Recent Posts

Sunday, December 20, 2020

Fiscal Dominance As Obfuscation

One phrase that keeps being used by mainstream economists is "fiscal dominance." This is a fuzzy term that is being used to express disapproval but yet appearing to be a neutral technical term. However, the simplest way to view the discussion is that neoclassical economics was ambushed by reality, and that they are hiding behind jargon to distract from this.

What is Fiscal Dominance?

Although there might exist a formal definition of fiscal dominance somewhere, in practice, usage in commentary is somewhat fuzzy. A recent example is in the speech "The shadow of fiscal dominance: Misconceptions, perceptions and perspectives", by Isabel Schnabel -- a member of the executive board of the ECB.
At the time of the Maastricht Treaty, high government debt was seen as a major threat to central bank independence, and it was feared that fiscal dominance could induce a central bank to deviate from its monetary policy objectives, endangering price stability.

This was not just idle speculation. History is full of examples of high government debt eventually being resolved through higher inflation and financial repression.[1]
The examples of "high government debt eventually being resolved through higher inflation and financial repression" is based on an article by Carmen Reinhart and M. Sbrancia. I have not seen the article in question, but based on Carmen Reinhart's track record in the area of fiscal policy (routinely conflating currency sovereigns with ones within a pegged currency system), I think the fact that no examples are given is telling. If we confine ourselves to relevant examples for developed currency sovereigns, we are stuck with the relatively small sample of developed welfare state economies in the post-World War II era. Although "financial repression" existed, inflation was generally minor until the 1970s -- when debt levels had been reduced by growth.

However, the precise definition of fiscal dominance has been left aside. In practice, it appears to be as follows: central banks are forced -- for some unspecified reason -- to set aside their inflation mandates because of fiscal policy. However, there seems to be an even vaguer version: central banks have to take into account fiscal policy when determining monetary policy.

Since no developed country has faced high inflation in the past decade, one would need to be wildly revisionist to believe that central banks have set aside their inflation mandates because of fiscal policy. In fact, central bankers have failed in their promise that they could hit their inflation targets solely with monetary policy: they generally missed to the downside in the post-Financial Crisis period. They came up with an excuse -- it's all the Zero Lower Bound's fault -- something that they neglected to warn anyone about in the 1990s when inflation targeting was being pushed.

So we are stuck with the weaker version: they need to take into account fiscal policy when setting monetary policy.

At the Operations Level, Monetary Policy and Fiscal Policy are Joined at the Hip

The first thing to note is that it should surprise nobody that monetary policy and fiscal policy are linked. At the level of monetary operations, the central bank is the central government's banker, and operations need to be coordinated.

The pre-2020 Canadian framework is the simplest one to comprehend this issue. I described a simplified model that captures its working in my handbook, Understanding Government Finance. (In 2020, the Bank of Canada undertook Quantitative Easing operations, which changed things.)

In that system, private banks have an end-of-day target of a $0 balance in the payments system, which are normally described as settlement balances. In systems with reserve requirements (typical in Economics 101 textbooks), these balances are called reserves. Meanwhile. the fiscal arm of government banks at the central bank.

The net result is that the central bank needs to cancel out all the net transactions between the private sector and the central government every single business day. Some of these transactions would be the purchase/sale of banknotes as they are taken into/out of circulation, but the rest would be net monetary transfers of the central government.

This means that the central bank has no choice to accommodate every single fiscal operation of the government: sending out payments to the private sector, and taking tax payments, and bond issuance. If the central bank did not coordinate with the fiscal arm (e.g., Treasury), the banking system would either be short of required settlement balances, and/or the central government bond auctions would fail. Although some economists appear to believe that central bank independence allows central bankers to arbitrarily force the central government to default by disrupting the payments system, there is no evidence that there is legal support for that position in any country with a sane monetary system. (The Euro area provides the exception.)

The implication of this is that it makes no sense to pretend that monetary operations can be independent of fiscal policy. However, this tells us nothing about interest rate policy.

Lender of Last Resort

The next area where central banks need to take into account "fiscal policy" is in credit policy (in a wide sense). Central banks are lenders of last resort, and one of their most basic tasks is to ensure the stability of the wholesale financial system. The central bank is a bank to other banks (hence the name), and it can only implement policy via a functioning banking system. 

Although many mainstream economists want to view monetary policy as twisting an interest dial to get optimal outcomes for society, monetary policy will have no effect on the economy if the intermediary banks are melting down and are unwilling or unable to borrow and lend. The paralysis of the financial system will immediately freeze the real economy, as 2008 demonstrated.

The most awkward problems come from sub-sovereign and private sector borrowers. Central banks can either liquidate the financial system, or engage in lender-of-last-resort operations. These were across the board in 2008, but in 2020 we saw policies such as the Bank of Canada launching extensive repo operations in provincial debt. The central government could have bailed out those borrowers with fiscal operations, but lender-of-last-resort operations make far more sense, since they can easily be wound down once the financial system regains its footing.

These operations can blur into operations with central governments. In the case of the ECB, there is no "central" government, and all of the "sovereigns" are in the same position as Canadian provinces. The ECB had no choice but to bail them out to keep the peg system from disintegrating. This reality is distressing for the neoliberal consensus, but is completely unsurprising for heterodox critics of the euro framework.

Even floating currency sovereigns see interventions in central government markets -- QE, the Great Treasury Repo Crisis of 2019. Although doom-mongers will claim that these interventions are needed to keep the central government from defaulting, the reality is that what was buckling was private sector government debt trading infrastructure. Even if neglected, the central government would get financed, but term premia might blow out in order to draw in risk allocations from end investors to reduce balance sheet stress of dealers. This would raise borrowing costs across the board, messing up the transmission of monetary policy to the real economy. The central bank has little choice but to stabilise the wholesale debt markets, or else its interest rate policy lever would end up compromised.

Taking Debt into Account When Setting Interest Rates

Once we put aside operations needed to stabilise the wholesale debt markets, we are left with a very vague notion: the central bank needs to take into account central government debt levels and/or the deficit when setting interest rates. 

(The linkage from debt stocks to the flows in the real economy shows up as the result of interest income flows, which are the average interest cost times the stock of debt. Otherwise, there are no "magic levels" for the debt/GDP ratio.)

The question is: why should the central bank care?

This is where mainstream theoretical squishiness about government debt shows up. Do they believe that floating currency sovereigns like the United States, Canada, and Japan can be forced into involuntary default? At present, when MMT discussions come up, the usual story is that "we knew it all along that those governments cannot be forced to default by bond vigilantes." However, if we believe this, there is no reason for central banks to "bail out" such central governments. (Once again, the ECB is not in this boat.) The government will always get financed at the rates implied by the risk-free curve, and the front of that curve is under complete control of the central bank. (I put aside issues of term premia, but we need to keep in mind that the average duration of issued debt is fairly short, since the amount of long-term debt that needs to be rolled over every year follows an inverse relationship with its maturity.)

To repeat: there is absolutely no reason for the central bank to believe that interest rates need to be set any level to avoid default. Anyone who suggests otherwise has to provide the mechanism for bond vigilantes to force a default. There is a deafening silence of plausible scenarios in which that happens.

What does the central bank need to worry about? We need to circle back to Functional Finance 101: the constraint on a floating currency sovereign is inflation.

"Fiscal dominance" just means that the central bank needs to take into account fiscal policy when setting rates. In particular, it needs to take into account the interest income channel, which implies that raising rates raises interest expenditures by the central government, and may be inflationary.

This is literally what heterodox economists have been pointing out for decades. It is only a surprise to economists who have effectively ignored fiscal policy in their models. "The government needs to obey the inter-temporal governmental budget constraint, so fiscal policy has no effect on the economy!" That was a ridiculous analytical assumption, and so it needed to be abandoned.

No Magic Wand to Wave

The whining about fiscal dominance is just a way of wishing that debt/GDP ratios reverted to relatively low levels, so that the interest income channel was negligible. However, there is no good way to get there from here.
  • The government could decide to default, creating the financial crisis conventional economists were worried about. This would be an insane overreaction to the analytical failures of mainstream models. Just get better models, and deal with it.
  • Rapid nominal GDP growth would shrink debt ratios, hence reduce the importance of the interest income channel. However, unless real GDP growth magically rises a lot, that means we need a good clip of inflation. In other words, we need to have high inflation so that central bankers can go back to pretending that they can control inflation solely by raising interest rates.
  • Austerity policies would take decades to reduce the ratios, as seen in the last cycle. Since this option conforms with the neoliberal ideology, it is a safe bet that this is what will be recommeded.
The sensible solution would be to accept that MMTers were correct, and that fiscal policy settings ultimately control inflation. This is unacceptable, since it means that technocratic central bankers can no longer pretend to be controlling the economy and handing out optimal outcomes.

Concluding Remarks

Fiscal dominance is going to be an extremely common buzzword in the coming years, mainly because its use provides a mechanism from discussing the relationship between monetary and fiscal policy in plain English. It would be quite uncomfortable if the fad of using clear language caught on, since it would be clear that heterodox critics of the monetary policy dominance doctrine were correct all along.

(c) Brian Romanchuk 2020

No comments:

Post a Comment

Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.