The increasing concentration of incomes among a small number of high earners has been documented at length here (, , ) and elsewhere. Any sensible response to this development has to be based on at least a partial understanding of how and why this trend began - and we still don't have a theory that seems strong enough to use as a basis for policy.
Here is a story that makes sense to me. Then again, I may be putting one and one together and getting eleven.
1) The hypothesis that the sharp rise in executive compensation was driven by an equally sharp rise in firm values. This study (pdf) concludes that "[t]he sixfold increase of CEO pay between 1980 and 2003 can be fully attributed to the six-fold increase in market capitalization of large US companies during that period." According to this story, the increase in high-earner wages isn't a problem that needs solving: CEOs are being paid their marginal product.
2) The hypothesis that the sharp rise in equity prices was driven by demographics. The story we'd tell here is that as the leading edge of the baby boom reached the age of 40 or so, the demand for financial assets increased, driving up firm values. (See here for a theoretical model that develops the point.) The timing certainly fits: the rise in asset prices began in 1985.
So the claim is that demographic pressures are behind the underlying cause of the increasing concentration of income in the right-hand tail of the income distribution. Clearly, the timing fits this story. The beginning of both the asset price boom and the trend to higher top-end salaries occurred when baby boomers started saving/investing.
As far as policy goes, this is very much a story in which those high executive salaries are best thought of as rents: those CEOs just happened to be at the right place at the right time.
But what happens now as the baby boom ages out of the work force? According to the second story, the wave of dissaving by retirees would provide downward pressure on asset values. Apparently this has led to talk of an asset price meltdown, but it seems more likely that the decline will be modest. Either way, this conjecture would predict that the trend to high incomes should level off as asset prices reach their plateau. And as asset prices stagnate, shareholders may start asking why they are playing high wages to people who are producing mediocre returns.
There's another point to consider: the fact that high-income concentration is largely a phenomenon of English-speaking countries. The explanation for this would be a claim that the baby boom was more pronounced in the Anglosphere combined with a claim that there is an Anglosphere investor home bias. I don't think either of these claims is clearly wrong, but I'm not sure they are strong enough to explain why we don't see high-income concentration in countries that don't speak English.
I don't recall seeing this conjecture before, so if you are aware of work that has been done on this point, please let me know in the comments.