America's Welfare State fails

[Admiral Yi]

Bubble days. When pretty much everyone who wanted a job got one. As it turns out, few didn't want one, even if they were eligible for benefits (which were better back then as well).

I got that.  What are you comparing the bubble days to, in order to come to the conclusion that UI does not disincentivize job searching?

[Iormlund]

Why would a comparison be needed? The author states that UI provides a disincentive to work so strong that policy should be significantly altered. I simply put forward an example, in real life, where full employment existed despite most of the workforce qualifying for 2 full years of benefits.

[Admiral Yi]

Why would a comparison be needed? The author states that UI provides a disincentive to work so strong that policy should be significantly altered. I simply put forward an example, in real life, where full employment existed despite most of the workforce qualifying for 2 full years of benefits.

Yeah, unfortunately that doesn't work as a rebuttal.  The thesis is not that UI will, under all conditions, keep people unemployed for X amount of time.  Rather it's that UI will keep people unemployed longer, everything else being equal.

[Iormlund]

Or in other words, my argument cannot apply because it is an untestable thesis. Alrighty then.

[Admiral Yi]

It can be tested very easily because it's one data point: unemployment in your town was 3% in the year X.  Your data point doesn't work to refute the thesis because it doesn't account for the other factors that can influence unemployemnt.

[Valmy]

It can be tested very easily because it's one data point: unemployment in your town was 3% in the year X.  Your data point doesn't work to refute the thesis because it doesn't account for the other factors that can influence unemployemnt.

Is there a way we could actually test something like this?  It is hard to find testable and repeatable conditions for economics theories.

[Admiral Yi]

Is there a way we could actually test something like this?  It is hard to find testable and repeatable conditions for economics theories.

Econometrics!

[Zanza]

Rather it's that UI will keep people unemployed longer, everything else being equal.

I wonder how the authors quoted in the article came to that conclusion. It's not like they could empirically test this. Especially not the ceterus paribus part of your argument, that is found nowhere in the articles by the way.

[Zanza]

Is there a way we could actually test something like this?  It is hard to find testable and repeatable conditions for economics theories.

They just looked at an alternative universe where unemployment insurance wasn't instituted, every other policy was exactly the same and compared how long spells of unemployment were. Simple really, you just need to master extra-dimensional observation first.

[Iormlund]

It can be tested very easily because it's one data point: unemployment in your town was 3% in the year X. 

Except all other factors are rarely if ever going to be equal. So there's no way to actually test this unless, as Zanza said, you can access alternate universes.

Your data point doesn't work to refute the thesis because it doesn't account for the other factors that can influence unemployemnt.

It puts limits on things. It shows that this particular factor could account, at most, for a minute part of the workforce refusing to search for jobs. And that's if all other explanations for that jobless 3% are assumed to be insignificant, which is rather unlikely since it includes things such as an overwhelming prevalence of home ownership instead of renting, people in transition between jobs, the presence of long-term unemployed without access to benefits or the disparity of demanded and available skills.

And yet this problem is so big it seems to be hurting the poor.

[Admiral Yi]

Econometrics. 

[Jacob]

Econometrics. 

What about them?

[Iormlund]

This is what springs to mind when I hear about an economist predicting the future:

[Zanza]

Econometrics. 

I found the paper the quote from Feinstein and Altman is from (http://www.nber.org/chapters/c0046.pdf). There is no econometric substantiation of the claim in that paper. I read the abstracts of the three papers they cite there to corroborate their thesis. These seem to indeed contain econometric analysis, but I cannot read them due to paywalls. However, none of the papers seems to analyze the pure existence of an unemployment insurance, merely the effect on when people search for a new job.

[Jacob]

Why does searching for jobs faster help the poor?