How to Lie With Statistics, Women and Child Care Edition

My latest post is up at Forbes, highlighting two research papers that look at the impact women’s earnings and the cost of child care have on women’s decisions on whether to have children and whether (or how much) to work. They are good papers, but they both make a critical error:

Both papers assume that men commit full-time to the labor force, and that the choices families are about the balance of women’s working hours and caring hours. It’s one of the most infuriating aspects of the work-life debates that the choice is so often framed that way. The reality is that in addition to earning potential and cost of child care, the degree to which male partners share in child care duties is a major factor driving women’s career and family choices.

Leaving working fathers out of the choice equation tarnishes the studies’ results, and can have a dangerous effect, if policymakers feel that the solution suggested by papers like these is to expand the choices available to women without expanding choices for men. Framing the work-life conundrum as a women’s issue only makes it more likely that it will remain women’s burden. The research error becomes self-fulfilling.

This case is a perfect example of the problem outlined by Darrell Huff in his classic book, How to Lie With Statistics. I’m a great advocate for inserting more data into debates about work and family, but it’s equally important to be skeptical of the data presented to us. Ask not just, ‘Does this data answer the question we’re asking?’ but also, ‘Are we asking the right questions?’ At the moment, I’m not convinced we are.

Read it all here.

Are Men Threatened By Women At Work?

Some of them are, according to a new study I’ve written up at Forbes.

Researchers at Harvard and UNC-Chapel Hill surveyed men in three kinds of marriages: traditional (wives who don’t work), neo-traditional (wives working part-time) or modern (wives working full-time). And they found that the more traditional a man’s marriage, the harder he was likely to be on the women he works with.

There is an obvious reason for this: that men who live in traditional marriages are more likely to have more traditional worldviews overall and less likely to have been exposed to feminist or gender-egalitarian ideas.

The more interesting suggestion is that these men are acting out of self-interest. We know that the earnings premium for married men is highest for those whose wives don’t work outside the home, and instead provide supportive labor in the home that enables their husbands to be better employees.

And so the authors of this paper suggest that men with stay-at-home wives are enforcing in the workplace an order that they know benefits them personally, seeing the women who work for them as proxies for what their wives could become. The values these men express – that women aren’t competent at their jobs, that marriages work better when women stay home– are actually rationalizations for a self-interested reaction to a perceived threat.

Read the whole post here.

Baseball and the Marriage Premium

At Foreign Exchange:

In honor of the World Series, which starts tonight*, I dug out a research paper I’ve been sitting on for a while.

The paper, ‘Productivity, Wages, and Marriage: The Case of Major League Baseball‘ looks at the wages of baseball players and identifies a 16% gap between the wages of married players and unmarried players.

Economists have been documenting the marriage premium – the income boost (anywhere from 10 to 40 percent) married men have over their unmarried counterparts – for decades. But researchers have historically gotten stuck when it comes to providing explanations for the phenomenon: Do married men perform better because their wives are doing more of the housework? Do married men perform better because women tend to marry high performers anyway? Do married men perform about the same, but employers discriminate in their favor because they come across as reliable?  Without data about productivity, it’s hard to say.

That’s what makes the baseball study distinctive: baseball is a geek’s sport, filled with statistics, and – in a post-Moneyball world – increasingly managed by the numbers. That allowed the paper’s authors, Francesca Cornaglia and Naomi E. Feldman, to control for productivity (using both Batting Average and On-Base Plus Slugging) and sorted players into groups by age (early and late career) and ability (low, medium, or high performers). They were not only able to show that there is a marriage premium, but able to test the prevailing theories about why it exists.

Go read.

Economists to America: “Honey, we broke the calculator!”

Once a month, the US government locks a bunch of Econ PhDs in a room to calibrate changes in employment, payroll, production and consumption. Today, the experts announced the results of their November gathering. Guess what? We’re in a recession, and we’ve been there since December 2007.

Well, duh. Americans have been feeling the hit for months.

But economists usually wait around for two consecutive quarters of negative GDP growth before calling a recession. That definition rests on the fact that production, consumption and income are interlinked: what a country makes must add up to what it buys. In theory, when GDP is rising, it should be because consumption (roughly 70% of the domestic product figure) is rising too. And when people consume more, it should be because their employers are making more and passing it on.

That means that most of the time, the line about the rich getting richer while the poor get poorer is just gas. Income disparity is growing, but that usually reflects the rich getting richer faster than the poor are emerging from poverty. Assuming that income, consumption and production all move in the same direction justifies A) the 2-quarter definition of a recession and B) the argument that getting people to make and buy more stuff must implicitly raise our standard of living.

That’s apparently no longer the case. Read the rest of this entry »