..... plus 142,000.

Total nonfarm payroll employment increased by 142,000 in August, and the unemployment rate changed little at 4.2 percent, the U.S. Bureau of Labor Statistics reported today. Job gains occurred in construction and health care.

Ah, more people who become sick (we didn't cause that did we?) and then we have to hire 10 people to make sure you pay for every one who fixes your busted-up body when you have a car accident.

The market appeared to be expecting a disaster and didn't get it; what had been a decently-red futures number before the print went positive and as I write this is just above zero.

The unadjusted household number is -690,000, however, which is certainly not a positive print, and what's additive to that is that 1.171 million more people said "screw this!" and started to couch surf (not-in-labor-force.)  What's incongruent with that is that all four of the subtables for educational attainment had either flat or positive change, so obviously there is a data discontinuity in there somewhere.  Exactly where is tough to suss out, but this is not all that unexpected in the area of turning points -- these are surveys and not exhaustive, so sampling error does occur.

Speaking of which the usual pattern of "errors" in one direction continues with both of the last two months revised downward -- seriously so.  May I remind anyone who has had a stats class that errors are normally distributed; if what is reported is abnormally in one direction it is not an "error" but rather either poor data (in which case you can't trust it at all) or deliberate.  It doesn't matter which if you're trying to make decisions based on said "data."

Of much more important note is the population-adjusted employment data, which has always been a key metric.  Its a lagging indicator and thus useless to predict the future, but trends are still useful to keep an eye on although obviously huge dislocation events (like the pandemic) skew it.  We're well out that zone now and it is trending negative, now on 12 months basis being -1.722 million employed having failed to pu tup a positive print since last November, and that only a small and expected result of holiday temporary hiring.  Discounting that one-month number its been negative since May of 2023.

A large part of this is demographic and there is no way to solve it with immigration despite what some will try to claim because otherwise the data would have inflected the other way by now in the current administration  This is the longer-term consequence of policy decisions going back two decades and continuing on an unbroken basis since irrespective of which party has control of Congress or the White House -- and there is no indication anywhere in the data that it has or will turn.  It should not be expected to do so until and unless policy changes.

This report looks to me a lot like good old Wile-E -- you know, the oft-repeated scene in Looney Tunes where there is no gravity for a short while after he runs off the cliff -- until he looks down.

Amazon thinks people will pay for Alexa?

I'll make a prediction: No they won't.

But this is, after all, the test -- right?  "AI" has to generate revenue somewhere or its just a bunch of people playing with very expensive computers that use a lot of power -- twice, once to run them and then again for the A/C so they don't melt.

Remember that when this stuff first showed up as "customer service" went more-and-more toward automated response systems (which was, at first, "supervised" and in some cases still is, which is an easy "carry" from a human to a robot; you're none the wiser if its a chat window) the prediction was that basically everything would go this way.

It has, kind of -- but only in terms of the "initial" customer service line stuff.  If its "which item are you returning?" then its pretty easy, but really not any different than the same interaction with a mouse on a keyboard when you get down to it.

What hasn't happened is widespread adoption of "digital assistant" stuff that is actually, well, worth something aka "Jarvis."

My prediction: Free works for the here and there, but a subscription will not unless it actually does useful things with near-100% reliability, and I don't believe either of those two metrics will be met.

This isn't like Adobe going to subscription software where they have a "hook" in that your 10,000 prior edited pictures suddenly become inaccessible if you don't pony up the next month's (or year's) fee.  Now you have to get continuing and daily value -- which means it has to be timely, save you time or effort and  be accurate or you're not going to pay.

Unlike a Starbucks coffee you can't drink this one.

My bet is that it fails.

Here we are again.

Once again an analyst asked the key question on the conference call: Where are the revenue applications from customers (not you selling chips, but the customers selling goods and services), when will they appear, and how do you know they're economic?

It wasn't said in exactly those words, but that was the question.

He didn't get a straight answer -- he got 10 minutes of word salad which amounted to "every business will have to have and use this."

Really?

Let's clear up some facts.

First, computers do not "think."  We don't know how humans think and thus we cannot build a computer to do that which we cannot describe.  Every one of us engages in what we see as a routine and trivial undertaking almost every day: We drive a car.  It does not require kilowatts or even megawatts of power to do this; in fact if you're reasonably good at it and have a Garmin you can drive eight hours in your car and your bodily stress level will never leave the "resting" state.  I do it all the time.

This is a pretty-extreme problem for a computer, however.  Tesla claims somewhere between 150 and 300 or so miles per "critical disengagement" which we can reasonably presume means if you didn't take control back you'd bend the car (and probably something else not owned by you) or worse.  That's a nastily-bad record for what humans think of as a trivial task -- clearly, for a computer, it is not.

But what computers are very good at is pattern analysis.  That is, mathematics.  Thus the current rage is all about applying mathematical "models" to data in order to try to answer questions.  This is done by "weighing" (applying mathematical values) to some set of data and attempting to produce a result.  But unlike a spreadsheet where 2 + 2 = 4 each and every time, the data set is extremely vast and precisely how a human would arrive at their answer is not so clear.  What we do know for certain, however, is that said machine is not mimicking the process that a human brain uses to evaluate a question.

It is quite easy, for example, to calculate the moment arms that exert force on a thing, what the load is at some point of that thing (e.g. over a fulcrum or other point of impingement) and then figure out what the material qualities are that you need to be able to withstand that without damage.  That sort of analysis has an exact physical answer and we do it all the time.  The same is true with CAD/CAM; you have your 3-D model and with it you can "trial" the assembly to make sure (1) the parts will go together in some order such that you can in fact assemble them and (2) the assembly, when complete, will take the loads designed -- provided you can describe them mathematically -- without failing.  You can then take that set of files and send it to a robot that will produce the parts with a given set of tolerances.

But that is not AI -- its computational science.

Where is the line between discernment of a deterministic process that has one correct answer and creation, which often does not have only one answer?  That is, how does a machine weigh the intangibles and what error rate, as determined by the humans involved, occurs?

Given a business policy can a machine take input and determine an outcome that has fewer errors than a human?

Maybe; that depends on what we want the machine to do.  What we do know is that the machine is faster -- for example, it can handle 100 calls with voice recognition and response at once, where a human can only handle one or perhaps two at once.  This means it can be 50 to 100 times as efficient but that is only true if its error rate is equal or better than the human error rate.

If its not it might do 50 or 100 times as much damage as the human does when an error is committed simply because its 50 or 100 times as fast!

Now are there tasks which today humans do but machines can do better and faster -- and at lower cost?  Sure. Take the ordinary "voice response" system; I put one in.  Why?  Because finding a human that would show up reliably and answer the frapping phone and direct calls to a department based on what the customer wanted proved nearly impossible so I spent the money.  Today that same task (which was quite expensive then) can literally be done on a $25 computer the size of a pack of cigarettes!  But as the complexity of the task goes up for the computer the price goes up exponentially.

Further, contemplate the premise of the "new hardware" Nvidia says is coming online.  Let's assume its faster, cheaper per-unit of processing and uses less power.  That's how computers typically work out, right?

Ok, so how about the billions of dollars customers spent on the last version?

You see, those are no longer viable on a revenue generating basis: They cost more to operate than the new ones and produce less -- which means if your competitor has the new one and you don't you get buried.  This is always the problem with emerging technologies: If you can't produce revenue with it right now, and so far nobody is in this space, when evolution in capacity occurs your "investment" becomes competitively worthless because your competitor who bought the new version has a lower cost of operation than you do, always has a lower capital cost of acquisition per unit of output the device can produce (or he'd never buy the new offering in the first place) and you've not defrayed any of your capital expense by producing revenue during the interim period!

Of course the markets have gone ahead and added a huge amount of valuation to this company despite these facts.  Never in the history of computing, however, has the above not been, in every case, the correct set of facts.  For forty years I've witnessed this occur countless times -- in CPU, RAM, storage, data transport (e.g. Internet connections) and more.  You've personally seen it yourself countless times -- in personal computers, laptops, cellphones, televisions (tube, 3-tube HD behemoths, DLP, LCD, OLED, 2k, 4k, etc.) video disc players (LaserDisc to DVD to BluRay to 4k) and more.  There has never been an exception across the entire modern electronics and computer age and in fact what has both made and destroyed every computer and electronics-related firm ever has been exactly that progression when the firm's cost of revenue gets undercut by a competitor who can deliver the same or better both faster and cheaper.

I know, I know: This time its different.

No it isn't.

Nvidia makes very fine graphics cards that serve people's needs for years.  But that your competitor can render out a movie in half the time doesn't mean you can't make the release date in the theater or on Netflix and so you don't have to throw away the one-year old card.

In a competitive market where you're all trying to put out "AI" services and the other guy has bought the faster and cheaper one, never mind if the new one consumes half the electricity to produce the same output your billions of "investment" get turned into smoke every time new silicon shows up because in this case time is the entire point and to beat him you need twice the resource which cost twice the money and worse you must use twice the power on a recurring basis which means your cost of delivery is much higher than your competitor's and he takes all the business away from you.  If the output quality improves with density the same thing happens; your "investment" in the former generation again gets turned into smoke.

This is inherent in all emerging technologies but unlike in the 1990s when I was selling Internet service today there's no revenue coming in on the current-generation devices.  We were getting revenue in by the truckload -- the trick in the 1990s was to make enough to pay for the gear before it had to be replaced lest your competitor hit you over the head with the newer hardware (and if he was writing it more-quickly than you could, software too) that was better and faster, burying you.  There was a roughly 18-month "turn cycle" at the time and it was brutal; one serious mistake and you were probably finished and as such I spent a huge amount of my time so as to avoid said mistakes -- and there were several close calls.

But during all of that revenue was coming in -- and today it isn't.  Yes people are "buying" the stuff, although one has to wonder -- how much of this is vendor financed rather than paid for in cash, and how much of that is at risk given the terms and, if the customer blows up what are the odds of meaningful recovery of anything close to the invoiced amount?  Repossessing used gear in a fast-moving technological field may as well be taking ownership of a warm bucket of spit as Lucent discovered when Winstar and others got turned into smoking craters during the tech wreck.

The key question I heard no meaningful answer to -- and this is now on two earnings calls in a row -- is this:

Where is the customer's revenue stream, what's it look like on a coverage basis among your customers on a consolidated basis (that is, how long does it take to pay for your stuff, plus the housing and operating costs, given the customer revenue), how does a new cycle of devices impinge on that from a competitive market point of view on their (your customer's) end, how much visibility do you have into all of that if any and what reason do you have to believe that said revenue from end users HAS AND WILL cover said acquisition and operating cost AT THE CUSTOMER'S END?

If I sell a lot of Beanie Babies it's great so long as people want to buy them -- but how do I justify a stock priced at forty times sales unless I can demonstrate that the people buying my stuff can turn a profit by doing so and that their investment in my products -- the entire reason my stock isn't selling for five bucks a share -- doesn't get reduced to smoke six months from now when the next generation of chips starts to ship?

You'd think after writing these columns for more than 15 years you wouldn't have to say something more than, oh, perhaps 10 times in one before the questions changed to new ones.

But the breathless anticipation of The Fed never does change, does it?  Its a great deflection tactic from the various political folks in Congress and Treasury, both of which are the real lever and the MTS, mandated by law, is where you find truth -- after all, it is sort of a transaction log at the summary level for the entire Federal Government much as anyone can produce for their household with Quicken.

Yet the latest Jackson Hole symposium produced the usual people sitting on the edge of their seats with a picturesque background (seriously; if you've not been there it really is very pretty as the Grand Teton National Park is right there) and, of course, all the money that is attendant.  Opulence, of course, is part of the image projection.

Reality is a bit different.  As I've noted many times over the years (and first noticed when I first started caring about such things; being a CEO has a way of doing that since monetary and fiscal policy have outsized influence on all businesses) the Fed, in nearly every case, just follows along with what the market has already done -- and takes credit for it.  This either produces loud cheers or guffaws; the latter are called "mistakes" and the former "wise decisions."

Thus our Fed Chair's Yoda-style declaration that "now the time is" when it comes to rates.

I didn't hear one person note that the IRX, the interest rate on the 13 week T-Bill which reflects what people actually pay in the secondary market (that is, you pay a discounted price on the bill when you buy it and that discount translates into an implied rate of interest) is now sitting right at 5% while the overnight Fed rate is at 5-1/4 - 5-1/2.  In other words the market has already set rates at 5% for short-term money.

But but but but.... you sputter, the Fed sets rates!

No they don't -- but they want you to believe they do, because absent that The Fed's only job is in fact what their job actually is, which is to clear all the transactions between banks and attempt to make sure that they have a marketplace (by providing one of last resort) so any given bank can have on hand what is necessary to clear your check to buy that car or house tomorrow -- and more-importantly, so your employer's bank can pay you and not have it bounce!

Now certainly it is true that The Fed's various governors, when they speak, influence rates.  The spoken word is quite powerful and that of powerful people built up by the media that produces a God-like aura around them can only amplify that influence.  Who remembers the "Briefcase indicator" from the Greenspan years -- before 24x7 Internet and streaming media?  So I do not discount, nor spit upon, the fact that such influence exists -- it clearly does, it clearly moves markets, and it clearly should be paid attention to, if for no other reason than all the other people in the marketplace do.

But the market in fact controls this issue.  And what the market is telling you right now makes no objective sense.  The MTS, for example (now available through July) says that thus far we have run a $1.517 trillion deficit.  That'll be $1.82 trillion by the time we get to September 30th, assuming of course nothing out of the ordinary happens and thus we're 10/12th of the fiscal year in at this point.  GDP at the most-current read is $28.629 trillion, so the government is running a 6.4% fiscal deficit, that is, the Federal Government is forcing 6.4% of real inflation into the monetary system by spending that much in excess of tax and other receipts.

These are mathematical facts; they're not subject to interpretation.

This in turn means the market is in fact, right now despite the claims of many in the media, operating with negative real interest rates all across the curve -- people are being paid to borrow.  That in turn is a boot on the scale of supply and demand because you can borrow $1 million and the cost of doing it is less than the nominal amount you'd have to spend in one year for the same basket of goods and services as a result of said federal government spending.

That is what is causing asset prices to not only rise but maintain their "value" today, and the further you go out the interest rate time structure the more negative the rates are.

This is an utterly nuts thing from an objective analytical basis and yet here we are.  It exists because people in the market have come to believe that this will always be the case and that there are no limits to the profligacy of Federal spending.

But all things in the physical world have limits.  Your life has limits, your credit card has a limit, every single physical thing obeys the laws of physics and entropy and there are no exceptions to this.

Acting as though there are is delusional -- yet here we are.

The real question is "how far away from a complete disaster are we nationally?"

That's a good question -- and I assure you the answer isn't "infinitely far, and it will never occur."

Yes it will, and there is no evidence we are going to change course.