Financial markets have started to behave fearfully once again, as the focus has shifted back toward both the likelihood and potential severity of an oncoming slowdown in economic activity.
Before we get started, Passover began last night, so to my readers celebrating their first full day in recognition of the Exodus, I bid you a Happy Passover. Today is also the start of the holiest four-day period on the Christian calendar, peaking in intensity tomorrow and culminating on Sunday. To my readers celebrating the Resurrection this weekend, I bid you a Happy Easter. May those recognizing no religious events whatsoever, have a nice three-day weekend.
Except for bond traders. Bond traders will still have to work (a little) tomorrow and will have to react to the BLS Employment survey results for March in real time.
Macro Do Da Day
I don't think that there is any longer a doubt about it. The U.S. economy is showing weakness. Does this lead to outright contraction in activity? We'll know the answer to that question when we know it, but it does seem likely.
On Wednesday morning, the ISM (Institute for Supply Management) went to the tape with their service sector or non-manufacturing survey results for March. I'll try not to get carried away, as this survey still shows headline-level expansion. That expansion, however, is apparently decelerating quite rapidly.
That headline (PMI) print of 51.2, was down from 55.1 in February and the weakest print at the top of this page since the pandemic shut down much of the economy in May 2020. Making an exception for the pandemic, March 2023 was the weakest month for the service-based side of the U.S. economy since January 2010.
Within the report, showing alarmingly rapid deceleration were the sub-components for New orders, Employment, Deliveries, Order Backlog, New Export Orders and Imports. Inventories did show accelerating growth, which is probably the one place we don't need to see such expansion.
That report followed the ADP Employment Report for March, which does not always line up too well with the BLS survey print for Non-Farm Payrolls or even Private Payrolls, even though they cover the same turf. Many economists, myself included, believe that the ADP report is probably more accurate than the BLS given that ADP results are based more on data and less on survey results, and also because the BLS's own two surveys (Establishment & Household) often disagree quite sharply with each other on the level of job creation for the very same months.
On Wednesday, the ADP Report for private payrolls showed job creation of 145K positions for March, well below the more than 200K that had been expected. Nela Richardson, Chief Economist at ADP, was quoted in the press release saying, "Our March payroll data is one of several signals that the economy is slowing. Employers are pulling back from a year of strong hiring and pay growth, after a three-month plateau, is inching down." I have appeared in the media alongside Richardson on a number of occasions, especially when she was on the way up. I like her and consider her judgement quite worthy.
The Result
Late Wednesday morning, the Atlanta Fed once again revised their GDPNow estimate for Q1 economic growth lower, this time to 1.5% (q/q, SAAR). This came after this estimate had been revised down to 1.7% from 2.5% on Monday partially in response to the ISM's disastrous March survey results for the manufacturing side of the U.S. economy.
On Wednesday, the Atlanta Fed was forced to reduce its input for real personal consumption expenditures, while also reducing the input for real net exports in response to the Bureau of Economic Analysis' February Trade Balance print that landed worse than projected. These negative revisions were slightly offset by an increase that was made to the input for real gross private domestic investment.
Atlanta is expected to revise its Q1 GDP estimate again on both Monday and Friday of next week. Monday will be in response to final wholesale inventories for February, and Friday will be in response to March data for both Retail Sales and Industrial Production. That will be a big one.
In perhaps the most overt signal of any coming contraction in the size of the U.S. economy, the Treasury yield curve continues to warp as traders bet on lower rates as a reaction to the poor macro and the Fed still props up the short end. I showed you the yield spread between the U.S. 10-Year Note and 3-Month T-Bill here yesterday. That spread closed even more inverted on Wednesday at -156 basis points. Let that sink in. U.S. 10-Year paper pays 3.29% early this morning. U.S. 3-Month paper pays 4.81%.
The 2- Year Note currently pays 3.75%. That's right, the yield spread between the 2-Year Note and 3-Month T-Bill stands at -106 basis points and is expanding rapidly. This is what these two yields look like when overlaid on the same chart over the past two months:
Something comes this way, soon. I gave you late Q1/early Q2 a while ago, as timing for when I thought the U.S. economy would go into recession. I may or may not be correct on my timeline. The bond market is telling you that the storm that cometh is probably far less than two years out. Maybe a year. Maybe not.
At least U.S. 30-Year paper out-yields the U.S. 10-Year. That storm won't last forever. It may outlast some of us though.
Marketplace
For a second consecutive day, defensive sectors dominated U.S. markets as four (three of them defensive) of the 11 S&P SPDR sector ETFs shaded green for the session. Utilities ( XLU) led again, adding 2.59% on Wednesday, with Health Care ( XLV) in second place, up 1.73%. Technology ( XLK) , Industrials ( XLI) and Discretionaries ( XLY) again populated the lower rungs of the daily performance tables.
While the S&P 500 held its own, down 0.25% for the session, the Nasdaq Composite did not, surrendering 1.07% and closing in the red for a third consecutive day. The banks, transports and smaller-cap indexes all gave up some ground on Wednesday, but not like the semiconductors. The Philly Semiconductor Index was taken for a beating of 1.8% during regular hours, led lower by large-caps Global Foundries ( GFS) , Advanced Micro Devices ( AMD) and KLA Corp. ( KLAC) .
Breadth remained poor, as it has been of late, on up days and down. Losers beat winners by more than 3 to 2 at the NYSE and by more than 2 to 1 at the Nasdaq. Advancing volume took a 35.9% share of composite NYSE-listed trade and a 30.5% share of that same metric for Nasdaq listings. Trading volume continued to dwindle for NYSE listings, but did show a (very) small day-over-day increase for Nasdaq listings, albeit at very low levels.
Funny Thing About the Semis
Semiconductor stocks were among the weakest on Wednesday. That was partially due to the weak macroeconomic data discussed above. This was also to a large degree due to a research paper released by Alphabet ( GOOGL) that showed that their in-house fourth-generation Tensor Processing Units (TPUs), which are meant to train artificial intelligence models, are more power efficient and faster than Nvidia's ( NVDA) A100 chips. Markets reacted.
The funny thing is that Google did not compare its in-house chips to Nvidia's H100 (most advanced) chips that have come to market since the creation of Google's chips and use later technology than do the A100's.
Readers may have noticed that Qualcomm ( QCOM) was the only large-cap semiconductor stock to close in the green on Wednesday. That was likely due to a report at ML Commons that showed that Qualcomm's AI 100 chips can outperform Nvidia's H100's in completing certain tasks such as classifying images.
I still believe in AMD and NVDA. I will admit to taking a little something off of both long positions on Wednesday with the intention of adding back on later. I love these two names, but let's be honest, I have very large victories to protect here and this could be a year where victories cannot be allowed to wither over time.
Not As Bad As It Looks?
On Wednesday evening, Costco ( COST) reported net sales of $21.7B for the month of March. This was good enough for year-over-year growth of 0.5%.
Now for the bad news. Same-store sales decreased 1.1% on the heels of a 3.5% increase for February. Comp sales were -1.5% in the U.S., and -2.4% in Canada, however, comp sales increased 2% internationally. Perhaps the real kicker was in e-commerce, where the comps were -12.7%.
Now, to be sure, a lot of this had to do with falling prices for gasoline. Comp sales ex-gasoline, and ex-FX were up 2.6%, or when broken out... +0.9% in the U.S., +7.4% in Canada, and +7.6% internationally. However, e-commerce comps were still -11.6%.
COST is trading lower overnight. Are these sales truly problematic? Probably not, if the U.S. dollar keeps softening and if OPEC+ has anything to say about gasoline prices, right? Still, those e-commerce sales might be an issue.
You know what else is probably an issue? A stock trading at 34 times forward-looking earnings going into a recession. Even for a discount bulk retailer. Costco may not issue a special dividend anytime soon, but how far away can a membership price increase really be?
Economics (All Times Eastern)
08:30 - Initial Jobless Claims (Weekly): Expecting 200K, Last 198K.
08:30 - Continuing Claims (Weekly): Last 1.689M.
10:30 - Natural Gas Inventories (Weekly): Last -47B cf.
13:00 - Baker Hughes Total Rig Count (Weekly): Last 755.
13:00 - Baker Hughes Oil Rig Count (Weekly): Last 592.
16:30 - Fed Balance Sheet (Weekly): Last $8.706T ($-27.845B).
The Fed (All Times Eastern)
10:00 - Speaker: St. Louis Fed Pres. James Bullard.
Today's Earnings Highlights (Consensus EPS Expectations)
(GOOGL and COST are holdings in the Action Alerts PLUS member club . Want to be alerted before AAP buys or sells these stocks? Learn more now. )
