Unemployment Benefits: Incentives Matter

I’ve argued time and time again that incentives matter.  And why people think this wouldn’t apply to the incentive not to work is beyond me.  But now there is evidence of just how strong that incentive is: via Dan Mitchell

The extension of UI [unemployment insurance] is found to have a positive and significant impact on the national unemployment rate…. The UI benefit extensions that have occurred between the summer of 2008 and the end of 2010 are estimated to have had a cumulative effect of raising the unemployment rate by .77 to 1.54 percentage points.

That seems pretty significant to me.

4 responses to “Unemployment Benefits: Incentives Matter

  1. The question would be whether or not the benefit of extending it is outweighed by the loss. At some point as jobs come back (as they finally are) there will be enough opportunity that extending unemployment makes no sense. When things are really tough, however, the fact some abuse the situation is a risk worth taking due to how many are really in difficult straits.

    • The question would be whether or not the benefit of extending it is outweighed by the loss.

      Hi Scott,

      I agree. If taken to the extreme the position could be that we pay no unemployment benefit; the incentive to work dominates right away. Or, on the other hand, we could say that we will pay out benefits for X number of weeks right away. This provides both the cushion AND the immediate incentive.

      At some point as jobs come back (as they finally are) there will be enough opportunity that extending unemployment makes no sense.

      I would feel more comfortable if we would actually try to do more measuring. I feel the whole process is too touchy feely.

  2. I am skeptical of studies (even like the one you cite) that tries to give precise numbers. They are all built on assumptions and methodologies that are themselves questionable. The precise numbers give an illusion of exactitude.

    For instance, the main method in the study you cite is to compare the duration of unemployment with the recession of 1980. Yet it could be that the nature of this recession (high debt, no sudden dip in oil prices, etc.) simply means the unemployed stay unemployed longer. If that’s the case, the conclusion of the study is false. In fact, the idea that somehow this recession is a mirror of 1980 is very dubious. That said, as a social scientist I appreciate the study and think it’s valuable — but only along side a lot of other studies and comparisons. What happens too often is people assume there is more certainty behind a study than there is. I’ve never seen a complex issue ever have a “silver bullet” study that proves something conclusively — rather, comparing a vast myriad of different and competing studies is important.

    Still, the logic – that longer unemployment insurance creates an incentive for some especially low income workers to stay unemployed longer than they have to – is pretty solid.

    • I am skeptical of studies (even like the one you cite) that tries to give precise numbers. They are all built on assumptions and methodologies that are themselves questionable. The precise numbers give an illusion of exactitude.

      Sure. We all have studies that we pull out to “prove” our point of view.

      But I like studies more than I like the pure brass knuckles of politics. For example, I’m willing to buy that the number “.77 to 1.54 percentage points” could be wrong. But I think it’s a LOT more right than the populist message that was used to pass 99 weeks of UI.

      comparing a vast myriad of different and competing studies is important.

      Hallelujah.

Leave a Reply

Your email address will not be published. Required fields are marked *