Arnold Kling  


Tabarrok and Caplan at Too many smart people?...

This rant struck me, mostly because it is strikingly consistent with my 22-year-old's experience.

Dear current Management-Generation of Cubicle Land, please understand that:

1. My generation was misinformed—by elders and fortune—about the value of our college degrees. $120,000 of your/our money now buys, career-wise, just a hair more than your free high-school diploma used to. As many of my peers now lament, “A law degree is the new B.A.” We’re the best-educated generation in American history, yet the job requirements haven’t changed...

3. Are you aware of how little time it actually takes us to do things?

4. If you’d let us, we could make the computer system work right.

My daughter's first employer kept giving her projects that he thought would take days but that she completed in less than an hour, including the time it took for her to find online tutorials in Excel that explained shortcuts. Most of the time, she sat around bored. She quit after two weeks.

For many college students, the first thing they discover upon graduation is how low-paying and low-skill the job market is for them. If that information ever filters back to high school students or parents, maybe they will think twice about paying top dollar for tuition.

TRACKBACKS (3 to date)
TrackBack URL:
The author at Future Tense in a related article titled The Time Conundrum for Managers writes:
    Arnold Kling at Econolog offers a few hypotheses on why 20 somethings aren't being used to their full intelligence level (see posts here and here to follow along) His last one struck me: It takes a lot of effort to... [Tracked on July 6, 2005 7:14 AM]
COMMENTS (29 to date)
spencer writes:

It sounds to me like there are several problems here. One is that the employeer does not really understand the power of the modern computer to do basic research and busy work. the more interesting question might be how much her boss used excell for. Probably like most managers they learn how to do a few things and that it it -- they do not know how powerful the program is.

I doubt if much has changed since 40 years ago when an uncle that recruited engineers told me that all a degree in engineering meant was that the person had the intelligence and self-discipline to learn to do the job he wanted done.

But isn't the real point that the true value of an education is not how you use it in your first job, but how you use it to advance beyond your first job?

Did your daughter stop to think about why the information she was working with was valuable to her boss. The next step for her should have been to think of other ways of presenting or organizing the data that would have helped her boss do a better job. If she only though about her work as busy work rather then as a problem solving opportunity she is not showing much
of her capacity to her employeer. But isn't that the difference between the person who gets ahead and the one that doesn't.

Randy writes:

Isn't this just abhorrent to economists? I mean, talk about inefficiencies? I see this all over the place, man. I see managers that should be pumping gas, not running a company. I mean, is this really news to you? It's true, a degree in Law is the new BA. Professional life is a farce, unless you're in a guild with Master->Journeyman->Apprentice such as the college & university system in America.

Basically, as I see it, there is an overabundance of people sitting in offices doing nothing for days and weeks. This is a disgusting inefficiency. Why not provide the basics to people and let them have their leisure? The only reason it happens now is because the basics of rent and food and transportation are so expensive that you have to work nearly-full-time to stay off the streets. What freedom we have!

Randy writes:

And don't get me wrong. I believe in open markets, but what we have is clearly not anywhere close to the theoretical open-market. Last I heard the government budget (federal only?) was about 20% of our GDP. At this time, politics are so intertwined with corporate business that most of our economy seems to operate outside of the free market.

Nathan Whitehead writes:

Yes yes yes! As amazing as productivity growth in the US has been, I honestly think that most workers could get 10x as much done each day if they had the correct incentives (at least in tech fields that I'm familiar with). I fully expect to see young workers negotiating for more and more flexible hours and pay based on performance. As it is now, tech workers have to start their own companies to effectively get performance-based pay.

Randy writes:


FYI, I've been posting here for the past few months as Randy. I didn't read the instructions first - oops. So if you plan on continuing to post here, we should probably agree on a naming convention. Do you want to change your name or should I change mine - or we can just confuse people if you prefer.


Randy1 writes:

I can't imagine complaining that it doesn't take me all day to get my job done. If I had to work all day, how would I have time to write to blogs?

Lancelot Finn writes:




(on time to write blogs...) There's usually down time while the programs are running. (I'm an analyst.)

I wonder if that is the reason for more outsourcing!

Timothy writes:

I am one of those under-used new college graduates. I finished school last year and have been at my first job since November. I'm an analyst, but really only in title. I did some analytics and some accruals for Sarb-Ox at the end of the year, but as I'm in an internal department that pays an incentive once a year the bulk of my job day-to-day is data management and reporting. Which would be fine, if it kept me busy.

The unfortunate thing is that we often have to wait on IT to get things done, or my boss is out, or what she thought would take me days only took me an hour. I think managers consistenly under estimate what new college graduates are capable of, thinking back to what they were capable of when they first started. I can do regression analysis (and I'm teaching myself time-series when I have time), but even after months of prodding they've yet to get my the tools I need to do any of that. I did some reports (it's amazing how often "analysis" means "graphs") and just stuck in an Excel polytrend for the heck of it, and when I answered that I'd used the 5th degree poly because it had the highest R^2 the department head it was for seemed shocked that I even knew that. He was further impressed that I made the usual caveats about R^2 not being all that good sometimes, about R-bar-squared being better, and about not having F or t scores and I thought he was going to faint. It's not like I'm a super genius or anything, but I did manage to learn a lot in the five bloody stats and econometrics classes I took. I can even do linear algebra.

It's frustrating, but I stick with it because I know in a couple of years I'll be easy to easily swing moving to a more interesting area where I might get to do some forcasting.

ErikR writes:


If they won't give you the tools you need, you might consider just downloading them. For example:

Andrew Whitacre writes:

Ha, you guys are wonderful. (I'm who wrote the piece/letter on Fungible Convictions, by the way.)

As comments have come in and clarified the debate, I'm thinking more and more--as you guys seem to--about the economics. There are too many smart people for too few smarts-required jobs, just like there've never been enough English professorships for all the English majors who want to teach.

It's definitely true that even the smartest college grad will be green for a year at his/her first job. But business structures seem to box people in, even after they've accelerated well along the learning curve.

Generally, at least in the publishing business, I'd like to see more collaboration. That's how people my age understand things best: by see how things interrelate. How about, for example, an entry-level worker gets a mentor not only in his/her own department but one from another department as well?

Andrew Whitacre writes:

Oh and I meant to say about your post, Arnold...when I wrote the piece, I was thinking of two audiences: people my age (25) and people my bosses' ages. I never thought about people with kids in their twenties, and how those parents would shade the argument. What did you tell your daughter when she wanted to quit?

Timothy writes:

ErikR: Thanks for that, we have a CrystalBall license and they want me to use that, but I'm an admin on my own workstation, so I can always install this in the meantime.

Robert Schwartz writes:

"If that information ever filters back to high school students or parents, maybe they will think twice about paying top dollar for tuition."

Neither they nor their mothers will ever listen to you.

dsquared writes:

when I answered that I'd used the 5th degree poly because it had the highest R^2

Oh sweet jesus, a baby who's been taught how to shoot a gun but not where to point it.

I have been, on and off, in various senses, an econometrician for money for the last ten years. I have never seen a relationship which was best modelled by a fifth degree polynomial and I have never met anyone else who has.

Adding polynomial terms to a regression is something that one should only do if a) one is doing a RESET test or b) one has a decent idea why they ought to be there. Adding polynomial terms so as to give you a line that twists and turns all the way around your dataset is more or less guaranteed to give you a fragile model which fits everything and explains nothing.

Even if you're just adding the trend as a smoother (and a moving average would almost always be a better way to smooth time series data, while a lowess would be the more normal way to smooth a scatterplot), selecting one by the r-squared criterion (or any other residual variance criterion) wouldn't necessarily make sense.

I don't want to kill your enthusiasm or patronise you, but as I suggest above, knowing how to use the tools is about 5% of analysis. Knowing what to use them on is the other bit, and that's most likely what your manager knows about.

With a little bit of thought and a couple of macros you can do almost anything you might need to in Excel by the way; it's not always the most convenient way to get things done but judicious use of the LINEST command will do for all your linear models and Solver will do for most of your MLE if you know how to set up the likelihood function (and if you don't you're probably doing something you shouldn't anyway). The only failure I have come across is trying to fit very flexible functional forms to small datasets (Easingwood's variable-influence diffusion model, to be precise).

LizaG writes:

I would have to disagree Robert,
"Robert Schwartz writes:

Neither they nor their mothers will ever listen to you."

I'm a listening mother of a 21 year old college junior and an 18 year old soon to be college freshman. Not only am I listening but passing it along.

spencer writes:

good comments dsquared

Roger McKinney writes:

Interesting subject and comments on it! I see two things going on here: One is diploma inflation; the other is poor management. The education requirements or skills for most jobs have not increased a great deal in the last 50 years, but as the government poured increasing amounts of money into getting more young people into college, the value of degrees diminished. Because employers have far more applicants than they can sift through, they use a diploma as a filter. So today, jobs that used to require a high school diploma, now require a bachelors degree. The masters degree is now the equivalent of a bachelors degree in the old days, but there are so many of those that you have to have a doctorate or law degree to stand out.

On the other hand, what passes for management in the US is taking highly motivated, energetic graduates and beating that motivation and energy out of them with stupid policies and arrogant managers. That's a paraphrase from Tom Peters from a long time ago. I'm in my 50s and have many years of experience in 4casting, time-series, regression, optimization and other sophisticated analyses. My boss is my age and was a liberal arts major. Because she doesn't understand any of the analyses I can do, she refuses to use any of it. I've had similar experiences at other companies.

Timothy writes:

D-squared: Thanks, came across as a bit patronizing, but not overly so. I mostly find the macro stuff in Excel frustrating, and I'd rather use a programable stats package. I tossed the poly trend in there as "marginally more informative than just a bar graph" not for any real modeling purpose. My direct supervisor knows jack about stats and estimation, her role is more operational and making sure that the incentive program I work on keeps running on the SQL side, they hired me to do analytics (in theory) but thus far it's been more reporting and less analysis. Anyway, thanks for the tip, I'll pull up last year's data and dink around with it a bit.

My pet project is building a forcasting model for each officer in the program based on the few years of monthly historical data we have, but I'm waiting on that until I can get a programable stats package. I lobbied for WinRATS, but they have CrystalBall here, so I'll use that instead. Although I'm swiftly learning that trying to teach oneself time series is a bit difficult, anybody have suggestions other than the Walter Enders book?

Timothy writes:

As an addendum, I'll agree that understanding the "why" of the data has been a lot more challenging than anything technical they might ask me to do. I can always ring up an old prof or two, or go to somebody in another department for technical help, but the "what on earth am I trying to explain here" question is a lot harder at times. I guess that's what I get for coming to work at a bank and not knowing anything about commercial lending. Ah well, at least it's interesting stuff to learn.

spencer writes:

Timothy --your last comment show that you are trying to get it.

The best advice for young people is not to complain about being bored. Use the time to learn about the business you are working in.
Prepare to take over your bosses job. That is the difference between an employee that gets ahead and one that just complains about stagnating.

Timothy writes:

spencer: Well, I have a lot of skills (some of which are even useful), but I lack both knowledge and experience in the industry. Easier to make up for that, and to gain both, if I recognize my lack going in. I've picked up a lot since I've been here, I'll learn more, it's a process. During downtime I try to teach myself something useful, SQL, time series analysis, Visual Basic for MS Access, but there's only so much of that I can do in a given day.

Roger McKinney writes:

For time-series analysis, StatSoft has a good intro at Select the Electronic Textbook link and then Time-Series. The software is greate, too.

Aren't we overlooking a couple of things? Namely,

* The bliss of being able to goof off. I understand that there are those who genuinely want to work hard at a job (not that I understand this compulsion myself). But certainly I'm not alone in enjoying (and having looked for) a job that's ... occasionally demanding. A little dull. That uses about 10% of what I have to offer. I do my work, go home at night, and have some energy and brains left over for my own interests. Actually, about five years ago, I switched out of a job that engaged me pretty thoroughly and into a job that's just a clock-punching kind of thing, officeworkwise anyway. Should have done it years ago! I hated swapping my actual interests and talents for money. I hated having to argue with know-nothings over things I cared about. I hated having to do compromised versions of things I wanted to do. I'm a much, much happier person these days, just showing up and executing whatever the hell they want me to do.

* And the question: Who led these kids to think that they would ever find "fulfillment" in the job market? I'm in my 50s, and I'm really struck by how brainwashed the current 20somethings have been. Evidently someone somewhere gave them the impression that jobs are supposed to be fun and glamorous and satisfying -- a kind of whirl, where you magically get to put all your gifts and training to work, and to watch your salary go steadily up over time. I guess I've known a couple of people whose job history this describes, but not many.

My own contribution to economic theory: No one wants to give you money in order for you to have fun. Instead, and at best, they're grudgingly willing to trade you money to do things they don't want to (or can't) do themselves.

danl writes:

Why are all these smart people taking jobs at companies where the work is boring? Didn't they do any research before accepting their job offers?

dsquared writes:

I'm in my 50s and have many years of experience in 4casting, time-series, regression, optimization and other sophisticated analyses. My boss is my age and was a liberal arts major. Because she doesn't understand any of the analyses I can do, she refuses to use any of it

Robert, the alternative would be either for your boss to decide to learn time-series econometrics (which is extremely unlikely, and which would in any case mean she wouldn't need you), or for your boss to make decisions based on analyses she didn't understand (which would be insane). What you appear to be complaining about here is the fact that your boss is your boss.

(There is a general point here; it is always a bad idea as well as bad manners to present people with things they do not understand. If the decision-maker does not understand ARIMA, present him with a seasonally adjusted exponential smoother and count the number of times that this very simple model makes a mistake that a more sophisticated one would not have done.)

Chris Bolts writes:

This is one of the truths (or lies, depending upon your point of view) that I always tell people who suddenly think that it is pertinent to get a college education. I recently graduated with a degree in economics, but am I doing anything that is related to economics? No, I'm working at Social Security helping people get disability income. Several of my co-workers had done this job right out of high school. But that's besides the point: I wanted to work in the federal government and this is the job that would hire and pay me a decent wage to boot.

Does that mean I should not have gotten a college education? No, because getting a college degree was a long-term goal that I was committed to achieving. Also I do believe that in the long-run many college educated persons will earn more in their lifetimes than will high-school educated persons. However, in the short-term none of that really matters. I know of several high school graduates that are making three times as much as I am making.

The way I see a college education is that it serves more as a screening tool for employers who are too lazy (or too busy, once again depending upon your beliefs) to interview a large applicant pool of potential employees.

As far as allocating resources more efficiently, I think that if resources were allocated more efficiently before some students had decided to enter college then we wouldn't have this problem of too-smart people working in too few smart-jobs. Which goes back to Arnold's last comment:

[quote]If that information ever filters back to high school students or parents, maybe they will think twice about paying top dollar for tuition.[/quote]

But that is a topic for a different day.

Zetjintsu writes:

education serves more as a screening tool for employers who are too lazy (or too busy, once again depending upon your beliefs) to interview a large applicant pool of potential employees.

Actually, they're not lazy, it's that the government has forbidden them from using IQ and many other kinds of tests which are the most efficient way of sorting people for raw talent. A shame too, because a cheap two hour test is orders of magnitude more efficient than years of school and hundreds of thousands of dollars. Of course, where would all the teachers go if they weren't fleecing us out of billions of our tax dollars?

psh writes:

The first time your employer tries to get you to break the law: that's what you really need the college education for. You imagine it will be an obvious opportunity for ethical heroics; actually it will be confusing, if you notice it at all. You will need the critical thinking skills to resist the organization's manipulation and cant.

Nothing wrong with techniques, but you've got to get the
education first. The right kind of education may actually reduce employability, since it's hard to dissemble a skeptical turn of mind. You've got to decide if the freedom from control is worth the risk.

Comments for this entry have been closed
Return to top