"there are tough trade-offs to be made, and that while there may be a role for redistribution, it comes at a cost: It reduces innovation and growth"

By Stan Veuger of US News & World Report. Excerpt: 

"The story Aghion et al. tell is as follows. There are workers and capitalists in their model economy. The capitalists receive income from the firms they own that derives from their proprietary cutting-edge technology if they recently innovated or from generally available technology if it's been a while since they or their parents did. The workers either receive wages or become capitalists themselves by innovating and replacing the incumbents. Mark-ups on products and services based on cutting-edge technology are higher than from generally available technology, so the more innovation there is, the more income goes to the capitalists. Of course, the more productive research and development there is, the more often new entrepreneurs join the capitalist class, and the higher social mobility is – unless barriers to entry keep entrepreneurs out. And if research and development is more productive, the economy grows faster, ultimately benefiting workers as well.

Now, that's the theory. What Aghion et al. do next is test it using data on the United States. To measure innovation, they look at patents granted per capita, between 1975 and 2010, in all 50 states and D.C. When they relate that measure to year-by-year measures of the income share earned by the top 1 percent, they find what their model predicts: More innovative states are also more unequal. And that's not just because the wealthy few file more patent applications. They exploit variation in innovation driven by whether states have members of Congress on Appropriations Committees and by innovation in other states that they interact with a lot to show that such types of quasi-experimental variation in innovativeness, not driven by the mere presence of wealth, also actually causes inequality.

But only when we measure equality as the share of income that goes to the 1 percent – broader measures of inequality are not much affected; those appear to be the product solely of some of the other trends I mentioned before, like skill-biased technological change.

That may sound sad: Can we really only have iPhones if we restart the Gilded Age? Oh no, the authors say, do not fret! Remember the social mobility stuff? Well that shows up in the data as well. Areas that innovate a lot show precisely the kind of Schumpeterian dynamics that make income inequality a lot more palatable to most observers who are not of the bitterly envious variety: There may be big gaps between rich and poor, but the poor today still have a reasonable shot at becoming rich tomorrow. And do you know what areas get a lot less of both innovation and mobility? Places with tons of lobbying activity, where the incumbents keep the innovators out.

All in all this research suggests that some 20 percent of the increase in the share of income going to the top 1 percent since the mid-70s was caused by innovation alone. That may not sound like that much – but of course, we can't just eliminate only the inequality we don't like and keep the inequality that incentivized people to innovate. Bad news, perhaps, for those who would like to spread the wealth around."