With traders hunkering down and freaking out ahead of tomorrow’s CPI report, which will either be a Valentine’s day massacre or happy ending (which we will preview shortly), another macro event is shaping up as potentially even more consequential.
As a reminder, the reason why so many are transfixed on tomorrow’s CPI is because it only feels like yesterday that US inflation prints were seen as last year’s news given the recent falls. In addition, forecasts and breakevens suggested we were on a glide path to normality over the next few months and quarters. However, as DB’s Jim Reid warned this morning, that view has been jolted in the last 10 days for 4 reasons:
- i) First we had payrolls print which raised the prospect that core services ex-shelter could stay stronger for longer;
- ii) Then we had lots of hawkish central bank speak that the market had previously ignored but was now slowly waking up to;
- iii) Then Manheim suggested US used cars (+2.5% mom in January) climbed at their fastest rate for 14-months, and finally
- iv) we had US CPI revisions on Friday that have rewritten the last year of history and in turn reduced core inflation by around a tenth each month leading up to June and have increased it by an average of around a tenth in each month since August. As such the trend in core CPI hasn’t fallen as much as expected and we now haven’t seen any month less than +0.3% MoM. In addition 3m annualised core CPI ran at 4.3% in December rather than the 3.1% reported at the January 12th release. So although year on year hasn’t changed the momentum is notably different.
But while a hotter than expected inflation print will surely have an adverse impact on risk markets as it will prompt concerns of an even higher for longer Fed (read the full JPM scenario analysis which sees the S&P sliding if the headline CPI prints at 6.4% or higher tomorrow vs median consensus of 6.5% of higher), a more tangible indicator of economic overheating is looming and will be revealed exactly 24 hours after the CPI print, in the form of January’s retail sales which, if the latest Bank of America card spending numbers are accurate, will be nothing short of a “knock your socks off” blowout scorcher.
According to BofA economist Aditya Bhave, total card spending per household (HH), as measured by BAC aggregated credit and debit cards, was up a blistering 5.1% year-over-year (y/y) in January, which would make it the biggest annual jump since the summer of 2022.
It’s not just the annual jump: card spending per HH also surged by 1.7% month-over-month (m/m) in January on a seasonally-adjusted (SA) basis; this leads BofA to forecast an above-consensus 2.2% m/m increase in the Census Bureau’s ex-autos retail sales figure for January.
Additionally, BofA’s economists expect core control sales (retail sales ex autos, gas, building materials and restaurants and which feeds directly into the GDP bean count) to rise by 2.6% in January.
As further shown in the table below, card data show a strong pickup in spending across most categories on both a y/y and m/m SA basis, including general merchandise, clothing and airlines.
Some of the YoY increase reflects “base effects” as the Omicron wave in January 2022 weighed on services spending, particularly travel. Consequently, BofA expects this part of the YoY rise to unwind fairly quickly. However, the bank economists also see signs of real strength in services spending in January 2023: Bank of America card spending per household was up 3.5% MoM SA on airlines and by 1.8% on restaurants and bars. International spend rose too, particularly in Asia, compared with the prior two years, due to further reopening and Lunar New Year celebrations.
So what’s behind the blowout surge in spending which would translate into a roughly 3-sigma beat to core retail sales? BofA attributes the expected strength in January retail sales to three factors:
The first is a statistical distortion: last month the bank flagged that since the start of the pandemic, holiday spending has become more front-loaded, and so the spike in spending in December has become smaller on a not seasonally adjusted (NSA) basis. As a result, the NSA decline in spending in January has also become smaller. However, seasonal adjustments are still largely based on pre-Covid spending patterns. Therefore, on a seasonally adjusted basis, January retail sales have been exceptionally strong since the start of the pandemic. Translation: just like the blowout jobs report was entirely due to seasonal adjustments, so too the red-hot retail sales report will be largely a function of various excel data transformations.
The second driver of strong spending in January was the increase in disposable personal income (DPI) due to robust labor market gains and various inflation-related adjustments. The most notable of these was an 8.7% cost-of-living adjustment (COLA) to social security benefits, which alone likely raised DPI by around 0.5% in January. The COLA was announced last October. Starting in November, spending has been meaningfully stronger among older HHs than younger HHs.
There is no similar pattern in the pre-Covid data
The outperformance of older households was concentrated in retail in November…
… but spread to some services categories in December-January.
In Bank of America’s view, it is likely that some older households boosted spending over the holidays in anticipation of receiving the COLA in January. Others might not have been aware of the COLA until it went into effect. Therefore, the uptick in spending for the older HHs could extend into at least February.
A third and final reason for the pickup in spending in January could be that consumers were holding back on spending in anticipation of large post-holiday clearance sales. This ties in with the fact that some of the weakest categories in December – furniture, clothing, general merchandise and department stores – saw significant payback in January.
In conclusion, there are good and bad news. First, the good news: as BofA’s Bhave writes, the factors discussed above support the bank’s forecast that consumer resilience will help the US avoid a recession in the first half of 2023 (if not in the second half). However, the bad news is that – similar to the January payrolls report – the underlying drivers of strong January data (in this case spending) are mostly one-off factors or level-shift effects that will not lead to an extended acceleration in economic activity. Therefore BofA still expects the economy to slip into a recession in the second half of the year.
The bigger question is how will markets respond: this may be contingent on what the BLS reports in tomorrow’s CPI report. If we get signs of inflationary overheat just days after a blowout payrolls report, and then couple this with a red-hot retail sales report, even if the Fed will be hard pressed not to make it clear that much more rates pain is coming to cool down the overheating economy. On the other hand, a cooler than expected inflation print Tuesday coupled with strong retail spending data may be just what the “soft landing” (or no landing) narrative needs to push stocks to a new 2023 high thanks to a benign macro environment where households are busy spending money without inflating prices.
More in the full reports available to pro subs here and here.