I was reading this article on planned obsolescence, which states

On a macroeconomic scale, the rapid turnover of goods powers growth and creates reams of jobs

In context, it's implied that this is good - it is good to have people working to design, market, make, and distribute goods to replace other goods that break after a short lifespan.

Naively, it seems very bad to me, compared to the option of having goods that last a long time. Those people working on making replacement goods are now not available to work at other jobs or use that time on other pursuits. If the goods lasted longer, we could spend far fewer resources, as a society, supplying everyone with goods. Then we would have many more resources (raw materials in the goods, human labor making the goods, time consumers spend comparing and purchasing new goods, etc.) available to solve other problems or take on other projects.

Yet, I almost always hear people refer to a project or industry hiring lots of people to work on it as an innately good thing, unlike all other forms of resource expenditure. (I never expect to hear that planned obsolescence has an upside because, "look at all the copper it's using!" or "it really keeps our power plants busy providing electricity for the factories!".)

Why is "jobs" treated so differently as a resource, and should I change my naive approach to thinking about them?


4 Answers 4


In general, it is not:

Despite the focus on ‘jobism’, jobs are a cost of production and consumption, not a benefit, and should be minimized

Mark J. Perry @ AEI

Public discourse tends to be carried out in terms of jobs, as if a great objective was to create jobs. Now that’s not our objective at all. There’s no problem about creating jobs. We can create any number of jobs in having people dig holes and fill them up again. Do we want jobs like that? No. Jobs are a price and we have to work to live. Whereas if you listen to the terminology you would think that we live to work. Now some of us do. There are workaholics just like there are alcoholics and some of us do live to work. But in the main, what we want is not jobs, but productive jobs. We want jobs that will be able to produce the goods and services that we consume at a minimum expenditure of effort. In a way, the appropriate national objective is to have the fewest possible jobs. That is to say, the least amount of work for the greatest amount of products.

Milton Friedman in a 1980 lecture

Under most circumstances, wages paid to people employed on a project should not be counted as benefits, although this practice is disturbingly widespread. As mentioned above, the opportunity cost of labor used on a project is the output that that labor would have generated elsewhere, either in paid work or in household production. Including wages paid to labor on a project as a benefit implies that that labor would have produced nothing otherwise and that the agency making the wage payments does not have standing.

Zerbe and Bellas (page 117-118):

I say in general, because there are some evidence that "jobs are good for you", and if increasing jobs diminishes the number of long term unemployed, that might be considered a benefit, even though it diminishes the quantity of leisure and home production they consume.


No economist believes that a job is a desirable end in and of itself.

Instead, a job is merely a means to an end, the end being consumption ("consumption" broadly conceived).

You are correct that "people" (i.e. laypersons and in this case some writer for the BBC) often assume that creating lots of jobs is automatically a good thing. You are also correct that these "people" are wrong to assume so.

If in the future automation, the singularity, and all that bring about an unlimited cornucopia of abundance for everybody even while nobody has a job, then this will be a very good thing.

Conversely, the Soviet Union and other communist states had, officially and technically, 0% unemployment. But life there was not necessarily good.


I agree with the previous posters that jobs themselves are not desirable in economic theory. Nevertheless, I believe jobs can be a good thing, in cases where theory does not apply to reality.

More specifically, the assertion that jobs are not desirable rests on an implicit assumption: Distribution of wealth is not a concern.

Many basic economic models do not have room for natural unemployment. Arguing that wages paid to workers should not be counted as benefits, like in the Zerbe and Bellas quote, implies that workers could have found employment otherwise (which they admit). However, I do not know how realistic that assumption is today for low skilled workers.

Unemployment itself, however, is not a concern, if lower labor grows the economy and we compensate the unemployed. This does not happen in reality. Nor is there the political will to do so in the future, with many countries looking to slash welfare payments and characterizing welfare recipients as parasites.

It is true that we don't want jobs, but actually want goods and services. This ignores distributive issues, as we often do in basic economic theory, which is why there is a difference between what laypeople think and what theory tells us. In reality, however, distribution is extremely important and a major driver of recent political changes around the world.

There is another theoretical reason we ignore distribution in many of our models. Based on the second welfare theorem, we can try to maximize total welfare, which implies minimizing jobs and then give the jobless a share of that welfare to compensate them. While the theory behind this is solid, this rarely ever happens in practice.

Automation will reduce jobs, which overall is good for the economy, as the previous posters said. However, there will be winners and losers. Naturally, if the losers are not compensated (which they rarely are in practice) then losing jobs is not a Pareto improvement and therefore cannot unconditionally be supported by economic theory. The fact that those losers could be compensated in theory is of little importance to those people not being compensated in practice.


People here have made good points, but there are reasons people like to see jobs created. Lower unemployment reduces the tax burden due to "scroungers" (who are always treated as deliberately unemployed, even though this as a reason to praise a rise in labour demand is self-contradictory); it might increase revenues for the treasury; and those who already have jobs may see theirs become more secure and their wages rise due to an increase in demand for labour. There may be reasons this cetetis paribus logic is flawed, but it's what people have in mind when they praise jobs.

  • $\begingroup$ It reduces the tax burden for those people, but someone is still paying those people, so the total burden hasn't changed! $\endgroup$
    – user253751
    Dec 16, 2020 at 14:08
  • $\begingroup$ @user253751 But at least it's felt like those people gave something back to earn the money, even if we don't really need it. $\endgroup$
    – J.G.
    Dec 16, 2020 at 14:14
  • $\begingroup$ Like the glazier who can't find any business so he pays some teenagers to smash windows so they'll pay the glazier for new windows. $\endgroup$
    – user253751
    Dec 16, 2020 at 14:19
  • $\begingroup$ @user253751 That old chestnut. My answer is about reasons people want job creation that don't necessarily hold up to scrutiny, but nonetheless influence either policy or who gets to make it. $\endgroup$
    – J.G.
    Dec 16, 2020 at 14:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.