3 ms·
Technically yes, although as a specific example $100k/yr is the 80th percentile in Austin, TX[0]. My point was, if you define middle class as the middle of the
by pault 7y ago
Technically yes, although as a specific example $100k/yr is the 80th percentile in Austin, TX[0]. My point was, if you define middle class as the middle of the income distribution, yes; if you define middle class more subjectively, like "being able to afford a house in the suburbs and pay for 2.5 college tuitions", that privilege is now out of reach for pretty much everyone except the 80th percentile and up.
[0] https://statisticalatlas.com/place/Texas/Austin/Household-Income https://statisticalatlas.com/place/Texas/Austin/Household-In...
- learc83 7y ago>Technically yes, although as a specific example $100k/yr is the 80th percentile in Austin, TX You're comparing household income to individual income. If you look at individual income, $100k/yr in Austin is much going to be much higher than 80th percentile.
- yakshaving_jgt 7y agoNitpick: When people talk in terms of class, that traditionally encompassed far more than monetary wealth. It's also your education, your status, your social circle, your distance from aristocracy, etc.
- ovi256 7y ago>you define middle class as the middle of the income distribution I don't think that's a good definition for the middle class. There's strong support for the fact that income distribution is an exponential distributions - just google "income distribution exponential". A exponential distribution is not symmetrical, so its mean or median is not a great centrality measure, it does agree with intuitions developed from observing, say, a normal distribution (like the weight or height of persons). The mean or median are very far from the mode (the most frequent individual) in an exponential, whereas they're almost identical in a normal.