5 ms·

By definition, the “cluster” below the median is exactly half, regardless of the shape of the distribution. > Also CPI doesn’t really account for housing price

by bhupy 6y ago

By definition, the “cluster” below the median is exactly half, regardless of the shape of the distribution. > Also CPI doesn’t really account for housing prices Per my source, the median income is the highest it’s been since 1981, and the total compensation the highest it’s been since WW2 regardless of the inflation index you use. Also, the median number of years to recoup the cost of a bachelors degree, adjusted for inflation, has gone down since the 1980’s, from about 22 years to about 10 years -> https://libertystreeteconomics.newyorkfed.org/2014/09/the-value-of-a-college-degree.html https://libertystreeteconomics.newyorkfed.org/2014/09/the-va...

nicoburns 6y ago
> By definition, the “cluster” below the median is exactly half, regardless of the shape of the distribution. Right, but that doesn't measure how for away from the median the upper and lower clusters are. > Also, the median number of years to recoup the cost of a bachelors degree, adjusted for inflation, has gone down since the 1980’s, from about 22 years to about 10 years That would be consistent with a hypothesis that real wages have increased for higher earners and decreased for lower earners, right?
- bhupy 6y ago
  > Right, but that doesn’t measure how far away from the median the upper and lower clusters are. Well, it doesn’t particularly matter how far from the upper the median clusters are if their absolute standard of living has improved — and we measure that by accounting for inflation. Regardless of which inflation metric you use (some include housing prices, some don’t), median income has risen. As for the lower cluster, the most reliable metric to use there is the poverty rate...which has also substantially fallen -> https://twitter.com/Noahpinion/status/1308790816937095169?s=20 https://twitter.com/Noahpinion/status/1308790816937095169?s=... You can also look at the historic income trends divided up by quintile -> https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-income-households.html https://www.census.gov/data/tables/time-series/demo/income-p... (Table H1, All Races, it's an Excel spreadsheet). Inflation adjusted income for the bottom two quintiles have also increased substantially. > That would be consistent with a hypothesis that real wages have increased for higher earners and decreased for lower earners, right? No, it would be consistent with a hypothesis that real wages have increased faster for higher earners than for lower earners, but demonstrates no evidence that it has decreased for lower earners. The empirical results show that real wages increased across all income levels, and between 2014-2019, it actually increased faster for lower earners -> https://www.cbsnews.com/news/low-wage-workers-are-getting-bigger-raises-than-bosses/ https://www.cbsnews.com/news/low-wage-workers-are-getting-bi...
peteradio 6y ago
And that cluster can be anywhere below the median that is the problem. It could even be at zero.
- bhupy 6y ago
  That's a separate argument, because at that point we're no longer talking about the standard of living of the "average American". That being said, across all quintiles (the bottom 2 quintiles include the clusters below the median), real income has also grown substantially -> https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-income-households.html https://www.census.gov/data/tables/time-series/demo/income-p... (click Income Limits for Each Fifth and Top 5 Percent, it's an Excel spreadsheet)
  - peteradio 6y ago
    I'm not sure what use it is to consider the "average american". I'm concerned about the cohort of people who are barely scraping by, you seem to paint a picture that these peoples lives have improved because their "real income" has grown substantially. Ok so they were barely making rent at 15k/year, and their real wages increase by 5k/year since 2010. Has their rent remained stable over the last then years? Perhaps these quotable numbers aren't all that useful except to further marginalize?
    - bhupy 6y ago
      Correct, which is why we adjust for inflation, and there are a ton of different inflation indexes one can use. Per the links in the source comment, regardless of which index you use, we have seen an increase in income. > Has their rent remained stable over the last then years? From the data, outside of expensive coastal metros, yes. If you don't have a college education, you're more likely to enjoy a higher standard of living in Little Rock, Arkansas or Minneapolis, MN than in San Francisco or Manhattan. > Perhaps these quotable numbers aren't all that useful except to further marginalize? No, these quotable numbers tell us that we're generally trending in the right direction. Therefore, any solution we employ to help those that are the poorest among us (which is important) doesn't necessarily need to upend the entire system.
kergonath 6y ago
The problem with using the median for income is that it does not tell anything about inequality. If the top 1% get their income doubled (or the bottom 30% their income halved, or both at the same time) with all else being equal, the median does not change. You really need a high percentile as well. Low percentiles are not that interesting, because the distribution is much more compressed there. In essence, a median that keep climbing nicely is no indication that the situation is not becoming catastrophic. Granted, using the average is arguably worse.
- bhupy 6y ago
  If the top 1% get their income doubled, and the median doesn't change (even after accounting for inflation / purchasing power), then it's debatable if we even have a problem. If the bottom 30% income halved, you're correct that we have a problem, but that hasn't actually happened. You can view the inflation-adjusted income data split up by quintile here -> https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-income-households.html https://www.census.gov/data/tables/time-series/demo/income-p... (Table H-1, All Races, it's an Excel spreadsheet). Even among the bottom quintile, inflation-adjusted income has increased.
  - kergonath 6y ago
    > If the top 1% get their income doubled, and the median doesn't change (even after accounting for inflation / purchasing power), then it's debatable if we even have a problem. I think we do, as these people have disproportionate economical and political power, which results in unbalanced economic policies that favour one specific socio-economic class and breeds discontent. I don’t have links handy, as I am on the phone, but there are works showing a correlation between political instability and economic inequality.
    - bhupy 6y ago
      By definition, the top 1% are a minority of voters, so their influence is only as strong as the votes that they themselves are able to cast, or that they are able to convince others to cast. Their impact on elections is overstated -> https://fivethirtyeight.com/features/money-and-elections-a-complicated-love-story/ https://fivethirtyeight.com/features/money-and-elections-a-c...

by bhupy 6y ago

nicoburns 6y ago

> By definition, the “cluster” below the median is exactly half, regardless of the shape of the distribution. Right, but that doesn't measure how for away from the median the upper and lower clusters are. > Also, the median number of years to recoup the cost of a bachelors degree, adjusted for inflation, has gone down since the 1980’s, from about 22 years to about 10 years That would be consistent with a hypothesis that real wages have increased for higher earners and decreased for lower earners, right?

bhupy 6y ago
> Right, but that doesn’t measure how far away from the median the upper and lower clusters are. Well, it doesn’t particularly matter how far from the upper the median clusters are if their absolute standard of living has improved — and we measure that by accounting for inflation. Regardless of which inflation metric you use (some include housing prices, some don’t), median income has risen. As for the lower cluster, the most reliable metric to use there is the poverty rate...which has also substantially fallen -> https://twitter.com/Noahpinion/status/1308790816937095169?s=20 https://twitter.com/Noahpinion/status/1308790816937095169?s=... You can also look at the historic income trends divided up by quintile -> https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-income-households.html https://www.census.gov/data/tables/time-series/demo/income-p... (Table H1, All Races, it's an Excel spreadsheet). Inflation adjusted income for the bottom two quintiles have also increased substantially. > That would be consistent with a hypothesis that real wages have increased for higher earners and decreased for lower earners, right? No, it would be consistent with a hypothesis that real wages have increased faster for higher earners than for lower earners, but demonstrates no evidence that it has decreased for lower earners. The empirical results show that real wages increased across all income levels, and between 2014-2019, it actually increased faster for lower earners -> https://www.cbsnews.com/news/low-wage-workers-are-getting-bigger-raises-than-bosses/ https://www.cbsnews.com/news/low-wage-workers-are-getting-bi...

peteradio 6y ago

And that cluster can be anywhere below the median that is the problem. It could even be at zero.

bhupy 6y ago
That's a separate argument, because at that point we're no longer talking about the standard of living of the "average American". That being said, across all quintiles (the bottom 2 quintiles include the clusters below the median), real income has also grown substantially -> https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-income-households.html https://www.census.gov/data/tables/time-series/demo/income-p... (click Income Limits for Each Fifth and Top 5 Percent, it's an Excel spreadsheet)
- peteradio 6y ago
  I'm not sure what use it is to consider the "average american". I'm concerned about the cohort of people who are barely scraping by, you seem to paint a picture that these peoples lives have improved because their "real income" has grown substantially. Ok so they were barely making rent at 15k/year, and their real wages increase by 5k/year since 2010. Has their rent remained stable over the last then years? Perhaps these quotable numbers aren't all that useful except to further marginalize?
  - bhupy 6y ago
    Correct, which is why we adjust for inflation, and there are a ton of different inflation indexes one can use. Per the links in the source comment, regardless of which index you use, we have seen an increase in income. > Has their rent remained stable over the last then years? From the data, outside of expensive coastal metros, yes. If you don't have a college education, you're more likely to enjoy a higher standard of living in Little Rock, Arkansas or Minneapolis, MN than in San Francisco or Manhattan. > Perhaps these quotable numbers aren't all that useful except to further marginalize? No, these quotable numbers tell us that we're generally trending in the right direction. Therefore, any solution we employ to help those that are the poorest among us (which is important) doesn't necessarily need to upend the entire system.

kergonath 6y ago

The problem with using the median for income is that it does not tell anything about inequality. If the top 1% get their income doubled (or the bottom 30% their income halved, or both at the same time) with all else being equal, the median does not change. You really need a high percentile as well. Low percentiles are not that interesting, because the distribution is much more compressed there. In essence, a median that keep climbing nicely is no indication that the situation is not becoming catastrophic. Granted, using the average is arguably worse.

bhupy 6y ago
If the top 1% get their income doubled, and the median doesn't change (even after accounting for inflation / purchasing power), then it's debatable if we even have a problem. If the bottom 30% income halved, you're correct that we have a problem, but that hasn't actually happened. You can view the inflation-adjusted income data split up by quintile here -> https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-income-households.html https://www.census.gov/data/tables/time-series/demo/income-p... (Table H-1, All Races, it's an Excel spreadsheet). Even among the bottom quintile, inflation-adjusted income has increased.
- kergonath 6y ago
  > If the top 1% get their income doubled, and the median doesn't change (even after accounting for inflation / purchasing power), then it's debatable if we even have a problem. I think we do, as these people have disproportionate economical and political power, which results in unbalanced economic policies that favour one specific socio-economic class and breeds discontent. I don’t have links handy, as I am on the phone, but there are works showing a correlation between political instability and economic inequality.
  - bhupy 6y ago
    By definition, the top 1% are a minority of voters, so their influence is only as strong as the votes that they themselves are able to cast, or that they are able to convince others to cast. Their impact on elections is overstated -> https://fivethirtyeight.com/features/money-and-elections-a-complicated-love-story/ https://fivethirtyeight.com/features/money-and-elections-a-c...