Arnold Kling  

The Job Assignment Problem

Market Discipline for Banks... Is Blinder Going Off the Reser...

Alan S. Blinder writes,

The last two quarters were even more extreme: Productivity in the nonfarm business sector grew at a shocking 8.1% annual rate. There are two possible explanations. One: The last two quarters were among the most technologically innovative and entrepreneurial in the history of the United States. Two: Fearful businesses pared payrolls to the bone. If the second is closer to the truth, payrolls are extraordinarily lean right now. Which means that firms will need to hire more workers as their sales and production grow. Which means that employment may start growing sooner than the pessimists think.

I agree that high productivity growth is very good news and that it portends increases in hiring. But I caution against thinking of the economy, as Blinder does, in conventional Keynesian terms, where we all work in the GDP factory making GDP units.

One way to describe the Recalculation Story is to note what I call the job assignment problem. Suppose that it were your job to assign every worker in the United States a job. You have to decide who will be a manicurist and who will be a dental assistant. You have to decide who will work in a biotech lab and who will work on a web site.

Obviously, this is a ridiculously complicated task for any one person. It has become more complicated in recent years, as the variety of goods and services has increased.

The last time we had anything close to a centrally planned economy was during World War II, when the government assigned a lot of jobs quite directly. My father's score on the mechanical aptitude test he took in the army was so poor relative to his other scores that they suspected him of deliberately hiding his ability. However, they did not make him a mechanic.

But those were simpler days, when men were either driving tanks or building them. In today's economy, the job assignment problem is much tougher.

The Recalculation Story is simply that the job assignment problem has become too difficult for the market to solve quickly, using its tools of price signals and incentives. Government could step in and assign lots of jobs to lots of people, but that would probably come at a loss of the productivity growth that really ought to be our focus.

What people are calling a "jobless recovery" is what I would call the market taking a long time to solve the job assignment problem. Where should the market assign this year's graduating college seniors, who have spent four years learning about the evils of capitalist oppression and the virtues of not working for a profit? I know a lot of parents with twenty-somethings living at home who would like to see that job assignment problem solved. Unlike Professor Blinder, I am doubtful that the solution will arrive quickly.

Comments and Sharing

COMMENTS (11 to date)
TA writes:

Well, if Blinder is right, and there is no productivity slack, and you are right about slow recalculation, then recovery of GDP will be slow. I think what a lot of the comment has been about is the possibility of GDP growth without employment growth, which is something else again.

Douglass Holmes writes:

What really scares me is that those graduating seniors are deeply in debt to pay for an education that has not prepared them for the job market. That's wasted capital that could have been better spent increasing our productivity.

Well, I guess that, since my recent college graduate has a job and no student loans, I had better spend more time worrying about finding work for my favorite 58 year old - me!

q writes:

if productivity is increasing this rapidly, then there are profitable investments not getting made.

the fact that productivity is increasing means exactly that capital is flowing to profitable investments, which means that there is no fundamental problem of 'recalculation'. the economy is showing clearly what constitutes a profitable investment. people are being employed in productive ways.

in short it is not at all clear that the wrong kinds of investments and jobs constitute the economy right now. if that were so we would not expect productivity to increase so fast.

the problem is simply that there isn't enough investment. the 'hurdle rate' for capital is set too high. the problem could be solved by more money going after the same kinds of things. the fact that productivity is going up so high now means that enough of these would be profitable.

manuelg writes:

> I agree that high productivity growth is very good news and that it portends increases in hiring.

Running a business, if I have productivity growth, and, for whatever reason, I wanted productivity to go back down, the fastest way to do it would be to hire more employees.

I am not supposed to admit this, but I feel compelled to be honest. Higher unemployment and higher productivity growth have no demonstrated inverse relationship.

When the social costs of unemployment come home to roost, we will see legislation to enforce lowered productivity. I can complain about it, but it would be foolish to not see it coming. Somehow, I will be compelled to hire more employees, and productivity will go down. Because employees bring with them the burden of mandated entitlements and the burden of their own inner sense of entitlement. Like it or not, it is foolish for me to pretend otherwise.

> What people are calling a "jobless recovery" is what I would call the market taking a long time to solve the job assignment problem.

The market has solved the job assignment problem by not hiring. And productivity has increased. During the "good times", the market was more tolerant of low productivity - now less so. Businesses act accordingly. Because, at this time, businesses are not compelled to bare a share of the current and expected social costs of high unemployment (including lost taxation and lost consumer demand, besides criminality induced by idleness).

8 writes:

Maybe it isn't a recalculation problem, but an input problem. If the equation is balanced, then we need to change a factor to lower unemployment.

Randy writes:

I'd replace "job assignment problem" with "assignment problem". That is, I see no reason to assume that people should be working just because they exist. Employment bias limits thinking on alternative solutions, e.g., family, friends, home work, communes or cooperatives, and yes, even government.

CJ Smith writes:

I second manuelg's second comment and expand as follows:

Innovations in technology have made technological and automation capital much cheaper than human capital in the long term. Other than the hurdle of initial investment, technology is either cheaper to replace, or lasts longer than similar human capital in most industrial applications. Spend $50,000 a year for a union spotwelder in your auto factory or spend $250,000 for a robotic welder that will last for 10 years, not demand raises, benefits, time off, the right to strike, etc. Technological capital is also becoming more prevalent in service industries - ATMs, online billing and collections, decision tree telephone routing systems, self-service checkouts, etc. Even McDonalds is going to customer touchscreen ordering, and God help the poor person trying to get through a live person at Microsoft, HP, or Dell.

Prior to the downturn, getting rid of human capital or converting from human to technological capital posed significant difficulties in terms of union pushback, discrimination lawsuits, etc. The downturn provided a perfect opportunity to clear shop floors and offices. Given the opportunity afforded by cleared shop floors and offices to rebuild using human capital versus technological capital, I'm betting the the jobs recovery will significantly lag GDP growth as compared to prior recoveries. Not because of a recalculation problem per se, but because the need for human capital in tasks that can be automated has dropped and will continue to drop significantly.

I disagree with manuelg' first comment regarding legislation to compel more utilization of human capital - that's been attempted by, among others, the Luddites, the Wobblies, and the Communists. None have succeeded in the long run. Instead, I anticipate a combination of the following: the workforce will either develop or rediscover markets that require human capital (R&D, personal service) or the supply will re-train to enter other labor markets a la buggy whip makers and farmers.

manuelg writes:

> I disagree with manuelg' first comment regarding legislation to compel more utilization of human capital - that's been attempted by, among others, the Luddites, the Wobblies, and the Communists.

I didn't say it would work as advertised! ;-) It will "work" to reward a particular voting block, as one would expect. High unemployment provides the "cover".

> ... the workforce will either develop or rediscover markets that require human capital (R&D, personal service) or the supply will re-train to enter other labor markets a la buggy whip makers and farmers

Or neither. New positions never opening up at an adequate rate is a distinct possibility. Take R&D - why not use 40 Indian PhD's to do the work of 60 US PhD's? Take personal service - a goodly portion of the work of a personal assistant can be done remotely - again employing someone without the same expectation of quality of life and compensation in a place with a much lower cost of living.

The majority of the work will be of poor quality, but the cost savings will be compelling. Like the craftsmanship of a pocket-watch compared to the time display on a plastic cell phone. People spend a smaller fraction of their income on telling the time (pennies out of the cost of the phone and service), and lose daily contact with items to true quality, and that prior level of quality becomes a boutique luxury item, and employs far fewer people.

> to enter other labor markets...

It may in fact be easier to "retrain" a US worker to become a ethnic Chinese worker (chopstick skills, walking stooped to pass through lowered doorways, reading/writing Chinese letter-forms) than it would be to retrain a US worker to have the skills and sense of responsibility and lack of sense of entitlement that the new employment positions opening up require. Look at the malaise of Japan's labor economy as the positions disappeared for lifetime employment by a single employer, never to return. Even after years of a desperate situation, it is not 100% certain that a human will change their attitudes to match a new reality.

Not to put too fine a point on it.

Mike Rulle writes:

While your intellectual construct of "recalculation" is very useful, it is difficult to accept your belief that modern complexity makes new job discovery/creation more difficult---or time delayed from a macro-perspective. One could argue that the very large diversity of jobs in a modern economy makes it easier for people of differing tastes and skills to match themselves with occupations, on average, than in prior times (WWII, for example).

Dirtyrottenvarmint writes:

I am late commenting as usual, but I still don't see how "annual productivity growth" is a meaningful number when productivity increases are measured in dollar values per unit, and the thick and greasy finger of Government is stuck way up the backside of the dollar.

Again I note that productivity per hour jumped 8.1% in the third quarter - the same quarter that the Government was spending significant amounts of dollars on such wonderfully Productive programs as "cash for clunkers". Coincidence? Or Causation?

Productivity is defined as the division of two variables, value of output and time. But we have a third variable, which is how we measure "value". That variable has been anything but stable. Therefore the numbers are bullshit.

This is getting a bit more subjective, but I much prefer the UK Marketplace. The interfaces seem to be more colorful, has more flair, and some cool features like 'Blog's that let you quickly see the owners point of views, stances, and opinions, allowing you to get to know the owners and then back someone you parrallel on a moral level. This also is letting you find others with shared tastes and becoming friends with them, thus, build up your online social community. Those concerned with privacy will be relieved to know you can prevent the public from seeing your personal listening habits if you so choose.

Comments for this entry have been closed
Return to top