3 ms·
One of the things I’ve learned as I started digging into wage inequality statistics is that performing this data analysis is extremely complicated. Demographics
by rayiner 7y ago
One of the things I’ve learned as I started digging into wage inequality statistics is that performing this data analysis is extremely complicated. Demographics have changed dramatically between then and now. The labor force participation rate for women has nearly doubled from 1950 to today. The US keeps income statistics on a per household basis, and marriage rates and household sizes have declined dramatically. The percentage of the population that is immigrants has tripled since 1970. The age distribution has changed, as well as college attendance rates (four years where young people are earning no income). Racial distributions have changed, which creates complex effects. (Phenomena that are better explained as racial discrimination show up as income inequality. Also, the median ages of people of different races varies quite dramatically, from 28 for African Americans to 43 for white non-Hispanic Americans. So some of what you might characterize as racial inequality actually ends up being age inequality.) The shift from hourly factory work to salaried service jobs, the create of tax income like the EITC, etc., all make these analyses extremely complicated.
To isolate what you might think of as income inequality due to structural economic factors, you’d want something like individual lifetime income adjusted by race, immigration status, and gender. I have yet to see anything that thoroughly adjusts for all of these confounding factors.
- refurb 7y agoDespite not always agreeing with Raynier, his posts are always well thought out. Glad to see this at the top. In terms of demographics, the baby boomers caused a big bulge in the distribution of the population. As they age, their earnings and overall wealth grow as well (take a look at income versus age). What happens when you have a disproportionate chunk of the population get wealthy? Income inequality grows. If this is indeed a factor, once baby boomers die off, we’d likely see a decrease in inequality.
- IkmoIkmo 7y agoRan into this a lot myself. So far I haven't been able to get very far to consolidate all these variables. Whenever I see threads pointing to economic growth or decline, I'm left saying there are tons of confounding factors, but without being able to offer the right answer. If you have any good reading to point to, please let me know. It's one of the things I'd love to have a good answer to, and that should be able to be answered to some extent, but isn't. Moreover I feel even many top economists routinely invoke extremely simplified single-variable perspectives (e.g. average wages per worker, or average household income) which indeed leaves out so many confounding factors that I've almost lost my expectation/hope that people even consider this kind of stuff... Only in academic papers which look to study something specific, are these confounders taken into account. But in general studies of economic growth or changes in or comparisons of economic wellbeing of various demographic subgroups, it seems the simple-perspective prevails. I get that many of these discussions are inherently political, thus susceptible to cherry-picking perspectives, but I'm still surprised there isn't some standard method so far that is more complex than charting 'household income for xyz'. Any pointers to good literature is most welcome, and thanks for posting!
- hhs 7y ago> leaves out so many confounding factors Indeed, that’s one of the key problems with retrospective analysis. Fieldwork studies are promising. I find John List exploring this area and think his work is interesting: http://www.fieldexperiments.com/papers/recent/ http://www.fieldexperiments.com/papers/recent/