The "The Two Income Trap" Trap

A surprising case for two being more than one

“femme maison” by louise burgeois, aka the goat

Yesterday I got into Twitter trouble for this tweet, about a campaign ad from two months ago for creepy venture capital fascist (strikingly common combination) and failed Arizona Senate candidate Blake Masters:

maia @maiamindel

The problem with this line of thought is that raising a family with a single income is possible, it's just that real wages for women are *high* enough that it's not worth it to stay at home vs invest in appliances and services

Blake Masters @bgmasters

You should be able to raise a family on one single income. But @shoe0nhead is right to ask — how do we get there? First, let’s talk about making more money. And later this week we’ll answer the next question: how do we get costs down so that your money has value? https://t.co/YT4deiJfYy https://t.co/04MTDkrt1O

It’s a trap!

The question here is what’s known as the Two Income Trap, which is a pretty simple explanation for why dual income families (aka those where two parents work) are inferior to single income families. The TIT (hehe) was first formally proposed by (checks notes) progressive firebrand Elizabeth Warren in a 2004 book.

The argument is pretty straightforward, and comes from Warren’s studies of bankruptcy as a bankruptcy law professor. Most families went from single income to dual income between the 70s and the 2000s, which made household real income go up. Meanwhile, things like housing, education, healthcare, and childcare got more expensive. Ergo, Warren argues, real income stagnated, and women had to pick up the slack to make ends meet, and consequently households had higher incomes. Housing plays a really big role, because people started spending way more on it during the 80s and 90s. And childcare in particular plays a big part in this, because adding a second earner means you have to pay for that.

A major point Warren and her daughter (who co-wrote the book with her) make is that even if families had higher incomes, they were also more exposed to adverse shocks if one of the two breadwinners lost their job - since they had higher expenses and lower individual incomes. If, before the 70s, the (male) breadwinner lost his job, then the woman stepped up and got a job herself, which meant that the family had overall lower expenditures and also had a big unemployment/sickness insurance. To quote the book:

If her husband was laid off, fired, or otherwise left without a paycheck, the stay-at-home mother didn’t simply stand helplessly on the sidelines as her family toppled off an economic cliff; she looked for a job to make up some of that lost income. (…) A stay-at-home mother served as the family’s ultimate insurance against unemployment or disability — insurance that had a very real economic value even when it wasn’t drawn on.

Well, do the book’s claims add up? No.

First, the boring reason: real individual wages went up during the 70s and 80s, for all low-level private sector employees, and for services sector employees specifically. This applies for basically all professions: manufacturing, transportation, leisure and hospitality, retail, construction, information services, and professional services. It also applies for women and men, separately. That women’s real wages went up pretty steadily is the most important part, for reasons I’ll detail later.

But her data is also wrong on other, brand new, interesting ways. Warren’s main claim for why families actually have less money to spend comes from the fact that fixed expenditures like mortgages, childcare, and healthcare got more expensive in real terms, aka adjusting for inflation, which is actually true. As a result, the book claims that discretionary income (aka the one you can spend in food, clothes, etc) was a bit lower and that families were in a more precarious situation. See this chart:

The issue here is that the BLS changed the methodology it used to measure inflation sometime in the 1970s, and Warren doesn’t use the CPI that measures with the same methodology, so she overstates inflation by a lot - resulting in purchasing power for the 1970s that is 13% lower than claimed, so even accepting literally everything else she claims, you still have that people in the 1990s were better off. A second issue is that the book uses the overall consumer price index for each category instead of the specific CPI for all of them - which becomes an issue when most of the stuff in the discretionary income part got cheaper in real terms.

This issue is pretty egregious considering that housing inflation specifically had a pretty big change in measurement in 1983, resulting in probably overstated housing services prices (because mortgage payments in particular declined significantly until 2005). In a related note, housing got much bigger because of the same regulations that increased costs - so people were paying more for more housing, which might be debatably good (or bad) but is a separate issue.

The piece by Matt Bruenig I linked also makes a separate point, that childcare is a temporary expense - so families have higher available income before and after having young kids, resulting in a more comfortable financial position over their lifetimes. I don’t think this is the most relevant criticism, but just putting it out there.

Domestic laborer or domestic engineer?

The housewife of the future will neither be a slave to servants or herself a drudge. She will give less attention to the home, because the home will need less; she will be rather a domestic engineer than a domestic laborer, with the greatest of all handmaidens, electricity, at her service. This and other mechanical forces will so revolutionize the woman’s world that a large portion of the aggregate of woman’s energy will be conserved for use in broader, more constructive fields.

Thomas Edison (1912)

One of the main claims that substantiates the TIT hypothesis is that household income was too low, so women had to enter the workforce to make up the difference. This is known as the Added Worker Effect, and it generally applies to recessions. In general, during hard times, people either join the labor market to help out (the AWE) or find it too difficult to find a job and drop out of the labor market (Discoraged WE). Since the DWE is out of the question, since labor force participation increased, then the issue is whether female labor force participation followed AWE patterns or not - and it didn’t, primarily because families that have wives as a reserve also had unemployment insurance and other welfare benefits that didn’t involve breaking up family dynamics.

The important question here is what the relationship between income and female labor force participation is. We can start by a pretty simple model from Nobel Laureate Gary Becker. Families can either produce goods and services they consume themselves, or they can purchase it from others through the marketplace. Which one they choose depends on the overall cost of the purchases, and on overall wages for each family member. With no change to prices, higher wages for all adults induces entry into the workforce, and lower wages reduce it. If prices increase, so does participation, and viceversa. So from the theory alone there’s clear evidence that, if real wages were increasing and the real cost of things like food and appliances were decreasing, then participation was up for good reason.

Does this pan out? Firstly, the increase in quality and especially the decrease in cost of appliances contributed to a higher opportunity cost for the same prevailing wage - since outside options were cheaper and, especially, less time consuming. It’s worth pointing out that childcare is the only expense that bucks this trend since it’s a labor-intensive task, and one that disproportionately falls on women, resulting in lifelong earnings penalties.

At the same time, a bunch of changes to the economy and its composition made it so that women could join the workforce in better positions. For starters, more and more women attended college, and the ones who did did so to advance their careers and not meet husbands like in previous generations. Secondly, things like the birth control pill, abortion access, and divorce allowed women to take more control over their life choices and therefore time their labor market entries and exits more optimally. Thirdly, social attitudes towards women in the workforce changed significantly, both among men and among women. Lastly, it’s possible that a shortage of young female employees due to the Great Depression and World War Two cohorts being unusually small resulted in longer duration female employment.

Evidence from US history

For ages woman was man’s chattel, and in such condition progress for her was impossible; now she is emerging into real sex independence, and the resulting outlook is a dazzling one. (…). Under these new influences woman’s brain will change and achieve new capabilities, both of effort and accomplishment.

Thomas Edison (1912, cont.).

So what’s the evidence in favor of higher, not lower wages being the reason why women entered the workforce?

Economist Claudia Goldin (the GOAT in this sort of thing) describes it as a “U-Curve”, with rich countries and poor countries having very high participation, and countries in the middle having very few women working. The explanation is that in extremely poor stages of development, everyone has to work; meanwhile, at middle income tier, one earner is enough and women have to stay home to provide goods and services for the family. Finally, as incomes get higher, female labor opportunities get better, and the opportunity cost of staying at home cooking versus working becomes higher than the cost of buying appliances or hiring a nanny. In fact, the fact that childcare was something that normal families paid for is a sign that incomes were higher, not lower, since if they weren’t, then mom would have just taken a very prolonged break to look after the kids. Of course, this points towards the TIT being nonsense, unless you think the US has been devolving into an agrarian society for the past 50 years.¹

In the US, female labor force participation was only higher than in the present during the Revolutionary Era, which was the most agrarian time in US history - even for urban women, who worked at even higher rates. Participation leveled off during the 1900s, behaved weirdly during the Depression, spiked in WW2 and went back down, and then steadily increased from the 60s and 70s onwards. The role of marriage is pretty relevant: until my parents’ lifetime, women left the workforce after marriage because it actually was illegal, until the 1950s, to hire married women in most industries, since it was expected that they become unpaid domestic laborers for the rest of their lives. Some professions, like teaching and nursing, were allowed to continue, however.²

Call me a strident blue-haired radical feminist who hates men and doesn’t wear a bra, but I don’t think women being legally conscripted into pretty much indentured service to men for their entire lifetimes is good, and that making people abjure their entire personality, preferences, and desires so some violent alcoholic can get a free dinner every day is a good model for society (though Zygmunt Bauman kind of thought it was, shots fired at extremely obscure thinkers, for an econ audience).

Something that I also believe provides evidence for the two income trap being absolute bunk is how the gender wage gap is structured in the present. The gap is actually widest among white, high income women, precisely because they lose out so much from looking after their kids. So I’d wager that low income women don’t drop out of the workforce because the opportunity cost is even higher for them, since the family would struggle - but this is the opposite of TIT! (another case is informal work in Latin America) This is especially important because the gender wage gap only really exists for women with children - women have to devote ungodly amounts of time to household chores and childcare, harming their workplace prospects and therefore their long-term earnings due to systematically fewer hours and less experience due to childcare breaks. This does not show up for fathers.

Conclusion

When the most ardent adherents of a concept are weirdo conservatives who want to RETVRN to 1950s advertisement and Edward Hopper pastiches, you know it’s a bad one. But the evidence and the theory all points to the Two Income Trap hypothesis being false, and the opposite being true - that women joined the labor force because it was beneficial to do so for family finances.

Tho I’d like to point out that childcare, healthcare, and housing are all too expensive due to unnecesary regulations, so that would be nice to fix.

Sources

There’s nothing here I haven’t said in previous posts about the gender wage gap, women’s workforce history in the US, and women’s participation across the developing world. Real ones read those.

1. This is the millionth example of political scientists coming up with an idea first (cough cough Acemoglu), because the idea was originated from Engels’ The Origin of The Family, although at least Goldin gave him credit for the thought.

2. Family note: of my two grandmothers, one was mostly a housewife but helped run my grandfather’s clothing store, and the other is an architect who’s still working to this day. That side of the family was weird because all three of her sisters went to university: two studied accounting to run the family’s business, and the other one was a biology teacher.