True, false, or uncertain?

Let’s start with a measurement, namely that the current rate of unemployment for individuals with a college degree is about 3.7 percent.

Therefore if we cut government spending on the jobs of those individuals, they will be reemployed reasonably rapidly.  We should not assign much weight to the aggregate rate of unemployment in making this judgment.  True, false, or uncertain?

Variant: If we increase government spending to hire these individuals, it will not much lower the rate of unemployment.  True, false, or uncertain?

Additional exercise: What percentage of the money spent on the labor of government military contractors is spent on individuals with a college degree?

I find it remarkable how infrequently these simple considerations are mentioned, much less analyzed.


"I find it remarkable how infrequently these simple considerations are mentioned, much less analyzed." Model that! as they say.

What is surprising about any of this if we apply the Public Choice model of man to the politicians, economists and journalists who fill the world with their wisdom on these matters.

I find it more remarkable that Tyler you would write this in the semi-passive voice, and not either conclude (or not conclude, and that is something I would like to see) that the party that waves the patriotism flag and keeps military spending higher (much higher) than it should be, is the Republican Party (Dwight Eisenhower notwithstanding?) Why don't we ask Dan Klein to replicate his work on political party affiliations of college professors, with political party affiliations of Lockheed Martin management? Of course, we could ask the counter-factual, and suppose you (as a person with credibility) could move the Republican Party to be more willing to take a peace dividend and redirect spending to low end of education spectrum (or even Heckmanian age range), perhaps the Democrats would sense a patriotism opening and start... I don't know... shooting clay pigeons?

What would really help "the low end of the education spectrum" would be not importing so many uneducated foreigners.

I find it more remarkable that neither of you consider the much, much higher numbers of degreed government employees in other sectors, whose services are arguably rather less valueable, but poke at Pentagon instead. Of course, Pentagon is an easy target to poke at, SWPLs will lap it up. Teacher unions, not so much.

And don't get me started on using the unemployment rate instead of actual employment rate (they don't add up to one at all) as well as putting degree-holders employed in positions which don't, or should not by any reasonable standard require, a degree (Griggs v. Duke Power Co FTW), in the same pot with those who hold positions where a degree (or rather the skills and knowledge which the degree is supposed to prove) is actually required.

Is there evidence that teaching is as over-funded? That's the kind of thing that might convince me, the moderate. As it is, I see much more ability to lop off defense spending without sacrificing safety than lopping off teachers without sacrificing learning.

There's quite a bit of research that indicates that lowering of class size has had no discernible affect on education results. The US has 30% more primary teachers per student than it had 50 years ago with little increase in student performance.

That's a lot of additional money for little return. On the other hand, The Teacher's Union.

Bullshit. Google it, Wikipedia it, I don't care. Class sizes in the US have gone up in recent decades, and there's always been a correlation between smaller classes and better performance. Why wouldn't there be?

Or, all government spending is too high. Your proposal is that in return for lower military capabilities we get what other than more daycare subsidies? Here is one of my developing rules of thumb, particularly for economics, in this case the case for more daycare- if an idea is named after a dude, it may not be fully vetted. For instance, are we sure that daycare subsidy does not cause more children to be produced? I'm no eugenicist, but neither am I a dys-eugenicist.

So make a deal. Cut $10 from defense, spend $6 or $7 elsewhere (with higher argued value).


One input piece of data you would want is what proportion of the job-holding college degree holders are employed in jobs for which a college degree is required, rather than having beaten out a non-degree-holder for, say, a job at Starbucks.

Tyler your contortions above do make it a little tricky to see what state of the universe you would prefer. Should articles about defence spending always mention employment levels by education band?

There are a lot of degreed people working in jobs that do not require degrees (like your Starbucks example). When the job market is tight, everyone gets pushed downward. Those with the least education get squeezed out the bottom with no where else to go.

So yes, get rid of those government employees and will they find other jobs? Yeah, probably, but they might be worse-paying jobs outside of their field, and may end up just pushing non-degreed workers out of jobs they otherwise might land.

I'd imagine that overall, total income would drop, and thus, so would total spending. And since their spending is someone else's income, the effect would be spread around. Granted, on the scale being discussed, I don't know what effect it would have on the aggregate measures.

Well, I see none of you have answered the professor's question. Therefore "F" for FAILURE for all you bums. You remind me of somebody saying in Physics 101, "well gravity is not really constant since it varies according to Newton's law by height, therefore, I cannot do the homework problem where gravity is assumed a constant". WRONG! Here's my crack at the questions, which seem quite easy: First one: TRUE. Easy one. TC told you 3.7% is for college, and therefore it has nothing to do with the general UE rate. Second one: (variant): TRUE. In the short term. In the long term, presumably the baristas at Starbucks will go into engineering and medicine and get a college degree (assuming they don't already have one of course). Third one: "What percentage of the money spent on the labor of government military contractors is spent on individuals with a college degree?" - aside from ROTC, very little. Most military personnel are like your average blue collar worker.

Pfui. This is not class and he's not our professor.

That's right, we would only be compelled to do his bidding if we were paying him!

Well, there is this thing called MRU.

Yeah, you're right - he's not our professor.

1. Hmmm, that 3.7% might stay constant as a kind of steady-state flow number, but the unemployment might trickle down as these degree-holders take a position away from a non-degree holder.
2. Hard to know. I wonder how much government hiring is perverse and how much they pay up for degrees. For teachers, they certainly pay up for credentials that have nothing to do with productivity. So, what is the excess spending?
3. Defense spending - I'd guess 100% of the excess spending is captured by people with degrees.

All that to say, repurposing people may be a way to re-price their college degrees and labor costs, but certainly could be a marginal economic disruption. It isn't like this is a choice, though.

Props for playing but you are wrong about most "military contractors" not having a collage degree. I think you are conflating enlisted (who now a days have a lot of collage in certain branches like the air forces) with civilian contractors (who are better educated then the general population.

Third one: “What percentage of the money spent on the labor of government military contractors is spent on individuals with a college degree?” – aside from ROTC, very little. Most military personnel are like your average blue collar worker.

No, he's not talking about military personnel, but about civilian employees of defense contractors -- that category includes large numbers of administrators, engineers, and scientists.

boeing is known for hiring the average ex-ROTC ,

Reread the 3rd one, it's asking about military contractors. I suspect most of the employees of these contractor companies have degrees, not all but more than half would be my guess. We're not talking about the military personnel.

I think Q1 is uncertain. We're making the assumption that degrees will make one a superior candidate for jobs not requiring them. Most college grads who have been working in offices will not put in a good day's work as a laborer, and many will get sick working in the cold because their immune systems and metabolisms are not ready for the demands that will be put on them. Similarly they will not generally have the skills for jobs as mechanics, machinists, punch machine operators, carpenters, brick layers, concrete pouring, or working as line cooks, sous chefs or chefs, waiters or bartenders. They cannot be electricians, plumbers or HVAC people. They won't be able to be painters, or even knock out work post construction.

They will be competing with the no skill jobs in most cases and there their degree will be viewed as a negative signal.

I fail to see how the unemployment rate of the group from which one is hiring makes any difference. If you create a new job, there is one more job available. Logically, that means one more person is going to have to work in order to fill all jobs. Maybe you hire someone who was already employed, and that person's job is also filled by someone who was already employed, and so on. But somewhere, someone new must enter the workforce, because the total employment has increased.

Now, I don't think that military projects should be justified as jobs programs. But neither do I think this makes sense as an argument against.

WRONG! F for your as well, "dan1111". You say: "Logically, that means one more person is going to have to work in order to fill all jobs" -yes, true, The 'one more person', in the short term, is going to come from the 3.7% college unemployed pool, not the general unemployed pool. So you failed to grasp this simple fact. See my answers for the correct answers.

"The ‘one more person’, in the short term, is going to come from the 3.7% college unemployed pool".

This is a faulty assumption, because these groups are not isolated from each other. Due to the economy, lots of graduates are working at jobs that don't actually require a degree. This has the effect of increasing unemployment among those who do not have a degree. Creating more jobs in graduates' fields means some of them will leave these low-level jobs, opening them up for others.

This rests on the faulty assumption that there is no overlap between the groups. Lots of graduates are working at low-level jobs because of the economy. If some of them leave for jobs in their field, it will improve the job market for those without degrees, as well.

Apologies for the duplicate, the first comment did not appear at first.

Clearly, it depends on how much of the unemployment is structural. If non-college grads have 10% unemployment, but it is all because of search and matching frictions, and college grads have 3% unemployment, but half of it is due to a sudden fall in the supply of skill-intensive jobs, then creating a job for a college grad would be more effective than creating one for a non-college grad. Creating one for a non-college grad would just put upward pressure on wages.

Moreover, (continuing to assume the above situation) if college grads want the non-college grad jobs too (if they can't get skill-intensive jobs), then taking a job away from a college graduate effectively puts her into the pool of 10% unemployment non-college grads.

Surely you have heard of a thing called structural unemployment...

Keep in mind the sequester exempts most standard military personnel and would fall heavily on *contractors*. They are overall a well-educated bunch.

So are teachers, if you cut grade school or high school funding or programs that keep kids in school.

On the otherhand, beltway defense thinktanks employ many college graduates.

Maybe we should ask: what has been the rate of growth of defense related businesses since 2001 and what is the rate of growth of teachers.

You don't make choices unless you pose alternatives.

Which should be cut?

"Which should be cut?"


The thing is, federal defense spending is a thing. "Teachers" are a bit more diverse, heterogeneous. Are you suggesting that something concrete, like federal education funding per student, is too high?

I suspect we could abolish the Department of Education without major repercussions. Once upon a time, before there was a Department of Education, people in this country got educations.

But most education money, of course, is at the state and local levels. And the states don't have the latitude of the federal government in kicking the can down the road. So, as we speak, tough decisions are being made on these issues, and state and local governments continue to shed jobs. All very salutary- since the public sector isn't subject to the regular pruning that the private sector has to deal with, these adjustments tend to be more disruptive and painful. And this produces eloquent demonstrations of the bubble in which so many public sector employees continue to live, such as teachers' strikes. Oh well.

But it's all connected: federal stimulus chiefly allowed states to defer cuts from 2009 to 2010- arguably, this was good macroeconomics.

The answer is not taxing rich people, or reforming Medicare, or cutting the Pentagon, or middle class tax increases, or cutting spending, but all of the above.

If we abolish the Department of Education we end net transfers from rich states to poor, for education. That seems an obvious way to create a negative feedback loop.

Talk about the elephant in the room is the title of this piece by the Project on Government Oversight, and it has CHARTS and GRAPHS to show you what has changed since 2001; here is the link:

"This week, Defense Secretary Leon Panetta is poised to deliver his recommendations on at least $450 billion in defense cuts over the next ten years, according to The New York Times. Although the paper outlines several important potential cuts (in a nifty multimedia chart!)-it fails to include on its list of targets something that should have a huge bull’s eye: Department of Defense (DoD) service contractors.

According to the Pentagon’s own analysis of personnel costs dated January 2011, service contractors are “increasingly unaffordable.” That same analysis goes on to say that “the savings are here” and growth in spending on service contractors “has been unchallenged.” The chart shows that former Defense Secretary Robert Gates’ short-lived insourcing initiative only touched a tiny speck of the hundreds of billions the DoD spends each year on service contracting.

POGO found in its “Spending Less, Spending Smarter” report, that reducing spending on DoD service contractors by 15 percent would save about $300 billion over ten years. In FY2010 alone, DoD service contractors cost taxpayers more than $200 billion—$50 billion more than the cost of all uniformed personnel. Also, POGO’s analysis of 35 comparable job occupations fulfilled by federal employees and by government contractors found that, on average, contractors cost significantly more than federal workers.

Talk about an elephant in the room."

According to the Census Bureau, elementary and high school populations were up about 3% (49.5M vs. 48M) between 2000 and 2011. That's cumulative, so I'd expect essentially no growth in teacher ranks year to year, but this would obviously vary from place to place. In fact, if teachers were getting more productive, we might even expect a decrease in absolute numbers, but it doesn't seem that we really expect that.

Most teachers have useless teaching degrees and would be absolutely unemployable if they were fired from their teaching jobs.

Oh we are talking about the sequester.

Does anyone actually think it is a good idea? I'm not sure who is on the other side of this one...

Tyler does...

I do too if that's the only way to get budget cuts. There's smarter ways to cut, but the sequester is better than not cutting.


Actually you are wrong, at least in 2013. While military service members are not affected, DoD civilian employees will take the brunt of this year's sequester because it's applied by line-item across the board, and salaries of federal employees are their own line item. Contractor salaries are not broken out but are an ordinary part of each contract's cost, so they'll be spread across both labor and materials. The result is that all DoD civilians will take a 20% furlough for six months (March through September 2013), while a few contractors lose their jobs and the vast majority continue on as usual. The contractors who lose their jobs will be the lowest-paid and credentialed ones at the bottom of the corporate food chain.

In 2014 when the cuts are applied at a higher level, I'd expect the usual pattern of pain (federal employees decide to cut contractors) to take effect and the cuts to contractors to be much more severe, although again lower-paid employees like mechanics and assembly line workers will fare much more poorly than cybersecurity specialists or contract managers.

If Pentagon employees are like State Department employees, they can't be fired anyway. They can be "reduction in forced," which entitles them to another job within the federal system. It's that person who takes the job loss.

Berke Breathed addressed this issue:

Let's look at the time unemployed by sector. I wouldn't be surprised that if you lay off a teacher he stays unemployed for a year or two, and that the 3.7% unemployed are long term unemployed. Therefore, a lost job in this sector will show up in the overall unemployment rate as much as any other job.

great idea! lets fire all the economics pundits!

Is government soaking up all the bullcrap majors like women's studies, etc.? Start cutting higher education subsidies and forcing government departments to cut down on non-core personnel and you may well see unemployment for people holding college degrees move higher.

So the crocodile tears to be shed over the poor folks that will be selling apples in the street due to budget cuts are as genuine as Clinton's?

This stuff is all about sinecure. Golly gee.

And the answer is no. The skills and abilities necessary to thrive in government circles, either directly employed or as contractor, are far different than required in the real world. One example; a local electrical contractor has as the highest paid employee on staff an industrial hygienist. The only reason he is there is because of regulatory requirements. Without, he wouldn't be, or wouldn't be paid as much.

Even skills that would translate would require substantial change in approach. The exigencies of a free market are far different.

Closer to home it may be more pertinent to ask yourself this, "if the state of Virginia deemed George Mason University superfluous and decided to shut it down, would Tyler and the other members of the Economics Department find comparable jobs at an equivalent university?" Unless you can answer this, your questions above are just silly.

One needs to know if the fiscal expenditures under consideration will lead to changes in the deficit or if they will be ofset with other changes in taxes or expenditures. Then one needs to know what monetary policy is and how it will respond to the fiscal action. Then these assumptions need to be put into a model to estimate the results.

Finally, if one is thinking about the sequester, specifically the DOD sequester, the most important thing to consider is the effect on defese.

I assume that the defense sequester will not be offset and that the Fed will not be able to fully ofset the decine in spending, but that the effect on defense is so marginal that it's worth allowing to happen. We need to shift those defense consutants into figuring out how to reduce health care costs, right?

I'd argue it's uncertain. We probably shouldn't assign much weight to the aggregate unemployment rate, but why is "college graduate" the right level of disaggregation? Not to get all Arnold Kling on you, but these are likely to be people with more specialised skills and therefore not necessarily easy to re-employ into a new "college graduate" role. If the sequester cuts funding for a lot of procurement, leading to massive unemployment among aerospace engineers, they might not retrain easily into the growing field of Affordable Care Act Compliance Monitoring.

I don't like the jobs argument for defence spending, but this isn't the right counterargument.

If you figure there are only a certain number of jobs that can be done without a college education, and even those jobs would prefer to hire someone college educated, we can see that this is incorrect. If you eliminate a college-education job, the college-educated worker might find a new job quickly, but it might be a no-college job. That would make the no-college workers even worse off, since they have to compete with more workers with better qualifications. When all the burger flipping positions are filled by former defense contractors, the unemployment rate among the college-educated won't rise, but it will become even worse among uneducated former burger-flippers.

Irrelevant? The unemployment rate makes no comment on the quality of jobs, so a masters degree holder flipping burgers or writing software are both "employed". That Atlantic article displays the obnoxious Thomas Friedman/David Brooks-level superficiality that tends to rear its head from time to time on blogs when people are trying really, really, hard to show everyone how clever they think they are.

The effect on unemployment rates for college graduates depends on how the spending is cut. The sequester really is about the worst way you could enact a counter-productive policy. I would say that if we prove our ability to restrain costs, or whatever nonsense, you'll see the unemployment rates for the least skilled rise the most as not only the multiplier (regardless of the sign attached to the second derivative) works in reverse, but as college educated people get jobs well below their skill level.

Sometimes, simple insights aren't commented on that often, because they're fairly obvious, and not really worth talking about.

My first comment. I agree with Ray's answers. I was a military officer. Required a degree to get a commission. Earned a Masters in the service. Proved to be a great tie breaker, perhaps the only value to me. Managed a number of contractors. Balance tilted towards blue collar employees, many if not most ex military. All in all, strongly believe that military personnel are cheaper. However, different pots of money and different political support.

Hey all right Earl, at least somebody agrees with me. I based my opinion on all the guys that are contractors who are ex-military. Insofar as the question posed by TC goes, I don't think we ever got an official answer in this thread from our teacher. Methinks TC I think is playing a game of intellectual chess with his economist peers and it's a couple of steps removed from common sense. These kind of debates are not for me, as oftentimes you'll find the debaters are debating a model (e.g., Neo-Keynesianism, which I learned incorporates old fashioned Keynesianism from the 50s and 60s with Lucas Rational Expectations of the 80s) that has a number of assumptions that may or may not be applicable in the real world.

Wouldn't some argue that college educated people laid off from the defense industry would displace less qualified employees in other industries.

Is the argument that the recession (a cut in consumption and reduction in aggregate demand) has not affected the college educated, and therefore further government cuts where the employment is mostly college education shouldn't affect them much either?

I find that unlikely. 3.7% is larger than 2%, the pre-recession employment rate for the college educated.

The recession basically doubled the unemployment rate for all educational attainment levels at it's worst point. Buried in that doubling of the unemployment rate are the ~150,000 teachers laid off via state and local budget contraction.

For details, see

Note that a college educated person serving coffee at Starbucks counts as employed.

So yes, the college educated are still feeling the effect of the recession. I don't see how cuts in government spending now while unemployment is still high will not make that worse.

"Therefore if we cut government spending on the jobs of those individuals, they will be reemployed reasonably rapidly. We should not assign much weight to the aggregate rate of unemployment in making this judgment. True, false, or uncertain?"

Very uncertain, as it depends on how many of those individuals are directly or indirectly employed by the government.

"Variant: If we increase government spending to hire these individuals, it will not much lower the rate of unemployment. True, false, or uncertain?"

See above, there isn't enough data to answer this question.

"Additional exercise: What percentage of the money spent on the labor of government military contractors is spent on individuals with a college degree?"

Not seeing the relevance, so skipping the additional assignment.

It's also often forgotten that the unemployment rate has little to do with how many people that would want a job are actually employed. People with college degrees are reluctant to see themselves as unemployed. They will consider themselves on a sabbatical, a mid-career re-orientation, a child-rearing break or whatnot. Unless pushed really hard, they won't ask for unemployment benefits either as it would be a personal blow.

I've worked in the Pgon, and I've been a contractor, so here's what I think. People who work in defense related activities fall into two basic categories -- civilians who have a career in the business, or prior military who have switched. They have very specific human capital -- they know technical issues related to defense, but not much else. The older they are, the more specialized they obviously are, and the greater the transition problems to alternate jobs. They are not recent graduates, and not easily reemployable -- draw down defense, and, in the short run, unemployment rates will go up. The same goes for short-term rates from cutting acquisition of new or the modernization of old weapon systems -- you will see layoffs of workers with very specialized human capital in areas heavily devoted to defense spending, and unemployment rates will go up. Obviously, over time, this will ease off, and the effect will be small or none; however, there will be permanent losses of personal wealth for these individuals as they are likely to lose the value of their specific human capital, and get wages that are at best equal to the value of general workers.. But, overall, any defense drawdown due to a sequester will not be that huge in relation to the total economy, so it won't matter that much for the total unemployment rate. Locally, it could be important, but for the economy, not so much.
But here's a question for Tyler: why does he link to a picture of the Pgon taken before 9/11? The building has changed quite a bit since then.

Comments for this post are closed