Most economists will tell you that workers displaced by technological progress and globalization should be able to find comparable employment elsewhere -- or at least that the income losses of those who lose their jobs are more than offset by the broader gains to consumers as a whole.

Yet you have to wonder if that's quite right. Is it possible that machines and trade might eventually polarize the rich world's labor force into two distinct classes: servants and elites? If we want to prevent that, we may need to reconsider how we think about work and income in an age of abundance.

Two recent articles in the Wall Street Journal provide some nice context. One describes how the recovery's meager gains have been concentrated among those who were already the best-off. The other describes the rising demand for high-end butlers, maids and housekeepers. These jobs pay as much as $200,000 a year and, unlike many other lines of work, odds are against butlers being replaced by machines. After all, the people who like having human servants probably won't want to part with the one luxury that never goes out of style. (This also suggests that few people will be able to get well-paying jobs as butlers to the super-rich.)

You might brush this off as a consequence of the recent recession, but it fits into an alarming trend that began about 25 years ago. According to recent research by the World Bank, members of the global upper-middle class (in practice, the middle-class in developed countries) were the only people who missed out on substantial real income gains from 1988 through 2008. Trade and mechanization obliterated well-paying manufacturing jobs, and the health-care and education jobs that replaced them paid much less.

Economists have found that there was basically no net job creation between 1990 and 2008 in those parts of the U.S. economy subject to foreign competition. The growth that did occur in the period was in finance, government, health care and education -- sectors where it's difficult to measure productivity, or where productivity is very low. No wonder anthropologist David Graeber was tempted to speculate that there may be someone "out there making up pointless jobs just for the sake of keeping us all working." Just think of finance, where many innovations mostly seem to be about devising clever new ways to extract fees. And how much health-care employment would really be necessary in a world where big companies didn't relentless ply children with sugar?

Consider that, in the not-too-distant future, we may not need very many people to produce most physical goods. Manufacturing employment has been in decline almost everywhere thanks to the rise of the machines. Even in China, the share of people working in factories seems to have been in decline since the mid-1990s. Perhaps, eventually, "working" will no longer be considered a necessary requirement for a basic standard of living, something Switzerland is considering. (On the other hand, many of us seem hardwired to need the stimulation of a job in order to stay sane.) Maybe we already have enough stuff and not enough leisure time to enjoy it. And if that doesn't work out, let's devote some human ingenuity to creating better pointless jobs.

(Matthew C. Klein is a writer for Bloomberg View. Follow him on Twitter.)