Archive
Your Revenue is My Expense
I saw a chart a few weeks ago, which someone sent me from The Daily Shot. It has been bothering me since then. Here’s the chart:
Well, it looks amazing and a great reason to invest in global equities, right? Global earnings are surging due to the AI revolution![1]
What bothers me is the simple phrase which is the title of this article. I completely understand that the leaders of the exploding AI industry are going to have drooling analysts projecting great earnings growth far into the future. That may even be correct. But if a company is making bank because their AI tools are tremendous, and their revenues are surging…it means that other companies are paying for that product and so revenues and expenses should net across the corporate landscape. This is not only true of AI, of course; big earnings of oil companies are offset by added costs for businesses and/or declining revenues for other companies selling to consumers with less ex-energy discretionary income.
That is probably the reason that, most years, forward estimates start out high because we can see the concentrated winners clearly but then decline gradually as the dispersed losers become plain. But that doesn’t seem to be happening this time. In fact, it seems like both sides are claiming gains from AI: the suppliers who are selling it, and the buyers who analysts think will show remarkable gains from using AI.
This is a great moment to review again Brad Cornell/Rob Arnott’s “The ‘Basic Speed Law’ for Capital Markets Returns“. I wrote about this most-recently in 2024, in “AI: Even a Big Deal is Smaller Than You Think.” The upshot of their ‘Basic Speed Law,’ which has held up remarkably well over time, is that real earnings cannot grow much faster (or slower) than the growth rate in real per capita GDP. So, unless we decide that AI is going to accelerate per capita GDP growth drastically – and as you can tell from the link to my prior article, I don’t buy that – we need to be skeptical about forecasts of huge earnings growth rates.
Now, a 100-year chart can hide a lot of deviations in the average. The basic speed law here is grounded by fundamental accounting identities. GDP is total expenditure on all goods and services, so it makes sense that total corporate revenues (and expenses) should be pretty highly correlated to that figure. So if the share of that expenditure that is profits increases – which is what the analysts are saying is going to happen – then something else needs to be true:
- Margins – and in this case global margins – need to expand as companies take more money from labor (if revenues minus costs go up, and revenues are constrained by GDP, costs need to go down and there’s only so much you can do with materials costs). But at least in the US, margins have been historically wide for some time. NIPA profits as a percentage of GDP are in the 12% range, which is already almost double of what it was pre-GFC. How much can this go up? Populism is already on the rise.
- Government spending can increase – this is another way to extract money from others, and that is to get the government to buy your stuff and sell debt to get the money to pay for it. Government deficits are part of the Kalecki equation. However, at least in the US the government deficit is shrinking. This might not be true this quarter since the US had to refund huge amounts of tariff revenues, but the general trend is towards better balancing of US trade and fiscal deficits.
- Consumers can spend a higher percentage of their income, and save less. This is also part of the Kalecki equation, and indeed personal savings have been declining. Over the last two years, the decline in personal savings has been the source of about 90% of the rise in corporate after-tax profits. See the chart below showing net corporate profits versus net savings (inverted, right scale). So…how much less can Americans save? Currently, savings is 2.72% of GDP. The last time it was lower (with the exception of right after COVID when people were spending their stimmy checks) was 2005-2008. And that was the all-time low of 1.5-2%. Before COVID, we were around 5%. Back in the early 1980s when interest rates were more attractive, savings was more like 7-8% of GDP.
- US companies can increase revenues and profits at the expense of non-US companies. That doesn’t really help us with the chart at the top of this article, which is global.
- Public companies can make hay at the expense of private companies. I totally believe in this one, since I manage a private company that is perpetually running into rule systems put into place to protect the big companies. But it doesn’t square with the AI promise, nor the fact that new business formation is robust and has been so for several years…and does not appear to be ebbing. These could all be profitless businesses – but now you have to explain why profitless businesses are exploding.
- Earnings per share could be increased by decreasing share count. That’s been the game for many a year, adding several percentage points to EPS growth by decreasing the “S”. There are some suspicions that that trend is reversing, with a lot of high-profile IPOs throwing a lot of shares onto the market. I don’t have good data on that. But I doubt that share buybacks are increasing.
- An increase in overall leverage, either financial leverage (debt/equity) or operating leverage (more fixed costs), could cause forward earnings growth to look better. It is instructive that over the last hundred years – looking at the ‘speed law’ chart – the gradual, pervasive increase in both forms of leverage has been associated with only a tiny acceleration in profits growth over time. I do think it has been a factor, but additional leverage is not freely available without bound. However, this might be part of the story – not because leverage is increasing broadly and suddenly, but because the composition of the indices is shifting to more high-margin and high-operating-leverage (but interestingly, very low-financial-leverage) businesses. Still, that doesn’t really explain the upward scoop in the global indices that cover a huge number of companies. Even accounting for the fact that those indices are generally market-cap-weighted, it seems far too large an effect to be mere substitution.
In short, while I think parts of these explanations have a kernel of truth to them…I think a lot of what we are looking at is analyst myopia. Analysts are focusing on the shiny objects of great worth, and are missing, for now, some of the offsetting effects on other companies. It may be possible for those earnings estimates to even be realized, if enough revenues are shifted temporally relative to costs (exciting companies realize the revenues now; boring companies capitalize the AI expenditures or defer the costs to later quarters). Sooner or later, though, it all must add up. Right now, in my mind it doesn’t seem to add up.
[1] As an aside, for much of the rest of this article I will refer to the US case. Not only is the US contribution to these earnings the largest by far but also – if the source of the surge is AI, then the US is where we would expect that surge to be sourced. I just mention this here because someone will noted the disconnect between this global chart, and the commentary based on US trends.
Inflation Guy’s CPI Summary (May 2026)
While the worst is probably over for the monthly CPI prints, the real question going forward is ‘how much better does it get?’ We know that energy prices will eventually retreat, but even if they merely flatline they will stop flattering headline CPI. But where does core and more importantly median CPI settle in? That’s the real question. For now, we just have one more month of data so let’s dig in.
The economist surveys had CPI at +0.51% headline and +0.27% core. The inflation swaps market was in roughly the same place, with +0.66% NSA the last trade.
The last month has seen a lot of volatility in markets (duh), but in particular a significant increase in real yields.
Nominal yields have risen a bit, and get all the ink when 30-year yields peek above 5%. That actually masks the real problem, which is that nominal yields have risen so little only due to the sharp decline in inflation expectations. In the front of the curve, that decline in breakevens is significantly a carry phenomenon (as we roll through months with solid NSA accretion) and so, therefore, is some of the rise in real yields. But a 40bps increase in real yields at the 5-year point gets one’s attention. And at 2.17%, 10-year real yields are near the absolute highs they’ve seen since the Lehman-related spike over 3% in 2008.
(Oh, and if any technician tells me this is a ‘flag’ formation projecting to 5%, I’m going to smack you. Real yields don’t go to 5%.)
The uptrend in real yields, as an aside, has also been bad for gold. Gold behaves like a long duration TIPS bond and as TIPS have sold off, gold has been a whipping boy. That won’t last forever. I like buying 10-year TIPS anywhere north of 2%, and if real yields get above 3% some day – back up the truck. Let’s hope that isn’t soon though.
The actual data comes in today at +0.473% m/m on headline CPI and +0.208% on core CPI. Both of those are below expectations, with core a meaningful miss. Here are the last 12 core CPI figures (keep in mind that last month’s jump was a payback for the 6-months-ago quirk in rents due to the government shutdown).
And here are the m/m, y/y, and prior y/y for 8 major subgroups. It’s striking that Apparel is +4.8% and “Other” is +4.9%. Those are not usually exciting categories! I’ll return to this a little later.
Here is the coarse breakdown of core goods (+1.06% y/y) and core services (+3.42%).
It isn’t surprising that core goods is decelerating. It’s actually somewhat concerning that it isn’t decelerating faster. The hook higher in core services bears some further investigation. Even though the overall numbers looked good this month, this breakdown isn’t all sunshine and roses.
Primary rents were +0.36% m/m, and 2.92% y/y versus 2.79% last month. OER was +0.3% m/m, 3.32% y/y. The jump in Rent of Primary Residence is a little concerning (although I will note that it puts the actual y/y number exactly on our model, which goes pretty flat near this level for the next year). Last month’s hook higher made sense because of that make-up month due to the shutdown/6-month lag effect. But that isn’t the issue here. Lodging Away from Home was alto +0.4% m/m, 5.2% y/y. Some people will say this is a World Cup effect, and possibly we are seeing a little bit of that in Lodging Away from Home. But this isn’t France. The US is a pretty huge country and there is no way that World Cup tourism is enough to move rents for the entire country.
But landlords are seeing higher direct and indirect costs. This is why rents are not going to go into broad deflation any time soon.
What might also be flagged as a World Cup effect, but more likely is just jet fuel pass-through, is the 2.69% m/m rise in Airfares after a 2.82% rise last month. I have airfares just slightly above the model given jet fuel prices, and within the error bars, so if there’s an impact there it’s pretty small. I think this is worse in Europe. Airfares are indeed part of the story in the core-services move higher that I noted above. Rents are too, but the airfares increase is easier to figure out.
The median category this month looks like Recreation at 3.51% annualized m/m. I may be slightly low, depending on where the regional rent indices get adjusted, but my guess for Median CPI is +0.287% m/m for a small acceleration y/y to 2.83%.
I have to say – if you shovel some of April’s jump back into October and November where it belongs – it still doesn’t look like deceleration to me.
Okay, here are the four pieces charts. Food & Energy +9.81% y/y. Core Commodities +1.06% y/y. Supercore +3.55% y/y. Rent of Shelter +3.33% y/y
The Core Services less Rent-of-Shelter (Supercore) is the one I don’t like, but again part of that is the Airfares/energy thing. None of this looks like super good news, though.
By the way, there were only 3 categories that declined at a faster annualized rate than 10% this month: Car and Truck Rental (weird) at -40% annualized, Motor Vehicle Insurance (very weird) -18%, and Misc Personal Goods -11%. Above 10%, and ignoring food and energy, we have Jewelry/Watches (41%), Misc Personal Services (28%), Communication (17%), Tobacco and Smoking Products (+13%), Infants/Toddlers Apparel (+12%). Motor Vehicle Maintenance and Repair just missed the cut at +9.95% annualized.
Here are a couple of interesting things I’m watching. This chart is Computer Software and Accessories, and it’s where AI tools land. Now, it’s a tiny, tiny part of the basket at the moment but I’ll bet it’s larger when they reweight next year. This is the NSA price index, not the rate of change – so prices for computer software and accessories are higher than they’ve been for years.
Like I said, this is a tiny category but it actually matters more for PCE. That fact annoys the Fed, who just published a research piece (https://www.federalreserve.gov/econres/notes/feds-notes/measurement-of-computer-software-and-accessories-inflation-20260522.html) explaining that this is partly ‘measurement error.’ Uh-huh. Boy are we getting picky.
Also, I happened to notice that computer prices are actually increasing, due to upward pressure on DRAM and other component prices thanks to AI demand. This sort of annoys me because I need a new laptop and the prices aren’t going down like they usually do if you wait. And yes, that’s the main reason I noticed.
This is one of those categories that animates conspiracy theorists because the BLS hedonically adjusts it…so since computers are always improving, there’s a general decline in the quality-adjusted-price over time. Well, not just a general decline, but a pretty large decline. “But computer prices haven’t actually fallen! Yes I get more computer but I don’t have the option to get the old one! I want my Windows 95!”
So prices going up, at least if it continues, is interesting. It’s a symptom of AI demand. It’s still not a large part of CPI, but I think AI is going to start showing up more here and there. Like here:
This is obviously not all AI – the upswing happened in the aftermath of COVID, possibly partly because work from home means power needs are broader throughout the day. But the continuation recently…I am pretty sure AI data center demand, if it isn’t yet affecting this, is going to!
And that is emblematic of the long-term story here. Not this month’s story per se. But these upstream pressures on petroleum and electricity are passing through more and more downstream. And that’s a hard dynamic to arrest. It is difficult to put that genie back in the bottle.
Now, this isn’t to say there is no good news.
This is Medicinal Drugs, aka pharmaceuticals, in core goods. It was -0.8% this month after -0.3% last month, and is -2.2% y/y. This looks like a real TrumpRX effect. On the other hand, Hospital Services is still rising at about 6% y/y, so while overall Medical Care in the CPI is +2.6% y/y, that’s being flattered because of the TrumpRX effect which won’t last forever.
Now, earlier I pointed out Apparel’s interesting y/y increase. The absolute price of apparel basically peaked in 1993 or so. This is a wonderful picture of the power of offshoring, as we went from producing a lot of apparel domestically to producing basically nothing, and saw apparel prices decline in real terms and even in outright terms for nearly 30 years. There was a sharp dip and recovery due to COVID, but in the last year or so the Apparel index has actually gone to new all-time highs.
Some of that is re-onshoring. Some right now is actually petroleum since many types of fibers are downstream petroleum byproducts. Think polyester, but it’s broader than that. But prices prior to the energy spike were already at 20-year highs.
The bottom line here is that the rise in the headline CPI is causing some people to shrilly declare that the Fed needs to raise rates. That’s ridiculous – the Fed looks through energy price increases. Although as I said before, those energy price increases, if they are sustained long enough, start to percolate through, and they appear to be…core and median and trimmed mean CPI won’t be heading back to target (not that there is a target any more) any time soon. As long as the economy stays pretty strong, the Fed has actually stumbled into what should be a comfortable spot for a while. Warsh will work on trimming the balance sheet, hopefully, but I wouldn’t expect rate changes for a while and that’s a change in my view from before when I thought the Fed would be easing (I thought growth would be weaker and I continue to be confounded on that). I’m saying that while the headline inflation data look ugly, that will pass as energy prices decline. But I do think it will be difficult for the Fed to get comfortable with Median CPI going back up, or just not going back down, while growth is strong.
“Mike, Mike, Mike…you’re making too much of these little things! Core doesn’t look too bad. Median is not alarming.”
Yes. But electricity, petroleum, the demographic pivot, re-onshoring, and let’s not forget money growth. These are not small things and they affect how difficult the future looks with respect to inflation. The tree’s leaves are pretty but the trunk is rotten. I am not optimistic about future shade.
A New Era of Positive Stock/Bond Correlations and What That Means
I read recently – I can’t find where – that stock/bond correlations in the US are the highest (most positive) they have been in decades. This of course is bad news for investors who commonly allocate to both stocks and bonds with the expectation that adding bonds will reduce the risk of a portfolio not only because they have a lower natural volatility than do stocks but also because the expectation of negative correlations between them have the effect of lowering the volatility of the portfolio further (since the variance of a 2-asset portfolio is equal to the weighted sum of the variances plus 2 times the product of the weights and the covariance between the two assets. So, when two assets are negatively correlated, total portfolio variance is lower than the sum of the weighted variances of the assets; when they are positively correlated, total portfolio variance is higher than the sum). That’s sort of Portfolio Management 101, but since most of my readers are not professional portfolio managers: think of one person pushing another person on a swing. If they’re pushing in rhythm with the swing (positive covariance), then the person on the swing goes higher and higher. But if they’re pushing in the opposite rhythm (negative covariance), then the swing goes up less and less and Dad is telling the kid it’s time to go home.
So, this matters a lot for portfolio construction and optimization, of course.
By the way, it isn’t like this just started happening. I’ve been warning about this (and showing the chart I am about to show) since at least 2019. In 2022 I even had a nice table to go with the chart (see this year-end piece, and scroll to the “Other Things” part at the end https://inflationguy.blog/2022/12/22/2022-year-end-thoughts-about-2023/ ). But let’s update it.
This heavy line in this chart shows the rolling 3-year correlation of monthly returns of stocks and bonds, going back to 1948 (I sourced equity returns from Ken French’s site based on CRSP data; bond returns I estimated based on Shiller’s lengthy series). You will notice that stock/bond correlations are not guaranteed to be negative – in fact, for the 35 years or so prior to 1998, correlations were positive. The shaded area illustrates the salient point, and that is that correlations tend to flip when inflation gets sustainably over about 2.5% (the shading is positive when 3-year compounded inflation is above 2.5%, and negative when it is below). That’s not coincidence. The simple way to explain it is that stocks and bonds react very similarly to the inflation factor and very differently to the growth factor. That is to say, when there’s news about good economic growth, then stocks tend to rise and bonds tend to sell off (yields rise because real yields rise). But when there is bad news about inflation, then stocks tend to fall and bonds also tend to fall (yields rise because inflation expectations rise). So, in periods where inflation is low and stable, the growth factor dominates and stocks and bonds move in different directions; in periods where asset markets perceive inflation risk, stocks and bonds tend to move together more often.
By the way, this shifting of correlations isn’t only true with stocks and bonds. The entire correlation matrix between many asset classes experiences a shift when the inflation-state changes. But since portfolios tend to be most heavily weighted in stocks and bonds, and because the math gets quite a bit uglier when we add more assets, we tend to focus this sort of discussion on stocks and bonds.
Again, the point of this is that portfolio optimization routines – which tend to be built on covariance matrices built from some recent window of historical data – will tend to completely miss this shift unless portfolio managers intervene, and portfolio managers are loathe to mess with the models.
How much does it matter?
Let me introduce another concept. A ‘risk parity’ portfolio is one in which the assets are weighted in such a way that they each contribute the same amount to the overall variance of the portfolio. So, since bonds are lots less volatile than stocks in general, a risk-parity allocation means that you’ll tend to hold a lot more weight in bonds.[1] Suppose stocks have over time a 15% standard deviation and bonds have a 7.5% standard deviation (which isn’t that far off, actually). Then the weight in stocks, ignoring the stock/bond covariance for now, is 7.5% / (15%+7.5%) = 33.33%; the weight in bonds is 15%/(15%+7.5%) = 66.67%. The 2/3 of your portfolio that is in bonds will contribute 66.67% x 7.5% = 5% to your portfolio risk, and the 1/3 that is in stocks will contribute 33.33% x 15% = 5% to your portfolio risk. That’s the ‘parity’ in risk parity.
Now, true risk parity is done with variances, not standard deviations, and also takes into account the correlation between the assets – and here’s where it gets interesting. If I assume stocks and bonds have a correlation of -0.3, then my weight in stocks is more like 26% and my weight in bonds 74%. But, if the stock/bond correlation is +0.3, the weight of stocks drops to 10.5% and bonds go to 89.5%. So that correlation shift should cause you to cut your holdings of stocks by 60%, from 26% of your portfolio to 10.5% of your portfolio!
[“Heck,” you say. “I gotta hold more stocks than that! I can handle the risk!” That’s fine. The risk parity proposition is merely that you get better returns per unit of risk if you equate the marginal contribution of risk. With stocks, since 1948 you’ve earned 11.74% annualized through the end of April (relax, we are ending this accounting in the middle of a bubble so of course it looks stupid), on annual risk of 14.85%. So every 1% of risk got you 0.79% return. On the other hand, the naïve risk parity got you 7.47% return on 7.15% risk, so you got 1.04% return per unit of risk. And that’s where the risk parity firms will lever up that portfolio so you get similar to equity risk or at least 60/40 risk.]
Again, my point though is not to argue for risk parity. My point is that shifting the correlation between stocks and bonds given even basic approaches to portfolio construction implies a significant reduction in equity risk is in order in an inflationary environment – and that doesn’t even consider the fact that inflation tends to lower market-clearing equity multiples so that prospective equity returns are lower in that kind of environment. So if the new higher-inflation era (and it appears ever more difficult to refute the notion that we are in one) means that investors either need to accept higher levels of portfolio risk or to shed equity risk…where is the stock market selloff?
Your guess is as good as mine. Either (a) investors still don’t believe that inflation is going to be persistent (although the flip in correlations suggests they do), or (b) investors are willing at least for now to hold more portfolio risk in order to harvest the fruits of the AI valuation explosion, or (c) portfolio managers are loathe to cut equity exposures because they don’t want to lose performance to their peers (since actual customers tend to look at returns, not risk-adjusted returns!). I think the answer is some combination of (b) and (c). But both of those reasons are ephemeral, and depend on continued momentum. Given the valuation levels in the equity market, a prudent manager will be at least trimming risks opportunistically these days.
[1] Since over time, stocks have better returns than bonds, people tend to hold more stocks than bonds and firms who deploy risk-parity portfolios typically employ leverage so that they aren’t sacrificing stock allocations so much as adding levered bonds. Anyway, a mean-variance optimization done correctly makes more sense than risk parity, but I’m just using risk parity as a way to illustrate the size of the effect a correlation shift can have on a portfolio.
