Stimulus Jobs

One major difference between now and the 1930's is production technology. Back then, infrastructure projects were very labor-intensive. You could accomplish something useful by gathering up hundreds of low-skilled unemployed people and putting them to work. But that is not true anymore. Most major construction today requires high-skilled workers, from project managers and senior engineers to people who know how to handle a backhoe.

Almost all jobs programs coming out of the government will do nothing to help unemployed people. Instead, they tend to throw money at sectors like health care and information technology where there is very little unemployment. This can be actively harmful, if the money displaces private activity and hires people away from firms where they were already producing useful things.

A Great-Depression style public works program will simply not work in the modern world. If we are going to do fiscal stimulus, we need projects where we can hire low-skilled people to do useful jobs. This article, written by an energy industry insider, identifies one such job:

Energy efficiency is generally more cost effective than pretty much anything else having to do with energy.  Current incentives should focus on that, we should develop energy efficiency jobs and go out and weatherize everybody's house to cut or emissions because the amount of energy we simply waste is pretty mind boggling considering how costly energy use is to us.  
...
once you start digging into old house, what you find that needs fixing can be never-ending.  The actual work of crawling through somebody's attic or crawlspace with a caulk gun is also especially miserable, dangerous, and temporary in nature, so we should keep in mind that the "green jobs revolution" we often hear touted is something of an idealization.

The work of crawling through attics and crawlspaces with a caulk gun is absolutely perfect for a government jobs program. It requires no special skills and accomplishes something useful. And since it is nasty and people will probably hate it, they will not want to remain on the government payroll for too long. Once they establish a record of steady work, private employers will be more likely to hire them. This kind of thing is the best possible scenario for fiscal stimulus.

Unfortunately, it still has not worked:

Federal and state efforts to stimulate creation of green jobs have largely failed, government records show. Two years after it was awarded $186 million in federal stimulus money to weatherize drafty homes, California has spent only a little over half that sum and has so far created the equivalent of just 538 full-time jobs in the last quarter, according to the State Department of Community Services and Development.
The weatherization program was initially delayed for seven months while the federal Department of Labor determined prevailing wage standards for the industry. Even after that issue was resolved, the program never really caught on.
"Companies and public policy officials really overestimated how much consumers care about energy efficiency," said Sheeraz Haji, chief executive of the Cleantech Group, a market research firm. "People care about their wallet and the comfort of their home, but it's not a sexy thing."

This does not mean that it would be impossible to have a good jobs program. I can tell from the article that this program was badly managed. Instead of letting the market determine wages, the government 'determined prevailing wage standards' or set price controls. That never works. The article later notes that there were 200 applicants for 16 job slots for a green jobs program, which is a sign that the wages it assigned are way too generous. It would be better to pay each person less and hire more people, meaning that less people have to suffer the effects of unemployment. Instead of having a bureaucracy delay things seven months to craft a damaging regulation, they should let individuals choose their own wages.

Also note the math. $93 million has been spent for 538 jobs. That is over $172,000 per job. That number is actually pretty low compared to the amount that tariffs cost to keep a job, but it is still not a good use of money. Also note that this money is just the amount of taxpayer money spent on this program; it does not include the money that other people had to pay, or the time and money spent writing the regulations.

$93 million is enough money to hire over five thousand people at minimum wage for a year and supply them with the basic weatherproofing tools. Instead of trying to get homeowners to pay money to have a stranger of questionable character come to their house, the government should just hire all the people to weatherproof government buildings. The energy savings in the future would have repaid the cost, and a lot more people would have been helped. Then, once people had the experience, they could have been hired by private companies to do this work or something else.

An important lesson from economics is the targeting principle. If you want to do something, you should do it in the simplest and most direct way possible. If you want unemployed people to have a job, then you give them a job. You pay them the money yourself and have them do something useful. Do not mess around with partnerships and training and bureaucracy. Trying to be tricky will almost always generate a lot of negative side effects or will simply not work.

Of course, the Davis-Bacon act and public sector unions would make such a simple targeted program impossible. All of the newly hired people would have to be paid way too much, which means that a lot less people could be hired and they would not want to leave the government payroll. The failure of California's program shows that, even with the optimal situation, with a useful job that needs doing and can be done by unskilled people, our government, with current laws, cannot do fiscal stimulus right.