I don’t love charts that purport to compare apples and oranges.
I say “purport” because you can’t compare apples to oranges. That’s the whole point of the “apples to oranges” idiom.
But when it comes to producing engaging macro-market content, the only thing that “sells” better than fear are apples to oranges charts. (A close third on that list are double y-axis charts with the scales adjusted such that two series which aren’t correlated appear to track each other 1:1.)
In the context of the AI bubble (and yeah, it’s a bubble), a popular technique for creating apples to oranges charts involves juxtaposing the market cap of America’s largest companies with national GDPs other than America’s and China’s.
You’ve probably seen such charts a hundred times before, and yet you probably enjoy them every, single time you (re-)see them, which speaks to the indispensability of apples to oranges charts for macro-market documentarians.
Without further ado, the figure below is perhaps the most exhaustive attempt at the market cap versus GDP chart I’ve seen to date. It’s from Goldman’s Peter Oppenheimer whose tomes are what you’d get if BofA’s “Flow Show” were written by someone who gives a damn. (That’s a funny joke, although most of you won’t get it.)
As you can see, Tesla’s “bigger” than Indonesia; Meta than Australia and Spain; Amazon than Canada, Brazil and the entire French and German equity markets (considered separately); Apple than the UK; Microsoft than India; Nvidia than Japan; and the top five US stocks than all of that, as well as Germany and the entire European equity market as proxied by the Stoxx 50.
“The biggest five technology companies in the US are worth around 16% of the entire global public equity market,” Oppenheimer remarked, driving home the point and adding that when you cast a wider net to include the top 10 largest US stocks, that cohort “account[s] for nearly 25% of the global equity market.”
Many investors, he went on, “have started to ask whether all of this is rational or if we are seeing the classic signs of an unsustainable bubble.”
The table above hints at Goldman’s answer to bubble inquiries. Taking the median, the Mag7 trades on a 27x multiple looking out 24 months. That’s considerably lower than where the largest seven companies traded during the dot-com boom, and nowhere near the 67x two-year forward multiple for Japanese shares before that bubble burst.
“It is true that valuations are high but, in our view, generally not at levels that are as high as are typically seen at the height of a financial bubble,” Oppenheimer went on to say.
Coming full circle, it’s worth noting that the top five largest US companies are now the size of China using the GDP comparison. And the top 10 are very close to trading 1:1 with the US economy.
Interesting. I have been alive through all of these and find this somewhat surprising. It would have been neat if the bank had put the rate on the ten year next to each set of columns.
Call me crazy, but I’m not sure it’s the PE comparison to dot com bubble that stands out. It’s the market cap of the Magnificent 7 that’s 8x the market cap of the 2000 tech titans.
If google can be trusted, cumulative inflation since 2000 is 88%. Let’s just say inflation doubled the last 25 years and now you have a Mag 7 market cap that’s roughly 4x what we saw in 2000 and we still have a PE of 27? Seems to me that it’s a lot harder to grow into a valuation when all these companies have already reached the stratosphere, but then again, I tend to be wrong more often than not.
On a related note, it sure would be nice if we’d actually tax wealthy people and corporations (I guess that’s redundant by the Supreme Court’s definition).
Between 1995 and March, 2000, the Nasdaq rose 600%. From March, 2000 through October 2002, the Nasdaq fell 78% and gave up all of those gains.
Here is what was going on with 10 year rates, deficit spending and the Fed’s balance sheet:
Between 1995 and March, 2000; the 10-year dropped from 7.78% to 6.26% (over the time frame that the NASDAQ went up 600%). So easing. As a side note: during that period when there was a long drop in the stock market, (March, 2000 to October, 2002), the 10-year dropped to 3.94%.
Compare that to last 24 months: 10 year has consistently been in the low 4’s%.
Here is what the federal government revenues and surplus/deficit looked like (fiscal year) during bubble buildup:
1995: $1.352T, $.164T deficit
1996: $1.453T, $.107T deficit
1997: $ 1.578T, $0.02T deficit
1998: $ 1.722T, $.069T surplus
1999: $ 1.827T, $.125T surplus
2000: $ 2.025T, $. 236T surplus
So (almost) balanced to slightly restrictive and nothing like what we are seeing today:
2023: $4.4T, $1.7T deficit
2024: $4.9T, $1.8T deficit
2025: $5.2T, $2.1T deficit
With no end in sight because none of our politicians can agree on anything; with the exception that they do all agree that they want to spend more money.
It is true that the Fed’s balance sheet has dropped from a peak of $8.9T in May, 2022 to $6.6T in October, 2025. For comparison purposes, the Fed balance sheet, prior to the crash in 2000, was a nonissue as it was pretty flat and well below $1T.
So, of course I am nervous about losing my tech gains. However, I absolutely don’t want to fight deficit spending/the Federal Reserve (trending towards being controlled by the same policy group).
I keep coming back to look at your art for this post. The composition, simplicity and dark background of just a few fruits on a table weirdly reminds me of my favorite, but not well known, Dutch still life painter, Adriaen Coorte.