Tuesday, December 11, 2018

Suspicious Economic Statistics

While evidence of increased inequality is overwhelming, other economic survey research looks more suspicious.

Inequality has generally increased during Republican presidencies and decreased during Democratic presidencies, but New Democrats don't advertise the fact that inequality also increased during most of Obama's presidency. New Democrats pull a slight of hand, trying to take credit for gains correlated with long gone presidents, having differing policies.

Democrats have been wedded to the narrative that wages for nonwealthy workers have "stagnated" for decades because evidence of decline would make their own neoliberal, free riding policies look bad, though Republican neoliberal policies are much worse.

Part of this seems driven by the intuition that median per hour wages have increased simply because people now have more toys, ignoring that individuals have more toys because values changed, more wives work now, individuals have fewer children to support, and the nation is older, meaning more individuals have had more years of collecting toys, not to mention wealthy economists simply generalizing from the small sample, availability error of their own circumstances.

Among Consumer Price Index errors, hedonic pricing appears to be a major contributor, especially when applied to housing. Technologies that are more harmful and habit forming should not be treated as hedonic pricing positives. This and other errors leads to alternative Consumer Price Indexes.

Turning to unemployment and some other issues, unrepresentative sampling is a big problem: Because of increases in caller ID and illegal telemarketer calls, a large percentage of working class individuals do not answer phone calls from strange numbers or numbers outside their area codes. Many have pre-pay, pay per minute plans. Almost every time such individuals answer a call it costs them a minimum of ten cents per minute. Since the top one percent today mostly engages in free riding and dark triad activities, they are also less likely to answer calls from strange numbers. Upper middle class individuals, individuals with middle class values, are more likely to be helpful on the phone, making such individuals over sampled.

We have neocolonial establishments that produced thousands of fallacious statistics that we were winning the various counterinsurgency wars, that nonwhite immigration is beneficial, that racial diversity is also beneficial, that genetic factors should be ignored or discounted when doing research, that epigenetics is massively important, that priming has big effects, that early childhood environmental factors have huge impacts on adult behavior, etc. Why should we simply simply assume their economic stats are accurate when economics is not far from sociology and criminology in junk science tendencies?

No comments: