes, etc as a share of wage income go down the more you earn unless your consumption rises faster than wage income.

Second, the EITC is a negative marginal tax rate, tho that does go down rapidly for the very limited population who get it. But higher wage income in the urban areas which are more blue than red pays for more consumptionn, not less.

I suspect that in red rural and urban areas, higher wage income probably cuts consumption at multiple points as wage incomes rise, sometimes disasterously. Ie, earn $10 more a week and consumption falls $100 leading to job and income loss. Eg, cross 100% poverty line and lose 100% medication needed to function at a job, or lose child care.

Ironically, health care consumption can go up when income falls - not paying doctors to see the poor annd paying for medication leads to unpaid ER visits at much higher costs. The conservative free lunch health care plan. When ERs are not paid for EMTALA, health care costs less, even if the cost is much higher.

So, when you count unpaid EMTALA care of $25,000-50,000 as income, earning enough to get $5000-10,000 in tax refunds for Obamacare insurance means lower income due to high marginal tax rates cutting income.

Evidence: FT using CBO figures 2017

Cross from under 138% of FPL to over and you get a Federal tax cut of of over $5000 to $10,000 depending on the number of dependents. Granted tax huge tax cut goes down as you earn more, but you need to increease your income to 4x FPL before that happens.

Obama and Democrats delivered a bigger tax cut to working class people than the GOP has, plus ensure more workers got jobs that pay well, even for low education cleaning people getting jobs in medical facilities.

I know conservatives do not value healthy consumption, preferring tax cuts get spent in food and booze, gambling, oxy, etc, things that privide pleasure. Then they blame consumers for bad choices, and then call for punishing them with higher taxes, etc.

To AdamC Responding to Mulp with data is normally a waste of time but thank you for link

doubt if this large black market wage economy shows up on the graph

Has the size of the black market changed dramatically over time?

I think there are equally plausible stories that suggest that the black market has either increased in size or shrunk in size since the 1970s.

One of the big black market industries in Boston is clinical trials. My impression from anecdotes is that a lot of people with high marginal subsidy-elimination do that for income since it appearently isn’t reported to the IRS. I’m sure it has grown a lot but maybe is just replacing God knows what.

"firms employing these workers have moved out of the city but could move back in with lower housing and land costs."

What kind of business that employs non-college educated workers can afford the costs of urban land? Food packaging, oil refineries, breweries? The largest Toyota plant in the world is in the middle of nowhere.... sorry, Georgetown Kentucky pop. ~30K.

A century ago manufacturing and other industries where 100% urban. I don't know how, when or why it changed, but today the world is different.

"I don't know how, when or why it changed, but today the world is different."

I work in Factory automation. Plants don't need to be in an urban environment anymore. The power grid is pretty good almost anywhere in the US. The telecommunications are fine. Most importantly the American transportation system is good enough that you can get just in time deliveries reliably almost anywhere close to a major highway or interstate.

Furthermore, the education gap between rural workers and urban workers is much smaller than it was historically. Rural workers can read and write and they are comfortable operating machinery. Indeed, since they are coming from high automation farming areas, they probably have more experience operation heavy machinery.

It makes sense to build factories in areas with cheap land and eager workers.

If factories left the city center, why people did not followed factories?

All the de-industrialization talk is about outsourcing to China or Mexico. I've never seen any research/good journalism on factories going rural.

"If factories left the city center, why people did not followed factories?"

New factories were built in sub-urban and rural areas. For the most part, they hired locally. In some cases people did follow the factories.

When GM built the new Saturn plant in Tennessee in 1990, a substantial portion of the workers relocated from Michigan. However, that was the exception rather than the rule.

" I've never seen any research/good journalism on factories going rural."

Have you seen stories on auto-plants moving to the South East? That's a subset of this case. The Warren County TN Bridgestone plant, Nissan plants in Smyrna, TN and Canton, MS. BMW in Greer, SC. Etc.

Nice to see the hyphen back in "sub-urban." It's not suh-burban, the parts actually mean something on their own.

Sometimes they just commute in from a nearby city, or on temp agency vans.

This place is just 40 miles from Chicago

Thanks for the link. Sad to read about the town that offered tax exemptions for 20 years and today has no money to repair the roads.

Returning to the topic of today, would that IKEA 1.5 million sq feet warehouse fit in an urban area? I think no, investment follows cheap land. Warehouses and jobs are far away from high density urban centers.

Commuting more than 15-20 miles is expensive. Also assuming they already have a car, no debt, a valid driver's license, etc. For such low salaries, they practically need to live in town to profit from the job.

That's why I was wondering about people following the jobs. Tyler once asked why people did not moved to South Dakota after the fracking boom. Moving to the exurbs of Chicago seems less drastic.

isn't south dakota the 13th fastest growing state in the u.s.?

SD may have slowed with prices, but previously it absolutely went nuts with people moving in for jobs. And the prices and schools and cops and roads and fire dept were massively overwhelmed.

which also contradicts the axa fello/as premise that
people did not move to south dakota with the fracking boom

I meant pop growth slowed with gas prices decline.

Is anyone seriously suggesting that the oil boom didn't bring influx of gas gypsies and camp followers? Hell they were recruiting strippers from across the nation.

"A century ago manufacturing and other industries where 100% urban. I don't know how, when or why it changed, but today the world is different."

Only if history started circa 1990.

I grew up when lots of factories were located in rural farm, logging, ranching communities, often tied to the seasonality of those industries.

Railroads enable distributed manufacturing, providing the means to access skilled labor found in rural areas. Farming, ranching, logging, require skilled joats (jack of all trades), so "custom" manufacturing gained access to workers with self direction.

In Indiana, the small town businesses I knew were:
Farm equipment
School busses
Milk cans
Compressed gas containers

Then there were the auto industry complements which timed model change overs to seasons which had demand driven by season. New models came out in the falls while factories laid off to retool, presidents day marked the start of car selling season as consumers started thinking about summer travel and replacing winter beat up cars, that were not being replaced with new that get beat up byy winter. Etc.

The destruction of farm jobs substantially by GOP policy, pluss destruction of rail service, especially for individuals and small businesses hurt rural town economies hard.

A rural business was appliances, eg, Maytag. Maytag was big enough it bailed out the railroad they needed to keep their small rural city alive.

A business requires lots of complements, especially transport, but also job training in skills. Conservatives reject interdependencies, so eliminate one dependent business, and a long list follow, each small, but in total, they destroy the rural economy.

A couple weeks ago we had some comments about the ability of low skilled labor to command a higher wage. It was my opinion that corporate policy was stronger than any ability of workers to walk away. This data may not confirm this belief, but it is consistent with it. However many "Non-College Educated Jobs" there are, there is no shortage of "Non-College Educated Workers" to force wage growth.

And so a minimum wage might be a necessary policy to protect the weak from the powerful.

Yes let’s make a few of them better off, and price the rest out of the labor market.

All while ramping up the financial incentive to automate.

"Hmmm" is a lazy and forgetful troll, but for anyone serious:

Yes let’s ignore the multiple research papers on this, and the underlying theory.

anonymous: “labor demand curves slope upwards!”

Reality: “no. They don’t.”

Another day of "I will ignore data on the internet and provide none of my own."

What else is new.

But if you scroll down this page it is a sadly consistent pattern. And at least one person seems to be rejecting data to be named later, the whole idea that data could be used.

Yeah the probability that you have JSTOR access which would let me link research to you is the equivalent of absolute zero.

You’ve never cited data, you link twitter and CNN articles. Which you didn’t bother to read, since the text of the article disproves your point. But it’s never about evidence for ideologues.

Hilarious. An appeal to expertise used as a canard.

Keep it up, you are debasing your position nicely.

I guess if you can’t argue the facts, you can slam your fists on the table. Congrats, you’re the intellectual equivalent of a two year old.

Arguing that labor demand curves have a positive slope is an impossible position. It’s not a giffin good. You could link academic articles about monopsony in labor, but that wouldn’t apply to minimum wage workers since the majority work for small businesses or franchises. It also wouldn’t work since you read journalists’ Twitter feeds, not the American Economic Review.

But it ain’t about the facts, is it ?

If you weren't so lazy this would be harder, troll.

And again you fail to read the literature or your own links.

I’m starting to think you’re a libertarian trolling liberals.

One study which says that more experienced workers "benefit" but they had to work more hours outside of Seattle without the higher minimum wage: "a non-negligible portion of the weekly earnings increases accruing to more-experienced workers can be attributed to increased work outside the city."

What's more: "Seattle’s minimum wage ordinance appears to have delivered higher pay to experienced workers at the cost of reduced opportunity for the inexperienced." Talk about disparate impact. Minimum wages are racist.

"Minimum wages are racist."

Thomas Sowell claims they were created for exactly that reason.

"All while ramping up the financial incentive to automate."

+1, intelligent comment.

Let's just do a little math to emphasize the point. Assume that a business is in operation 120 hours per week and employees 3 workers to cover that time frame.

If the wages go up $4 per hour, that's roughly $4.32 after adding in FICA. Plus you'll need to add another 10% to cover vacations, holidays, sick days, family leave, etc. (We'll just ignore the costs of health care for this analysis) So $4.75 per hour.

$4.75 X 120 x 52 = $29,640

If the Hurdle rate is 10%, that implies an extra $296,400 to fully automate the job.

At minimum wage a machine would need to have been cheaper than:

$7.25*1.08*1.1 (ignoring health care)
$7.83 x 120 x 52 = $48,859 x 10 = $488,590

If you bring minimum wage up to $11.25 per hour: $784,990

Bear/anonymous has no use for facts, math, or logic. He’s an idealogue. He’s the left wing version of the idiot holding a snowball in the Senate talking about global warming.

Things anonymous has stated and then subsequently became enraged when confronted with facts proving said things incorrect:

1. Air quality was better in 1970 than 2019. He states he lives in LA, making this even more absurd.
2. There are fewer trees in the US in 2019 than 1970. No.
3. There is more pollution in general in the US in 2019 than 1970. No.
4. Average worker makes less in 2019. This confuses household with worker. More single parent households with dependents.
5. Federal government collected massively less revenue in FY 2018 than FY 2017 due to tax cuts. No, $14 billion more in revenue was collected. Inflation adjusted federal receipts went down ~1%. Spending increased by almost $200 billion in a rare bipartisan spending bill that appeased everyone except Rand Paul.
6.Immigrant soldiers were being kicked out and deported by Trump. No, this was a part of a program designed to get soldiers with mission critical language skills. Many could not get security clearances for obvious reasons, and were not “kicked out” they were never in-processed.
7. Demand curves slope upwards. No, labor is not a Giffin Good.
8. Democrat politicians are concerned about global warming. No, they are trying to tear down every nuclear plant in the US.

The list grows....

You have been reading, but alas only selectively.

Like a little kid peeking around the pillow at a scary movie, you only allow yourself to see ..


I think it is fair to say you are only having a discussion if you can frame your opponent's position in a way which he agrees is accurate.

And if I have you none of these is accurate, all of them as recast, that might tell me that what I believed I was saying was correct.

Haha. If that isn't a pot/kettle comment. Man, you have some real self-awareness issues.

He’s clearly about to win the Nobel Prize (clockwork, don’t even, yes it’s technically called something else) in Economics.

He has “internet data” showing labor demand curves slope upwards.

Another claim I never made, troll.

A better man would understand that claims about the pricing power of low education workers are very multi-variable models, and are going to be true or not true depending on a *lot* in practice.

But certainly strong job growth, if you can show that (in opposition to the graphs up top) would show pricing power by workers.


It’s multi-variate. You’re showing your communications major credentials. Even the fool can hide his imbecility by keeping his mouth shut. Some wisdom for you sweetheart. Maybe stop commenting about things you’re not qualified to weigh in on darling.

Go ahead, point me to where I misrepresented someone.

Look above. I only have time or interest for one, but iirc the question was what (job?) was better 50 years ago. I said something like "park ranger, more trees clean air."

People pushed back that tree count (plantations) and cleaner air (urban centers) refuted that.

It was a light silly topic anyway, and park ranger was a corner case, but WTF.

Tree plantations and urban air don't affect that narrow job category. Park ranger

Look at what Hmmm the troll does with it above, in his opening argument.

Do you see "park ranger" in item 1?

Of course not, he's a troll pushing you off what I said to what he wants you to see.

He's trolling you, sure, but you deserve it for your constant self-congratulations and inability to read or write very well.

Are you self aware enough to notice that you moved your claim (to him trolling, thank you) but that you continued to accuse me without evidence?

I have made two separate claims, not revised a previous one.

Everyone here knows that you are the troll anonymous. You can't go around ignoring peoples arguments, hand waving away facts and then expect people to ever take you seriously.

Revenge of the losers, maybe.

I think he may be trying to strawmen liberal positions by being ridiculously stupid, innumerate, and ignorant.

Liberals have some serious areas of concern, and they do justice by advocating for them. We should take them seriously. His inanity and ignorance are doing those beliefs a disservice.

Maybe we should consider him a conservative troll trying to make any rational wage subsidy position or carbon tax seem untenable by association.

"there is no shortage of "Non-College Educated Workers" to force wage growth."

Doesn't this imply the opposite?

There's no surprise here.

Low skilled workers have been hammered by:

a) high amounts of low skilled immigration
b) increasing amounts of international trade
c) increasing automation

Being in a major city doesn't prevent that from happening. It just brings all of the low skilled workers down to the same level.

Okay, I know a certain political movement has rested on all wage problems being immigration based, but is that what the data shows?

Empirical research in recent decades has produced findings that by and large remain consistent with those in The New Americans. When measured over a period of more than 10 years, the impact of immigration on the wages of natives overall is very small. However, estimates for subgroups span a comparatively wider range, indicating a revised and somewhat more detailed understanding of the wage impact of immigration since the 1990s. To the extent that negative wage effects are found, prior immigrants—who are often the closest substitutes for new immigrants—are most likely to experience them, followed by native-born high school dropouts, who share job qualifications similar to the large share of low-skilled workers among immigrants to the United States. Empirical findings about inflows of skilled immigrants, discussed shortly, suggest the possibility of positive wage effects for some subgroups of workers, as well as at the aggregate level.

So probably not the whole non-college segment, no.

I mean think about it, the fraction of non- college jobs that require fluent English is quite large.

This is why economics is absurd. The magnitude of the effect of immigration in the US is unknowable from the the data. All we can go on is common sense and theory.

Economics might be .. data intensive for things like this, but worthwhile.

Say a company adopts e-verify, what happens to it's compensation? Relative to competition not using e-verify, etc. There are lots of things to look at.

And we'd be better off with data driven policy than assumption (possibly rooted in prejudice).

Both common sense and basic theory says that if you increase the supply the demand (and wages will drop).

However, people will come up with elaborate reasons and torture the data to avoid confronting an uncomfortable fact.

" (and wages will drop)."

Technically that should be put downward pressure on wages. Generally speaking, inflation combined with sticky wages will prevent direct drops in most cases, but it's absolutely clear that low skilled wages have been stagnant for decades.

Or wages will fail to rise, or will rise less than they otherwise would have.

We have not seen levels of immigration this high in over 100 years. That was not some ideal era. Violent labor conflicts that often turned deadly. Political graft and the rise of urban political machines. High crime in urban areas. Vice ran rampant in many urban areas. Urban gangs. TThe rise or organized crime. The rise of socialist parties. Sweatshops. Cheap labor with few workers rights. Race riots. Slums that were much worse than the current version. Overcrowding. We came very close to turning socialist, etc

Huge mass immigration had an extremely negative impact on the country.

BTW isn't it amazing that the immigrants who lived through that experience and their children voted to restrict immigration and we didn't again reach those levels for over 100 years

... "hammered by"... the transition to temp-based contract workforce.

True that. Metastability, man...."the nature of which is to differentiate between two similar things." To take a Persian critique of Aristotle.

"I also suspect that the urban wage premium for the non-college educated is endogenous–firms employing these workers have moved out of the city but could move back in with lower housing and land costs."

I suspect not -- I think the old model of 'multi-story urban factory surrounded by worker housing' that died ~60 years ago is never coming back. Not in the U.S. anyway. It's not just the price of land, but the fact that what's available is in a patchwork. Not to mention the urban congestion, crime rates, leftist politicians, local minimum wage ordinances, higher property taxes (and possibly city income taxes). Against that, what are the advantages of putting a production facility in a dense urban environment? I'm coming up empty.

Actual real world investment decisions over the last 20 years appear to confirm that; few if any new auto plants, for example, in urban areas.

And I see nothing on the table that’s likely to lead to lower urban housing and land costs - quite the opposite in fact.

There are a number of cities where at least some urban land has gotten inexpensive relative to their suburbs and exurbs (Detroit, Cleveland, Buffalo, St Louis, Baltimore, etc). But that hasn't drawn industry back in, and I don't expect it to. There are just too many other disadvantages with those urban zones (which is precisely WHY the land has gotten cheap).

probably not a coincidence that the cities you listed
Detroit, Cleveland, Buffalo, St Louis, Baltimore
have pretty much the highest murder rates in the country

all cities with no illegal immigrant populations. also, census tract areas that contain just 1.5% of the country’s population saw 26% of America’s total gun homicides.

You also have to ask yourself where the best artists are in the country? Detroit and then LA? NY? Wait, why aren't the best artists in the LA and NY?

Not only would there have to be lower housing and land costs to entice these companies back in, but they'd also have to be willing to pay higher wages than they're paying now. Otherwise, those graphs wouldn't change. Not sure why they'd want to do that.

Hard to know if this means much, since not many people live where the population density is 1000 per sq mile or less.

"not many people live where the population density is 1000 per sq mile or less."

1000 per sq mile is the US census threshold for being classified as Urban in the US.

so nobody lives there anymore because it is so crowded?
you say "not many people live where the population density is 1000 per sq mile or less."
census bureau says
"Roughly 80 percent of Americans live in urban areas"

over 25 percent of women in small rural and isolated areas live more than 40 miles from the closest Intimate Partner Violence Program, compared with less than 1 percent of women living in urban areas

so that means about 75 % (the vast majority) live less than 40 miles away from the closest Intimate Partner Violence Program?

that's an odd sorta sociology statistic you came up with
sorta like the whole food desert thang (narrative)

Yes, the point is that urban areas are like a deli grocery shopping trip in the first place. NYC has no Walmart but an Ikea and Home Depot and Target each about 3 miles from each other. But try driving from Ikea to Home Depot.

so now the sociology dept has determined
nyc is a walmart desert?
if something is 3 miles away in nyc
ur options are drive, take a bus, subway ,walk, run, taxi
uber, lyft
swim ,kayak etc. there are a lotta options if your not
wearing high heels

so many options you aren't actually ever home unless you're a native New Yorker. The drive from Ikea to Home Depot is harrowing, man. But you still got to see that triumph out the window, right?

Implies the minimum wage is set too high? It's not that the right-hand end of the curve has fallen... it's that the left-hand end has risen?

The graph isn't broken down by gender. Women have increased their share of wages both through more education and government mandates. This 2013 post from Dr. Michael W. Kirst's blog at Stanford tells some of the story:

According to data from the Department of Education on college degrees by gender, the US college degree gap favoring women started back in 1978, when for the first time ever, more women than men earned Associate’s degrees. Five years later in 1982, women earned more bachelor’s degrees than men for the first time, and women have increased their share of bachelor’s degrees in every year since then. In another five years by 1987, women earned the majority of master’s degrees for the first time. Finally, within another decade, more women than men earned doctor’s degrees by 2006, and female domination of college degrees at every level was complete. For the current graduating class of 2013, the Department of Education estimates that women will earn 61.6% of all associate’s degrees this year, 56.7% of all bachelor’s degrees, 59.9% of all master’s degrees, and 51.6% of all doctor’s degrees. Overall, 140 women will graduate with a college degree at some level this year for every 100 men.

What to make of the fact that inequality is lower in rural areas but incomes are higher in urban for both groups, albeit only slightly so for non-college educated 2015? Does that mean that (1) progressive urban policies help primarily the rich relative to egalitarian conservative rural policies, (2) income inequality leads to higher incomes, or (3) inequality statistics don't tell us very much about well being?

An article such as this might lead a high school student (or his parents) to conclude that he or his kids should attend college. However, I suspect that the devil is in the details, which the actual paper might glean. I still think far too many people are going to college for degrees that they will never use.

I wonder how the numbers work out for Mike Rowe's skilled workers.

Yeah this is what I've been wondering about for some time now. Mike's been talking about skilled trades for years now but I'm hearing mixed things about people looking to get into those careers.

Now also low-skilled want to live in the cities, so the job location in a city is a wage premium, unlike before when cities were more polluted. So low-skilled city workers are still better off.

You're probably dealing with some degree of saturation effect for urban workers, where low education urban workers really are disadvantaged in "human capital" in 2015, while in 1970 much less so.

Overall graduation effects haven't changed *that* much since 1970, but you have the effect of older low ed, high talent (high "human capital") pre-Boomer workers dropping out of the pool, combined with secular saturation of education among urban workers who are even halfway skilled.

Urban schools *really* push further education hard in my experience while rural are a bit more laissez faire - different composition of the labour market.

I'd guess a high school grade rank version of the same graph probably wouldn't show as dramatic changes at the low end and more changes at the high end....

How much could this be a function of rising commute distances? That is, lower skilled workers are pushing down the marginal rate on their own jobs by minimizing their living expenses by living in lower cost areas further away from urban areas. Taken another way, lower skilled workers are innovating on living cost optimization in a way that higher skilled workers are not, or are doing less, with real estate arbitrage?

Commuting costs have increased substantially. The average daily one-way commute has increased from just under 22 minutes in 1980 to nearly 27 minutes today.

That's the equivalent of adding an entire additional work week for the typical worker. A week that we don't get paid for AND have to pay for out of our own pockets in terms of gas/vehicle maintenance/etc.

College education is probably a proxy for IQ -- dubious that college education in and of itself has much to do with wages.

How does the minimum wage figure in this? Taxes? Blue states? The ability to write a complete sentence...?

Back in the day, the cleaning lady used to work for the same business that the lawyers and engineers. Nowadays the city firms hire a business, which hires a cleaning lady, for their cleaning needs. So, even if you are doing janitorial work in the JP Morgan building you don't work for JP Morgan.

Comments for this post are closed