In 2011, Jacob Hacker wrote a ground-breaking paper in which he coined the phrase predistribution. Under Hacker's definition, predistribution refers to measures governments take to reduce or eliminate inequality in market incomes. This differs from redistribution, which Hacker uses to mean measures states take to reduce or eliminate inequality after market incomes have been distributed, for instance through taxes and government benefit programs.
As far I am concerned, there is no moral or political difference between the two. Predistributive institutions and redistributive institutions are both just institutions. What matters is achieving greater economic equality, not so much the precise institutional regime that we use to get there. If anything, I tend to find so-called redistributive institutions more attractive because they are easier to fine tune and strike me as more liberating.
But, as Hacker correctly points out, my view is almost certainly an outlying one. For cultural or other reasons, Americans tend to be more supportive of equality-producing measures that get baked into paychecks than they are of equality-producing measures that go through more overt government channels. As a result, the US has a very stingy welfare state and delivers much of its government spending through opaque, submerged mechanisms like tax credits.
If we are going to live in a society that shies away from after-the-fact redistribution, the only way we can hope to achieve greater economic equality is through the predistributive channel. It is in that context that the recent Demos report Underwriting Bad Jobs is most salient. Authors Amy Traub and Robert Hiltonsmith find that the federal government is, in one form or another, involved in the employment of at least 2 million low-wage workers, a figure that is higher than the number of low-wage workers at Walmart and McDonalds combined. The authors arrive at this figure by adding up the number of low-wage jobs that are funded by federal contracts, federal health care spending, infrastructure spending, and a few other sources.
Generally, reworking our predistributive institutions so as to deliver more equitable outcomes is a tricky affair. Getting private employers to do what you want them to do takes some serious trying. But when the federal government is the buyer of the services, it should not be nearly so difficult. The federal government could demand improvements in worker compensation and other workplace standards as a condition of receiving federal contracts, leases, and other monies. In some areas, the federal government could even do away with private subcontracting altogether, and replace it with public delivery by public employees.
By itself, winning higher wages for these 2 million jobs won't deliver the kind of equitable society that we should be aiming for, but it would make strides in that direction. A wage bump for those jobs could also have spilllover effects for other workers in related industries because employers in those industries would have to compete with these higher-wage jobs when trying to hire their own workforces. Of course, the potential spillover effects from said competition will only become a serious possibility when we get closer to full employment, which we are not anywhere near at the moment.
To be sure, boosting the wages of these 2 million workers through these kinds of measures will be somewhat messy and imperfect. But that comes with the territory of living in a society that puts so much value on reducing economic inequality at the paycheck level. So long as ours is a culture that fetishizes the pre-tax income over the post-tax income, and it is, these kinds of predistributive reforms are one of the few politically-viable mechanisms capable of putting a dent in our skyrocketing inequality.