I should say that the housework we do ourselves is not captured in GDP.

Sure it is. It's captured by its effect on our overall productivity.

A significant share of the growth in U.S. GDP since the 1970s can be attributed to women leaving the home and entering the workforce. It's not that they were doing any less work beforehand, but a woman who sucked at housework but excelled at piloting an aircraft could work more productively, improving overall efficiency of the economy.

GDP didn't grow simply because her work was now measurable in dollars and cents. People lament that to have a similar standard of living today as 60 years ago, a family needs two wage earners. Yes, and part of the reason is because women entering the workforce _depressed_ wages. In other words, as women earned more money men began earning less. (Except it happened slow enough that men's absolute earnings never declined, they just grew more slowly.) Nonetheless, we're all better off when a woman can make better use of her unique skills. Our economy became more efficient overall, so overall GDP went up commensurately.

Many of the criticisms in that article aren't pertinent to a large, diverse economy like the United States. No economist believes that the GDP figures for a natural-resource intensive economy are as meaningful as for the U.S. They can absolutely be meaningful in some contexts, but are obviously less meaningful in others. Kudos to the journalist for admitting his young naivety, but he's assuming that the his elder contemporaries at the time were similarly naive about the limitations of GDP. The better assumption is that economists in the 1990s that had the same amount of experience as he enjoys today were similarly wise to the limitations of GDP.

Analogies about personal GDP don't work very well. GDP is a macroscopic view that is predicated on various behaviors averaging out. If you had an economy of a single person who only did housework, GDP is obviously useless. But so what? That doesn't say much of anything about it's utility for measuring the overall productive capacity of a huge, diverse economy.

Regarding the relevancy of GDP to Japan... that's why we have metrics like per capita GDP and metrics like the Gini coefficient. Japan is shrinking thus their rate of growth in GDP will slow if not shrink. As a population grows you can asymptotically approach a more efficient allocation of labour and resources, but if its not growing or shrinking you lose out on that avenue for improved productivity. If your entire economy is a single, two-person household, someone will invariably have to specialize in doing housework all day while the other works outdoors. If you have an economy of 300 million, you can out-source both mundane housework and mundane outdoor work and everybody gets more work done overall; certainly more so than an economy of two, but also more so than an economy of 200 million.

But population growth isn't the only opportunity to improve productivity. Japan has a high quality of life partly because of low inequality and thus a low Gini coefficient. Higher inequality tends to signal a more inefficient allocation of labor and resources. Which isn't to say that Japan doesn't have room to improve efficiency (sex inequality is still a big thing), but in many respects they're in a better place than the U.S. to begin with.

Like the the case with natural-resource intensive economies, the Gini coefficient tells us something about the accuracy and meaning of GDP, except in a quantified way. (And if you can't quantify something, you can't easily apply that knowledge.) A high Gini coefficient (and high inequality) suggests that we should be more careful about the conclusions we draw from GDP. In particular it suggests that our presumptions about things "averaging out" are more questionable. Inferences about particular details of the economy will be less accurate, and predictions more volatile. So it's not so much the case that GDP is a poorer way to measure quality of life in Japan, it's that because of our higher Gini coefficient it's a poorer way to measure quality of life in the U.S.

OTOH, it also follows that over the period where GDP growth in Japan has slowed relative to U.S. GDP growth, quality of life in the U.S. has probably in many ways improved faster. The effects mitigated, of course, by our inequality. Likewise, we can find all sorts of fault in how accurately GDP reflects particular details about China, but in relative terms if quite obviously reflects (as expected) a rapidly improving quality of life.

Do you actually do “housework”? You have some strange statements in here from the perspective of someone who lives in a dual income family with shared responsibility in the home. All the time-consuming chores of yesteryear take much less effort and time today. It takes me roughly 20 minutes to do a load of laundry, and we have to do 3-4 of those a week. It takes me 30 minutes to wash dishes once a day and the dishwasher is effectively a few minutes per day at most. Grocery shopping takes an hour or two, and that is getting serious tech applied to it lately, to the point that we may well not need to spend more than 10 minutes a week on it in the near future. Also it takes me about 15 minutes to make tea the way I like it. ;)

There are a lot of factors that contributed to women entering the work force, and one big one is home automation that left them with more free time.

I ask this as a general question, as someone with a passing interesting in history, but no formal training: how do we know that free time caused women to enter the workforce, rather than women entering the workforce gave them enough money to trade money for time in household work, thus craeting a housework automation economy?

In history, things seem so... fuzzy[0]... that identifying the cause from effect seems so arbitrary.

[0] More formally, we never really know the full history, we generally rely on unreliable/biased narrators, and we don't know what information we don't know.

In general we can establish causality for socioeconomic theories using natural studies. For instance your question can be answered by comparing the timing of introduction of technology into the home and the rise of women in the workplace. You can ask people how much time they spend doing housework and compare to what people with technical alternatives spend. This is a relatively easy natural study since so much of the data was well collected at the time. Harder ones that are more recent ask things like whether or not raising the minimum wage increases unemployment. You’ll be able to find studies that used Seattle’s minimum wage law to determine the answer to that question.

In the narrow case of women entering the work force, the biggest factor was the employers being willing (forced) to hire women because they were the only workers available quite suddenly. Women would have entered the work force in more substantial ways before that but they were perceived as taking jobs from men. WW2 changed all that. One reason they were able to keep working after the war is because they had money and the products were available to allow them to get the housework done in less time. It was a status symbol in the 50s (and probably still today) to have a very bored wife at home.

FWIW, my point isn't predicated on why women were able to enter the workforce. I was merely arguing against the claim that informal labor isn't reflected in nominal GDP. See my reply elsethread.

IIRC, I first came across the claim about women disrupting the labor economy in one of Francis Fukuyama's books, probably either "Trust: The Social Virtues and the Creation of Prosperity" (1995) or "The Great Disruption: Human Nature and the Reconstitution of Social Order" (1999). It's been awhile since I read it, however; but I've read many books since then and while I don't know if any other source I've personally read made such a straight-forward claim, I think the contention is largely accepted and, especially more recently, widely acknowledged that women entering the workforce (for w'ever reason) substantially suppressed the nominal wages of men.

It doesn't matter _why_ women entered the workforce. I was arguing against the idea that GDP only reflects nominal wages, and doesn't reflect the value of labor outside the formal labor market. My point was that when women were primarily homemakers, the average salary of the man indirectly reflected the work of the entire household and thus was actually reflected in the GDP. Not perfectly, of course, because she wasn't as free to participate in the labor market as men, and thus society systemically undervalued her labor. But to a large extent it was indeed reflected in nominal wages.

Likewise, when the article discussed how the per-capita GDP of Kenya, at only $2, clearly didn't reflect the "real" amount of labor and wealth in Kenya, it's wrong and misleading. Much of that value is reflected in $2 in that $2, in nominal wages, is all you need to subsist. Remember, there's no such thing as "intrinsic" value, not in economics. If $2 is all the Kenyan labor market is willing to pay, then $2 reflects rather well how the Kenyan labor market is structured, and reflects what value any investments can leverage in the short-term. (Would you invest $100 million in Kenya in exchange for help harvesting your fields and babysitting your kids?) We don't need some ad hoc system to quantify the potential of the Kenyan labor economy because the potential is implicit in the huge difference between the Kenyan GDP and the GDP of richer countries.

Look at it another way: if I sit here and pour myself into writing HN posts all day rather than participating in the formal labor market, can I complain that GDP isn't accurately reflecting the value of my labor? It's nonsense. I would be neither wrong nor right. My complaint would only beg the question, value by who's reckoning?

There's a debate to be had, at the margins, about the limits and accuracy of metrics like GDP, but the notion of trying to quantify wealth outside the formal economic system is fundamentally and hopelessly flawed. Rather than trying to create some framework to guesstimate the value of supposedly hidden wealth, people should be working to plug these systems of labor and exchange into the formal economy. A free market economy based on supply & demand and a fungible currency is fundamentally a pricing system. It's a giant calculator with a built-in unit for determining and expressing "value". For all its many flaws, it's foolish to believe anybody can do any better by trying to tally behaviors individually. Such systems are invariably worse in terms of systemic biases. It's why communism failed--at scale no system of pricing the "value" of objects and labor is more honest, accurate, and consistent than what people manifestly are willing to freely accept in exchange; and no system of quantifying that value is more honest, accurate, and consistent than a nearly perfectly fungible, nearly perfectly substitutable commodity--currency. It's not perfect, but it represents the best that is achievable.

If you find flaws in how the system prices those things, it usually has something to do with 1) the lack of freedom of accepting or rejecting exchange or 2) the medium of exchange (i.e. currency). Thus debates about whether, e.g., healthcare should be considered some sort of right (forcing government or employers to express the cost, previously hidden by silent pain and suffering, in nominal currency) so that people would be more free to change jobs. The issue and task is to figure out a way so that these things can be more transparently priced into the formal economy, rather than spending time trying to quantify these things in some ad hoc, contrived model. What good would such a number be, anyhow?