How did the Bank of England get its migration forecasts so wrong? (In addition to all the other things it got wrong)
Bryce Elder JUNE 13 2023
The Bank of England is having a bad decade. Not only has its credibility been undermined by egregious forecasting errors, the debate has gone mainstream around whether its independence mandate is a useful fiction or an obstacle.
Energy costs help illustrate one problem. Last summer’s surge in wholesale gas prices went straight into Monetary Policy Reports based on the retail price cap methodology and financial support packages of the time, rather than accounting for a widely-expected state intervention.
Independence locked the BoE into forecasting based on announced government policy rather than the likely path ahead.
But independence also makes the BoE’s unforced errors, such as around UK population growth, harder to overlook.
The bank has “very materially underestimated the supply-side potential of the UK economy because it failed to update its migration assumption — despite a body of evidence already in the public domain that net migration into the UK was poised to come in well ahead of the 2020-based population projections,” says Panmure Gordon chief economist Simon French in a note published today.
The projection he refers to is an Office for National Statistics estimate for net migration of 692,000 over the three years from 2021/22 to 2023/24. When the figure went into the BoE’s February 2023 supply-side stock take it was already stale. An ONS update from November 2022 was disregarded and a January revision apparently arrived too late for inclusion.
As a result, based on recent data, the projection used by the BoE was out by nearly 100 per cent.
Net migration will probably be about 1.2mn over the same three-year period, more than 70 per cent of whom are working age.
“Given the materiality of the difference, and the pessimism of the broader supply side stock take, this was a poor judgment from the BoE,” French says:
We have no evidence — and are not suggesting — that there was political pressure brought to bear on the bank given the salience of migration to UK public policy.
However, it is either that or a poor attempt to use the latest data to accurately estimate the supply side capacity of the UK economy. Whichever way, it does not look good.
When defending its record the BoE tends to highlight that price predictions come from the market, so failures are the fault of gas futures traders and forex dealers rather than its own economists.
As governor Andrew Bailey said last month, forecasts are “conditional on commodity prices, they’re conditional on government policies. So, as those conditions change, we change our forecasts.”
It’s an approach that looks increasingly flawed, says French, as it “introduces the potential for the market path and the expectations of Monetary Policy Committee members to decouple — with obvious challenges in standing, rhetorically, behind their central economic forecast”:
The result of this is that communications resulting from [Monetary Policy Reports] have been frequently undermined as the MPC scrabble to disown or qualify their own forecasts.
For the BoE, the path of interest rates should not be presented as a conditional assumption, says French, who argues in favour of adopting Federal Reserve-style dot plots. “That the MPC should know more than the market on the most likely forward path for UK interest rates should be a feature, not a bug, of their economic forecasts.”
Berenberg economist Kallum Pickering was arguing something similar last week around policy uncertainty, and how using market forecasts “blurs its reaction function and contributes to often unreliable guidance about the policy outlook”:
Pickering also wants dot-plots introduced, as well as some deeper reform around forecasting and guidance:
The BoE should no longer base any forecast on the market curve assumption and instead produce one central forecast based on the assumption of no change in monetary policy. [ . . . ]
The BoE should temporarily introduce state-contingent forward guidance with “knockouts” to commit policymakers to keeping the bank rate at least at the current level until inflation is brought back under control on a sustained basis.
But there’s also a question of whether the BoE is even listening to itself.
Bailey told Jackson Hole in August 2020 about the value of “going big and fast” with quantitative easing, then kept buying bonds in what French calls “autopilot volumes” until December 2021, when financial conditions were exceptionally loose.
The same speech now gets cited to explain why a short, sharp £80bn a year of quantitative tapering won’t make financial conditions tighter.
Not only has the decision to keep adding to its balance sheet aged badly, it “looks like making policy that is at odds with the bank’s own research on the efficacy of asset purchases,” Panmure tells clients.
All in all, the BoE “has managed to dent a well-deserved reputation for competence” in ways that can’t be blamed on fuel inflation alone. A functionally independent yet politically constrained central bank cannot be a market-leading forecaster because it’s compelled to apply policy positions that lack credibility; this “cannot be a sustainable position”, says French:
The reputational road back will require difficult conversations with lawmakers, but it is very clear to us that those conversations need to happen.
No comments:
Post a Comment