Yesterday, commodity weakness in metals was widely attributed to Goldman Sachs, which cut its copper forecast because of projected weakness in China. Then there was concern that the August jobs report, out this Friday, was going to be weaker than expected (stop me if you've heard this before). Citigroup led the downside charge, putting out a call out for 125,000 jobs, far below consensus expectations for a gain of 161,000. This happens all the time now. Bulls cynically call it "the recessionista crowd," those folks who insist there is a serious slowdown coming. You can't blame the bulls for being cynical. Bears have been insisting we are going into a recession for the last two years. Now we are heading into another seasonally weak September-October, accompanied by new cries that a recession is imminent. Someday, the "recessionistas" are going to be right. But they've been terribly wrong for two years, and anyone trying to time a recession would have lost a lot of money. Every jobs number moves markets now In the meantime, every data point on jobs is a potentially market moving event. Even the weekly initial jobless claims, out every Thursday and not traditionally a market mover, is watched carefully now. This week, markets are expecting roughly 230,000 in jobless claims. What was it a year ago? About 234,000, almost exactly the same. We know the most likely playbook: if the data comes in better than feared, the Fed will cut 25 basis points and the market will rally. If the jobs data is weak, the market will panic again, but bulls have not abandoned the reasonable assumption that there is indeed a Fed "put" that will provide some kind of floor under the market, even if it is 5%-10% below the current level. Tom Lee from Fundstrat, who has been directionally correct all year, is in this camp. He told CNBC yesterday that stocks could see a 7%-10% pullback this fall, but then added that investors should be ready to "buy that dip." Nvidia's multiple is falling, not its earnings Meanwhile, there was much gnashing and wailing around Nvidia 's big 9% drop yesterday. An old friend, veteran trader Joe Zicherman, called me near the close and said this may be the biggest one-day dollar loss ever for a single stock, and he appears to be correct. Nvidia lost $279 billion in market value Tuesday. This looks like a record for a single day dollar loss in one stock. Previous big losses include Meta and Amazon in 2022. Big one-day losses Nvidia $279 billion (9/3/24) Meta: $232 billion (2/3/22) Amazon: $207 billion (4/29/22) These $200 billion dollar down days are getting to be a bit of a habit with Nvidia. CNBC's Robert Hum tells me this is the 5th time Nvidia has lost more than $200 billion in one session. But remember: they are taking down the multiple on Nvida, not the earnings. The forward multiple for Nvidia's current fiscal year (February 2024-February 2025) is now at 38.3, well below the 48 multiple in June. The forward multiple for Nvidia's next fiscal year (February 2025-February 2026) is 28.8, well below the 36 multiple in June. Not so the earnings estimates, which have been rising for all quarters for the foreseeable future, though as we have been noting, at a far more modest pace than last year. Nvidia is down again early Wednesday after reports of a DOJ subpoena around an antitrust probe . While Nvidia is certainly not a value stock, it is much more reasonably priced than it was two months ago.
