I often cite Baumol, et al's Good Capitalism, Bad Capitalism, and the Economics of Growth and Prosperity for the simple reason that many critics see only the Bad and suggest that this is all there is. Some writers cite crony capitalism. The irony is that many of the prescribed "fixes" open the door to more cronyism. Why is the IRS tax code, lengthy, bulky, complex, always growing and likely to remain so? Because it is chock-full of special treatments.
Increasing (reported) inequality is everywhere red meat for demagogues. It is time for a companion volume that describes "good" vs. "bad" inequality. First, the big three: (1) Not all gains or riches are ill-gotten; entrepreneurial success must be rewarded; it is indispensable because it assembles new ideas and new products that enrich others; the rule of law means that ill-gotten (and only those) gains are pursued at great risk; (2) We do not live in a zero-sum world; the gains by some are not at the expense of others; (3) Increasing poverty is not equivalent to increasing inequality; magically double all incomes and poverty declines while inequality increases.
And three more: (4) Most reports of increasing inequality are flawed; comparing statistical groups over time is meaningless; the key measure is economic mobility as seen via panel data; the rare panel data we have are inadequate; the most upwardly mobile, the immigrants, are excluded; (5) It's a safe conjecture that more Americans aspire than envy; economic growth provides more upward mobility possibilities; redistributionist interventions are likely to stunt growth -- and create more cronies and more inequality; (6) All of the acknowledged national income measurement problems and shortcomings make their way into the inequality discussion; not everything that counts can be counted; the greatest inequalities involve longevity, health, general well-being -- all beyond the scope of traditional national income accounts.
There are surely more. The problem we have is that the mere waving of the "inequality" flag often ends the discussion.