Committee size is the same as efficiency: the central bank versus the legislature

Any part of the government can be efficient or inefficient, depending on its size and how many conflicting responsibilities it has

Let's talk about different ways you can arrange important government responsibilities in your country.

Imagine your central bank is run by a committee of 5 people, but your legislature has 500 people.

That's a ratio of 100 to 1.

Roughly speaking, your central bank is 100 times more efficient than your legislature.

What if a recession afflicts the national economy?

Economics teaches us that we can fight a recession with with either fiscal policy or monetary policy.

During a recession:

Fiscal policy means the legislature mandates that the government takes on debt so it can build more stuff: roads, bridges, airports, military equipment. Building all this stuff will give more people jobs, and then those people can use their money to buy more stuff, and thus stimulate the economy, lifting it out of recessison. (Fiscal policy can also mean expanding unemployment benefits, an important tool to fight income inequality, in which case the government gives extra money to people who are unemployed, so they go buy more stuff and thus stimulate the economy and thus lift the economy out of recession.)

Monetary policy means the central bank will lower interest rates, so people are able to take on more debt, so they can buy more stuff, so they can stimulate the economy, and thus lift it out of recession.

Suppose a crisis occurs, who can respond first, the legislature or the central bank? Obviously the central bank. Why? Because it is 5 people. By contrast, the legislature is mired down in intractable, exhausting infighting, because it is 500 people, who have many competing interests and priorities, and they have huge egos and many of them are angling to get up into a leadership position.

Therefore we tend to fight recessions with monetary policy instead of fiscal policy. And for a mild recession this works. If the economy was running hot, and interest rates were 7%, and then a mild recession occurs, then maybe cutting interest rates by 4% is enough to get back to a strong economy. So the interest rates drop down to 3%, and at that lower rate businesses are able to start borrowing again, and building, and hiring people, and so the economy revives.

But what if the economy was already weak, and interest rates were 2%, and then a recession strikes? Suppose you again need to cut interest rates by 4%? But that would lower interest rates to -2%. That can't work. No bank will loan you money at -2%, since that would mean that they have to pay you money, instead of you paying them. So the banks stop lending. So the economy screeches to a halt. We are up against "the zero lower bound." Interest rates cannot drop below 0%, so we have reached the limits of what monetary policy can do.

At that moment, only fiscal policy could save us. And at that moment we will wish we had a had a smaller, more efficient legislature. (And this is not hypothetical, since we ran into the zero lower bound for several years after 2008,)

Let's flip this around now.

Let's imagine a country where the central bank is run by a committee of 500 people, but the legislature only consists of 5 people.

When a recession strikes, how do they fight it? Obviously, they fight it with fiscal policy.

Why is that? Well, the central bank is mired down in intractable, exhausting infighting, because you have 500 people, who have many competing interests and priorities, and they have huge egos and many of them are angling to get up into a leadership position.

By contrast, the legislature is just 5 people, so it can quickly and efficiently cut taxes while taking on debt to build more stuff: roads, bridges, airports, military equipment. Building all this stuff will give more people jobs, and then those people can use their money to buy more stuff, and thus stimulate the economy, lifting it out of recessison. They might also offer things like extra unemployment benefits, to help take the edge off the worst kinds of income inequality.

To a first approximation, the size of the committee will tell you the efficiency of the committee. If two committees have an overlapping concern (maintaining the health of the economy) then whichever committee is smaller will tend to take the lead. The second most important variable, which determines the efficiency of a committee, is how many conflicting responsibilities it has.

Many will argue that the larger committee is more democratic since it will contain more people from more diverse backgrounds, offering a greater diversity of opinions. And yet its size will keep it from taking action, and so its democracy will be meaningless, since it will cede leadership to the smaller and less democratic committee, which at least as the advantage of being able to turn opinions into real-life action. And in the end, real-life action is the goal of government.

One thing to note about the above is that monetary policy has a definite limit: the zero lower bound. By contrast, fiscal policy has no hard boundary. It is clear that very high levels of debt are probably bad and should probably be avoided, but right now Japan's government has total debt equal to 263% of GDP and Japan is doing fine. By contrast, the USA only has debt of 120% of GDP, so it has a long way to go before it gets to Japan's level, and even if it got to Japan's level, it would probably still be fine.

Therefore, fiscal policy is more flexible than monetary policy, it can be pushed further, it has less hard limits, it can also be better targeted to limit income inequality. Therefore we should prefer to use fiscal policy, rather than monetary policy, if only the committee that oversees fiscal policy was as small, and unconflicted, as the committee that oversees monetary policy.

But lets say you think debt is very, very bad. What if you feel that bringing down the debt should be our number one priority? In that case, when the economy is strong you will want to cut spending while raising taxes, to run a surplus, and thus bring down the debt. Is this easy to do when the legislature has 500 people? Of course not. Would this be easier to do if the whole legislature consisted of just 5 people? Obviously.

So whether you are the type of person who favors deficit spending to fight recessions, or you are the type of person who is eager to bring down the debt during good times, or you believe in both of these things, you can clearly see the benefit of having a legislature that only has 5 people in it.

But wait, what about all of the other responsibilities of the legislature? We have so far only talked about a committee of 5 people in the context fiscal policy.

Would this committee of 5 people be diverse enough to represent the racial diversity of a multicultural society? Absolutely not.

Would this committee of 5 people be diverse enough to represent the religious diversity of a multicultural society? Absolutely not.

Would this committee of 5 people be diverse enough to represent the professional diversity of a highly specialized, technologically advanced society? Absolutely not.

Would this committee of 5 people be diverse enough to represent the geographic diversity of a widespread society? Absolutely not.

So this committee of 5 fails some basic tests of democratic governance. It is too small to function as the legislature. But could it function as one of many legislatures? Yes, and if you understand this, then you understand what lead me to the idea that I now call "demodexio."

But wait, you say, it's important to have all legislative power concentrated in one legislature. To the extent that we need a small committee that is in charge of fiscal policy, we will simply set up a committee inside the legislature, as part of the legislature, and it will set all taxes and spending, and thus achieve what has been suggested above. Won't that work?

For sure, during the mid to late 20th Century the USA's legislative chambers had committees that oversaw the budget process, and for several decades they did a fairly good job of managing taxes and spending. But that process has broken down. And I believe it has broken down permanently, it will never be restored. The old system was referred to, in the general case, as “regular order”, and in the specific case of the budget, a process first specified by the Budget and Accounting Act of 1921, it was sometimes described as the regular order budget process.

We covered some of the deep background history in other essays, but let me recap the highlights here:

1642-1688 — Britain goes through a civil war, a revolutionary period, then a reactionary backlash, and then finally the successful completion of what has been called the first bourgeois revolution. By 1688 Britain consolidates a new type of government, where the executive (the monarch) has their power limited by the legislature.

1688-1748 — the early phase of the Enlightenment sees many philosophers consider and discuss the new government of Britain, learning lessons from it. Finally, in 1748, Montesquieu publishes "The Spirit Of Law" in which he says a free society can be best maintained if the government has 3 branches, the judiciary, the legislature, and the executive, and if each branch is independent of the other. Montesquieu invents the phrase "checks and balances" to describe how the branches should limit each other.

1748-1883 — there followed a century during which government was simple enough that the executive and legislature could function without many internal specialized bodies. But after 1850, the growing complexity of society, and the need for specialists who have deep knowledge of particular areas, lead to the expansion of the bureaucracy. Perhaps the USA 1883 Pendleton Act (establishing a non-political, merit-based civil service) can be said to mark the boundary between eras.

1883-1946 — the legislature becomes more reliant on a system of internal committees to handle specialized work. The committees develop subcommittees as the process of specialization proceeds. The legislators intuitively and empirically understand that small committees are more efficient than big committees, and therefore if something is important it should be handled by a small subcommittee. The committee system reaches the peak of its prestige, and intellectual coherence, with the 1946 Congressional Reorganization Act.

1946-1994 — the committee system works for several more decades. But it only works when the larger legislature is willing to act as a rubber stamp for the decisions that are made in the specialized committees. Increased partisanship means the whole system falls apart. In the USA, the Democrats held the House for almost the entire time from 1932 to 1994, but when the Republicans regained control, the committee system was in trouble.

1994-2024 — With increasing frequency, a good decision would be made by some subcommittee, but the decision was then sabotaged by culture-war partisanship in the general assembly. This rendered the whole system of committees increasingly meaningless.

Our society continues to become more complex and more specialized. The need for highly specialized committees continues to become more urgent. The quality of governance will decline until such highly specialized committees can become the one and only way that our government creates real-life action.

What is the path forward?

A possible path is to move beyond the “3 branches of government” architecture that Britain pioneered in the 1600s and which Montesquieu analyzed in 1748. Too many conflicting demands have built up on each branch of government. Instead of 3 branches of government, perhaps we need 300, each highly specialized.

In "How to save American democracy: end the Imperial Presidency" we talked about ways we could break up the Executive branch into smaller, more specialized branches of government.

The same basic set of ideas also applies to the legislature.

Who would design this system? Clearly, the upper house of the legislature, which should lose the ability to pass ordinary laws, and instead retain only the ability to amend the constitution (see what we wrote in “There is one correct way to amend a Constitution”). Thus the sole job of the upper house would be to play the role of architect, and design a beautiful system of specialized committees. And each committee would have the power to create law. Each committee would be its own branch of government. Perhaps the public would vote for each branch independently.

Each committee would be small, and therefore efficient at its task, but collectively all of the committees together would be larger than the current legislature and therefore better able to represent the real racial, religious, professional and geographic diversity of a widespread, complex, and multicultural society.

There might be other architectural decisions that would work, but in writing this essay I have been influenced by the long years that I have spent working as a software-architect. I know that complex data systems will tend to start off as a monolith, but later evolve into "microservices" meaning dozens of small, highly specialized apps. (A basic tutorial on this ideas says “The decomposition of microservices architecture is a strategic approach to breaking down complex systems into manageable, autonomous services.”) And it is likely that our legislature needs to follow this evolution. We saw the legislature evolve from a monolith in the 1700s to a system of highly specialized committees and subcommittees by 1946. The next step would be greater independence for the subcommittees so they cannot be sabotaged by rancor in the general assembly.

This much is certain: the government will continue to grow bigger and more complex as our society grows bigger and more complex. And as the government grows bigger, it's individual parts need to be better decomposed into smaller and more specialized parts. (Here again, I am influenced by my experience as a software architect, since for any growing complex system we will give the advice “it's individual parts need to be better decomposed into smaller and more specialized parts.” This seems to be a general pattern that pervades many aspects of our existence. Even in biology we see the same pattern: the Eukaryotes needed to internally develop many high specialized organelles before the Eukaryotes could evolve into bigger life forms. Meanwhile the Prokaryotes remain internally unspecialized, and therefore they’ve never grown into anything more complex.)

Indeed, even the emergence of 3 independent branches of government, back during the 1600s, was already a step towards decomposition and specialization, when compared with the earlier system where the monarch simply contained within themselves all the functions of government. And we are much further advanced now. We need our government to evolve to meet the demands of our modern levels of specialization and complexity.