Governor Walker signed “right to work” (RTW) legislation earlier this week, which it is fair to say has led to mixed reactions among the electorate. A Wall Street Journal piecetouts the “right to work advantage,” whereas Slate.comteaser says “It has never been more painful and humiliating to be a Wisconsin Democrat.” Owie.
(Curiously, the sign on the table in the photo is “Freedom to Work,” rather than “Right to Work”).
Right to Work laws generally allow employees to work in unionized workplaces without paying union dues. In principle, the free-rider problems caused by the elimation of compulsory union dues mitigates union bargaining power, hence lowering wages (ceteris paribus), and increasing employment. Clearly, then, this legislation potentially has fundamental implications for employment, wages, output, and probably a whole lot of other stuff. How is one to sort all this out?
The bottom line is that the economic effects of RTW laws are not nearly as clear-cut as their advocates and opponents make them out to be. Correlations of RTW laws with wages and employment are economically small even when they are statistically significant. Most problematic of all is the question of causation—does RTW cause observed differences between states, or do pre-existing differences cause the passage of RTW laws?
It’s almost too bad that the effects are so benign, as RTW is a genuine political dynamo. In response to Walker’s signature, President Obama took to his Twitter feed today to “blast” the Wisconsin legislation and encourage the 25 states that don’t have RTW legislation to keep it that way.
Unemployment insurance programs are often criticized because they encourage various forms of shirking: the unemployed may not try hard enough to look for a new job or may turn down reasonable job offers. Also, the taxes that finance such programs are thought to decrease the labor supply. This talk will look at an alternative way of insuring against unemployment events through personalized unemployment accounts. We will discuss their advantages but also warn against potential pitfalls. The discussion will be backed up by simulations performed on the labor markets of Oregon, Austria and France.
At one level, the story is transparent. You can see over time that the wage premium for people with college degrees and graduate degrees have grown substantially, while high school graduates and dropouts actually seem to be losing ground since the early 1970s. That is the basic message.
The purpose of this post is simply to unpack the elements of this particular characterization of “changes in wages” to give an idea of how some truly great economists have addressed the problem. I hope that you will see why this characterization is likely to be more compelling than the plotting of raw data that we often spattered about these days.
The topic is one that you have probably heard before — “women only make 77 cents for every dollar men make.” Now why would that be? Is it because of discrimination?
Many economists discount the idea that discrimination is the driver, because bigotry is such an expensive vice. Consider the following: Suppose Bigoted Bob’s hires only men and has annual labor costs of $100 million per year. If the difference in male and female earnings is due solely to discrimination, then it should be possible to hire a staff of women who are exactly the same quality and produce exactly the same quality and quantity of output for only $77 million per year. So, it hardly takes benevolence to hire women — simple greed, er, profit maximization will do — the “benevolent” employer can presumably pocket the $23 million in labor savings! In other words, a business that wants to exercise its discriminatory preferences for men over women for whatever reason will have to pay a steep price on the labor market.
So, perhaps it’s some other factors, and this is partly true. If you control for human capital accumulation (education and experience, for example) and industry choice, the gap is less than the largely purported, but there is still a gap of about nine cents on a dollar. In other words, controlling for what we control for, women only make 91 cents for every dollar men make.
Last Friday, the Bureau of Labor Statistics reported that the unemployment had fallen to 9.0%, a sizable drop from December’s 9.4% and November’s 9.8%. The BLS also released the payroll survey which indicated that the number of employed based on the non-farm payroll survey only increased by 36,000 in January. Such a small increase fails to keep up with trend labor force growth which averages about 125,000 per month (based on a civilian labor force of roughly 150 million that grows at about 1% per year.)
So is this good news or bad news? Actually, these two results are largely unrelated to one another. The unemployment rate is derived from a household survey which also revealed that the number of employed people rose by 117,000, that is almost the trend rate. The payroll survey tends to look at relatively older companies, which typically do not drive employment growth. James Hamilton, in a recent Econbrowser piece, lays out some of the relevant details. The chart below highlights key labor market patterns. The relatively low level of labor force participation puts downward pressure on the unemployment rate, which is why that indicator is not particularly informative with regard to the JOBS, JOBS, JOBS agenda. The details will entertain those of you who will take Econ 320 next term.