Self vs effective

This post on Best Of A Great Lot is a part of a series on the subject of designing a new form of governance. Each piece aims to stand alone, but fits together on the Table of Contents.

Self-governance is the idea that we rule ourselves, and the main implementation of it in our current society is the act of voting for Congress and the President (or in Parliamentary systems, voting for your MP). Effective governance is the idea that we aim to get rules and systems that work well for us. The original aims of democracy in the United States were focused on self-governance and stability. In part this was a response to the previous governments that had been monarchical or unstable or both, but in part it was simply that they believed in their leaders.

In 1798, John Adams wrote.

We have no Government armed with Power capable of contending with human Passions unbridled by morality and Religion. Avarice, Ambition Revenge or Galantry, would break the strongest Cords of our Constitution as a Whale goes through a Net. Our Constitution was made only for a moral and religious People. It is wholly inadequate to the government of any other.

In other words, when we have venal, corrupt, or stupid politicians, we will have a venal, corrupt, or stupid government. More importantly, if we want to solve the problems of society, under democracy we need to elect people capable of solving them.

Somewhere in the 20th century we shifted to a belief that self-governance via representation wasn't enough to solve the hardest problems. The birth of the administrative state was and is built on the idea that experts know more, especially about the details, than politicians.

You can tell different stories about the administrative state, depending on whether you have a philosophical objection to it (it's not self-governance) or preference for it (decisions are made by experts), or you like or disagree with the particular outcomes of it. For those who are opposed to it, the administrative state is fundamentally a power grab by bureaucrats, and the power is taken away from Congress. Or it's a cowardly avoidance of responsibility by those elected. For those who are in favor of it, the administrative state is where people can be insulated from the political winds while they work out the actual best policies, or it's where the government can hire the actual experts to be responsible for the details that Congress can't focus well enough on.

With each passing year we both create and discover more complexity in the world. Today it's entirely reasonable to expect someone earning a PhD in a particular subject to take 5 years just to fully understand their small niche of the subject area. The idea that politicians can both be politicians and learn enough about many different subject areas to be able to craft and select great policy seems more and more unlikely.

I'm not sure who we should point to for the rise of the Administrative state. Heritage Foundation claims it was Woodrow Wilson's idea but they're somewhat biased, as they strongly oppose the whole idea and are partisan on the subject. Regardless, somewhere in the early part of the 20th century, Congress decided to start deferring the details of a policy area to an agency staffed with bureaucrats and experts. Since then, this pattern has become common and (at least mostly) generally accepted.

In writing about how the courts are (in 2022) re-evaluating whether this governmental design is constitutional, Matt Levine sums up the design of the system pretty well.

Some federal law is made by Congress, but quite a lot of federal law is made by government departments and administrative agencies. In many cases, Congress passes fairly general laws, and those laws instruct the relevant agency to write rules implementing the laws, and then the agencies write more specific rules. Sometimes these rules just fill in details in a comprehensive statutory scheme. Other times the agencies have pretty broad mandates to write rules that are in the public interest, and they get to set their own agendas and decide what that means.

He also describes the pros pretty concisely.

There are obviously good reasons to do things this way. Congress does not have time to write all the rules, so delegating rulemaking to agencies is efficient. Congress also has limited subject-matter expertise: The SEC knows more about securities law and financial markets than the average congressperson, so it makes sense for the SEC to write most of the securities rules. The rulemaking process is often both more flexible and more thoughtful than the legislative process; agencies have to consider public comments and explain their reasoning in a way that Congress does not. Agency rules can be overturned by a court if they are "arbitrary and capricious," which is not generally true of laws passed by Congress; Congress can be as arbitrary as it wants.

His list of cons:

There are also objections though. Congress is elected, and the SEC isn't. Letting agencies write rules is more technocratic but less democratic; the agencies might be more captured by industry or just by longtime staffers who are less politically accountable.

I think he's missing or underselling some of the key challenges by quite a bit, especially his brief description of them as "less accountable", but I'll return to this in more detail later.

This is fundamentally a tradeoff between self-government and effective government. In this (implicit) system design, if it were working well, Congress is the place where the big disagreements between sides of our culture get worked out to the point that a broad directive can be handed to a regulatory agency to work out the details. We agree that investing is dangerous and there are a lot of people out there who are hoping to defraud others (for example, most of the cryptocurrency industry), and so the SEC is given the mandate to set up rules and systems that protect people from fraud. The people who work at the SEC are, in theory, experts who have studied this particular topic deeply and are constantly paying attention to it, while Congress is bouncing back and forth between high profile issues and cannot stay focused on any one of them, let alone develop the level of expertise necessary to write great law.

As a goal for a system, this is pretty good. We want self-governance, and we also want the best decisions, i.e. effective governance. Is it actually necessary that we have to tradeoff between the two? Well, until someone designs a system that maximizes them together, this is the best we've come up with. We have this idea that experts know more than the rest of us, and for those of us who have read the comments threads on the internet, it seems like a plausible idea in aggregate.

Unfortunately, as we'll return to as a theme, this model isn't delivering.