The figures have been adjusted for inflation using the CPI-U-RS, which (according to Bureau of Labor Statistics) was only applicable from 1978 on. So I’m wondering how they managed 1968? Also, I’m always uneasy about inflation-adjustments since they include everything in the market basket. Lower income people are subject to inflation on food and housing, but people earning under $25K (2011) were probably eligible for Medicaid, so medical inflation wouldn’t be applicable to them. In short, I think I’d take this with a large barrel of salt.
What would you say if income decreases across all brackets?
That’s an excellent point, as the marginalization of all except the highest income tiers out of the political process is a very real phenomenon. And the absence of meaningful regulations on campaign and political advocacy spending ensures that this situation only worsens even more, sort of a final kick in the pants to the disenfranchised middle class. Studies have shown what should already be intuitively obvious: that relative rather than absolute wealth also matters to many aspects of both real and perceived well-being, so that people in countries with lower overall national wealth but more equitable income distribution are generally happier than those that have to deal with extreme income disparities, and are not just less politically marginalized but less socially marginalized.
Here’s more proof of that, and a thought for those who think the OP graph represents a utopia of upward mobility.
When one looks at both the trends and the US position in the Gini coefficient (an index of income disparity in which 0 is perfect uniformity and 1 (or 100%) means one guy has all the money), one can observe that after a slow but steady decline in the post-war era, it started a fast rise toward greater inequality around the 80s and has so far increased by a rather stunning 20% (while at the same time the coefficient in many other industrialized countries has declined). This is remarkably correlated with the rise of Reaganomics but no doubt has other causes, too, including the prevalence of offshoring and a general trend toward the increasing influence of money in politics.
What is most notable and should be disturbing is that in absolute terms, US income disparity is far worse than that of Europe and the industrialized Commonwealth countries like Australia and Canada, and on a par with China, Mexico, and Russia – countries which can be categorized as having a poor population but with wealthy ruling oligarchies. You can clearly see this in the current Gini distribution.
Well, the point I’m trying to make is that money is merely a proxy for standard of living: typically, a high standard of living is correlated with a high income. But this doesn’t necessarily mean that a high income implies a high standard of living—that merely amounts to affirming the consequent.
It’s a widespread problem of human nature: we tend to chunk hard-to-quantify, ambiguous or vague matters into more easily measurable or definable factors, which at least in a certain regime we take to be well correlated with the original quantity of interest. This is, of course, a reasonable strategy: the latter quantity will typically give us some information about the former, thanks to said correlation. But all too often, we then forget about the original question, concentrating instead exclusively on our proxy quantity—thus, trying to optimize standard of living, we instead optimize income, occasionally even going so far as to conflate the two. But even if a high standard of living typically implies a high income, the reverse may simply be false.
A similar example is given by ‘quality of research’ and ‘number of citations’. High-quality research will typically generate a high number of citations; but not all highly cited research is of high quality. In fact, there’s a growing realization that by optimizing citation count, one may in fact end up hurting research quality: encouraging research in line with current trends, rather than the pursuit of more ‘risky’ lines of inquiry, and so on, leading to self-reinforcing loops such as: publishing in journals with a high citation index/impact factors leads to research being judged as being of higher quality (it’s in a high-impact journal after all), meaning it’ll be cited more often, which leads to the journal acquiring an even higher citation count, etc. In the end, citation count may not mean what it used to regarding quality of research.
So I guess the answer to your question is that I don’t know: it is entirely possible that a higher standard of living exists on a lower level income across the board, even though typically, a higher standard of living is correlated with a higher income. The point is simply that the inference ‘income –> standard of living’ isn’t sound in general, and that thus, the algorithms we implement as a society to raise income as opposed to the ‘standard of living’, even though they may work perfectly, simply might miss the point.
But of course, it’s far from clear how we could work to raise the ‘standard of living’—the nebulous nature of that concept is after all what prompts us to concentrate on the easy-to-quantify ‘income’ proxy.