OMNILIBRIUM
  Rational Discussion of Controversial Topics


GO TO THE MAIN THREAD Sort By:


Barry 3 July 2015 12:53 PM
72%

1a) If one study finds "MW raises unemployment," and another study finds "no evidence that MW raises unemployment," then - all else held equal - we should not make policy decisions based on the belief that the MW raises unemployment. A finding that is not consistently supported by the best empirical studies, is not a reliable finding, and therefore not a good basis for policy.

1b) However, the paper you link to is a literature review paper, not a meta-analysis. There's a big difference. The meta-analysis I linked has MUCH more statistical power than any of the papers cited in the literature review, and therefore should be given more weight.

1c) The meta-analysis I linked to used much better methodology than most of the studies emphasized in the paper you linked.

By and large, the papers said to be "more reliable" in your review paper are making state-level comparisons: So if (for example) Oregon raises MW but Alabama does not, and if Oregon's unemployment rises afterwards, that is said to show that the minimum wage caused an increase in unemployment.

The problem with that is, the differences between Alabama and Oregon's economies are too numerous and difficult to quantify to be controlled for. In other words, garbage in, garbage out. What's needed is a comparison between two areas are are extremely similar other than having different minimum wages.

Better studies, including the meta-analysis I cited, look at two economically similar counties, located next to each other but in different states. Because neighboring counties will have much more in common than (say) Oregon and Alabama, this methodology is much less subject to spurious correlations.

1d) Evidence trumps intuition.

1e) And although "generally" increasing price reduces demand, there are plenty of exceptions to this - for instance, if demand is inelastic, or if the quality ... read more

stars0
Reply


melian 3 July 2015 07:22 PM
72%

1a) “If one study finds "MW raises unemployment," and another study finds "no evidence that MW raises unemployment," then - all else held equal - we should not make policy decisions based on the belief that the MW raises unemployment.”

But does not it work both ways (i.e. we should also not make policy decisions based on the opposite belief)?

1c) “What's needed is a comparison between two areas are are extremely similar other than having different minimum wages.”

This makes a lot of sense. However, doing this creates another problem. Comparisons between different states and countries are very straightforward and easily verifiable. Trying to select two extremely similar areas is neither. A researcher would have a lot of freedom to decide which parameters can be neglected. Given a very partisan nature of the question, it is almost certain that a conservative economist will “play” with the data to prove that MW raises unemployment. The same can be expected of a liberal economist.

1e) I'm not sure those two particular exceptions apply to MW.

If the quality of the supply is improved by increased price (i.e. paying employees more results in much better service and increases demand despite raised prices) then I would expect employers to pay higher wages without being required to do so by law.
Regarding elasticity – demand may indeed be pretty rigid in some sectors, but how can it be rigid in all of them?


stars0
Reply


Fwiffo 17 July 2015 07:35 AM
55%

for 1e) the axioms of the theory under which the result is derived is known to be often violated and thus the result is not valid in those situations. We have increased reason to acertain that the thoery indeed applies in this case or if we can't do that in effect calcualte the probaility that the axioms are not sufficiently disturbed. Even if the reason for the disruption would be any previously catalogued one we can still kinda expect one migth be lurking around. Contrast this with CP- or CPT-symmetry breaking.

One plausible story would be that emploerys have to use employees in more demanding jobs ie that the new improved results are not from doing the same work with better quality but genuinely different work. This is like lifting too light weigths. If you just weight up you migth get better returns on your time at the gym which you might not have gotten if you did the thing with the light weight with better technique.

If employers can't capture more worth from "heavy duty" than from "light duty" work they might be pretty indifferent to the total efficiency. That is if you are making the same profits does it matter if your business is small or large? However the societal impact of people getting more bang for their buck is interesting even if the employer doesn't have any incentive to do that but on the flip side he doesn't have anything to object for either.

stars0
Reply


Barry 3 July 2015 12:59 PM
69%

Two related links - both are peer-reviewed research on this subject. (Also, both are pdf links, so don't click if you hate pdf.)

Minimum Wage Effects Across State Borders: Estimates Using Contiguous Counties Arindrajit Dube

Do Minimum Wages Really Reduce Teen Employment? Accounting for Heterogeneity and Selectivity in State Panel Data Sylvia Allegret

stars0
Reply