The Good Government Trilemma

What is the respective role of democratic and other means of holding a government to account in a well-ordered polity? In one way or another, this question is the subject of live―and lively―debates in many (perhaps all?) democratic societies. In Canada, it manifests itself especially in controversies about the use of the Charter’s “notwithstanding clause”; in the UK, about the role of judicial review (especially of ministerial decision-making) and the Human Rights Act 1998.

At the risk of generalizing, my impression is that these debates tend to present themselves as clashes between the values of, for lack of better terms, democratic government and accountable government. One side thinks that the important thing is that elected officials get to run the show as they think best, subject to eventually being booted out by the voters. The other thinks that what matters is that the government be kept in check and made to answer for its actions on an ongoing basis, through some mix of elections, judicial supervision, and other accountability mechanisms, either internal to the government (such as ombudsmen and auditors) or external (NGOs and media).

To be clear, the democracy camp does care about accountability ― especially, that provided, or at least thought to be provided, by regular elections. For its part, the accountability side doesn’t deny the value of democracy, though it might argue that it’s a mistake to think of democracy in purely electoral terms. But there is, or so people think, a tradeoff between a focus on democracy, which calls for limiting the ability of non-electoral accountability mechanisms, especially the courts, to interfere with the work of government, and that on accountability, which requires these mechanisms to get in the government’s way with some regularity.

However, I think that the debate framed in this way is incomplete. It ignores a third factor that needs to be taken into account: the size of the government in question. This tends to go unnoticed because, whatever relative values they attach to democracy and accountability, virtually all participants in the debate are committed to keeping government big, by which I mean (substantially) bigger than a classical liberal nightwatchman state, let alone a Nozickian minimal state. I’m not sure quite where the boundary of big government lies, but I am sure that all governments in democratic states in 2022 (and for all I know the non-democratic ones as well) are on the big government side of it.

I would suggest that the apparent need to trade off between democracy and accountability is in fact only special case of what I will, again for lack of a better term, call the good governance trilemma. Of democracy, accountability, and big government, you can have two ― if you do things well; many polities won’t get two, or indeed even one ― but you cannot have all three. It is possible to satisfy the trilemma by choosing fractions ― a dose of democracy, a measure of accountability, a government not quite as big as one might dream of ― but the total cannot go above two, and it will certainly never go anywhere near three. You can’t have it all.

How does the trilemma work? Let’s start, as most people do, with big government a given. A government so big it takes scores of ― or, in the UK’s case, close to a hundred ― ministers of various sorts (or, in the US, agency heads) to run itself, to say nothing of the tens or hundreds of thousands of civil servants. This, of course, is not a Kornbluthian dystopia, but our present reality. A citizen who wanted to keep track of what the government is getting up to at a rate of, say, half an hour per minister per week would have a full-time job on his or her hands. And for at least some departments (think treasury or foreign affairs, for example, but there almost certainly many many others), half an hour per week hardly seems like it would be anywhere near enough to know what’s going on. Never mind ordinary citizens: even members of Parliament would struggle mightily to keep the tabs on the administration by virtue of its sheer size, to say nothing of the partisan and career incentives weighing on backbenchers, and of government obstructionism vis-à-vis the opposition.

Realistically, voters are in no position to keep such a government accountable (a point that Ilya Somin makes in Democracy and Political Ignorance: Why Smaller Government Is Smarter). This is why taking big government as a given, as most people today do, leaves you with a necessary trade-off between democracy and accountability. If such a government it is going to be accountable for more than an infinitesimal fraction of its innumerable decisions and actions, it will have to be made accountable to, or at least through, non-democratic or indeed counter-majoritarian institutions: courts, tribunals, ombudsmen, NGOs, and journalists. Alternatively, a big government can be made answerable to voters alone, with no judicial and other interference. But then it would be foolish to expect it to answer for even fairly major screw-ups, let alone the small-scale indignities a large administration visits on those subject to it every day that ends in-y ― not because it’s necessarily evil or even especially incompetent, let alone corrupt; but because it is run by fallible human beings. And these human beings, too, are the more likely to be pressed for time or out of their depth the more tasks the administration has been given.

If, however, one were willing to sacrifice government size, one could at least hope for a government held accountable primarily through electoral means. For one thing, as the government does less, there is simply less for courts and other non-democratic accountability mechanisms to sink their teeth into. (I have written about this here: if, for instance, government didn’t take it upon itself to regulate who can enter the country, we wouldn’t be debating the merits of judicial review of immigration decisions, which are a big annoyance to the UK government in particular.) But, less cynically, if government only does a few things, it is easier for citizens to keep track of those few things, and the odds of their using their vote to reward things done well and punish things done badly improve. Admittedly, I personally would not be all that optimistic about the degree of the improvement; but there ought to be some. By trading away government size, one could get more accountability and democracy, because democracy would be (more) sufficient to ensure accountability.

At the risk of making this post even more off-the-wall, I will add that a (very) small government system would make it possible to improve the quality of democracy and accountability further in another way. As Bastiat points out in The Law, so long as the government sticks to protecting people’s natural rights instead of being an expedient through which everyone hopes to live at the expense of everyone else, it doesn’t matter all that much whether suffrage is universal or equal: “If the law were confined to its proper functions, everyone’s interest in the law would be the same. Is it not clear that, under these circumstances, those who voted could not inconvenience those who did not vote?” Some form of epistocracy, or at least a minimal test of political knowledge, could be applied without causing the same problems it must under a big government. And a more knowledgeable electorate would likely be better at holding the government to account.

Of course, I don’t expect many people to share my interest in radically smaller government. Fair enough. But I think that it would be good if they recognized the reality of the trilemma I’ve outlined in this post. Its cause ― the difficulty for voters and even their representatives to keep track of a large administration ― should not be a matter of partisan controversy. It’s a reality that needs to be acknoweldged and responded to, whatever values will inform each person’s response.

And, as I said above, the possible solutions to the trilemma are not all-or-nothing matters. Government size, obviously, is not a binary choice. A government that withdraws from some areas of activity, or abjures some forms of regulation, could be more amenable to political accountability and less in need of non-democratic accountability at least to that extent. Conversely, a government that expands in some new direction may require the creation of entirely new accountability mechanisms to address this specific development. All this should be borne in mind even if the boot of big government as I have (sort of) defined it here remains firmly planted on our faces, and other body parts, forever.