Central banks look likely to be late in cutting rates

Stay informed with free updates

Simply sign up to the Central banks myFT Digest — delivered directly to your inbox.

The latest data shows Germany’s economy was weak in 2023, but a 0.3 per cent contraction is not a recession. Elsewhere, US-led attacks on Houthi rebels in Yemen are still having no lasting effect on oil or gas prices, despite massive ship diversions from the Red Sea. Is this about to change? Email me: chris.giles@ft.com

Crippling uncertainties

Last week, I dissected the Federal Reserve’s uncertainties in its forecasts, concluding that if these are genuinely held, the world’s most powerful central bank is likely to be more timid in cutting rates than markets expect. I presented evidence showing that some of the high levels of uncertainty reported by the Federal Open Market Committee’s members was unreasonable. The Fed is not alone.

Bank of England cushions

I am not talking about the antique furniture in Threadneedle Street, but the extensive padding that the Monetary Policy Committee adds to its forecasts to represent uncertainty.

In what is an excellent piece of transparency, the MPC agrees on an “uncertainty parameter” for all of its forecasts, which is essentially a standard deviation of expected forecast errors. It measures how far on average the committee expects actual inflation to deviate from its predictions. (To be accurate, this is based on a two-piece normal distribution as outlined in this 1998 paper, but this is not an important detail.)

In its November forecasts, for example, the MPC expected inflation two years ahead at the end of 2025 to be 1.9 per cent with an uncertainty parameter also of 1.9. Roughly, this means the committee thought there was a two-thirds chance that inflation would lie between 0.0 and 3.8 per cent. It goes without saying that this is quite a wide target, which the committee expects to fail to hit almost a third of the time. It’s not exactly darts.

The BoE has also become much less certain about its forecasts over time. The chart below shows how the MPC at first objectively changed the uncertainty parameter according to past forecast errors and, more recently, subjectively raised it. The story is not a happy one. According to the chart, in 2020, the BoE was four times as uncertain about its two-year ahead inflation forecast as it was in 2006.

The chart shows that the BoE used historical forecast errors to calibrate its uncertainty parameter until the global financial crisis, but then decided the world was much more uncertain so added a lot of padding in. Then it plumped up the cushions again during the pandemic. The question is whether the uncertainties are reasonable.

The chart below shows the same two-year uncertainty parameter alongside the standard deviation of absolute inflation errors from a constant 2 per cent forecast. This is a dumb mechanical forecast and shows that apart from the most recent inflationary episode, BoE forecasts two years ahead have been on average about only 0.6 percentage points out, nothing like the 1.9 percentage points assumed by the MPC. The only way you can get the current figure is to start the clock in 2004 and include the latest very large errors. This is what the BoE thinks is reasonable. I don’t.

There is therefore a lot of padding in BoE forecasts with the central bank crippled by unnecessary uncertainty. This is corroborated by speeches given by its officials. In December, for example, Sarah Breeden, the new deputy governor, repeated many times that the economic landscape now was “more uncertain than usual”. Given her worries that the outlook is especially uncertain, it is difficult to see how she will achieve her other ambition — for “policy . . . to be nimble as we learn more about which state of the world we might be in”. The best way to resolve the contradiction is that the BoE is likely to be slow in making a move, then will seek to act decisively.

This view was corroborated by BoE stalwart, deputy governor Ben Broadbent, who said in December that “additional uncertainty” meant “policy is likely to be somewhat more delayed”.

In speeches and with the exaggeration of uncertainties, the BoE does not look likely to be rapidly responsive to data in 2024. It appears set to be playing catch-up.

“Too early” in the eurozone

The European Central Bank explicitly states that its forecast uncertainty range depends on “past projection errors” not including “increased uncertainty due to the recent financial market tensions”.

This is a more reasonable forecasting approach than that of the BoE. The result is that officials in Frankfurt produce uncertainty bands much narrower than those in London as the chart below shows and probably closer to what it believes is achievable.

But, if you thought that the ECB having more confidence in the accuracy of its forecasts would make the central bank in Frankfurt more decisive, you need to think again. At the moment, officials are having a debate about whether it is too early even to talk about interest rate cuts. President Christine Lagarde said in November that discussions about interest rate cuts were “totally premature”, adding shortly before Christmas that none had taken place. She was backed up last week by executive board member, Isabel Schnabel, who said it was “too early” to discuss rate cuts in her question-and-answer session on the social media platform X.

Chief economist Philip Lane, was willing on Monday to talk about “a sequence of rate cuts”, but later in the same interview stuck to the party line saying it was still “too early” to move to the implementation phase of loosening monetary policy. Lane wanted to wait “to make sure that the inflation problem is fully defeated”, which is likely to make the ECB also late in cutting rates.

BoJ: what is uncertainty?

Over in Tokyo, the Bank of Japan does not really believe in outlining the certainty of its forecasts or examining its past record. Although it describes “extremely high uncertainties” in its quarterly outlook for activity and prices, it merely highlights the small differences in forecasts between members of the BoJ’s policy board. These are generally about 0.4 percentage points wide two to three years ahead and therefore do not represent true uncertainty at all.

So what?

Fed officials say the world is more uncertain than the past 20 years, BoE officials produce forecast uncertainties wide enough to serve as a bus park and ECB officials worry whether it is too early even to talk about rate cuts. None of this suggests central bankers will be speedy with action in 2024.

What I’ve been reading and watching

• Lots of economists say things are looking bad in the Red Sea for the global economy, but I’m pleased Alan Beattie over at Trade Secrets has put things into perspective — it’s a big deal but not a major crisis for globalisation. That’s what financial markets are also pricing

• In the US, New York Fed president John Williams said the Fed would need to be “confident that inflation is moving toward 2 per cent on a sustained basis” before cutting rates, just before a slightly disappointing CPI report. Mohamed El-Erian thinks the figures signal a difficult year ahead for inflation

• In Poland, the difficulties of reforming autocratic institutions continues. The country’s constitutional court has shielded the central bank governor from being put on trial

• Stephanie Kelton talks to Brooke Masters and is characteristically unorthodox. She does not like demand management, thinks governments can just improve supply, wants zero interest rates and says inflation has come down in spite of the Fed rather than helped by it. All I will say is that sometimes constraints are real

A chart that matters

The Bank of Japan is looking closely at wage data to see if it can foster a virtuous wage price spiral. The latest data does not look that encouraging.