EDWARD LUCE interviewed Charles Murray over a very expensive lunch for the Financial Times, and the key topic was Mr Murray's contention that the plight of America's working class is due to their cultural decline, rather than external economic factors. Mr Luce suggests that surely some of that cultural decline itself ought to be chalked up to economic shifts. Mr Murray responds:
“OK, let's try this,” he said. “If you get a rising economy, for example, if Barack Obama could say we are going to bring on seven years of incredibly low unemployment, then he would argue that this would do a lot of good to the working class, wouldn't he?” I agree. “But we already had that in the 1990s, and yet the dropout from the labour force continued to go up, people on social disability went up. Divorce went up. We have no evidence that a robust economy has much to do with these problems at all.”
Right. Let's start with labour force participation. It rose from about 66% in 1993 to an all-time high of 67.3% in the first quarter of 2000 (for 15-to-64-year-olds). Men's labour-force participation fell slightly, but women's rose faster. It's very hard to find breakdowns of labour force participation (LFP) by income, but this 2005 paper by the Dallas Fed directly assessed the impact of GDP growth on LFP, which is exactly what Mr Murray is talking about here. It concludes that:
[P]articipation rates are pro-cyclical—positively correlated with economic output—and that the strongest correlation for males and females is between GDP today and participation two and three quarters from today. This supports the contention above that labor force participation decisions respond to changes in economic output with a slight lag.
Most importantly, the report correlates LFP to educational attainment, and finds that the strongest positive correlation is for workers without a high-school degree:
So it actually looks like the 1990s data support the opposite of Mr Murray's thesis: low unemployment and GDP growth have a strong positive effect on labour force participation for the working class, and less effect on high-income earners.
But what about divorce? Well, from 1993 to 2000, the divorce rate fell steadily from 4.6 per 1,000 to 4.1 per 1,000. Again, I'm afraid I can't find data that break this down by income. But here's the most interesting and, I think, most relevant way to put this: Justin Wolfers and Betsey Stevenson re-analyse Census Bureau data and find that as time goes on, people who got married in the 1990s have a significantly lower cumulative likelihood of divorcing than people who got married in the 1980s or people who got married in the 1970s.*
Does this mean economic improvement led to people staying married? Who knows? It's a correlation, not necessarily a causation. But at a minimum, Mr Murray's claim that "divorce went up" as the economy improved in the 1990s is wrong. Moreover, if there is any causal connection, Mr Wolfers and Ms Stevenson's analysis suggests that not only did better income opportunities have positive effects on marriage stability in the 1990s, but that the effect had a lasting impact on people who married in that decade, rendering them less likely to divorce even in the less economically rosy 2000s. Though this doesn't much help Mr Wolfers and Ms Stevenson personally, who live together and have a daughter but, despite having PhDs from Harvard, aren't married. Which really puts them off Mr Murray's map.
That's all the research I can do on Mr Murray's claims here; I don't have time to get to social disability. Maybe he's right on that one. One last note, though: it's really easy to find data on labour-force participation by state, gender and race. Ditto for marital status and divorce. But it's almost impossible to find data on either one by income. The closest you can get is educational attainment, and even that is hard to find. I think this shows just how taboo it's been to talk about class divisions in America until very recently. We're actually more willing to ask the seemingly loaded question of whether blacks are more likely to divorce than whites, than we are to ask whether poor people are more likely to divorce than rich people. I think Mr Murray has made a valuable contribution to the national debate by helping to bring such crucial discussions of class division into the popular arena.
* This post originally used the wrong version of the cumulative divorce rate figure, that is, the very one Mr Wolfers and Ms Stevenson re-examined the data in order to correct. In the adjusted version shown here, their conclusion that the divorce continued to fall from the 80s to the 90s is clearer.
