I definitely don't understand the link between unemployment and inflation. Could you elaborate a bit how full employment causes inflation? I'm not being purposefully dense; I just don't get it.
Anyway, I just think it's important to keep in mind how much the macro environment affects which bad decisions get punished and how much. The same mistake that might have got you a nasty email from the boss in the late 90s might have got you fired in the late 00s. The sorting effect you hypothesize might happen, but there's significant variability in how harsh it is, at least a portion of which we have control over.
As for inflation and unemployment, basically all mainstream economists agree that there is something called NAIRU, the non-accelerating inflation rate of unemployment. It's the rate of unemployment below which inflation starts to accelerate. The Wikipedia article isn't very helpful but here it is:
http://en.wikipedia.org/wiki/NAIRU
and related:
http://en.wikipedia.org/wiki/Phillips_curve
Here are Excel files of historical NAIRU rates targeted by the Federal Reserve:
http://www.phil.frb.org/research-and-data/real-time-center/g...
Here is a paper on NAIRU (opening section is a reasonably good general introduction) by Greg Mankiw, who is not my favorite economist but very widely respected [pdf]:
http://scholar.harvard.edu/files/mankiw/files/jep.ballmankiw...
The section on why NAIRU exists basically concludes "nobody knows." Unfortunately this is macroeconomics so there's a lot of just-becausing.
I'm not sure if it's a good theory or not but it's widely believed by economists, especially central bankers. So if unemployment goes above a given rate, which changes depending on circumstances (I think the latest estimate is 6.5%), they're going to tighten the money supply anyway so that growth and especially job growth slow down.
So essentially, it's an unstated national policy in almost every country not to have full employment.
At any rate thanks for helping me understand the NAIRU rather than just calling me an idiot.
I agree completely that the macro environment is hugely influential. Saving and frugality are beneficial at one interest rate or naive and foolish at another. It's really strange how the "common sense" idea that you shouldn't spend money you don't have is 100% true at 10% interest but only about 5% true at 1% interest. Especially with inflation at a couple percent.
"[W]e examined the cognitive function of farmers over the planting cycle. We found that the same farmer shows diminished cognitive performance before harvest, when poor, as compared with after harvest, when rich. This cannot be explained by differences in time available, nutrition, or work effort...Instead, it appears that poverty itself reduces cognitive capacity."
http://www.sciencemag.org/content/341/6149/976.abstract
Buying or not buying a soda seems like a plausible example arithmetically, but in practice any personal budget that does not include a few small luxuries is not a personal budget that anyone is going to follow. If it's not a soda it's going to be a magazine or something else eventually.
Anyway, even if we concede arguendo that those now presently in poverty got there by being stupid and drinking soda they couldn't afford or not finding cheaper lodgings etc., if we somehow removed bad decision makers from the population it wouldn't mean an end to poverty. It would just mean good decision makers would find themselves in poverty. That's because we need at least 5% unemployment to keep inflation at bay, with some number above that underemployed and some number below that dropping out of the workforce entirely. While those states could theoretically be transitory for everyone, in practice they're going to be sticky for a large number. So even in a population of 100% good decision makers you're still going to have some poverty just so the central banks don't run up against NAIRU.
And then, if this paper is correct, those new poor are also going to start having a hard time with congnitive tasks.