3 ms·
> In 1985, the typical male worker could cover a family of four's major expenditures (housing, health care, transportation, education) on 30 weeks of salary. By
by dcolkitt 7y ago
> In 1985, the typical male worker could cover a family of four's major expenditures (housing, health care, transportation, education) on 30 weeks of salary. By 2018 it took 53 weeks.
The author has clearly never heard of Baumol's cost disease. The primary reason that households now expend more on housing, healthcare and education is because they expend a significantly smaller fraction of their income on food, electronics, entertainment, energy, clothes, appliances, telephone service, automobiles, and furniture. All of which have seen strong deflationary pressures.
Households ultimately need to expend their wages on some category. As the cost of certain goods fall relative to incomes, most consumers will allocate more of their expenditures on the goods that fell at slower rates. For example the average Nigerian or Indian spends a much lower percent of their wages on education or healthcare viz-a-viz the average Swiss or Singaporean. Only a fool would conclude that Switzerland's wealth has made it "fall backwards" relative to Nigeria. Instead the explanation is much simpler: the Swiss can feed and cloth themselves using a small fraction of their income. That leaves more left over to spend on comparative luxuries like better medical care and longer educations.
This dovetails into the second fundamental flaw with the author's analysis. He's simply comparing the total expenditures on housing/healthcare/education/transport between 2020 and 1985. That's making the implicit assumption that people in 1985 were consuming the same basket of these goods as people in 2020. He explicitly eschews measures of inflation, because inflation reveals the fact that the items themselves have not seen price increase. Expenditures rose because households are using their higher incomes (including more two income families) to buy more of these goods with more premium qualities.
For example one major reason housing expenditures have gone up is because the average home is 50% larger than it was in 1985[0]. Another reason is because of major quality improvements like central AC, attached garages, higher ceilings, better fire safety, higher capacity electrical circuits, and swimming pools. Adjusted for CPI, the median cost per square foot for housing only rose by 5% between 1990 and 2010[1] (which doesn't even adjust for significantly lower mortgage rates).
Similarly expenditures on transportation has gone up because compared to 1985, Americans travel 25% more miles per capita, and have 20% more vehicles per capita[2]. Automobile fatality rates were nearly 200% higher in 1985[3]. And I don't think I even need to tell you how much more reliable and comfortable cars are today than in the 1980s.
Healthcare and education probably have shown more true cost disease. However without a doubt Americans have also seen significant increases in their consumption levels. College enrollment rates have increased by more than 100% since 1985. Healthcare consumption has soared over the same period, not least because the country is significantly older and more obese, both of which require many more medical interventions. Compared to 1985, 100% higher proportion of the population works in the healthcare sector.
All of which goes to show that the author is plain wrong about his conclusions. The trends he shows are easily explainable by the fact that real household incomes have grown significantly since 1985 (not least because many women entered the full-time workforce). It now takes a significantly smaller percent of household incomes to satisfy basic needs with regards to food, apparel, utilities, appliances and furniture. So people have much more money left over to spend on comparative luxuries like bigger homes, safer and more reliable cars for both adults in the household, more medical treatment with advanced techniques, more home healthcare, and longer post-secondary educations. That's not a story of American households struggling, that's a story of the incredible consumer growth.
[0] https://www.propertyshark.com/Real-Estate-Reports/2016/09/08/the-growth-of-urban-american-homes-in-the-last-100-years/ https://www.propertyshark.com/Real-Estate-Reports/2016/09/08...
[1] https://www.census.gov/const/C25Ann/soldmedavgppsf.pdf https://www.census.gov/const/C25Ann/soldmedavgppsf.pdf
[2] https://www.ssti.us/2014/02/vmt-drops-ninth-year-dots-taking-notice/ https://www.ssti.us/2014/02/vmt-drops-ninth-year-dots-taking...
[3] https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in_U.S._by_year https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in...
[4] https://www.statista.com/statistics/183995/us-college-enrollment-and-projections-in-public-and-private-institutions/ https://www.statista.com/statistics/183995/us-college-enroll...
[5] https://randomcriticalanalysis.com/why-conventional-wisdom-on-health-care-is-wrong-a-primer/ https://randomcriticalanalysis.com/why-conventional-wisdom-o...
- adventured 7y agoOne of the best responses I've ever read on HN. Bravo. You demolished the outlandish, propaganda-laden premise of the author.
- jdmichal 7y agoI can get behind your theory. But I don't know how well it holds for housing, because of the technique described in the longer article posted in another thread. [0] Specifically, housing is 40% fair market rent on a three-bedroom apartment. So while it's certainly possible that the things you talk about housing are true for apartments, it's a different argument to make. Certainly at least some of the things you discuss are applicable, like central AC. Some others, like size, are probably less applicable. > Adjusted for CPI, the median cost per square foot for housing only rose by 5% between 1990 and 2010 Minor nit, but it seems like poor form to take someone who's arguing against CPI, and then to use CPI-adjusted numbers in a counter-argument. [0] https://news.ycombinator.com/item?id=22378311 https://news.ycombinator.com/item?id=22378311
- dcolkitt 7y ago> Minor nit, but it seems like poor form to take someone who's arguing against CPI, and then to use CPI-adjusted numbers in a counter-argument. That's fair enough. But let's use the author's preferred denomination of nominal median household income. From 1985 to 2018 this number rose 21% faster than CPI.[1] Given this it means the median cost per square foot fell by 15% over the period. > Specifically, housing is 40% fair market rent on a three-bedroom apartment. Thanks for linking that. That article found a 172% increase in nominal terms for its "typical example" of a 3 bedroom apartment in Raleigh, NC. In nominal terms, median real household incomes at a national level rose by 196%[1]. Data for Raleigh specifically doesn't go back to 1985. But since 2006, Raleigh incomes have risen 0.5% faster than the national average.[2] Therefore using the author's preferred metric rental housing has actually become 8-19% more affordable. Finally focusing on rent ignores the fact that 65% of American households are owner-occupied. Over the same period the cost of rent has risen substantially faster than the cost of ownership, not least because mortgage rates fell 12.5% to 4.5%[3]. If you take a weighted average of renters and owners, you'll find housing costs grew at substantially slower rates than simply picking a single metro's rental costs over time. [1] https://fred.stlouisfed.org/series/MEHOINUSA672N https://fred.stlouisfed.org/series/MEHOINUSA672N [2] https://www.deptofnumbers.com/income/north-carolina/raleigh/ https://www.deptofnumbers.com/income/north-carolina/raleigh/ [3] https://www.valuepenguin.com/mortgages/historical-mortgage-rates https://www.valuepenguin.com/mortgages/historical-mortgage-r...