September 2022

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30  
Blog powered by Typepad

« Some Wisdom From Mr. Buchanan Late at Night | Main | On The Lighter Side: The Prychitko Stimulus Plan »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Casey Mulligan has been repeatedly pounding the drum on this. The means-tested mortgage modification is also a massive implicit marginal tax.

But you've got it all wrong. BOTH blue lines are wrong. Just imagine how unimaginably awful it would be if there WEREN'T recovery plans. See, the problem is that the blue lines were simply too optimistic. Clearly we need more stimulus. Brad DeLong agrees: http://www.theweek.com/bullpen/column/101511/We_can_afford_a_second_stimulus

Steve,

You said,

"Apparently Congress does not have the intelligence of college first-years. Or does and doesn't care."

I know you well enough to vouch for your sensitivity to the human side of things, so it's weird to see you talk like that. I think you fell (most uncharacteristically!) into economism or some similar trap. Let's leave aside the "doesn't care" point for a moment and assume the good will of Congress. They might reasonably choose to let the unemployment rate stay up a bit in order to cushion the unemployed from the perils of a bad job market. Yes, more benefits means more unemployment. But mostly we're talking about people who are struggling with modest incomes and a weak job market. Killing unemployment benefits will not, of course, magically fix the economy. So you have to trade off mitigating the troubles of the unemployed vs. encouraging unemployment. I think we should err on the side of helping the unemployed and not worrying too much about the unemployment rate. Being unemployed is lousy so it's not that we have folks just lolling around on their yachts collecting their checks

At least you're consistent Roger: you want to extend benefits and not worry about the unemployment rate. My point, not expressed clearly, was really: "You [Congress] cannot simultaneously pat yourselves on the back for extending those benefits and then complain to the hills about how high unemployment is."

I'm with Roger Koppl on this one. On the margin this is sure to increase unemployment or at least extend the length of time that someone stays unemployed, but that is balanced by the fact that nobody WANTS to stay unemployed on a not-so-generous benefits regime.
In Western Europe and Canada unemployment benefits can continue indefinitely, and that is a huge problem. But simply extending the time to a few more months seems to be right trade off to make. Now if they had increased the eligibility requirements, or the amount of money given, that would be way worse than simply extending the timeline I think.

Steve,

Your point about consistency is completely right. Elected representatives are not big on consistency!

But why should we do anything at all or even worry about unemployment? Hasn't Mulligan informed us that all these people are just lazy bums unwilling to work at existing wages?

The benefits could be scheduled to decline after a person has been on them for more than X number of weeks. The problem is that Keynesians (and the public) believe unemployment is INVOLUNTARY which means: you can't get a job by looking harder or adjusting your wage demands,etc. It is deficient aggregate demand that causes the unemployment, according to this view. Another issue is that people might hate the government even more if they had to bear the full consequences of bad policy decisions.

People may not want to be unemployed, but extending benefits lowers the cost of additional job search. The empirical relationship is well established. If empathy drives the policy, there are better ways of addressing unemployment. Bob Litan's wage insurance is one such idea. I like Mario's last point.

Whatever happened to tough love?

This is not the beginning of greater unemployment welfare. Let's not forget that the Federal minimum wage was raised early this year, as well. I come from a country with a myriad of unemployment benefits (Spain), and my country suffers almost 20% unemployment. One thing that has helped those who cannot find a job in the city is the nearly free market in labor in the countryside.

If anybody is interested I have espoused my personal opinion (based my experience in the industry) on the regulation-free agricultural labor market in Spain: http://www.economicthought.net/2009/09/agricultural-employment-in-spain/

Look here:

http://www.nytimes.com/2009/11/07/business/economy/07econ.html?_r=1&hp

So according to an expansive measure of unemployment, it is up to 17.5%. Putting all measurement problems aside -- including comparisons with other recessions, I am amazed. I am amazed that 17.5% can be unemployed and yet not be starving in the street. What is going on?

How many of these unemployed work under the table? How many are young adults who live with their parents? The article mentions "unemployed or underemployed"; what does the latter mean? There has been an increase in homeless people, though. I live in an area of southern San Diego that has never had homeless people roaming its streets. Now, it's not unusual to see one.

Steve:

Don't you teach about frictional, structural, technological and institutional unemployment? Unemployment insurance is a source of institutional unemployment. Raising the benefits or their duration will raise raise the natural unemployment rate. But there are people who remain unemployed after their benefits run out. Tightening up unemployment insurance by reducing the benefits or their duration would lower the unemployment rate.

It is a trade off. More extensive unemployment benefits, higher unemployment but less suffering among those who cannot find quickly.

Of course, if the context of this argument is getting people to adjust the level of wages and prices so the real quantity of money can rise up to meet the demand--well, I think I have a better solution than trading off the adjustment in the real quantity of money with the suffering of those who can't find jobs after an extended period of time.

There was 25% unemploymet in the Depression before there was unemployment insurance, right?

The October dot has in fact been added.

See here: http://tinyurl.com/yfnhcxg

Yup, got it Jeff, thanks! In fact, the new one will make an appearance in my talk at AU Tuesday night.

The comments to this entry are closed.

Our Books