Grey Value Management

𝗖𝗼𝗻𝗰𝗲𝗻𝘁𝗿𝗮𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗠𝗼𝗺𝗲𝗻𝘁𝘂𝗺, 𝗔𝗱 𝗡𝗮𝘂𝘀𝗲𝘂𝗺

Jon Sindreu recently did a great piece for Reuters Breakingviews on the impact of risk concentration on valuation in the equity markets (link below).

More than half of the S&P 500's market capitalization is represented by companies driving the AI wave.

Korea and Taiwan, central to the tech supply chain, account for 51% of the MSCI Emerging Markets index.

This has nurtured a 𝗴𝗶𝗮𝗻𝘁 𝘁𝗿𝗲𝗻𝗱 𝗼𝗳 𝗯𝘂𝘆𝗶𝗻𝗴 𝘀𝘁𝗼𝗰𝗸𝘀 𝘁𝗵𝗮𝘁 𝗮𝗿𝗲 𝗮𝗹𝗿𝗲𝗮𝗱𝘆 𝗴𝗼𝗶𝗻𝗴 𝘂𝗽.

The MSCI USA Momentum index outperformed broader equities to an unprecedented extent between March and June, and has now plummeted just as quickly (see below).

Yet 𝗮 𝗹𝗼𝘁 𝗼𝗳 𝗶𝘁 𝗿𝗲𝗳𝗹𝗲𝗰𝘁𝘀 𝗶𝗻𝘃𝗲𝘀𝘁𝗼𝗿𝘀 𝗿𝗼𝘁𝗮𝘁𝗶𝗻𝗴 𝘁𝗵𝗲𝗶𝗿 𝗺𝗼𝗻𝗲𝘆 𝘄𝗶𝘁𝗵𝗶𝗻 𝗮 ‌𝗰𝗹𝗼𝘀𝗲𝗱 𝗰𝗶𝗿𝗰𝘂𝗶𝘁 ⁠𝗼𝗳 𝗵𝘆𝗽𝗲𝗿𝘀𝗰𝗮𝗹𝗲𝗿𝘀, 𝗰𝗵𝗶𝗽𝗺𝗮𝗸𝗲𝗿𝘀 𝗮𝗻𝗱 𝘀𝗲𝗿𝘃𝗲𝗿 𝗳𝗮𝗿𝗺𝗲𝗿𝘀, which are furiously competing to spend as much as possible in the hopes that all of them will simultaneously make good on the investment.

Investors aren't blind to this risk: the Nasdaq 100 trades at a lower multiple of its constituents’ earnings than it did at the start of the year.

It’s just that valuations in the rest of the world and among lower-growth "value" stocks are also flat or lower. Neither benefited from the recent outflow from AI themes, suggesting that 𝗶𝘁'𝘀 𝗺𝗼𝘀𝘁 𝗼𝗳 ⁠𝘁𝗵𝗲 𝗺𝗮𝗿𝗸𝗲𝘁, 𝗻𝗼𝘁 𝗷𝘂𝘀𝘁 𝘁𝗲𝗰𝗵, 𝘁𝗵𝗮𝘁 𝗵𝗮𝘀 𝗵𝗶𝘁 𝗮 𝘃𝗮𝗹𝘂𝗮𝘁𝗶𝗼𝗻 𝗰𝗲𝗶𝗹𝗶𝗻𝗴.

Read the full article, it's very good:
https://www.breakingviews.com/columns/considered-view/theres-no-escape-ai-whipsaw-2026-07-31/

𝗥𝗲𝗺𝗲𝗺𝗯𝗲𝗿 𝗪𝗵𝗲𝗻 𝗕𝗲𝘁𝘁𝗶𝗻𝗴 𝗼𝗻 𝗡𝗼𝘃𝗼 𝗡𝗼𝗿𝗱𝗶𝘀𝗸 𝗪𝗮𝘀 𝗮 "𝗡𝗼-𝗕𝗿𝗮𝗶𝗻𝗲𝗿"?

GLP-1 was undeniably transformative science.

But here's the thing:

𝗨𝗻𝗱𝗲𝗻𝗶𝗮𝗯𝗹𝘆 𝘁𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝘃𝗲 𝘁𝗲𝗰𝗵𝗻𝗼𝗹𝗼𝗴𝘆 𝗶𝘀 𝗻𝗼 𝗴𝘂𝗮𝗿𝗮𝗻𝘁𝗲𝗲 𝗼𝗳 𝗽𝗿𝗼𝗳𝗶𝘁𝘀. 𝗜𝗻 𝗳𝗮𝗰𝘁, 𝗺𝗼𝗿𝗲 𝗼𝗳𝘁𝗲𝗻 𝘁𝗵𝗮𝗻 𝗻𝗼𝘁, 𝗶𝘁 𝗹𝗲𝗮𝗱𝘀 𝘁𝗼 𝗹𝗼𝘀𝘀𝗲𝘀.

Why?

It seems counterintuitive, but whether it was railroads, aviation, electricity, the internet, or (prospectively) AI, 𝘁𝗲𝗰𝗵 𝘁𝗵𝗮𝘁 𝗶𝘀 𝘁𝗿𝘂𝗹𝘆 𝘁𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝘃𝗲 𝗶𝗻𝗲𝘃𝗶𝘁𝗮𝗯𝗹𝘆 𝗮𝘁𝘁𝗿𝗮𝗰𝘁𝘀 𝗮𝗻 𝗲𝘅𝗰𝗲𝘀𝘀 𝗼𝗳 𝗰𝗮𝗽𝗶𝘁𝗮𝗹 𝙥𝙧𝙚𝙘𝙞𝙨𝙚𝙡𝙮 𝙗𝙚𝙘𝙖𝙪𝙨𝙚 𝙞𝙩'𝙨 𝙪𝙣𝙙𝙚𝙣𝙞𝙖𝙗𝙡𝙮 𝙧𝙚𝙫𝙤𝙡𝙪𝙩𝙞𝙤𝙣𝙖𝙧𝙮.

Excess capital drives competition and excess capacity, which leads to losses when the supply ultimately exceeds demand.

𝗖𝗮𝗿𝗲𝗳𝘂𝗹 𝗶𝗻𝘃𝗲𝘀𝘁𝗼𝗿𝘀 𝘁𝗿𝘆 𝘁𝗼 𝗮𝘃𝗼𝗶𝗱 𝗰𝗼𝗻𝗳𝘂𝘀𝗶𝗻𝗴 𝗶𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗶𝗻𝘃𝗲𝘀𝘁𝗺𝗲𝗻𝘁 𝗿𝗲𝘁𝘂𝗿𝗻𝘀.

$SPCX

𝗗𝗶𝗿𝗲 𝗦𝘁𝗿𝗮𝗶𝘁𝘀

Bloomberg journalist John Authers has posted a really good two-minute video explaining the significance of Iran’s attempt to add a threat to blockade the Bab el-Mandeb passage to the Red Sea to its ongoing campaign to halt traffic in the Strait of Hormuz:

https://www.youtube.com/shorts/LGLdxo51VE4

‌

A closed Hormuz is bad. Bab el-Mandeb could be worse is leaning on its allies in to try closing off...

“𝗧𝗵𝗲𝘆’𝗿𝗲 𝗿𝗲𝗮𝗹𝗹𝘆 𝗴𝗼𝗼𝗱 𝗮𝘁 𝘀𝗼𝗺𝗲𝘁𝗵𝗶𝗻𝗴 𝘁𝗵𝗮𝘁 𝗶𝘀 𝗴𝗼𝗶𝗻𝗴 𝗼𝘂𝘁 𝗼𝗳 𝗳𝗮𝘀𝗵𝗶𝗼𝗻 𝗳𝗮𝘀𝘁."

As the author writes:

The long-term dilemma for Volkswagen and other German automakers is that their complete mastery of internal combustion engine technology does not count for much in a global market that is racing toward electrification.

Read Keith Bradsher's excellent The New York Times article here: https://www.nytimes.com/2026/07/10/business/volkswagen-china-problems-restructuring.html

For Those Playing the Hardware Angle of AI

China’s LineShine supercomputer just won the #1 spot on the latest Top500 supercomputer ranking released this week by calculating a full 22% faster than the #2 contender, El Capitan at California’s Lawrence Livermore National Laboratory.

LineShine relies on domestically developed central processing units, instead of graphics processing units, or GPUs, which power most of the leading supercomputers today. The machine also uses other homegrown technologies behind its memory, networking and cooling systems.

Washington curtailed Chinese supercomputer developers’ access to Intel chips and other American hardware in 2015.

The Biden administration later blocked the country from accessing powerful GPUs, mostly developed by American chip giant Nvidia, and the tools required to produce them.

In response to these foreign technology restrictions, China built an independent hardware and software ecosystem.

AI of doesn't require supercomputing capacity, but many have been assuming that it does depend upon cutting-edge GPUs.

But if China can build the world's fastest computer without GPUs, what does that mean for GPU demand going forward?

𝗛𝗲𝗱𝗴𝗲 𝗙𝘂𝗻𝗱𝘀 𝗥𝗲𝗼𝗽𝗲𝗻 𝗣𝗿𝗲-𝗪𝗮𝗿 𝗣𝗹𝗮𝘆𝗯𝗼𝗼𝗸

Last night I had a very interesting exchange with Ruth Carson at Bloomberg re. the market response to the - agreement.

It was inarguably good news, but at the same time it would clearly be premature to start hanging the bunting. Arriving at a final, functional agreement will be fiendishly complex, and maintaining the ensuing peace won't be any easier.

If we have indeed entered a post-war phase, the investment opportunity set doesn't look all that different than it did pre-war.

US equity valuations are stretched, and as I discussed with Ruth, “𝘄𝗶𝘁𝗵 𝘁𝗵𝗲 𝟭𝟬-𝘆𝗲𝗮𝗿 𝗻𝗼𝘁𝗲 𝗼𝗳𝗳𝗲𝗿𝗶𝗻𝗴 𝗼𝗻𝗹𝘆 𝗮𝗯𝗼𝘂𝘁 𝟰𝟬 𝗯𝗮𝘀𝗶𝘀 𝗽𝗼𝗶𝗻𝘁𝘀 𝗺𝗼𝗿𝗲 𝘁𝗵𝗮𝗻 𝘁𝗵𝗲 𝘁𝘄𝗼-𝘆𝗲𝗮𝗿, 𝘄𝗲 𝘀𝗲𝗲 𝗻𝗼 𝗿𝗲𝗮𝘀𝗼𝗻 𝘁𝗼 𝗿𝗲𝗮𝗰𝗵 𝗳𝘂𝗿𝘁𝗵𝗲𝗿 𝗼𝘂𝘁 𝗶𝗻 𝘁𝗵𝗲 𝗰𝗮𝗹𝗲𝗻𝗱𝗮𝗿 — 𝗼𝗿 𝗳𝘂𝗿𝘁𝗵𝗲𝗿 𝗱𝗼𝘄𝗻 𝗶𝗻 𝗰𝗿𝗲𝗱𝗶𝘁 𝗾𝘂𝗮𝗹𝗶𝘁𝘆 — 𝗳𝗼𝗿 𝘆𝗶𝗲𝗹𝗱.”

𝗜𝗻 𝗼𝘂𝗿 𝘃𝗶𝗲𝘄, 𝘄𝗲'𝘃𝗲 𝗲𝗻𝘁𝗲𝗿𝗲𝗱 𝗮 𝘀𝘁𝗮𝗴𝗲 𝘄𝗵𝗲𝗿𝗲 𝗶𝗻𝘃𝗲𝘀𝘁𝗺𝗲𝗻𝘁 𝗼𝘂𝘁𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲 𝘄𝗶𝗹𝗹 𝗱𝗲𝗽𝗲𝗻𝗱 𝘃𝗲𝗿𝘆 𝗺𝘂𝗰𝗵 𝗼𝗻 𝘄𝗵𝗮𝘁 𝘆𝗼𝘂 𝙖𝙫𝙤𝙞𝙙 𝗼𝘄𝗻𝗶𝗻𝗴.

Moreover, we think 𝙩𝙝𝙚 𝙤𝙥𝙩𝙞𝙤𝙣𝙖𝙡𝙞𝙩𝙮 𝙤𝙛 𝙘𝙖𝙨𝙝 𝙞𝙨 𝙦𝙪𝙞𝙩𝙚 𝙪𝙣𝙙𝙚𝙧𝙥𝙧𝙞𝙘𝙚𝙙.

Investors with the liquidity to pounce when dislocations occur will be glad they had the financial freedom to do so. And per my comments on yield spreads above, with so little additional compensation for stretching further out in the calendar, why wouldn't you stay at the short end of the curve?

Here's the link to the article written by Ruth, Winnie Hsu, and Abhishek V. - read it, it's very good:
https://www.bloomberg.com/news/articles/2026-06-15/hedge-funds-reopen-pre-war-playbooks-as-iran-risks-recede

𝗖𝘂𝗿𝗶𝗼𝘂𝘀 𝗮𝗯𝗼𝘂𝘁 𝘁𝗵𝗲 𝗨.𝗦. 𝗧𝗿𝗲𝗮𝘀𝘂𝗿𝘆 𝗠𝗮𝗿𝗸𝗲𝘁?

This podcast from The Economist is an excellent, very in-depth but accessible overview of how the UST market functions (and occasionally falters).

It features Jay Barry, head of global rates strategy at JPMorgan, Nellie Liang, senior fellow at the The Brookings Institution, and Mike Bird, Alice Fulwood, and Ethan Wu, co-hosts of the “Money Talks” podcast.

Here's the link:
https://x.com/TheEconomist/status/2065115619234603114?s=20

𝗜𝗣𝗢𝘃𝗲𝗿𝗯𝗼𝗮𝗿𝗱?

Spencer Jakab shared some interesting observations in his letter today (I encourage you to get on his distribution list - link below).

It's a statistical fact that 𝗲𝘃𝗲𝗻 𝘄𝗵𝗲𝗻 𝗶𝘁 𝗰𝗼𝗺𝗲𝘀 𝘁𝗼 𝘁𝗵𝗲 𝗺𝗼𝘀𝘁 𝘁𝗿𝗮𝗻𝘀𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝘃𝗲 𝘁𝗲𝗰𝗵, 𝗮 𝗳𝗲𝘄 𝗼𝗳 𝘁𝗵𝗲 𝗲𝗿𝗮’𝘀 𝗵𝗶𝗴𝗵𝗳𝗹𝗶𝗲𝗿𝘀 𝗺𝗶𝗴𝗵𝘁 𝗯𝗲 𝘄𝗼𝗿𝘁𝗵 𝗼𝘄𝗻𝗶𝗻𝗴, 𝗯𝘂𝘁 𝗮𝗹𝗺𝗼𝘀𝘁 𝗰𝗲𝗿𝘁𝗮𝗶𝗻𝗹𝘆 𝗻𝗼𝘁 𝗺𝗼𝘀𝘁 𝗼𝗳 𝘁𝗵𝗲𝗺.

It's also a fact that 𝘁𝗵𝗲 𝘃𝗮𝘀𝘁 𝘀𝘂𝗺𝘀 𝗯𝗲𝗶𝗻𝗴 𝗶𝗻𝘃𝗲𝘀𝘁𝗲𝗱, 𝘄𝗵𝗶𝗰𝗵 𝗮𝗹𝗿𝗲𝗮𝗱𝘆 𝗲𝘅𝗰𝗲𝗲𝗱 𝘁𝗵𝗲 𝗿𝗮𝗶𝗹𝗿𝗼𝗮𝗱 𝗼𝗿 𝗶𝗻𝘁𝗲𝗿𝗻𝗲𝘁 𝗯𝗼𝗼𝗺𝘀, 𝗮𝘀𝘀𝘂𝗺𝗲 𝘀𝗮𝗹𝗲𝘀 𝗴𝗿𝗼𝘄𝘁𝗵 𝘂𝗻𝗹𝗶𝗸𝗲 𝗮𝗻𝘆𝘁𝗵𝗶𝗻𝗴 𝗶𝗻 𝗵𝗶𝘀𝘁𝗼𝗿𝘆.

OpenAI’s projection last year that its revenue would rise to $145 BN by 2029 requires compound annual growth of 108% for five years. That’s never happened before (based on a sample of almost 19,000 large companies).

Palantir Technologies, Broadcom, NVIDIA, Intel and Alphabet Inc. were among nearly 70 U.S.-listed companies that fetched a multiple of at least 10 times trailing revenue at the start of this week.

According to GMO, 𝗮𝘁 𝘁𝗵𝗲 𝗲𝗻𝗱 𝗼𝗳 𝗹𝗮𝘀𝘁 𝘆𝗲𝗮𝗿 𝗮 𝗵𝗶𝗴𝗵𝗲𝗿 𝘀𝗵𝗮𝗿𝗲 𝗼𝗳 𝘁𝗵𝗲 𝗨.𝗦. 𝗺𝗮𝗿𝗸𝗲𝘁 𝗳𝗲𝘁𝗰𝗵𝗲𝗱 𝗮𝗯𝗼𝘃𝗲 𝘁𝗵𝗲 𝟭𝟬-𝘁𝗶𝗺𝗲𝘀 𝘁𝗵𝗿𝗲𝘀𝗵𝗼𝗹𝗱 𝘁𝗵𝗮𝗻 𝗮𝘁 𝘁𝗵𝗲 𝗽𝗲𝗮𝗸 𝗼𝗳 𝘁𝗵𝗲 𝘁𝗲𝗰𝗵 𝗯𝘂𝗯𝗯𝗹𝗲.

Looking at the performance of stocks that traded above that threshold across 40 years, buying just that group would have left an investor with 80% less money than an S&P 500 index fund, taking inflation into account.

Here's the link to Spencer's article: https://trk.wsj.com/view/67364bcc111e4b664c0c2a48rgb19.1w5k/35e362e8

Sign up for Spencer's Markets A.M. newsletter here: https://www.wsj.com/newsletters/markets-am?mod=djemMarketsAM

SpaceX

𝗙𝗼𝗼𝗹𝘀 𝗥𝘂𝘀𝗵 𝗜𝗻?

Those considering the SpaceX IPO consider the figures below, courtesy of Brent Donnelly of Spectra Markets.

They show the post-IPO performance of the companies listed by the indicated interval, with the final column (red box) providing the maximum drawdown experienced during the first 12 months of trading.

Note how many lauded companies got cut in half (or 𝘮𝘶𝘤𝘩 𝘮𝘰𝘳𝘦 - Robinhood dropped 90%).

𝗔 𝗧𝘀𝘂𝗻𝗮𝗺𝗶 𝗼𝗳 𝗦𝗽𝗲𝗰𝘂𝗹𝗮𝘁𝗶𝘃𝗲 𝗖𝗮𝘀𝗵 𝗙𝗹𝗼𝘄𝘀

Tsunamis initially pull water back from the shoreline because the water pulled offshore is what feeds the advancing destructive wave.

Per the first chart below, 𝗚𝗼𝗹𝗱𝗺𝗮𝗻 𝗦𝗮𝗰𝗵𝘀 𝗲𝘅𝗽𝗲𝗰𝘁𝘀 𝗹𝗮𝗿𝗴𝗲 𝗰𝗼𝗺𝗽𝗮𝗻𝗶𝗲𝘀 𝘁𝗼 𝗿𝗮𝗶𝘀𝗲 𝗮 𝗿𝗲𝗰𝗼𝗿𝗱 $𝟮𝟮𝟱 𝗯𝗶𝗹𝗹𝗶𝗼𝗻 𝗶𝗻 𝗜𝗣𝗢𝘀 𝘁𝗵𝗶𝘀 𝘆𝗲𝗮𝗿, 𝙖𝙡𝙢𝙤𝙨𝙩 𝙨𝙞𝙭 𝙩𝙞𝙢𝙚𝙨 𝙖𝙨 𝙢𝙪𝙘𝙝 𝙖𝙨 𝙡𝙖𝙨𝙩 𝙮𝙚𝙖𝙧.

𝗧𝗵𝗲𝘆 𝗲𝘅𝗽𝗲𝗰𝘁 𝘁𝗼𝘁𝗮𝗹 𝗶𝘀𝘀𝘂𝗮𝗻𝗰𝗲 (which includes follow-on offerings, SPACs, and convertible bonds) 𝘁𝗼 𝗵𝗶𝘁 $𝟲𝟳𝟱 𝗯𝗶𝗹𝗹𝗶𝗼𝗻 𝗳𝗼𝗿 𝘁𝗵𝗲 𝘆𝗲𝗮𝗿.

𝙒𝙝𝙚𝙧𝙚 𝙬𝙞𝙡𝙡 𝙩𝙝𝙚 𝙢𝙤𝙣𝙚𝙮 𝙘𝙤𝙢𝙚 𝙛𝙧𝙤𝙢?

There are of course many sources of liquidity, but 𝗶𝘁 𝗺𝗮𝗸𝗲𝘀 𝘀𝗲𝗻𝘀𝗲 𝘁𝗵𝗮𝘁 𝗿𝗶𝘀𝗸-𝘀𝗲𝗲𝗸𝗶𝗻𝗴 𝗶𝗻𝘃𝗲𝘀𝘁𝗼𝗿𝘀 𝘄𝗼𝘂𝗹𝗱 𝗯𝗲 𝘄𝗶𝗹𝗹𝗶𝗻𝗴 𝘁𝗼 𝘀𝘄𝗮𝗽 𝗼𝗯𝗷𝗲𝗰𝘁𝗶𝘃𝗲𝗹𝘆 𝘄𝗼𝗿𝘁𝗵𝗹𝗲𝘀𝘀 𝗰𝗿𝘆𝗽𝘁𝗼 𝗳𝗼𝗿 𝘀𝗵𝗮𝗿𝗲𝘀 𝗼𝗳 𝗲𝘅𝘁𝗿𝗲𝗺𝗲𝗹𝘆 𝘀𝗽𝗲𝗰𝘂𝗹𝗮𝘁𝗶𝘃𝗲 𝘃𝗲𝗻𝘁𝘂𝗿𝗲𝘀 𝗳𝗼𝗰𝘂𝘀𝗲𝗱 𝗼𝗻 𝗔𝗜, 𝘀𝗽𝗮𝗰𝗲 𝘁𝗿𝗮𝘃𝗲𝗹, 𝗲𝘁𝗰.

𝗧𝗵𝗲 𝗹𝗮𝘁𝘁𝗲𝗿 𝗮𝗿𝗲 𝗮𝘁 𝗹𝗲𝗮𝘀𝘁 𝗽𝗿𝗼𝗯𝗮𝗯𝗹𝘆 𝘄𝗼𝗿𝘁𝗵 𝙨𝙤𝙢𝙚𝙩𝙝𝙞𝙣𝙜.

Per the second chart, it feels like a crypto sell-off is already in motion.

Souring Bitcoin sentiment has triggered almost $1.5 billion in crypto liquidations over the past 24 hours, as the largest digital asset sank back to a two-month low.

The price of the token is down almost 50% from an all-time high of around $126,000 reached in October.

𝗛𝗼𝘄 𝗺𝘂𝗰𝗵 𝗰𝗮𝘀𝗵 𝘄𝗶𝗹𝗹 𝗳𝗹𝗼𝘄 𝗼𝘂𝘁 𝗼𝗳 𝗰𝗿𝘆𝗽𝘁𝗼 𝗮𝘀 𝗶𝗻𝘃𝗲𝘀𝘁𝗼𝗿𝘀 𝗹𝗼𝗼𝗸 𝘁𝗼 𝗿𝗶𝗱𝗲 𝘁𝗵𝗲 𝘄𝗮𝘃𝗲 𝗼𝗳 𝗜𝗣𝗢𝘀 𝗮𝗯𝗼𝘂𝘁 𝘁𝗼 𝗵𝗶𝘁 𝘁𝗵𝗲 𝗯𝗲𝗮𝗰𝗵?

Hard to say.

But we're moving to higher ground...

Read the article Matt Phillips wrote about IPOs for Axios here: https://www.axios.com/newsletters/axios-markets-d3c8e8e2-bde0-4052-a3c8-aa70bd463bae.html?stream=business

The Bloomberg article on crypto by David Pan and Suvashree Ghosh is here: https://www.bloomberg.com/news/articles/2026-06-02/bitcoin-tumbles-below-70-000-for-first-time-in-two-months?sref=1lDnBweI

$BTC