Policy and Income Inequality

Dean Baker weighs in on the dispute over the sources of income inequality  (original Krugman article, follow up email, DeLong summary, Mankiw response) :

Income Inequality: Missing Mechanisms, by Dean Baker: There has been a raging blog debate, following in the wake of some recent Paul Krugman columns, as to whether the rise in income inequality is due to policy or the natural workings of the economy. While Krugman indicated that he believed the policy view (promising details later), many of the economists weighing in have said that they don't see any policy mechanism(s) that could explain the rise in inequality.

Perhaps I have different eyes (or maybe I don't have sufficient training in economics), but I see the mechanisms almost everywhere. There is a nice example in the news today. A judge ruled that Northwest's flight attendants can't go on strike to oppose the wage cuts that the airline is unilaterally imposing following in the wake of its bankruptcy.

In other words, a U.S. judge is telling workers that they will go to jail if they refuse to work for the wages that Northwest wants to pay them. (I know, I'm skipping some steps here.) Judges don't have to threaten workers with jail for refusing to agree to employers' demands. This is a policy decision.

For those who find the labor-management framework difficult to understand, imagine that Northwest purchased flight attendant services from the Flight Attendant Services Corporation (FASC), which is a corporation wholly owned by the workers it employs. Suppose that FASC tells Northwest that it will not provide services for the lower fee it is now offering. Would any judge threaten FASC with jail if it didn't agree to offer its services on Northwest's terms?

There are many other examples of rulings that have gone against labor in the last quarter century. For practical purposes, it is now legal to fire workers for organizing a union (the penalties are a joke). This is not the whole policy story; I have much more in my book, The Conservative Nanny State. It's short and free (and the summary is even shorter), so you have no excuse not to read it, unless you want to remain as ignorant as an NPR reporter your whole life.

Update: If "My sense" is a valid econometric estimator, then Brad DeLong believes 40% (one fifth plus two tenths) of income inequality is attributable to changes in policy and power relationships:

I agree with Dean: this is a bad decision. ... This is a judge who is not doing his job properly. But in a big country this is a (relatively) little decision. My sense (and it is just a guess) is that declining unionization and union power might account for perhaps a fifth of the widening in income inequality; that reductions in the value of the minimum wage might account for a tenth; and that legal changes that have shifted the balance of power within the corporation toward CEOs might account for another tenth. I have a hard time finding other policy changes that have a big impact--and only a portion of declining unionization and union power is due to changes in government policy since the 1970s.

Brad says "it is just a guess," so, with "My sense" plus or minus two wetware standard errors, my sense is that this could account well over half of the variation in income inequality.

Update: An email talks about the econometric evidence that does exist:

I'd add that estimates based on the union-nonunion wage differential, or on the time series effects of local changes in union density, aren't reliable ways of inferring the effects of a change from an economy in which unions are marginal to one in which they're dominant and back again.