Privatization and the Law and Economics of Political Advocacy, Part 8:

This post is the second to last in my series on my upcoming Stanford Law Review paper on Privatization and the Law and Economics of Political Advocacy (see here for the technical paper). The next post will be my conclusion (where we'll be back to nontechnical talk), so the end is in sight!

In the last post, I explained two features of the theory: Why I concentrated on private firms and public-sector unions, and why I talked as if the private sector firms colluded with each other but didn't collude with the public sector (and to what extent this matters).

The previous posts have all elaborated on the simple model I put forth in the third post of this series. That model was characterized by a severe free-riding problem: The "dominant" actor (as defined in the fourth post — basically measured by that actor's proportion of total industry benefits) does all the industry-increasing lobbying, and the lesser actors free-ride off him. Because the public-sector union seems to be the dominant actor (I explained this in the fifth post), some amount of privatization — greater than zero, but not enough to change who's dominant — would always reduce industry-increasing lobbying, because the public sector would lobby less, while the private sector would continue to free-ride.

But there are more complicated possible models. In this post, I'll explain what happens when we relax various strong assumptions from the simple model. Sometimes, relaxing an assumption doesn't change the basic qualitative result. And sometimes, instead of unambiguously predicting that lobbying would fall, it predicts that the effect of privatization on lobbying is ambiguous — lobbying could go up or go down, depending on various empirical facts. Either way, this conflicts with the critics' view that privatization would affirmatively make lobbying worse.

First, I drop the assumption that money only buys victory for a given reform or candidate, and introduce the possibility that money can also change the substance of the reform or the candidate's position. This does not significantly alter the conclusion. Second, I drop the assumption that anti-incarceration political advocacy is fixed. I find that the effect of privatization on anti-incarceration advocacy is ambiguous (though pro-incarceration advocacy still falls with privatization).

The third and fourth sections show how privatization may have an ambiguous effect even on pro-incarceration advocacy. In the third section, I relax the assumption that all money is fungible and that all that matters is the total amount of money in the pot. Once we allow public-sector money and private-sector money to have independent effects, privatization has an ambiguous effect on pro-incarceration advocacy: Private advocacy rises, but public advocacy falls. In the fourth section, I introduce the possibility that the pattern of privatization, as we observe it today, is already the result of a political process where strong unions have successfully opposed privatization while weak unions have not. I find that exogenously increasing privatization in such an environment would likewise have an ambiguous effect on pro-incarceration advocacy, as it depends on the correlation between actors' influence in privatization politics and their influence in incarceration politics.

The bottom line is that, if one wants to argue that privatization will increase pro-incarceration advocacy, one must argue either, from outside the model, that the model is wrong, or, from inside the model, why privatization would increase private-sector advocacy more than it would decrease public-sector advocacy.

* * *