Are 50% of college graduates unemployed or underemployed?

House Republicans yesterday held a “press conference” (less than 11 minutes) in which I heard two crazy statistics. Each quote is paired with the unidentified Republican Congresswoman who said it:

Recent college grads still are having a very difficult time finding a job. In fact their unemployment is nearly 25%.

Today 50% of college graduates can’t find a job or are underemployed, that’s one in every two graduates.

The first one is just completely wrong. The second one may be just completely misleading (except insofar as 50% is one in two).

I doubt it, but it’s possible they were referring to this paper by Thomas Spreen in the February Monthly Labor Review. Spreen used data from the 2007-2011 October CPS surveys. That’s the month the CPS collects education information in detail, and he calculated the unemployment rates for people who had graduated colleges in the same calendar year. The rates were very high, and did at one point – for men only, in 2009 – reach more than 25%, as shown here:

By 2011 it fell to 16% for men and 11% for women. We should interpret these number cautiously, however, as they are based on quite small samples. According to Spreen, in 2011 the October CPS only included 440 people ages 20-29 who completed their BA degrees that year. Figure about 200 of them were men, and you’re talking about roughly 30 unemployed male recent college graduates. Granted 2009 was the worst year so a spike is plausible, you still need to put a pretty wide confidence interval around that number.

Anyway, because the Republicans used this phrase “or are underemployed,” which is not in the Spreen paper, I suspect the source of these talking points was this 13-month-old AP story, titled “Half of recent college grads underemployed or jobless, analysis says,” or some other version of it. “Underemployed” here means working at a job that doesn’t require a college degree. The number “unemployed” is not given. Those are two pretty different things and should probably never be combined. But Jordan Weissman at The Atlantic, trying to read between the lines, wrote,

Unfortunately, I don’t have all of the data the AP was working with. But their analysis implies that about a quarter of the post-collegiate population is outright unemployed.

That’s not crazy if you were writing about just men for 2009 (and remember most college graduates are women), but Weissman was writing in 2012 about 2011. He might not have all the data the AP had, but he – and you – have what we need to check unemployment rates using the IPUMS CPS data extractor. That will give you March CPS survey data (not the October survey, which identified graduates in the last year, but good for ballparking). It’s pretty easy:

Choose “Analyze data online,” then “Analyze all March samples 1962-2012,” then fill out your table request. Based on the definition given of “recent” college graduates as people under age 25 with a BA, this is what I did:

That gives me employment status, by sex, for years 1993-2012, among people with BA degree (no more, no less), age 15-24 (hardly any are under 20), who are not currently attending school. Here are the percentages unemployed from that:

Okay, so nowhere near 25% unemployed. The worst it was for men was 10.9% in 2012, for women 6.4%. (And note these are based on samples of more than 500 men and 800 women in recent years.) Shockingly high unemployment for college graduates, of course. And it’s interesting that it’s higher for those who graduated within the last few months (what the MLR paper showed) than it is for those who graduated sometime within the last few years (the under-25 grouping I used).

The underemployment thing may be important, but there’s not enough information here to evaluate it.

Listening to the press conference on the radio, I naïvely expected one of the reporters to ask, “Excuse me, did you just say the unemployment rate for recent college graduates is 25%?”

Anyway, thanks for listening.

That economists’ paper about gender inequality, marriage and divorce

I was planning to write a note about this paper by economists, “Gender identity and relative income within households,” which got a lot of play two weeks ago. But I forgot about it until today, and then noticed that in the New York Times Catherine Rampbell, economics writer, dropped it in her story on the Pew Report about women as breadwinners. In the cautionary part of the article, Rampbell mentioned “A new working paper by economists” that showed:

…perhaps even more tellingly, couples in which the wife earns more report less satisfaction with their marriage and higher rates of divorce.

Maybe reporters like what’s new, or maybe it was just on her radar because she reads Freakonomics, the Economist or the Financial Times, which all uncritically wrote up the paper when it came out. But it’s really a shame in a story about current trends to cite a “new” paper which (for this part of its analysis) used data more than 20 years old.
Anyways

Here is a brief critique I was going to give when the paper came out. Just taking two lines from the abstract, I offer a few suggestions:

1. Couple matching

The distribution of the share of household income earned by the wife exhibits a sharp cliff at 0.5, which suggests that a couple is less willing to match if her income exceeds his.

Suggestion: It’s not a good idea to use the relative incomes within couples years after they got married to discuss how relative income affects mate choice decisions. People move, change jobs, have children, etc., in the first few years after they get married. You need to look at income before marriage to study mate selection.

2. Divorce

Couples where the wife earns more than the husband are less satisfied with their marriage and are more likely to divorce.

This part of the analysis uses data from Waves 1 & 2 of the National Survey of Families and Households NSFH), which were collected in 1987-88 and 1992-94. I don’t always insist that everyone use data from this minute, but at some point — around two decades — a study becomes historical. That judgment depends on the context and the question being asked. In this case, relative earnings of spouses (as we just saw in the Pew report) has seen an order-of-magnitude change over this period. And the paper is about norms! That is, the authors speculate that couples with high-earning wives divorce because they are outside the mainstream. So if, 20-25 years later, they’re not outside the mainstream anymore, the paper might not be salient.

Secondly, this is well-worn territory, and the specific hypothesis offered here has been tested and found wanting in several award winning papers using more thorough measures and testing competing hypotheses. (The NSFH, one of the most productive data collection efforts ever, maintained a bibliography up to 2004, which lists 180 papers under the category “union quality and stability.”) For those interested in the fuller story, I recommend these:

• “Women’s Economic Independence and the Probability of Divorce A Review and Reexamination” (2001), by Liana Sayer and Suzanne Bianchi (winner of the Reuben Hill Award from the National Council on Family Relations for best article of the year on family issues). Their critique of the research existing prior to 2001 applies specifically to the 2013 Bertrand et al. paper:

…[M]easures of marital commitment and satisfaction are better predictors of marital dissolution than measures of economic independence. This strongly suggests that the independence effect found in prior research, which did not include controls for marital quality, may have been measuring the role of wives’ economic independence in exiting bad marriages, not in exiting all marriages.

• “She Left, He Left: How Employment and Satisfaction Affect Men’s and Women’s Decisions to Leave Marriages” (2011), by Liana C. Sayer, Paula England, Paul Allison, and Nicole Kangas (winner of the Rosabeth Moss Kanter award from the Purdue Center for Families and the Boston College Center for Work & Family, for the best work-family research article of the year). Their key finding is that employment within couples does matter for divorce, but it depends on marital satisfaction:

We find that when men are not employed, either husbands or wives are more likely to leave. When wives report better than average marital satisfaction, their employment affects neither their nor their husbands’ exits. However, when wives report below average marital satisfaction, their employment makes it more likely that they will leave.

• “Wives’ Employment and Spouses’ Marital Happiness,” by Robert Schoen, Stacy Rogers and Paul Amato (2006). They report that employment effects on divorce depend on marital happiness. Specifically:

…shifting into full-time employment is more likely for unhappily married than for happily married wives. … [C]ontrary to frequently invoked social and economic theories, wives’ full-time employment is associated with greater marital stability.

This provides a followup to a previous study using the same data which found…

…clear evidence that, at the individual level, women’s employment does not destabilize happy marriages but increases the risk of disruption in unhappy marriages.

The reason these marital satisfaction controls matter so much is that how happy women are within marriage affects their employment, and therefore their earnings. So what looks like an earnings effect is often an unhappy-marriage effect. Careful sequencing of longitudinal data (which these papers do) is required to sort this out.

I only mention the awards because I was shocked (shocked!) to see these major sociology papers in top journals, using the same dataset and asking the same questions, published over a decade, which have been cited hundreds of times in the academic literature, go unnoticed in this economics working paper, which — not-yet published, not-yet peer reviewed — would be quoted all over the place.

Married women learning that paid work pays

The economist Raquel Fernández has a new paper out called, “Cultural Change as Learning: The Evolution of Female Labor Force Participation over a Century” (published version here, free version here). If I understand it, though, “female” labor force participation really only refers to married women. Correct me if you understand this better than I do and I’m wrong, but I think that’s a problem for the theory.

The basic point is that married women learn from the experience of others, producing a generational change in employment rates as positive experiences transmit to younger cohorts. As it became more culturally acceptable for married women to have jobs, the cultural effect accelerated, but it reached a saturation point resulting in the stalled progress toward higher employment rates among (married) women. Here are the trends she uses:

The normative survey question she relies on is about whether it’s OK for a woman to work “if her husband can support her.” The S-shape of labor force participation rates is supposed to be consistent with the cultural transmission theory (rather than being caused by, for example, anemic work-family policy, anti-feminist backlash, or hollow anti-discrimination enforcement).

But I don’t see anything in the paper about increasing non-marriage (now about twice as common as in 1960), or about labor force participation rates for single women. Shouldn’t economists be concerned about that kind of selection issue? In fact, labor force participation rates for single women have stalled, too, as my figure shows:

I don’t think attitudes toward married women’s work — or anything about marriage alone — are going bear the burden of explaining two decades of stalled progress into the labor force for both single and married women. I’m happy to have cultural explanations as part of the mix here, but I don’t think this one will do it.

 

 

Gender wage gap, 2012 edition

Gender inequality stagnation continues apace.

The Bureau of Labor Statistics has released the wage report for 2012, which shows women’s earnings relative to men’s falling back to the 2005 level. The gender breakdown is available here (the content at that link changes when new data come out), and the historical series from 1979 to 2011 is available here.

The usually-reported number is the median weekly earnings of full-time wage and salary workers. These are the gender ratios (women’s earnings divided by men’s):

 

Follow the gender inequality tag for updates and previous posts.

Quick book review: The Price of Inequality

The Price of Inequality: How Today’s Divided Society Endangers Our Future, by Joseph E. Stiglitz (W. W. Norton, 2012)

My economics training as a sociologist — with a background in American Culture studies — has been spotty and roundabout. I got a healthy dose of Marxist economics in college, and then some feminist economics, a little human capital theory and some dated econometrics in grad school and since.

All that made reading made it interesting, and also frustrating, to read The Price of Inequality, by Joseph Stiglitz – a winner of the Nobel Prize for economics and an “insanely great economist,” according to Paul Krugman.

On the plus side, I am glad to see someone within mainstream economic theory freely discussing all the ways that common assumptions simply do not predominate in the modern economic scene. Especially helpful in this category is his discussion of how “rents” accumulate vast resources at the upper end of the income distribution, with perverse effects on economic development and politics alike. At the very top — in the finance sector especially, but also in energy and big manufacturing — there is nothing like free-market competition. And the beneficiaries of those distortions are the most powerful players in the economy and political system.

It is refreshing to see this concentration of wealth described as waste and distortion, as their vast profits provide little gain to anyone else. In fact, dumping vast wealth on the 1% creates a drag on the macroeconomy while fueling the historic run-up in economic inequality. This is all very timely and takes you right through the financial crisis up to early 2012.

So if you want to understand from an economic perspective how “the market” in America isn’t the way it’s supposed to be, this book may be for you.

Top 1% income shares, including capital gains, for the U.S. and Sweden. From the World Top Incomes Database.

The other good thing about the book for many readers will be its cogent and comprehensive economic rationale for the liberal reforms that many of you probably supported already. Stiglitz makes the case that a suite of reforms – an agenda Rachel Maddow, Elizabeth Warren and Robert Reich probably agree on – would, by (directly or indirectly) increasing taxes (or reducing subsidies) on the wealthy and redistributing wealth downward, reduce the federal debt, increase economic growth, and reduce economic inequality all at the same time.

Round numbers: if the richest 1% earn about 20% of all income, then taxing them another 10% would generate government revenue equivalent to 2% of GDP. (And it wouldn’t hurt anything, since they just hoard or waste their extra cash anyway rather than “creating jobs” with it, and they’re so greedy they wouldn’t be discouraged by the disincentive effect of higher taxes.) That’s an amount of money that could actually be useful for poor people.

The frustration I feel reading the book is more amorphous. I think there have to be better ways of describing this whole system than using the language of mainstream economics, which ends up painting a picture of an entire system that does not work according to the rules as imagined. Concepts like power, social class, social networks, elites and reification do not figure heavily in this story. In fact, Stiglitz’s apparent ignorance of sociology is sometimes funny as in this passage:

Social sciences like economics differ from the hard sciences in that beliefs affect reality: beliefs about how atoms behave don’t affect how Adams actually behave, but beliefs about how the economic system functions affect how it actually functions. George Soros, the great financier, has referred to this phenomenon has “reflexivity,” and his understanding of it may have contributed to his success.

I guess after what people like me have made of econometrics it’s only fair that economists would attribute the idea of reflexivity to Soros. (The discussion of reflexivity in Anthony Giddens’s book The Consequences of Modernity is very approachable.)

Anyway, the book is easy to read and informative, and has lots of footnotes and references.

Googling racism, votes for Obama, and population composition

This post contains racially offensive language.

Seth Stephens-Davidowitz, a PhD student in economics at Harvard, has analyzed Google searches for racially offensive terms across metro areas, and tested for a “racial animus” effect on the vote for Obama in 2008.* The results are pretty strong:

 The baseline proxy that I use is the percentage of an area’s total Google searches from 2004-2007 that included the word “nigger” or “niggers.” … A one standard deviation increase in an area’s racially charged search is associated with a 1.5 percentage point decrease in Barack Obama’s vote share, controlling for John Kerry’s vote share. The results imply that, relative to the area with the lowest racial animus, racial animus cost Obama between 3 to 5 percentage points of the national popular vote. … The statistical significance and large magnitude are robust to numerous controls including local unemployment rates; home state candidate preference; Census division fixed effects; changes in House voting over the same period; prior trends in Presidential voting; and a variety of demographics controls.

This is a creative way to measure racism — not perfect, but nothing is. And he did a fair amount of experimenting and tinkering with the measures to make sure it wasn’t fluky. Very nice.

Racism at the population level

Another thing that jumped out at me in the paper, however, was the finding among the control variables that racist searches are more common in markets with higher proportions of Black residents. This raises a potentially difficult issue with the whole Google-search method, since we don’t know who is doing the searching. Does his finding suggest that Blacks are doing racist searches? I don’t think so. I previously looked at state-level correlations between race/ethnic composition and search terms, and it looks to me like the most correlated search terms are indeed being performed by those groups. For example, Americans Indians live in states where people Google “Indian Health Service” and Blacks live in states where people Google stuff about historically Black colleges and universities (and Whites apparently Google AC/DC songs).

But at lower levels of correlation, I would expect the presence of one group to affect the search behavior of others. An obvious example is how Southern states mostly vote Republican in national elections — more Blacks equals more conservative voting, even though the great majority of Black voters vote Democratic. The higher rates of conservatism among Whites in those places outweighs the presence of Democratic-voting Blacks. (The effect on Whites was discovered before Blacks could vote in the South, but remains true.)

We also know from way back that inequality between Blacks and Whites is greater where Blacks are more highly represented in the population, and there’s good evidence at least some of this is due to increased racism by Whites. I’ve found this for earnings for both men and women, for middle and working class workers; and, with Matt Huffman, for occupational segregation and access to managerial positions. Only some of that research has actually measured racial attitudes, however. Google gives us a chance to look from a different angle — at the private behavior, not expressed attitudes, of populations.

Here’s one take, jumping off from Stephens-Davidowitz’s paper: searches for “nigger jokes.” This seems like something Blacks are unlikely to be looking for on Google.** But the searches are more common in states with larger Black populations:

Removing West Virginia, which is an extreme outlier on the jokes variable (more than 3 standard deviations from the mean), the correlation between searches for “nigger jokes” and Black population percentage is .48. Here’s the scatter plot (the non-Southern states have the pink centers).

And here’s the regression numbers for the relationship:

That positive relationship, tapering off, fits the long-standing pattern, as seen for example in this 1998 paper, which tested the percent-Black on common attitude measures in the General Social Survey (the figure estimates are net of a variety of controls):

All adding to the accumulating evidence for search behavior as a valuable research tool.

* Thanks to a tip from Rachel Lovell.

** Some searches seem even better for this purpose, such as “funny nigger jokes,” but fortunately there isn’t enough searching for that to get state-level frequencies, according to Google.

Twin peaks of intergenerational mobility

There is a lot of news about economic mobility from recent weeks. Some of it draws from Pew’s Economic Mobility Project. Not as recently, there was an excellent review and analysis by Emily Beller and Michael Hout in the Future of Children a few years ago. In between, I somehow missed a collection of economic analyses in a book titled, Unequal Chances: Family Background and Economic Success, edited by Samuel Bowles, Herbert Gintis, & Melissa Osborne Groves.

The first chapter is posted free, and it includes a good introduction to the statistical and conceptual issues that arise when trying to understand patterns of mobility across generations. It includes a discussion of heritability, genetics, IQ and the like, which is quite approachable to the reader who is ready to think about decomposing correlations.

One good example regarding genetic heritability of traits that determine income: race in South Africa, which is almost entirely inherited (since there’s very little interracial marriage) and has a huge effect on income, but the effect of which is still social/environmental, not “natural.”

Anyway, I like this “twin peaks” figure, which shows the relationship between parent and child family income decile:

Probability of offspring attaining given income decile, by parents’ income deciles, United States. Based on total family income for black and white participants in the Panel Study of Income Dynamics who were born between 1942 and 1972, and their parents. The income of the children was measured when they were aged 26 or older, and was averaged over all such years for which it was observed. The number of years of income data ranged from 1 to 29 with an average of 11.5; the median year of observation was 1991. Parents’ income was averaged over all observed years in which the child lived with the parents. The number of years of income data ranged from 1 to 27 with an average of 11.9; the median year of observation was 1974. The simple age adjusted correlation of parents’ and children’s incomes in the data set represented in the figure is 0.42.

So, 30% of children from the top decile stay there (point D), 32% of children from the bottom decile stay there (C), while the odds of making it from the top to the bottom, or vice versa, are both less than 2% (A and B).

There is a nice symmetry to the figure, but it’s important to know that what’s happening up and down the distribution is highly varied, according to the analyses in the book. For example, at the top there is a lot of transmitted wealth. At the bottom there are a lot of health crises and premature deaths, including from violence. And the bottom is much stickier for Black children than for Whites.

Marriage promotion, farce or fraud?

Sometimes I wonder if the National Marriage Project is just gaslighting me.

Did you know that married couples with children spend 50-times more on childcare than single adults without children? Well, if you didn’t you might not realize how good marriage is for “the economy.”

The other day I discussed a bizarre essay and the naive NPR report promoting it. I saved for today the additional contribution in the same Bradley Foundation-funded document by W. Bradford Wilcox and Kathryn Sharpe which claims that, “Strong, sustainable families pay long-term dividends to the entire economy.”

Unsurprisingly, the article provides no information about “strong, sustainable families” or “long-term dividends.” What it does is produce a simple table showing that married-couple-with-children households spend more on various things than single-childless households. If you’re thinking, “but there are more people in married-couple-with-children households,” then you may  already have done more thinking than the report’s authors.

To explain why this spending pattern occurs, they offer several reasons, the first of which is “household size.” Wait — you’re still thinking — if household size explains the difference in spending, then it’s not a difference in spending, it’s a difference in accounting, just pooling the spending of several people and calling it a spending increase. So how does this help “the economy”? Believe it or not, this is their reasoning:

To serve the needs of all the adults and children in their homes, they are more likely to buy many brands in bulk, from Bounty to Tide, and to fill their shopping carts at the local grocery store.

I must be doing something wrong, because I thought I spent less in the end when I bought in bulk.

The data abused in this report are from the Consumer Expenditure Survey, which is the authoritative source on household spending in the U.S., and used to calculate inflation. It’s something I’ve used before to study household spending (here and here). And if you use it, all I can say in a sentence is: you better account for household size, since all the spending is reported for households, not individuals.

To illustrate this, I took the data from their little table on (a) married-with-children households, and compared it with the sum of the spending for (b) childless-singles plus the (c) single parent households. If every adult needs one beer a day, and every child needs one glass of milk, then the level of spending in (a) and (b+c) should be the same, if they have the same number of children. This is not good science, but it’s appropriate for a blog-scale debunking. And the results:

This graph is for weekly expenses on small consumer items;* the graph for bigger ticket items looks about the same. If the dots fell along the dotted line, my beer-and-milk hypothesis would be supported. It’s pretty close — but tipping a little the way you would expect it to — toward bigger households spending less, since they have economies of scale (“buying in bulk”).

Anyway, the numbers are all junk. The more interesting question is: Is this farce or fraud? Maybe they really don’t know what they’re doing, in which case the foundation funding makes it a farce. Or maybe it’s fraud.** Maybe they are deliberately misleading the public, the foundation and the major corporations they are hoping will spend their “philanthropy” money on such “public education” projects, their actual recommendations!

Companies whose fortunes are linked to the health of the family, such as Procter & Gamble, spend billions of dollars each year on advertising. … Executives with oversight across brands should ask themselves a simple question: Do the messages used in our advertising make family life look attractive? Or do they exalt single living? Obviously, it’s in their long-term interest to do more of the former.

If you have another 3 minutes, consider watching this hilarious video they link to as an example of “family life = attractive.” It’s from Proctor & Gamble’s 75th anniversary in the Philippines (unless it’s a spoof, too), which includes images like this:

* cereal, baked goods, beef, pork, other meat, poultry, seafood, eggs, dairy, fresh fruit, fresh veggies, processed fruit, processed veggies, sweets, non-alch bevs, oils, misc. food, alchohol, tobacco, personal care products and services, and household products and services.

** colloquial use of the term “fraud” in this blogosphere context is not meant to express or imply legally criminal fraud.

How much have you got to spend?

Kids cost more for parents who have more.

The U.S. Department of Agriculture has released its annual report on the cost of raising children. Using data from the Consumer Expenditure Survey, they estimate the costs for one year, then extrapolate out to 17 years with inflation (thus not including college and beyond). The methods are complicated, involving the costs of an extra room, economies of scale, food needs, and so on. It’s not just an exercise — the numbers are meant to be used as a guide for child support, foster care and education program budgets.

But the costs depend on how much money the parents have. That’s natural in a descriptive sense, but should it also apply prescriptively? In other words, are the needs of children really determined by the wealth of the adults who care for them, so their wellbeing is relative? In further words, who is children’s wellbeing for?

The report breaks families into married versus single, and then breaks the married-couple families into three pre-tax income groups. Here’s how the 17-year projections cost out.

I sorted the categories by something like their elasticity, that is, the ratio of rich-parent to poor-parent spending. So, food spending varies the least, so I put it on the bottom. Childcare/education spending varies the most — which is mostly because higher-income families have wives who work and earn more, and they spend more on childcare — so I put it on the top. Miscellaneous includes your video games, entertainment and so on.

Consider housing: rich parents spend more than twice what poor parents spend to add a room per kid (and in the model kids all get their own room). This is morally OK in a world where parents have a right to produce children of their own social status. But where does that leave the rights of children?

Canadian inequality squashed

In Canada, like in the U.S., there has been an increase in inequality over the last few decades, as those with better jobs have pulled ahead of those with worse jobs — partly because the value of education has increased, increasingly separating those with more from those with less. This all according to a recent article in the Review of Economic Dynamics by Brzozowski and colleagues.
As a result, the Gini index for family earnings, which measures inequality on a scale of 0 to 1, increased from about .31 to about .38 over the last 30 years.

Some of the trend in Canada, like in the U.S., is from higher-earning people increasingly marrying each other, too. However, more than in the U.S., the tax and transfer policies of the Canadian government have squashed that run-up in inequality. In this figure, the top line is the amount of family inequality in pre-government income, and the bottom line shows the inequality in disposable income.

The authors describe it like this:

Not only does disposable income exhibit much less inequality than pre-government income, but the degree of inequality is also much less variable than that of pre-government income. This suggests both that Canadian policy has been and remains redistributive, and that it smooths cyclical shocks to pre-tax income inequality.

Gini estimates differ. The CIA World Factbook has a list for all countries that puts Canada at .32, with the U.S. at .45. I don’t have a directly comparable estimate of what policy does to inequality in the U.S., but the Census Bureau came close with a report on 2005 inequality, which found that government transfers and taxes reduced the household Gini from .450 to .418. So that’s something of a comparison.