Josh Barro on Raj Chetty’s UI Research

Josh Barro writes (Part 1 and Part 2) about my earlier case describing Chetty’s working on unemployment insurance. Barro:

I can think of a very good reason: unemployment benefits are capped at a fairly low level. In New York State, where I live, the maximum weekly unemployment benefit is $410, equivalent to an annual salary of $21,320. So, the need to forego further UI payments shouldn’t be much of a disincentive for an unemployed lawyer to return to work, but could be significant for an unemployed fast food worker. And there is surely a strong link between a worker’s savings and earnings potential. So, I’d say that the limitation of incentive effects to those with the worst liquidity is entirely consistent with the incentive story about UI benefits and unemployment….

One of the possible advantages that is touted for more generous UI (including by Mike Konczal) is the idea that it allows for better job matching—people can wait to find the right long-term job opportunity instead of taking the first job that becomes available. Under this theory, longer unemployment duration may actually be a feature of unemployment insurancerather than a bug.

We should be careful about this, as it assumes rational expectations on job seekers’ part about what they can expect from a new job. Is greater liquidity really allowing job seekers to take the time to find better jobs? Or is it just letting them prolong a vain search for an unattainable salary before settling?

Especially in a deflationary economy, it may be important to encourage workers to settle. In many cases, a worker’s best choice today sadly is to take a job that pays less than the one he or she lost. But people have a strong psychological resistance to doing this, which is why wage stickiness is a problem and why deflation hinders economic recovery. If generous UI fosters more wage stickiness, that’s an argument against generous UI.

To back up, let’s explain what is going on in this graph:

According to the paper: “Eligibility for severance pay in Austria is based on a simple discontinuous rule that applies to nearly all workers: people with 3 or more years of job tenure are eligible, whereas those with shorter tenures are not.” So those on the left side of the break get nothing, while those on the right side get a lump-sum.

The lump-sum shouldn’t have a work-disincentive effect. It shouldn’t alter the relative price of work versus leisure. But it does increase the length for which people look for a job. This is consistent with a liquidity effect; if you have cash on hand you’ll take more time to find a job. We should assume on a first pass that this is an optimal amount of time, increasing social welfare. People are putting what they consider to be the best amount of time, energy and job tradeoffs into their job search. Unemployment duration shouldn’t be seen as a bad thing here. If they take longer than normal that isn’t because of a work-disincentive effect. Let’s remember what is causing the longer unemployment duration here: it’s having cash on hand, it means you aren’t going broke and need to get working immediately.

Barro notes that under deflation this could be consistent with price-stickiness. I’m not convinced by the “structural unemployment” camp, finding arguments similar to Heather Boushey’s much more convincing, but to the extent that job searches have increased because of sectoral and geographic shifts then this liquidity boost is a good thing allowing for the optimal search time.

For what it is worth, they look at data broken by mortgage holders. Mortgage holders are, in general, richer and better educated than non mortgage-holders, and a similar mechanism shows up:

From The Chetty Paper: ” Figure 4a shows that UI bene…ts have a clear, statistically signi…cant effect on job …nding rates among households that are paying of mortgages prior to job loss. In contrast, Figure 4b shows that the e¤ect is smaller for households that are not paying of mortgages and are hence less constrained. Results are similar for the spousal work proxy: UI bene…ts have a much larger effect on job fi…nding hazards for single-earner families than dual-earner families.” So this isn’t a proxy for class or income as much as it is a proxy for being liquidity constrained. Mortgage holders take jobs quicker in low UI states versus high UI states.

This entry was posted in Uncategorized. Bookmark the permalink.

4 Responses to Josh Barro on Raj Chetty’s UI Research

  1. purple says:

    A restaurant worker wouldn’t qualify for max unemployment benefits. It’s an insurance program that pays out based upon previous salary, usually 60%. Therefore the opening paragraph that quotes Barro is nonsense.

  2. purple says:

    One wonders how much economists actually know about the real world when they don’t even understand the basics of UI. All the mathematical modeling in world can’t cure lack of intellectual curiosity(on the part of Barro).

  3. Jon H says:

    The assumption some seem to have, that if you made $X at your prior job, then you can easily get any job that pays less than $X, and if the difference is large, then you’re practically entitled to the lower-paying job, is not realistic.

    Employers offering lower-wage jobs are often uninterested in hiring people who formerly made more, because they assume that the hire will leave ASAP.

    So even if lower wage job openings exist, they may not be available to an unemployed person coming from a higher wage job.

  4. Pingback: Human Rights Facts (196): A Right to Unemployment Insurance? | P.a.p.-Blog, Human Rights Etc.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s