Superminority

This is part 2 of my legislative series.

I argued in part 1 that the traditional legislative chamber is fundamentally flawed. Nevertheless, it would not have hung around so long if it did not have some benefits. So, while its rules are opaque and its operations are fraught with dirty tricks and obstruction, it has the virtue of being a reasonable picture in miniature of a country’s ideological diversity, at least to a first approximation.

And yet any diversity it has gets destroyed by the need to achieve consensus. In practice, consensus means 50%+1, so groups that fall below that threshold are effectively discarded, unless they can form alliances to get above that magic number. Why is this the case? It doesn’t have to be this way; there is a simple method for expressing the diversity of a chamber to a far greater degree.

I call this the superminority method, it simply involves lowering the threshold of acceptance to get more than one output. Want two outputs? Set the threshold at ⅓+1. Want three outputs? Set it at ¼+1. Want 99 outputs? Set is at one-hundredth + 1. I call these a superminority of n; i.e. a superminority of 2, 3, and 99 in these examples. Notice that a majority becomes a superminority of one, indicating its monolithic and destructive nature. This technique must always be paired with a downstream decision making process, such as a citizen jury.

Let’s return to the French Citizen’s Convention on the Climate (CCC) for a moment. In my last post, I argued that the structures under which the CCC operated should have been more pluralistic. Now we have the mechanism for doing this. Instead of a single set of managers, we go to the French National Assembly and have them appoint three sets of managers, using a supermajority of 3. Each set of managers would organize the CCC for one third of the allotted time.

More generally, the superminority structure can handle any proposal, whether laws or appointment of government officers—though always in conjunction with a citizen jury. Just have the assembly produce some number of alternatives (never less than 3) and then have the citizen jury choose from among them. But what does this change do to the chamber itself? Returning to the considerations of part 1, does the chamber now resemble its historical predecessors a little more?

It does indeed. The superminority makes the chamber more like the original English Parliament, which was created to advocate for non-royal stakeholders in dialogue with the Monarchy. Let’s say an issue is raised under a superminority of 3. In that case, members self-organize into groups according to ideology and common interest. There is tremendous incentive for them to do so, as failing to self-organize means irrelevance. At that point, these groups operate independently, bound only by a deadline to submit their proposals. They are freed from the great pitfall of consensus-based assemblies: that getting to a conclusion—any conclusion—depends on the good faith of other politicians. In essence, the chamber is now three distinct chambers, each of which has a particular viewpoint. But neither are they unconstrained. They have to internalize the common good to a better degree than their competitors, otherwise the jury will reject their offering.

We can see the dark side of the traditional assembly in the U.S. response to COVID-19. All sides agree that the U.S. is currently in dire need of another relief bill, but as of this writing we don’t have one. This is not an accident—the consensus-based approach favors those who act in bad faith, by allowing them to shut down the process. The superminority method eliminates this tactic. Once an agenda item is adopted, multiple options will be produced, those options will be advanced to a jury, and a winner will be chosen.

This guarantee of procedural completeness has a powerful effect on the members of the chamber. They must now internalize political conflict rather than externalizing it. A consensus-based system encourages public demonstrations of political zeal for a wide variety of reasons: to demonize one’s opponents, to put pressure on other players in the system, and ultimately to cause political rivals to fail regardless of the merits of their actions. All of this is based on the fact that an agenda comes with no guarantee of reaching a procedurally satisfactory outcome.

With procedural completeness in place, members must now internalize political conflict. Screaming that the other side is negotiating in bad faith is out: there are no negotiations. Just write your own bill and submit it by the deadline. In that sense, the different groups in the chamber need not interact at all. That’s a good thing. Direct interaction of politicians does not require good faith, it assumes good faith. In so doing, it gives a massive advantage to bad faith actors, who have no stake in the public good and are usually quite skilled at embarrassing their opponents.

This dichotomy—between internalizing and externalizing conflict—is a deep property of political systems. We can see this, for example, in plea bargaining for criminal cases. The vast majority [1] of criminal convictions in the U.S. federal system are obtained through plea deals. This allows prosecutors to externalize their duties, placing the public question of guilt or innocence in the hands of the accused. The solution? Get rid of pleas. Every defendant is effectively required to plead not guilty. If you want to put someone in prison, convince a jury or go home.

In the legislative context, members will now be judged mostly based on whether their viewpoint wins with the jury, just as prosecutors are (or should be) judged by success with traditional juries. The best legislators will internalize the policy preferences of the general public (as expressed by citizen juries). That’s not to say that screaming and shouting will cease, it just won’t mean as much.

This begs the question: what views do legislators internalize now? The answer is obvious: their big donors. But the why is also interesting. Our system (at least in the U.S.) is set up in such a way that big donors act as a kind of jury to politicians. When a politician does something a rich donor likes, that politician gets positive feedback immediately, and similarly for undesired actions. Further, rich donors have the resources to look at policy in a way that others cannot.

The electoral system as it exists now does not provide the same signal. Voters can only express themselves at election time, and cannot segment their vote based on issue. Voters without an existing interest in policy are simply responding to social cues and noise—in other words, the tools of politicians who externalize political conflict. The endless calls to “get money out of politics” can improve the situation by reducing the bad signals from donors that politicians tune into. But that does nothing to improve the signal coming from ordinary citizens. Only a system based on internalizing political conflict and resolving it with citizen juries can do that.