The Failure of Central Bank Models

The Failure of Central Bank Models

The Bank of England. Established in 1694.

Every month since the start of the current inflation rally, central banks’ workhorse models did stubbornly predict that inflation would rapidly go down. However, only to be proven wrong the next month (see graph below).

As recently reported in the Financial Times:

“It was more or less impossible in our models to produce any inflation that would not be temporary.”

Belgium central bank governor Pierre Wunsch, an ECB council member, said last week, adding that they always showed price rises falling to the 2 per cent target, regardless of the assumptions.

“We have come to the conclusion that we know much less about inflation drivers than we thought.”

Well, indeed. There are unfortunately many inflation channels that cannot be captured by standard general equilibrium models. In equilibrium, supply = demand in both good markets and in the labor market, so the “obvious” channel by which prices (wages) increase because demand exceeds supply is just absent.

In standard (dynamic stochastic) general equilibrium models, inflation is driven by a positive output gap (overheating of the economy), itself only possible when prices cannot adapt immediately. Inflation expectations are assumed to be rational and hence only affected by true economic shocks and by the anticipated policy of central banks – there is no room for animal spirits, overreactions, or feedback loops from past realized inflation into inflation expectations. 

Models based on the rather silly assumption of equilibrium failed in and after 2008, when JC Trichet declared:

“Macro models failed to predict the crisis and seemed incapable of explaining what was happening to the economy in a convincing manner. As a policy-maker during the crisis, I found the available models of limited help. In fact, I would go further: in the face of the crisis, we felt abandoned by conventional tools.”

They failed again in 2021. And led central bankers to declare that inflation was transitory. Because the real mechanisms behind the current bout of inflation were not properly identified. And absent from “cutting edge” macroeconomic models. Time to pay attention to Pierre Wunsch’s remarks and introduce genuine non-equilibrium effects of various “kinds and behavioral/cognitive biases into macroeconomic (agent based) models.

I used this graph below ten years ago to illustrate how standard models struggled to capture what was going on after 2008. 

As for inflation today, models then predicted every year that the situation would improve, only to become disproved by reality. 

As a result, I was therefore very excited to read Olivier Blanchard’s 2014 piece “Where Danger Lurks”. Moreover, in which he called for a rekindling of macroeconomic models. He wrote in particular 

“We in the field did think of the economy as roughly “linear”, constantly subject to different shocks, constantly fluctuating, but naturally returning to equilibrium over time. […] The problem is that we came to believe that this was indeed the way the world worked. […] The main lesson of the crisis is that we were much closer to “dark corners”—situations in which the economy could badly malfunction —than we thought.”

The crucial sentence here of course is “naturally returning to equilibrium”. All these classical models can only predict that the system will return to equilibrium – there is nothing else it can do! This contrasts with more realistic frameworks (like Agent Based Models) where multiple equilibrium states can coexist, with tipping points separating them (probably what O. Blanchard calls “dark corners”, although only implicitly). 

Alas, when I met Olivier Blanchard a little later in his IMF office, he told me – somewhat dispirited:

“well, the economy is now back on track, and the appetite for new economic thinking is already fading away. Macroeconomics will revert to business as usual”. 
Sealing of the Bank of England Charter (1694). By Lady Jane Lindsay, 1905.

The problem, in my view, lies in what Willem Buiter (ex. Bank of England) said in his own 2008 piece:

“The unfortunate uselessness of most ‘state of the art’ academic monetary economics”
“Research tends to be motivated by the internal logic, intellectual sunk capital and aesthetic puzzles of established research programs. Rather than by a powerful desire to understand how the economy works. Let alone how the economy works during times of stress and financial instability”

A power desire to understand how things really work: this is what theorists should strive to achieve.

Written by Jean-Philippe Bouchaud

See more work by Jean-Philippe:

Game Theory in a Highly Complex & Radically Uncertain World

How Does The Fama French Model Work? Does It Really Work?

Jean-Philippe Bouchaud

The Failure of Central Bank Models


Paul Romer
Central bank independence versus inflation. Often cited research published by Alesina and Summers (1993). Shows why it is important for a nation’s central bank (i.e.-monetary authority) to have a high level of independence. This chart shows a clear trend towards a lower inflation rate. As the independence of the central bank increases. The generally agreed upon reason independence leads to lower inflation. Is that politicians have a tendency to create too much money. If given the opportunity.

The Failure of Central Bank Models