Pew on Marriage

Via the Facebook page of the National Sexuality Resource Center, I found a report from the Pew Research Center on marriage and how wives are rising (whatever that’s supposed to mean). Reading the first paragraph, I decided that I needed to change my evening plans and respond to this.

The institution of marriage has undergone significant changes in recent decades as women have outpaced men in education and earnings growth. These unequal gains have been accompanied by gender role reversals in both the spousal characteristics and the economic benefits of marriage.

The first part of the first sentence is correct: The institution of marriage has undergone significant changes. They have been very well documented by historians like Stephanie Coontz. But these changes were not driven by the outpacing of women’s eduction and earnings. These changes go back a bit further to the mid 1700s when love entered as the primary reason for marriage. Most of the changes in the last century were driven by changes in the laws that, first, set husbands and wives on equal footing (prior to that wives were covered by the husband) and, then, by changes in divorce law.

The rest of the paragraph suggests that women have made big strides in income. The wage gap between men and women has primarily closed because men’s earnings have stagnated or went down. Additionally, the Institute for Women’s Policy Research reports that the closing of the earnings gap has “slowed considerably since 1990.” Hardly something to celebrate.

Pew compares data from 1970 and 2007. In 1970, 4% of husbands were married to women with higher income. In 2007 that percentage had risen to 22%. Although these numbers are based on incomes adjusted for inflation, they should also be adjusted for labor force participation rate and the wage gap. In 1970, a woman earned less than 60 cents for every dollar a man earned; by 2007 that gap had narrowed to 78%. Or put differently – and this is based on a table in the full Pew report on page 8 – in 1970, a women who graduated college barely made more than a man with less than a high school degree (about $2,500 more). By 2007, a female college graduate made slightly more than a man with some college (less than $4,000 more; whereas a male college graduate made more than $20,000 more than a woman with the same level of education). Hardly something to celebrate.

Also, did they adjust for the change in age at first marriage? Pew is looking at 30- to 44-year-olds. In 1970, this age group was more likely to be married than their 2007 counterparts, especially the women.

But wait, here’s another interesting paragraph (from page 10 – as I was trying to get to Appendix B):

For married adults with some college education or married adults with a high school diploma, men’s median household incomes also grew more than women’s from 1970 to 2007. Recall that during this same period, median earnings of men in these education groups declined, while those of women in those education groups grew.

Huh? Let’s unpack this a bit. The first thing we need to find out: Exactly what is a household? Pew used IPUMS data, which is based on Census data. According to an IPUMS report, households can be made up of married couples and people living alone. And on page 7 of the Pew report is this note: “’Household income’ refers to household income adjusted for the number of members in the household.” Appendix B (page 35) reveals that this is not a simple calculation since there are economies of scale. Okay, fine. But a household can still consist of one or more people. Are we comparing household income to individual earnings? If so, maybe the household adjustment factor skews the data… (It’s not clear to me if the earnings are also based on a household).

What about income vs. earnings? Okay, I got tired of trying to find definitions on the IPUMS site and went to the Census bureau instead. According to definitions that were a bit easier to find, earnings are what you & I get paid on the job (see page 53). Income is earnings plus things like interest, dividends and public assistance. So, income is likely to be larger than earnings for a person who has at least some money in the bank in an interest bearing account. Maybe men made up the gap between earnings and income with investments? Whereas women relied more on earnings? And note that the whole wives rising celebration is based on the hair splitting between income and earnings. If we include all money (potentially) available to a person, men are still ahead of women. Hardly something to celebrate.

Now the table I find most interesting is in the Pew report on page 28. If women have made such terrific strides, outpacing men in education and earnings growth, wouldn’t they start contributing more to a household income? Yes, they would – and they do! In 1970, they contributed 6% and in 2007, they’re contributing 36%. Men’s shares are 99% and 72% for the two years. Okay, that’s roughly 1/3 from women and 2/3 from men in 2007. The rise of wives?!?

Incidentally, the chart on page 3 shows another interesting story that Pew totally missed because they were so set on presenting the myth that women are doing so well in marriage: Median adjusted household income for our 30- to 44-year olds has risen steadily IF these folks were married. For unmarried men and unmarried women, the income growth came to a halt in 1990 when both men and women had lower incomes when they were unmarried (note that this could be a factor of several things, not just marital status discrimination: Age – younger people are more likely to earn less and be unmarried – and the impact of the other income in a household – maybe married women’s household income is so much higher because men earn so much more).

There is lots of interesting stuff in the Pew report but, as happens so often, the rising wives story seems to be small changes blown out of proportion to bolster marriage. So what if wives earn more than their husbands? In an equal world this would happen about 50% of the time – it happens 22% of the time (and note, too, that $1 more could tip the balance…). Most of the touted rise by wives is simply catching up with men… Hardly something to celebrate. The real stories in the data – such as the gaps between men and women, black and white, married and unmarried – don’t make the headlines.

UpdateHere is a really cool video put together by that gives a summary of news coverage on this study. It’s interesting that they say that the number of wives outearning their husband has more than quadrupled. Well, yes 22% is more than four times 4% but it still means that more than 2/3 of husbands earn more than their wives… But the matrimania award goes to ABC news!

Another update: The Council on Contemporary Families has also done some myth busting on this study.

(Let me add a quick disclaimer: I did not read the report word for word, so I might have missed other gems or bloopers. I read the intro paragraph and got mad. After that, I looked at the charts and tables. Having worked in data analysis for over a decade, I know that numbers are harder to manipulate than their interpretation. I did read Appendix B on the methodology and looked in vein for the critical definitions, which I then located elsewhere. From what I’ve seen in the report, the claim that wives are rising is unfounded, as I have outlined above.)

Share this post with others:
Facebook Twitter Linkedin Email


Pew on Marriage — 1 Comment

  1. Pingback: Lisen's Blog – An Angle of PrismWork

Leave a Reply

Your email address will not be published. Required fields are marked *