2014 March 06 Thursday
Staples Closing 10% Of Stores With Nearly Half Of Sales Online

A sign of the times. They've also already cut average store size in half.

As retail store fronts continue to shrivel up what will the lower skilled and less bright members of society do for a living?

I am struck by my declining need to buy goods and services from the left half and even the middle of the IQ Bell Curve. It has been many months since my last visit to a bricks and mortar bank. I think I went to Target and Wal-Mart maybe once or twice in 2013. Still going to grocery stores for fresh fruits and vegetables. It has been years since I bought clothing or shoes in a store. I get all electronic devices online. Ditto bed sheets and pillows. Unlike the residents of New Jersey I pump my own gasoline. I do still shop in physical furniture stores though.

What can the less skilled do? Trash collection, but the trucks reach out and grab the cans without the driver leaving his seat. So not as many people collecting the trash. They can still work in food preparation. But automation will come to food prep in time. I'd be surprised if the fast food restaurant of 2034 needs even half the labor of the same sort of restaurant in 2014.

Not much productivity improvement has come to fast food restaurants so far. The rise in fast food restaurant productivity has been slow with a productivity increase of about 12% since 1987. But ordering and payment will get automated with touch screens and smart phone apps. Eventually automation will come to food prep as well.

The decline in wholesale and retail employment has not been as sharp as the decline in manufacturing. That page has a lot of useful graphs on employment trends per sector. Look at those graphs and see if any sector looks like it has the potential to absorb a lot of low skilled workers as employment in other sectors shrink.

Also see: College Grads Taking Low-Wage Jobs Displace Less Educated.

Recent college graduates are ending up in more low-wage and part-time positions as it’s become harder to find education-level appropriate jobs, according to a January study by the Federal Reserve Bank of New York.

My standard advice: Upgrade your skills. You've got to go up if you don't want to go down.

The economic analysis finds that Millennial college graduates ages 25 to 321 who are working full time earn more annually—about $17,500 more—than employed young adults holding only a high school diploma. The pay gap was significantly smaller in previous generations.2 College-educated Millennials also are more likely to be employed full time than their less-educated counterparts (89% vs. 82%) and significantly less likely to be unemployed (3.8% vs. 12.2%).

Part of the effect reported here is due to a larger fraction of smarter people going to college. A bright high school grad of 1970 was probably less likely to go to college than a bright high school grad of 2000. So the people who remain in the ranks of high school grads aren't as smart and ditto for high school drop-outs. Though at the same time IQ of the average college grad has dropped as more people have been encouraged to go to college. That would tend to depress the job performance of the average college grad.

Large demographic changes due to immigration further muddy the water. What we need: income by IQ level for each generation. That would let us see more clearly how much the wage premium for higher intelligence is rising.

Share |      By Randall Parker at 2014 March 06 05:05 PM 

refuguist said at March 6, 2014 7:10 PM:

A few quibbles about this as a follow up to my earlier comments on inequality.

--As you point out, the college data means nothing because the bar has been lowered every year. The article about grads displacing the less educated is coming from a blank slate framework.

--While interesting in itself, retail closures resulting from automation only matter to the income distribution if those workers can't find anything else to do. The CBO data on individual income by quintile suggests that, so far, they are doing as well as ever on the market. Their decline on a household by household basis is a reflection of poor family planning, not the result of automation.

--Consequently, there is no evidence that the bottom quintile has yet fallen into a zero marginal productivity trap on the basis of its market value. Rather, the lowest quintile has negative productivity because the government is redistributing money to them through the corrupt eds and meds sectors well in excess of their actual earnings.

--Lastly, if we have not yet arrived at the twilight of the unskilled worker, then the advice to invest in skills needs to be taken with many limitations, as you know. This is because automation can make skills worthless, destroying the expected return on investment. Since the unskilled have invested nothing, they are less vulnerable to automation as long as there remains a continuing need in some sector for unskilled labor.

There is a persuasive argument that the value of unskilled workers on the labor market will decline in the near future. I am not yet convinced that the future is now. That said, it is true that labor participation rates have declined. Possibly that is a sign that the twilight of the unskilled is indeed here. But it could mean a lot of things--hangover from the great recession, growing (covert) welfare benefits, more people in school.

What data would we need to thoroughly prove that the zero marginal productivity worker trap is nigh?

Randall Parker said at March 6, 2014 8:28 PM:


I was thinking about your links when I went looking for the materials in this post. I found a post by economist Mark J. Perry on income per quintile per earner and per household. I am struck by how in the bottom quintile 61% of the household have no earner! Wow. They have zero productivity. This ties in with another article that claims the average welfare recipient gets $40+k per year.

So there is an issue with bottom quintile incomes: how much of it was earned? How many people have already fallen into the ZMP state?

There are lots of different kinds of skills. If, say, you go get a masters degree in statistics robots aren't going to put you out of a job. Rather, greater computer power is making the statisticians more valuable.

The declining value of unskilled workers is not a future event. Factory wages have gone way down and most factory workers have fairly low levels of skills.

refuguist said at March 6, 2014 9:08 PM:

Mr. Parker,

I think I am in agreement with most everything you wrote in your comment. I am a little unclear on what it all means. The decline of factory wages, retail wages, etc., is irrelevant if those workers can maintain their income elsewhere. The CBO data suggests that they can and have, at least so far.

However, if I am reading you correctly, you are pointing out that whatever the income of the bottom quintile, the bottom decile is already unproductive. This could be a sign of a new era of ZMP workers. But if the bottom decile is falling and the bottom quintile is still not diverging, this means that the upper half of the bottom quintile, also presumably unskilled workers, is actually doing better than it was in the past! This suggests to me that the problem, if there is one, is entitlements and not automation. Still, there are so many causal inputs working here that I can't see how to get at a definitive answer.

I will have to give this some more thought.

Kuduz Bob said at March 6, 2014 10:37 PM:

If, say, you go get a masters degree in statistics robots aren't going to put you out of a job.

But who says that that statistician needs to be in America? Aren't radiology jobs being outsourced to India?

Barnabas said at March 7, 2014 9:47 AM:

Much easier to oursource a statistician than a radiologist since malpractice suits aren't a concern.

Wolf-Dog said at March 8, 2014 1:06 AM:

If the average IQ of the average citizen increases, this will actually make the job market even more competitive for skilled people. As automation software and hardware becomes more advanced, a small elite will be able to run the entire economy, as even software will be written by higher level software. So regardless of how smart you are, you should pray that the competition is not smarter than you because you would be without a job otherwise.

1) If your IQ is higher than the average IQ of your family, this means that in some sense you "stole" some high quality DNA from your siblings or cousins. This means that you owe something to your family if your IQ is higher than the average IQ of your relatives.

2) This economic system cannot survive in the future, unless the government builds robot factories and services instead of directly taxing the rich and using that money to make the poor buy from the rich, so that goods and services are direclty given by the government to those who are unemployable. But this would also mean that the government will have to find a way of obtaining fuel and raw materials to feed the robots that will support the impoverished majority, which means international tension to get these raw materials.

Namenotanon said at March 8, 2014 9:26 AM:

That 17.5 is more than enough to pay off student loans that everyone is whining about, Muffy and Tod would just rather live a fat lifestyle than honor their debts.

Randall Parker said at March 8, 2014 6:54 PM:


Labor force participation rates are declining at the bottom. Go to that Bloomberg article above and read the section titled "Participation Rate". That drop in labor market participation for high school grads is far too rapid to be caused by more of the smarter high school grads going to college.

Kuduz Bob,

I know very well paid statisticians in America. Heck, I know very well paid people with a number of kinds of advanced degrees in very mathematical disciplines.


My guess is the upper classes will spend enough money to keep the economy going even if most of the lower classes are unemployed.


17.5? What are you talking about?

Kudzu Bob said at March 8, 2014 8:17 PM:

I know very well paid statisticians in America.

So do I, as it happens. But what is it about their jobs that makes them safe from outsourcing to India or China, where there are lots of people who are good at math and work cheap?

shiva1008 said at March 9, 2014 11:37 AM:

Kudzu Bob,

The quality of work done by Indian companies is just not up to American standards. They fill a niche in the market where people want something done cheap, but not up to the highest standards. A lot of times a higher standard is required, or it's just easier and cheaper in the long run to get it right the first time, ie with high quality workers. That type of work which doesn't require high quality and high levels of cognition will be automated sooner than the primo stuff. A top-50 University researcher will farm out their statistics work to a known colleague or someone in their network, but farming it out to India? just lol There is definitely a lot of work in statistics because it's the part of research not a lot of people want to do or know how to do at a high level. Of course if peak oil or a debt tidal wave hits and the budgets for research and technological advancement are slashed, then all bets are off.

Tom said at March 9, 2014 3:40 PM:

"If, say, you go get a masters degree in statistics robots aren't going to put you out of a job.

But who says that that statistician needs to be in America? Aren't radiology jobs being outsourced to India?"

Not just that, but to get a master's degree in statistics or some similar field, you usually need a bachelor's degree in statistics or something similar. So in order to get the master's, you have to commit much earlier by getting the bachelor's. So there's greater risk.

AM said at March 10, 2014 5:13 PM:

Randall, you frequently recommend programming as a career, but according to programmers themselves, it's not a very good or stable career:

"Software Developers Are Terrified Of What Happens When They Hit 30"


The saying “30 is the new 20” doesn’t always hold true — in fact, it seems to be the opposite for those in the software development industry. While some professionals are just beginning to hit their stride at age 30, the future for an average programmer can appear pretty bleak after 29.

The problem sparked a lengthy conversation in a recent Hacker News thread, where developers voiced their concerns about what happens once they hit age 30.

One even suggested that the solution to being "too old" at 30 was simple: developers should retire at 40.

Developers often skew young because older staff can grow tired of relearning their skills each time a new platform comes out. Second, and more importantly, companies frequently hire younger, inexperienced programmers to perform the same work for a cheaper salary.

This leaves veteran engineers with the option of moving up to a managerial role, which may involve overseeing other developers rather than actually programming firsthand. That work can be less creatively fulfilling than building an app with your bear hands.

“If all you do is ‘write code’ then you have to be prepared to ‘write the same code’ in a new paradigm several times,” one commenter, ChuckMcM, wrote.

“Relearning most of your job skills every few years starts to get annoying the [twentieth] time you’ve had to do it,” another commenter known as "bane" wrote.

The issue isn’t necessarily new to the software development community. Vivek Wadwha, an entrepreneur that has given lectures at both Stanford and Duke University on entrepreneurialism and public practices, wrote back in 2010 that tech companies prefer to hire younger programmers with less experience. Citing statistics from the Bureau of Labor, professors Greg Linden and Clair Brown wrote in their 2009 book titled “Chips and Change” that salary increases for those working in the semiconductor industry slowed at age 40.

But the problem extends beyond the hiring preferences of today’s tech companies. Some developers feel stifled by the time they hit age 30, as game designer Michael O. Church wrote in a blog post from 2012.

“By 30, most of us have decided that we want to do something else: management, quantitative finance, or startup entrepreneurship,” he wrote.

It’s unclear if this trend is expected to change, but aspiring developers are advised to prepare for the future.

One ancient — ie 35 year old — developer advised that the best thing to do is plan to retire by the time you're 40...

Randall Parker said at March 12, 2014 10:31 PM:

AM, There is an important clue in this sentence fragment:

Relearning most of your job skills every few years

If your job skills amount to understanding some APIs then you do have to relearn when you switch to new APIs. If your job skills amount to knowing lots of powerful algorithms and techniques that sort of knowledge has value for decades.

You can make things worse for yourself by choosing niche languages or dying languages and APIs. You can make things worse by working in an area with few companies that hire software developers. You can make things worse by staying at a company that is gradually dying.

Software developers are not a big homogeneous mass. They are on a wide variety of career trajectories.

AM said at March 13, 2014 11:13 PM:


But the article notes that an even more important factor is companies getting rid of people over 30 and seeking younger workers:

Developers often skew young because older staff can grow tired of relearning their skills each time a new platform comes out. Second, and more importantly, companies frequently hire younger, inexperienced programmers to perform the same work for a cheaper salary.

Yes, a minority can do well, like in anything, but in general it doesn't seem to be a very stable or good lifetime career for the majority.

Randall Parker said at March 15, 2014 9:49 AM:


Someone can find a forum where some people will complain. Not hard. But how representative are they? The people who are doing well are less likely to show up to complain.

We need longitudinal studies of software developers to know what happens to them in their 30s, 40s, 50s. I know a lot of software developers who have continued in their jobs for decades. I've seen companies hire contractors who are in their 40s and 50s because they could not find anyone else. But I can't say I have a large enough pool of people to judge.

What I've seen: The weaker ones drop out or move to management or completely other types of jobs. Some of these people should have been fired sooner.

refuguist said at March 22, 2014 12:52 AM:

As an update, Gregory Cochran recently wrote a post about the income inequality issue and I asked him to comment on the individual/household income question. He made a helpful response (near the beginning of the comment thread) which is worth reading, and contradicts some of what I was writing earlier.


AM said at March 24, 2014 12:48 PM:

"The Brutal Ageism of Tech Years of experience, plenty of talent, completely obsolete


Silicon Valley has become one of the most ageist places in America. Tech luminaries who otherwise pride themselves on their dedication to meritocracy don’t think twice about deriding the not-actually-old. “Young people are just smarter,” Facebook CEO Mark Zuckerberg told an audience at Stanford back in 2007. As I write, the website of ServiceNow, a large Santa Clara–based I.T. services company, features the following advisory in large letters atop its “careers” page: “We Want People Who Have Their Best Work Ahead of Them, Not Behind Them.”

And that’s just what gets said in public. An engineer in his forties recently told me about meeting a tech CEO who was trying to acquire his company. “You must be the token graybeard,” said the CEO, who was in his late twenties or early thirties. “I looked at him and said, ‘No, I’m the token grown-up.’ ”

In talking to dozens of people around Silicon Valley over the past eight months—engineers, entrepreneurs, moneymen, uncomfortably inquisitive cosmetic surgeons—I got the distinct sense that it’s better to be perceived as naïve and immature than to have voted in the 1980s.1 And so it has fallen to Matarasso to make older workers look like they still belong at the office. “It’s really morphed into, ‘Hey, I’m forty years old and I have to get in front of a board of fresh-faced kids. I can’t look like I have a wife and two-point-five kids and a mortgage,’ ” he told me.

painlord2k said at March 27, 2014 3:26 PM:

The big problem of joblessness is due mainly to regulations and taxes.
A lot of low skilled workers could be profitable if regulations and taxes would be lower and more sane.
Not all of them, but a lot of them.

Without the dole a lot of people would find a job, any job.
With time, they would become skilled in something.
Employers would have an incentive to train and teach them, if the cost to hire them would be lower and they could be paid no more than they are able to produce in profits.

AM said at March 29, 2014 9:49 AM:

"A 55-Year-Old Developer Tells Us What It's Like To Face Homelessness In A Youth-Obsessed Silicon Valley"

[Parapundit says: when you excerpt too much from an article I will just delete the entire excerpt. Also, learn how to make a link into an a href link as I did]

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Web parapundit.com
Go Read More Posts On ParaPundit
Site Traffic Info
The contents of this site are copyright ©