Central Banks and Fiscal Policy

Or, how central banks become passive

You are reading Economic Forces, a free weekly newsletter on economics, especially price theory, without the politics. Economic Forces arrives weekly in the inboxes of over 11,000 subscribers. You can support our newsletter by sharing this free post or becoming a paid subscriber:

There is a lot of talk on the internet about whether monetary policy or fiscal policy is the dominant factor in the determination of the price level and inflation. Today, I’d like to write a bit about how to think about what monetary policy is. I would also like to discuss how particular views on the role of monetary policy might lead to fiscal dominance even if it is not the intention of the central bank.

---

What is it that central banks do? People often say that central banks are lenders of last resort. In a sense, the central bank serves as a bank for other banks. When banks cannot find anyone else to lend to them, they can go to the central bank. But central banks do other things as well. Some central banks have a role in regulating commercial banks. Yet, most of the attention that central banks receive is centered around the fact that central banks have a monopoly over currency issuance and that central bank use their monopoly power to conduct monetary policy.

But, what is monetary policy? And how does this differ from these other things that central banks do? Or does it differ at all?

Some people might answer that monetary policy has something to do with interest rates. Others might say it has something to do with the money supply. But that doesn’t really answer what monetary policy is. Those answers are more about the tools of monetary policy.

To think about monetary policy, let’s first imagine a world without monetary policy. Imagine a world in which gold is money, but banks can issues claims to gold. We will call these claims bank notes. In this sort of a world, the unit of account would typically be defined as a particular quantity of gold. For example, one “dollar” might be defined as 1/20th of an ounce of gold. This would necessarily imply that the dollar price of an ounce of gold is fixed at $20 per ounce. Of course, gold is just some commodity and like all other commodities, its value is determined by the market. Within this system, however, the dollar price of gold cannot change. One ounce of gold costs $20 by definition. Although the dollar price of gold cannot change in response to supply and demand, the real price of gold can. Thus, fluctuations in the supply and demand for gold cause fluctuations in the price level.

In a competitive environment with banks issuing their own bank notes, they are effectively just issuing their own brand of claims to gold. If you bring the little piece of paper back to the bank, the bank will buy it back from you with gold. Since each brand is distinct, over-issuance of bank notes by one particular bank isn’t going to affect the price level. Instead, the notes are going to end up being redeemed for gold.

In this sort of a world, you can imagine the types of problems that banks might run into. The most obvious problem is not having enough gold. Part of the bank’s business is issuing claims to gold. If they issue too many claims, there are going to be more people than usual wanting the redeem those notes for gold. The bank might start to run out. One thing the bank could do is sell some of its assets, directly or indirectly, for gold. Another thing the bank could do is try to borrow gold from other banks.

But what if a lot of banks end up with this problem? The scramble to buy gold might cause a significant deflation. Since loans are typically in nominal terms, this deflation could be very costly in terms of increasing aggregate defaults. And if this problem is true of banks generally, it will be very hard for one bank to convince another bank to lend them some gold.

Some people look at a scenario like this and say, “ah ha! This is why we need a central bank.” A central bank could do one of two things in this scenario. One thing the bank could do is serve as a lender of last resort, lending gold to commercial banks that need gold. Another thing a bank could do is conduct what we call open market operations. The central bank could go out and buy assets from the commercial banks. The central bank could buy assets from the commercial bank in exchange for gold, or the central bank itself could issue a claim to gold to the commercial bank. As long as commercial banks believe the central bank can make good on this claim, or that their customers believe the central bank will make good on this claim, that can work.

There are a couple of things we might say here. First, it is still not clear that this requires a central bank. This seems like something that some sort of clearinghouse could do. Second, this example suggests that the purpose of monetary policy is about liquidity. When banks have liquidity problems, this can lead to fire sales that turn an illiquid bank into an insolvent bank. When banks have liquidity problems under a gold standard, this can also lead to a significant, unexpected deflation as banks and their customers scramble to get gold.

An astute observer might say, “hey, the problem here is the gold. If you just replace the gold with the pieces of paper issued by the central bank, you would solve this problem. The central bank could just manage the supply of those pieces of paper to match demand and this solves the problem associated with deflation.” And yeah, sure, why not? That certainly sounds like something a central banker would say.

This makes monetary policy at modern central banks more consistent with the narrative I have weaved here. The settlement layer of the contemporary monetary system is about the supply and demand for liabilities created by the central bank and the central bank can create as many or as few of these liabilities as they want. Since these central bank liabilities are the most liquid asset and this asset is only provided by the central bank, the central bank is in the liquidity business. What we call monetary policy is the central bank’s attempt to supply enough of its liabilities to match demand, but not too much. The way that it judges whether it is providing too much or not enough is by looking at prices. If prices are rising, on average, over time, then the central bank is providing too much liquidity. If prices are falling, on average, over time, then the central bank isn’t providing enough liquid liabilities.

Some people might talk about central banking in terms of the interest rate or the money supply or the inflation rate, but fundamentally, it is about the supply and demand for liquidity.

---

With that as a backdrop, I would like to discuss a recent speech by Dallas Fed president Lorie Logan. In the speech, Logan clearly thinks about monetary policy as managing liquidity. Paying interest on reserves enables the Federal Reserve to supply far more of its liabilities to the banking system than would otherwise be possible and without the effect on the price level and inflation that a typical enlargement of the balance sheet would produce. Logan sees this as a good thing because it has led to significant declines in things like daylight overdrafts of commercial banks.

But two things were concerning to me about the speech. The first was that it seemed to include an implicit argument that the size of the balance sheet is not at all about macroeconomic implications or outcomes. The size of the balance sheet is entirely about liquidity. As my previous long-winded discussion detailed, however, what seems to be important about thinking of monetary policy as liquidity management is that the objective of this liquidity management is to minimize the costs associated with financial instability or macroeconomic instability that results from liquidity problems in the commercial banking system. (Of course, whether central banks can actually do this well is a separate issue.)

The second concerning thing about the speech is that she emphasized not only liquidity within the banking system, but also liquidity at non-banks. In particular, she argued that the Fed needed to pay attention to the liquidity needs of both banks and non-banks. The Fed’s liabilities are bank assets. The U.S. Treasury’s liabilities are non-bank assets. Thus, in Logan’s view, if the Federal Reserve is going to be in the liquidity management business, it must make sure to manage liquidity for both banks and non-banks.

It is not clear to me why liquidity differences wouldn’t show up in prices and to the extent these differences create arbitrage opportunities, these differences should be arbitraged away.

More importantly, the crucial point here seems to me to be that focusing on the Treasury market and liquidity in the Treasury market makes it more likely that fiscal policy will be the dominant factor in the determination of the price level and inflation. For example, if the Federal Reserve is trying to ensure liquidity in Treasury markets, when there is an excess supply of U.S. Treasury securities, the Federal Reserve is likely to be the buyer of last resort to ensure that the market is well-functioning. In doing so, however, the Fed risks expanding its balance sheet with the net effect of this behavior being that the Fed essentially monetizes a proportion of the debt through these actions. Thus, if the Federal Reserve becomes sufficiently concerned with the liquidity in the Treasury market, then the Federal Reserve risks becoming the passive player and fiscal policy becomes dominant.

It seems to me that the role of the Federal Reserve and any other central bank is liquidity management, but only because central banks are the sole supplier of the relevant liquid asset. It does not seem to be the role of the central bank to perform liquidity management for all other relatively liquid assets. To do so not only opens the door to fiscal dominance, but also to other issues related to political economy. The greater the role that the Fed begins to play, the greater the role that will be demanded by political forces.