The most dreaded moment of all Microsoft Excel spreadsheet efforts is a popup error box warning of the feared “circular reference”. You can’t go on until you find and fix the problem. Unlike inflation mandates, it just won’t take “transitory” for an answer.

This happened to me during a university exam. We had to design a model in Excel which would calculate the probabilities of various outcomes of a gambling game. We were given the rules of the game and had to build from scratch the model in Excel that would calculate what the probabilities of various outcomes must be.

A circular reference (for those who don’t want to skip this explanation for emotionally scarring reasons) is when your calculation is dependent on itself somehow, thereby making it impossible.

For example, if we add up three numbers in a column, but ask Excel to display the result of the calculation in the same place as one of those three numbers, you get the “circular reference” error. Excel can’t display a result of a calculation where it also displays part of that calculation. A + B + C = C cannot be calculated because the Cs clash with each other.

When Excel models get complex, the chances are that, somewhere, you’ve made a mistake and there’s a circular reference.

My contention to you today is that central bank monetary policy is inherently a circular reference, especially when it comes to claiming that inflation is “transitory”. And this is why high inflation will continue to be “transitory”, until consumers lose their temper and pull the plug out, or punch the screen.

Here’s why…

Central bankers are arguing that the inflation they are supposed to keep on target is only temporarily higher than this target. High inflation is “transitory”, as they put it.

They base this claim on the prediction of financial markets, which are of course assumed to be perfect. According to financial market inflation expectations, which are calculated in complex ways, inflation will return back to the target level of central banks in time. Thus, tightening monetary policy now would be a bad idea. The problem will solve itself.

The problem with this approach is that markets don’t just forecast inflation. They also forecast what central bankers are going to do to rein it in. This creates the circular reference problem.

An output – a central bank’s decision on monetary policy – is also an input in the decision-making process, because expected central bank actions influence inflation expectations.

You can’t have central bank policy as both an input and an output in a calculation. The whole thing doesn’t compute and Excel would’ve kicked central bankers out long ago, refusing to accept their “transitory” argument.

We learned about how important this is over the last few weeks.

The Reserve Bank of Australia (RBA) was supposed to keep its loose monetary policy going by pegging Australian 3-year bonds to 0.1% as promised. The market’s expectations and predictions, based upon which the central bankers make their decisions, believed in the central bank’s promise.

But this belief delivered surprisingly good economic performance and surprisingly high inflation data. And so central bankers decided to unpin the 3-year bond yield by not bidding on it when the yield spiked.

This caused havoc in the Australian bond market. And it may undermine the very economic data that had enticed central bankers to tighten monetary policy unexpectedly in the first place.

Do you see how the RBA’s decision depends on data that depends on the RBA’s decision, making it a circular reference?

If not, don’t worry, we’ve got the UK’s example too.

The Bank of England governor warned about the need for tighter monetary policy repeatedly because inflation is too high and rising. And so the market began to price in tighter monetary policy, which would lower inflation again.

But, when the time came, the Bank of England didn’t deliver on the warned about interest rate hikes. The Bank’s latest thinking turned out to be something along the lines that inflation was expected to come down again, so why bother…

This caused havoc in the UK bond market. And it may undermine the very economic data that had enticed central bankers to avoid tightening monetary policy as expected.

Inflation could overshoot even more now, instead of coming down as expected by markets under a scenario of tighter monetary policy.

Again, the data and the decision are intertwined in a way that makes it a circular reference.

The situation we are in is bizarre because central banks are pointing to inflation expectations coming down as the justification for not raising interest rates. But the market’s expectations are expecting the interest rate increases that generate the falling inflation expectations in the first place.

There is a very real possibility that this circular reference spirals out of control.

How?

There are several ways.

At some point, central bankers could wake up to the fact that they were supposed to tighten monetary policy in order to bring down inflation. At that point, it might be too late, allowing inflation to spike.

Or the rate of interest rate increases might need to be very fast and very far to claw back inflation, triggering a sudden and severe tightening.

As in 2007, central bankers might overdo it. They might begin to promise such tight monetary policy that they trigger a debt crisis. Or they might actually impose such tight monetary policy!

Alternatively, inflation expectations could remain one step ahead of central bankers all the way, even as monetary policy is tightened a little.

In this scenario, probably the most likely one, each new tightening will be seen by investors as proof that there’s plenty more tightening to come. This will bring inflation expectations back down again, leading central bankers to forgo the very monetary policy tightening that led to lower inflation expectations.

And so inflation would spike again, and the cycle repeats, with central bankers always one step behind inflation. Over time, inflation would continue to rise, mystifying central bankers each time the cycle repeats. Inflation would forever be “transitory”, after all.

The point is, whatever they do, central bankers are stuck in a circular reference – an incomputable and meaningless situation.

Of course, just as central banks know what inflation expectations are, they also know what the market expects them to do on monetary policy. But this creates another bizarre situation where the only way to actually conduct monetary policy is to surprise the market. And surprises are nasty, as we learned from the UK and Australia recently.

Perhaps all this is why central banks refrained from all the communication policy and transparency that is fashionable today. If central bank policy is signalled, it conflates an input with an output. It turns monetary policy decisions into being dependent on inflation expectations, which are in turn dependent on monetary policy decisions, thus creating the circular reference.

Perhaps secrecy and opacity are a required part of central banking.

If you’re confused, imagine being a central banker!

Nick Hubble
Editor, Fortune & Freedom