The more I read about the inside political debates during the Great Recession and its aftermath, the more I doubt whether the current U.S. political system puts a top priority on boosting employment among the long-term unemployed.

For example, based on Michael Grunwald’s book The New New Deal, which focuses on the politics and impact of Obama’s fiscal stimulus package, a tax credit program to expand job creation for the unemployed failed several times due to lack of support from the political system. Democratic Senators opposed it because they preferred to expand various spending programs as part of the stimulus. Republicans opposed it because they opposed Obama’s entire stimulus program.

Both Democrats and Republicans agree with Rahm Emannuel’s well-known statement that politicians should never let a serious crisis go to waste. If it’s a recession, and jobs are needed, Republicans use this as an opportunity to advocate for permanent tax cuts for all. Democrats use this as an opportunity to advocate for expanding a wide variety of spending programs. The politics are irresistible: let’s use the need for job creation to advocate for policies we favored already, for other reasons.

Tax cuts or spending increases during a recession do create jobs. But the problem is, they do so at such a high cost per job that during a truly “Great” Recession, it is politically difficult to have a big enough stimulus to create a sufficient number of jobs to re-employ the long-term unemployed.

For example, estimates suggest that spending increases have a cost per “job-year” created (one job for one year) of $92,000, whereas tax cuts have a cost per job-year created of $145,000. Although spending increases are a bit more cost-effective than tax cuts, a cost of close to $100,000 per job year makes it difficult to address our large employment problems. The latest estimates from the Hamilton Project suggest we are short 11 million jobs of getting back to pre-recession employment conditions. At a cost of $100,000 per job-year created, a fiscal stimulus of $1.1 trillion would be needed to make up this entire job gap.

This job gap has significant long-run costs. A lost job will lead to persistent losses in wage rates and reduced employment rates for the unemployed. This signifies a loss of job skills, job connections, and self-confidence among the unemployed. The monetary cost of this alone to the unemployed has been estimated, for example in research by Steve Davis and Till von Wachter, to be over $100,000 during recessions. CBO has identified the loss of job skills among the unemployed as a significant cause of a decline in the U.S. economy’s long-term growth potential. In addition to these costs for the unemployed and the economy, persistently high long-term unemployment has costs for our tax and transfer system, for example in increased costs for the unemployment insurance system.

Based on the Hamilton Project’s current projections, if U.S. job growth averages 208,000 jobs per month for the foreseeable future, which would match the performance in the best year (2005) since 2000, the U.S. will not return to pre-recession employment conditions until 2020.

If we really want to help create jobs on a large scale for the long-term unemployed, we need measures that target such job-creation. Only targeted measures, which are conditional on the job creation we’re seeking, can be sufficiently cost-effective that they can meet an appreciable share of our job needs at an affordable cost.

For example, I have advocated in the past for a strategy that would include two components:  a job-creation tax credit, which would incentivize job creation of all types by small business and small non-profits; a wage subsidy program run by local workforce agencies that would provide large subsides for new jobs targeted at the long-term unemployed. The former program targets job creation in general; the latter program targets a portion of those jobs at the long-term unemployed, to minimize the erosion of their long-term earnings ability.

I estimate the gross costs per job created from either of these two measures to be around $30,000 per new job. But the created jobs will increase tax revenue and reduce welfare spending. The net costs per job created, after accounting for these budget feedbacks, would be reduced to less than $20,000 per job created.

These targeted job creation programs can be more cost-effective in creating jobs precisely because they are targeted. Tax credits are provided or wage subsidies are paid conditional on jobs being created. This reduces budget costs that are not associated with job creation, and provides greater incentives for job creation.

Why isn’t there more of a political constituency for such job creation initiatives? In my view, because such job-creation initiatives are targeted only at one political goal: job creation. This job creation principally benefits the long-term unemployed, who are not a politically influential group. The targeting that helps job creation programs to be economically cost-effective is precisely what makes such programs politically ineffective.

What is the solution? I think the only ultimate solution is political: the long-term unemployed need to have more of a political voice. But that is easier said than done.