Rob Wood

UK inflation expectations are modestly deanchored.

5-year ahead expectations higher than when inflation was in double digits and the survey is biased vs. history down by a method change in 2020. Paging MPC.

You've got to the love the UK CBI retailing survey. Volatile much? But for what it's worth, sales for the time of the year reach the strongest in 14 months in July.

Pay slowdown this year is also mainly driven by firms passing NICS into weaker wages, rather than weak demand.

UK payrolls driven by NICS rather than weak demand. Sectors with high share of part-time workers—whose employment costs rise most after NICS hike—and self-employment—opportunity to cut tax liability by switching from employee status—see payrolls most. Latter means payrolls exaggerates job falls.

It isn't income-driven in consumer confidence. The most optimistic income group, and the one seeing smallest fall in confidence since election, is 50K plus, the age group with the biggest rise in optimism since election is 16-29.

UK political views increasingly seem to be driving responses to consumer sentiment—a phenomenon seen for some surveys in the US too—making them a less reliable leading indicator.

I think thats broadly right but huge difference between say -10k a month (still consistent with surveys) and -35k for monetary policy. Former is a correction, the latter a proper recession. While taking some of your points, I think weight of evidence points to smaller jobs falls than payrolls.

Most of the improvement in LFS sample size had happened by the base, so on that basis there will be little bias in latest 3 on 3

Payrolls still overstating job growth relative to surveys and one of best performing models points to improvement. REC is one of the poorest performing surveys for forecasting job growth. Best surveys point to more like 15k falls in employees and DMP now suggests rises.

Quite so, and its the latest data that is the key. Is employment falling or not. Payrolls says yes—although half as much as payrolls said a month ago so far this year—WfJ and LFS say no. LFS would be lower absent sample changes but unlikely to be showing falling jobs like payrolls.

Or more likely the LFS would have been stronger in 2023 and 2024. If it is now representative, and used to be in 2019, the change over the period in the chart is fine.

yes and that ONS comment was on data to March, so we already have 2 of the 6 months of estimates, and its not like its a switch that goes from biased to unbiased. The previous four months of data will also have been cohering. Besides, whats wrong with WfJ too?

The ONS say the low LFS vs PAYE in 2023 was due to poor sampling and LFS *was* catching up. ONS say that process has worked through by now (see below). Question remains, do you believe payrolls more, which were just revised job falls so far this year in *half*.

More volatile yes, not upward biased

In any case, what's wrong with WfJ?

Which do you believe? Payrolls that just got massively revised, or the other two series which are also revision prone and in the case of the LFS not firing on all cylinders yet.

Jobs falling or rising solidly?....What the MPC would give for accurate data.

Surprise, surprise, UK payrolls were revised up bang in line with their typical pattern. The question is, why would anyone believe the dodgy first estimates from this series showing tanking jobs? With a typical revision to June data, job falls are easing now.

Shall we call it the estate agents' PMI?

Here's the thing about the UK PMI. The higher uncertainty goes the worse an indicator of growth the PMI becomes. Policy uncertainty was nearly 5 standard deviations above normal in April/May, so the PMI is giving a far too pessimistic steer.

Cautious consumers you say? Retail sales growing the fastest in 3 years and now liquid asset accumulation dropping sharply. For my money the ONS saving rate data are wrong—they often revise it down sharply. Consumers will keep GDP growth ticking along.

The MPC will struggle to cut twice more this year after inflation surged to 3.5%, well above consensus of 3.3% but close to our call of 3.6%. Administered prices drove the surge, and there was a small Easter boost, but underlying pressures are stubborn. 'Skips' are likely now.

It's difficult to reconcile the data around the cautious consumer - not least another pick up in the savings ratio - with the strength in the retail sales. There are probably good reasons to put more weight on either set of data, but the discrepancy does muddy the true picture of consumer activity

The current economic conjuncture is hard enough to diagnose without being unable to trust the data. The saving rate will make good headlines today, but be very wary about data held together with double-sided sticking tape. Those Blue Peter creations didn't last long either.

The saving rate depends on measuring income accurately, but that is infected by the unreliable LFS jobs data. Statisticians put sticking plasters over the problems by using payrolls to 'adjust' income, but how much should we trust that after PPI and trade data had mistakes?(7/n)

Back in early 2023 the headlines would have reported a household saving rate surging to 9.3% in Q4 2022, from 6.8% in Q1 2022... now the ONS think the saving rate *fell* in 2022, to 6.5% in Q4 from 7.0% in Q1. From rocketing to falling is quite some change in view (6/n)

What if I told you that first estimates of the saving rate have been revised down 1.1pp on *average* since 1997,. The ONS have already revised down the 2023Q4 saving rate by 2.1pp (5/n)

Granted the quoted 2-year mortgage rate is about 70bp higher now than in early 2010, but the proportion of households reporting that it was somewhat or very difficult to pay the mortgage or rent is falling quick; it dropped from 42% in October 2023 to 31% in February 2025 (4/n)

Does this pass the smell test? Are economic conditions as bad as during the worst global economic crisis in a century? People reliably save more when unemployment is higher yet in Q1 2010 the unemployment rate was 8.0%, nearly *double* its current level (3/n)

Excluding forced saving in Covid the saving rate reached its highest since the peak after the financial crisis in 2010Q1. Headline saving includes incomes that households don't see day-to-day—imputed returns on pensions—but the 'cash' saving rate hit its highest ever (2/n)

Are more bad data leading us to misdiagnose the UK economy. Put another way can we really trust that the UK cash saving rate rose to an all-time high in Q4 2024 as the ONS reported this morning? The chart is startling, with households saving 12.0% of their income in Q4 (1/n)