SMCI, specifically, Supermicro.

I know this firm's products quite well and have used them personally since the mid 2000s.  They are a motherboard manufacturer and have some very nice kit with features that make them particularly suitable for server machines (not so much for desktop computers simply because the extra cost doesn't bring you much except in very-specific circumstances.)

Specifically, most have "out of band" management so you can not only do things like toggle the power, check fan and temperature status, reset the box and such but in addition they have a "virtual console" capability using what is called "IPMI"  over a dedicated network interface (preferred for security reasons since that interface never leaves your data center "raw") and their boards typically support ECC memory, which is very important in larger server configurations with processors that also support it (specifically the Intel Xeon series) since random memory errors (e.g. cosmic ray hits, which can happen although they're very rare) could otherwise cause either undetected data corruption or even crashes.

But in the last couple of years they've been on a tear because they uttered the words "AI" and were buying quite a lot of Nvidia chips.

Just recently their auditor quit with a rather-amusing letter to the SEC (which of course is public) and drew a number of snickers, implying that they didn't like what they saw in the firm's financials.  The stock crashed, and is now trading around $30 (after trading as low as $17) -- off a fairly recent high of around $140!

This has brought many people out thinking its a screaming buy -- provided, of course, the books are not entirely-fraudulent.

As recently as mid 2022 the stock was trading at a split-adjusted $4.00/share!

So exactly what has changed in the company's forward prospects selling server boards that leads someone to believe that it is was ever worth thirty times what it was worth in 2022 (note -- this was not a "startup" that just started making money!) and why is it worth seven times what it was in 2022 right now?

Again -- this is a company that makes motherboards, a commodity product that has decent operating margins and it is an established player in that space.  Note that it was a strong player in the space the entire time "cloud" was taking over the world and while they made a nice amount of money and have excellent products how did the market turn around and revalue them up by thirty TIMES in the first place when their essential product is a commodity item in the computing space?

Does anyone really think we will be putting thirty times as many servers in datacenters over the next few years?

Just contemplate that in the market as a whole folks -- look at basically everything and you will find this sort of pricing in what equities go for these days.  How is that defensible?

I argue it is not.

The Tyson fight, that is -- but not just on the fight.

I'll leave most of the analysis of the fight itself to those who follow boxing more-closely, but will note that Jake Paul allegedly said that he "was nice" to Tyson (more or less), which is silly.  You were paid to knock him out if you could, and he was paid to knock you out if he could -- or at least that's what the viewing public was sold.  If it wasn't what you were going to do then like so much else in today's society the watching public was sold a bill of goods.  I hope you and boxing in general pay for that in the future (in the form of a destroyed revenue stream.)

I'll focus my attention instead on the production and distribution which was, to put it mildly, horrible.

We'll start with audio.  I've done this for people before.  Its a bit of an art, as is (of course) camera operation, direction and video mixing.  To say that it was done "poorly" would be an insult to those who do it poorly.  It was horrible; the balance was wildly off between background and announcer volume, EQ was bad, micing (whether selection or placement on the announcers) was hideous, mixing was worse and the result was often unintelligible garbage.  A middle school play would sound better.

Video production wasn't much better.  I've done that too.  Camera angle selection, the jump cuts, just plain poor.  Shooting live sports is again an art, but one that plenty of people are pretty good at.  This wasn't.  Critical angles were missed during punches that shouldn't have been.

Production stunk like skunk, to be precise in four words, and that's leaving off the trashy stuff everyone expects in so-called "professional" boxing (pre-bout preening, grand entrances, etc.)  That sort of looked like "professional" wrestling..... I thought one was a sport, the other entertainment?  Maybe today, not so much.

When it came to content delivery, however, it was clear that Netflix's "production" lack of acumen was actually the high point of their competence.  If you recall many years ago I went after them in spades during the so-called "net neutrality" wars because their strategy (and the foundation of their "earnings" during those times) was to force last-mile carriers to eat their long-haul bandwidth requirements.  They ultimately lost that war despite Obama's Net-Neut position and have stuck racks of disks and servers all over the place very close to local distribution centers, such as at the head end offices of cable companies.  That's as it should be; there's no "peering" argument to be had for a service like Netflix because the other guy gets no value in the other direction, which is the entire premise of no-charge peering -- you get value from my customers, and I get value from yours; the differential is small enough that its not worth trying to figure it out and generate a bill in either direction.

But when you want to do live events that doesn't work.  At all.  Now you have to deliver the entire content stream to everyone watching all at once from wherever it originates to every single consumer.  The cable company still has their requirement but yours is now not to send down the new movie at 2:00 AM when everyone is sleeping -- nope, you have to send it right now to each of those head ends and the capacity has to be there to do it.

Well, it wasn't.  Whether that's because Netflix can't do math or they have deliberately not purchased it previously (because you don't need it when the content is on hundreds or thousands of servers in said cable headends) doesn't matter.  It wasn't there and that was obvious as the pixelation and then, for a huge number of people, complete failure to deliver the streams occurred -- and what made it worse is that while you could exit your Netflix app and restart, and usually get it back, automated reconnection by their app failed basically every time so if you didn't manually exit and restart the app you eventually got either a spinning "loading" thing that never moved or just a black screen -- or an error saying there was a "connection problem" (and implying it was on your end, which it wasn't.)  Worse, when you reconnected they were trying to cheat those bandwidth requirements by resuming from where you lost coherence (presumably they were locally caching in their racks at carrier head-end locations to do so) which meant you were now minutes behind the actual action and if you scrolled back to "live" you got hosed again immediately.

"Live" means live guys, not "well, you can watch from the start where we've cached it at your local cable company headend when we can't deliver what we told you we could."

I tried to switch to my phone from a Sony 4k player that I usually use for streaming and that was fruitless as well -- and it certainly wasn't my Internet link, which has enough capacity to handle 40 such streams at once and was working perfectly fine to everything else. 

Good thing I don't pay directly for this hot mess -- I get "basic" Netflix "free" because I'm a T-Mobile customer -- because if I had been paying for it that'd be the end of me as a paying subscriber.  Oh, its just one time people will say -- yeah, but guess what -- competence matters, this was a seriously promoted event that many people really did want to watch and, well, a huge percentage of them didn't, at least not usably.  In other words everyone, including Netflix, set a high bar themselves -- and failed.

I don't know if Netflix charged silly money for bars and other commercial establishments (as is common for various other things like MMA fights) but if they did the howling from them is going to be quite interesting, along with what I expect to be more than a few attempts to charge it back and, if resisted, I bet it draws lawsuits as well -- and should.

Netflix may be a perfectly-reasonable way watch shows and such "on demand" these days -- something I don't enjoy and thus have no reason to pay for -- but they clearly do not have the chops to handle live events whether in the production or content delivery side of the equation.

Period.

Well well, here it is....

The Consumer Price Index for All Urban Consumers (CPI-U) increased 0.2 percent on a seasonally adjusted basis in October, the same increase as in each of the previous 3 months, the U.S. Bureau of Labor Statistics reported today. Over the last 12 months, the all items index increased 2.6 percent before seasonal adjustment.

The index for shelter rose 0.4 percent in October, accounting for over half of the monthly all items increase. The food index also increased over the month, rising 0.2 percent as the food at home index increased 0.1 percent and the food away from home index rose 0.2 percent. The energy index was unchanged over the month, after declining 1.9 percent in September.

The index for all items less food and energy rose 0.3 percent in October, as it did in August and September. Indexes that increased in October include shelter, used cars and trucks, airline fares, medical care, and recreation. The indexes for apparel, communication, and household furnishings and operations were among those that decreased over the month.

The all items index rose 2.6 percent for the 12 months ending October, after rising 2.4 percent over the 12 months ending September. The all items less food and energy index rose 3.3 percent over the last 12 months. The energy index decreased 4.9 percent for the 12 months ending October. The food index increased 2.1 percent over the last year.

That bolded and particularly the bolded and underlined text is the important one, and is why I included the text of that entire section.

But for the energy index coming down core would be smoking hot, likely close to 4% or double the claimed "target" and within one percent of wage increases which in turn means that any idea that consumers can "earn" their way out of the inflation hole is fantasy-land nonsense.

Yet this premise is absolutely key to the idea that we can "keep" the sort of pricing that is in the marketplace (and thus the sort of stock and other asset prices) that are currently present.  On the number release the market decided 20 handles on the /ES futures must be added.

A caution: All the energy decrease was in gasoline and fuel oil (diesel, along with home heating use.)  Both natural gas and electricity were up on a 12 month basis, the latter more than the former, which strongly implies that policy decisions are largely responsible.  Electricity prices going up 4.5% on the year are a serious inflation risk as electrical power goes into literally everything -- its not just your light bill, it is also the light bill of every store, every warehouse, distribution terminal and manufacturing facility.  This is precisely the sort of "sticky" inflation influence that cannot be absorbed.

Oh, and car insurance?  Up 14% annualized.  You don't need to buy that (if you own a car), right?

CNBC is talking about inflation being "sticky": Yeah, that would be due to electricity costs among other mandatory spending such as health care and shelter.

PS: The PPI is now out as well and, well.... it is confirming this view with almost-exactly the lead time that I expected and worse, well, here's the quote:

Over 80 percent of the broad-based increase in October can be attributed to a 9.9-percent jump in the index for unprocessed energy materials.

Here it comes....

Remember "a permanently high plateau"?

Of course you don't; you weren't alive when it was spoken.

I've recently written on the yield curve.  What most people commenting in that thread failed to recognize (and our various folks in the investing and money business, including Powell) is that these swings are a Kondratiev cycle and there is no nation in recorded history that has managed to evade them.

Tamper with them, try to get around them, make excuses for them, yes.

Actually change them?  No.

As I pen this the market has been on an absolute tear since election day.  The irony is that many of Trump's policy pronouncements, such as removing EV mandates from carmakers (whether explicitly or by setting fuel standards to a level impossible to attain under the laws of thermodynamics any other way, as the Biden path in fact did) means Tesla no longer has a government-driven sales quota.

Now that's not to say that people won't want to buy Tesla vehicles standing on their own in a field of vehicle choices.  The obviously have but Tesla has benefited mightily in cash flow from those "credits" that others have had to buy (from Tesla) for the last several years which has been reflected in both the price of Tesla vehicles (downward) and Tesla's P&L statement (upward.).  That cash flow, if the mandates are removed, will disappear and the cost of them in the price of fuel-powered vehicles will also disappear while at the same time the subsidy in favor Tesla car prices will also go away.  In other words all other things being equal Tesla vehicles become more expensive and fuel-powered vehicles becomes less expensive because the forced payment from other companies to Tesla disappears.

How will all this balance out?  I have no idea but it is flat-out nuts to believe that it will be accretive to Tesla's earnings and thus should be reasonably-reflected in a higher stock price.  Yet that is exactly what has happened.

May I remind everyone that removing deficit spending by whatever means it occurs is good in the longer-term, particularly from an inflation standpoint (since all inflation is in fact caused by creation of credit) but a large part of the last several years' stock market price rise has been driven by that inflationary impulse.

Perhaps the market is telling us that despite all the bluster and claims nothing will be done about any of that and yet somehow we'll evade the march higher of interest rates over the next couple of decades, and evade the already-occurred turn in that cycle.  I've heard a couple of podcasts asserting that in fact rates will continue to come down from here, implying that we're headed back to a 2% long term bond.

No we're not.

Yes, there will be periods in the next twenty or thirty years when rates will fall but the general trend for the next two to four decades, with the median expectation of three decades of time, is upward.

The firms that will fare best in that world are the ones who have no operating leverage (that is, debt) on their balance sheets because you can't be involuntarily exposed to higher financing costs if you have no debt.  You don't have to deal with a ratchet job that isn't present and every firm is different in this regard in terms of its exposure.  A key element any investor in a firm should be looking at right now is the amount of debt on the firm's books compared with its free, unlevered cash flow -- that is, can the firm pay it off if necessary from operating income rather than roll it over or are they stuck with whatever rate environment they are presented when that time comes?  As an example Nvidia has an operating cash flow of about $49 billion with $10 billion out in debt.  Push comes to shove they could pay that off and survive.  It would hit their operating earnings severely but it could be done.

Now look at a firm such as Moderna.  You'd think they're mostly ok because the only have $1.3 billion in debt. The problem is that their operating cash flow is negative so they can't pay it off if necessary.  Whatever the market deals them in terms of a rollover rate they have no choice and must accept it.

How about Tesla?  Their gross operating cash flow is $14.5 billion but they have nearly $13 billion in debt.  While in theory they could pay all that off it would ruin their cash flow for that year.  Nonetheless they'd probably survive such a crunch, although not easily.

How about Amazon?  $112 billion in cash flow but $159 billion in outstanding debt.

Oh by the way this analysis has to extend to suppliers you can't replace easily too.  In the 1990s this was a huge problem as Internet access evolved; the DSL folks were all wildly levered and existing on Wall Street "another shovel of cash please" money, unable to generate free cash flow.

They all blew up when that dried up and if you were dependent on them you blew up too as your customers got cut off.

The usual answer to this sort of analysis is that the firm has plenty of time to figure it out and/or increase their operating cash flows before the debt must be either rolled or paid.  In a generally-declining rate environment this is almost-always true because tomorrow its cheaper to finance something that it is today, everyone knows this and therefore the odds of a dislocation and squeeze are small.

In a generally-rising environment the exact opposite is true: Everyone knows the price will be higher to roll over the debt tomorrow and thus the risk of a squeeze and bankruptcy rises vertically as the debt outstanding .vs. actual cash flow (not including leverage) goes up.

Read balance sheets and key financial statistics folks.  One of the greatest risks in a concentrated market like today's is that an utterly enormous amount of the index price (and thus ETF prices) is concentrated in just a handful of firms, and its very easy to just buy the ETF based on its current price change without looking at all at the concentration of risk or what those firms' balance sheets look like.