There's No Escaping An AI Bubble

GMO had an interesting paper about what is "probably" an AI bubble with a comparison of how they managed through previous bubbles and what to do about the AI situation. 

Getting right to the title of this post, what I mean is that whatever your equity allocation is, that sleeve won't be able to avoid an AI bubble. Here's an extreme example to make the point.

NVDX and NVDS are different inverse Nvidia ETFs. If there is an AI bubble and it pops, there's no escape for NVDA common stock. But a portfolio that was half the common stock and half an inverse fund would realistically be spared the full brunt of whatever might befall NVDA. Depending on the compounding maybe the blend would even go up a little. No one should put 50% into a stock and then 50% into an inverse variation of that stock.

This is the effect that GMO is getting to in their paper. What can be added to a portfolio, in our words, to help avoid the full brunt of a large decline? GMO refers to 2022 as the Duration Bubble. They say that during the Duration Bubble they were able to sidestep the worst of the fallout with equity long/short, merger arbitrage and global macro.

This is pretty much the exact conversation we've been having here for many years. Client/personal holding BTAL is of course long/short with a short bias, clients have owned the Merger Fund since the financial crisis and maybe a weaker association but in some circles, managed futures gets labeled as systematic macro so it might be in the same neighborhood as global macro even if not exact. 

The paper seemed to be in support of the GMO Benchmark Free Allocation Fund (GBMBX) which as of its most recent reporting allocates 48.5% to equities, 23.6% to fixed income and 27.9% to alternatives which are listed as equity "dislocation" and "alternative allocation." So that's pretty vague but gives an idea of how they size into alternatives in 60/40 terms, they take a little from equities and a lot from fixed income to build out the alternative sleeve. 

GMO also talks about deemphasizing predictions in favor of what I'd describe as making observations which is an ongoing conversation here as well. They talk about the Internet bubble not in terms of crashing but some sort of mean reversion. We talked about the "Duration bubble" not in terms that rates must go up (which of course they did) but in terms of inadequate compensation for the risk taken. 

Both instances are about making an observation and then avoiding or underweighting the risk posed by that observation. With the way the S&P 500 has evolved to be so heavy in a handful of stocks, many of which are part of the AI theme, there's probably no realistic expectation of avoiding any AI bubble fallout that might come along. If you have 10% in domestic equities, that 10% would get hit hard. If you have 90% in domestic equities, that 90% would get hit hard. What could spare the bottom line of your portfolio is what you do with whatever percentage is not in domestic equities. 

Since there's no way to know if or when there will be any consequence for the current AI market excess, it would be a bad idea to put 40% into an inverse S&P 500 fund against 60% in the S&P 500. If there's never a decline, that 40% will eventually evaporate. What if stocks go down and interest rates go up due to concerns about inflation? That was part of the story in 2022 and currently reported price inflation is simmering persistently above the 2% target. Many would argue that reported price inflation greatly understates what's really going on so some sort of stocks down, yields up scenario would not be a black swan from here. 

I concede that most people won't go to the extreme avoidance of duration as I do but underweighting it seems like a pretty good idea based on current observations. 

One way we've positioned certain alternatives is to say they do what people think/hope fixed income will do which is have very little volatility and trend gently higher.

The outcome of the Merger Fund is what I believe people want their bonds to do and in line with what people should expect from this fund. Other alts seek different outcomes of course. BTAL is essentially an inverse fund, managed futures seeks to be closer to all weather but doesn't always live up to that. Managed futures was great in 2022, a tough hold in 2023 and 2024 and doing pretty well (not great) this year. 

With each successive adverse market event there have been more tools available to help avoid the full brunt and that trend will likely continue. The importance here, said more directly, is that bonds with duration are no longer in the 40 year one way trade they were in, that's over. If bonds can now trade in both directions that makes them less reliable as diversifiers as we saw in 2022 and repeating for emphasis, some sort of run where stocks go down and interest rates go up is far from an impossibility. 

The information, analysis and opinions expressed herein reflect our judgment and opinions as of the date of writing and are subject to change at any time without notice. They are not intended to constitute legal, tax, securities or investment advice or a recommended course of action in any given situation.