Liminal Independence

Central bank independence is a great case study in the complexities of democracy

Institutions matter. That seems almost axiomatic (everything is obvious in hindsight), but it’s a crucial observation and won some folks a Nobel prize—it is widely accepted today that good institutions are foundational for the economic development of nations. Central banks are institutions of government, and independent central banks are widely viewed today as necessary for the development of modern financial markets and good governance.

Yet central banks were not always independent; even in the US, Fed independence only really came to be in 1951 with the Treasury-Fed Accord. Recently, president-elect Trump has talked about more or less doing away with central bank independence, which has resulted in much controversy. Plenty of research (e.g., World Bank) has shown that central bank independence is associated with, e.g., lower inflation and more balanced budgets.¹ In emerging as well as developed markets, central bank independence is seen as a safeguard against naked monetary financing (the central bank printing money for the government to fund itself).

But this is not a note about that. This is a note about how we—society, collectively—solve problems with institutions, and how independent central banks, as an institution, are a case study in the strengths and vulnerabilities of modern democracies.

Institutions as solutions to problems

Institutions include government bodies and laws, but are generally conceived (by economists and political scientists alike) to be broader than that. Most would agree that institutions are inclusive of social norms, attitudes, and even international norms. In a seminal piece, North (1991) defines institutions as:

the humanly devised constraints that structure political, economic and social interaction. They consist of both informal constraints (sanctions, taboos, customs, traditions, and codes of conduct), and formal rules (constitutions, laws, property rights).

Some folks tend to think of external control of the individual—via government usually, but also potentially by oppressive social norms—as the result of some “other” power ruling over society. This tends to be the general view of how dictatorships work: some power sits above the rest and imposes its will onto them (for the famous exposition of this idea, see Hobbes 1651, The Leviathan). That certainly happens, and this framework is probably useful for thinking about the way authoritarian regimes work to some degree.

In reality, though, notwithstanding the existence of abuse of power writ large, institutions have often developed organically between members and classes of society as solutions to two basic and common sets of problems: principal-agent problems and commitment problems. In the other words, the binding constraints are issued horizontally across a flat social structure instead of vertically down a hierarchical social structure. It’s the players of the game writing the rules (because rules make games more enjoyable and fruitful—and less risky—to engage in).

The principal-agent problem is basically a fancy economists’ term for not being able to trust your hired help. This typically rears its ugly head in three classes of problem:

1. moral hazard (you hire someone to sell something for you; they steal it);

2. adverse selection (you hire someone to sell something for you quickly; they have some incentive to do it slowly);

3. information asymmetry (another econ buzzword!) (you hire someone to sell something for you in market A; they know you should be really be selling in market B, but don’t tell you because they don’t feel like making the commute to market B)

You end up bamboozled.

Then there are commitment problems, which are basically when you can’t credibly promise something. Maybe you want to buy a trinket from someone, but you can’t really do that if you don’t know that they have rights to said trinket (in which case you’d be stealing from some third party, which could end poorly for you depending on where you are). You’d like to do the transaction, but the promise that the trinket belongs to the seller just isn’t credible (solution: create and enforce property rights—voilà). Maybe you want to take payment to deliver some good from Boston to Barbados; the buyer might not be comfortable with that because you can’t promise the delivery credibly (because of Black Beard and other pirates of the high seas, or lack of maritime insurance when a storm sinks the ship).

North lays this out cogently in his piece by leveraging some clear historical examples. For example, we got bills of exchange (medium-term commercial credit) from trade fairs in twelfth and thirteenth century Europe. We got printed exchange rates and postal systems because of the challenges posed by trade between Europe and the rest of the world. Importantly, both sides of the trading voluntarily agreed to these rules, because it made life easier. If you subscribe to the view that regulations, rules, and governing agencies are always and everywhere the result of dictatorial imposition, you’d have to explain the ballooning number of international organizations and rules (given the absence of such an international hegemon, nonetheless government), such as the law of the sea, commercial international arbitral groups, and the International Organization for Standardization (you’ve heard of them because of ISO codes), to take a few examples.

The time-inconsistency dilemma in monetary policy

Commitment problems—specifically of the time-inconsistency dilemma variety—are clearly present in central banking, and always will be. If you’re an elected leader of government, so long as inflation isn’t absolutely crushing (and sometimes even then—see Argentina prior to Milei), you have a pretty clear preference in monetary policy: you’d like rates to be cut (or held very low) so that you can get reelected. In other words, you have short-term political compulsions (because you’re a rational actor responding to an incentive structure, not because you’re evil). If you care about the country indefinitely, though, you have a different mandate: you need to care about long-term sustainability and therefore stable inflation, even when that means short-term pain. In situations in which inflation is higher than it should be, there is a clear tension here: short-term political compulsion versus long-term sustainability.

Very easy solution here for political actors (who might all equally find themselves in such a situation): just make the central bank independent of political influence. Now it’s not your problem and you can blame the central bank when something goes wrong. Plus, per the previously cited research, markets like your country more, monetary policy is more effective, etc. Everyone’s happy.

On not birthing a Leviathan

But to all the liberal (in the small-L British sense) democrats in the room, you’ve also created another problem: the central bank, a steward of the people (ideally), is significantly more insulated from democratic governance. In other words, we’ve solved a commitment problem, but created a principal-agent problem.

Shouldn’t we feel uncomfortable about this? How do we square this circle?

There are obviously safeguards, and political independence is not synonymous with lacking accountability. Central bankers are, of course, ultimately answerable to elected representatives (or at least government broadly defined, if not a democracy). This is a perennial problem in democracies, and tends to become even more complex as the domains we govern become more complex; indeed, in this particular domain, a vibrant debate has recently emerged about the extent of Fed accountability in the US, with the newly nominated head of the Council of Economic Advisers for the incoming Trump administration highlighting concerns over the Fed’s accountability.²

Votes for the unborn?

In some sense, democracy in general has a time-inconsistency commitment problem. Assume the weighted average duration of a voters’ time horizon is 70 years (i.e., life expectancy, less a few years of cognitive dotage). Theoretically, the government’s time horizon is indefinite (unless they have some exigent reason—like a really severe war—to believe the country won’t be around indefinitely). The government then faces a paradox: it is at once meant to represent the will of the people—necessarily short-term (in a relative sense)—while also governing in the long-term interest of the nation, which outlives any given generation at any given point in time.

In essence, the government must govern for the unborn, who do not have votes. If tradition is the vote of the dead, what is the vote of the unborn? There would be, at minimum, some guesstimating as to what two generations in the future might want. Some things are obvious (like having a non-bankrupt social security/pension fund, or the absence of getting nuked), others less so.

This note poses no answers, only questions. I only wish to highlight that the status quo is a policy choice too, and modern democracy is more a balance beam routine than a ballet: as the world gets more complex and governing requires more expertise and more information than ever before, do we live in a world more defined by principal-agent problems than commitment problems?

Central banks might offer us some guidance. They represent a balance (of varying degrees of imperfection) between shielding important policy from short-termism and political meddling, while still providing accountability and democratic legitimacy. That balance—between insulation and accountability—will vary from country to country, as the legitimation mechanism differs in different societies, but it seems to be a broadly popular model, at least for now. Will other bodies of government go in this direction? Would it be good if they did, providing better policy held less hostage to the whims of politicians seeking short-term gains? Or would we slowly put more distance between the vote and the policy, grinding away at democratic legitimation and accountability?

Only time will tell.

You can follow me on Twitter (X) @ArnoldVincient, or connect with me on LinkedIn.

This work is independent from and not endorsed by the Yale Program on Financial Stability or Yale University; all views are my own.

1. This is not without contestation, I should note (see, e.g., the People’s Bank of China, hardly independent, but highly effective). But on balance, the data suggest that independent central banking has lots of benefits. For a dissenting view of the value of central bank independence, see Posen 1995.

2. In the name of brevity, I will not cover them here, but these critiques are eloquent and well worth reading.