Aschenbrenner’s AI Fund Collapse Is Just The Beginning

But when you start running strategies that are running that much leverage against this much volatility for the individual securities, unfortunately, a blowup becomes inevitable. And it really looks like what happened within Leo's portfolio is that he created conditions under which a small decline in prices would force him to sell to reduce his leverage, which in turn caused prices to fall further, which caused him to be forced to sell to reduce leverage further. And ultimately that cascaded into an event that sent both his longs and his shorts against him. Once you become that large, the street actually identifies you as a target. You effectively become a wounded shark in a feeding frenzy.

Well, you know, the quick answer is that when you look at somebody who is engaged in the behaviors that Leo has, there's really no mechanism for him to have learned not to do this. And so he had a very strong thesis. He expressed it with the extraordinary use of leverage. His initial exposure was largely to non-public entities and he had grown his business under that framework which has a component of much lower volatility framing to it because non-public entities don't reprice themselves in the same manner. But when you start running strategies that are running that much leverage against this much volatility for the individual securities, unfortunately, a blowup becomes inevitable. And it really looks like what happened within Leo's portfolio is that he created conditions under which a small decline in prices would force him to sell to reduce his leverage, which in turn caused prices to fall further, which caused him to be forced to sell to reduce leverage further. And ultimately that cascaded into an event that sent both his longs and his shorts against him. In particular, he very much had the thesis that traditional software companies would be heavily disintermediated by the growth of AI and in particular the software sector that obviously contributed to the underperformance of that sector for a period. His selling actually contributed to the underperformance of that sector. And as he began to be forced to unwind his portfolio, that forced prices to move in the opposite direction of his underlying positioning and created conditions for the rapid collapse of the fund and the need to delever it in as quickly a time as he did. Nobody in their right mind should give a 25-year-old $20 billion at 4x leverage, but you actually can't blame the 24-year-old, right? The reality is he had a very strong view. He had a very strong conviction on his view and everything in his experience base up to that point had told him that this was the right strategy to pursue. Once you become that large, the street actually identifies you as a target. You effectively become a wounded shark and a feeding frenzy emerges.

Well, so a levered ETF carries the same characteristics as Leo's portfolio which is obviously running at 4x leverage. And the difference between the two is that a levered ETF because it has a prospectus that requires it to maintain that levered exposure has to rebalance every day. And this is where volatility creates a phenomenon called volatility drag. If you imagine a series in which I make 10% today and lose 10% tomorrow, many people would assume that the answer to that is I now have a zero return. But the reality is I start with $1, I now have $1.10 and I lose 10%, I have $0.99 or 99 cents. I've lost a penny. If I add four times leverage to that, you actually end up with a 2 to the 4th power impact on that volatility drag. Instead of investing $1, I've now invested $1 of equity and $3 of borrowing. I'm up 10%. Therefore, I suddenly have $4.40, meaning my equity has risen because I only owe $3. My equity has risen to $1.40. That is a 40% gain on a 10% change in the underlier, exactly as you would anticipate. But if you do the exact same math for what happens now if I fall 10%. The compounding effect of that leverage and the need to rebalance it creates the conditions that cause these sorts of catastrophic losses. If you then add the additional layers you do with the ETFs that they need to rebalance every single day, it's not like they went from $1.40 equity with $3 of borrowing, they actually have to lever up that $1.40 four times. So 4 times $1.40 is going to be, you know, $5.60, right? So that actually means you were at $4.40 in terms of your exposure the day before. Now I have to increase my position sizes by nearly 50% to maintain the leverage that I've promised my investors. That means that it creates what's called endogenous flow. It actually forces buying even without new investors adding money into the system and contributes to the sort of runup that we have seen unless investors harvest those gains. So the piece that I wrote about it, it's called A Semi Theory of Everything, explaining how this phenomenon plays out when you have large series of complexes that have historically run on this. Most professional investors would run them the way I described as a volatility harvesting strategy taking advantage of the fact that that compounding creates a loss. You actually short both sides of the trade and harvest the volatility loss associated with the volatility drag. It creates a very stable return profile as long as your volatility characteristics are maintained. Unfortunately, in the excitement of the post-March recovery in markets, early April to be more precise, we actually saw retail investors step into these types of products because they were seeking out a Leopold-like experience. They were actually buying these 3x levered ETFs or 2x levered single stock ETFs and then holding rather than harvesting their positions. We actually saw a behavior that suggested people were trying to dollar cost average into these strategies. If you run through the math on this, it is just a terrible way to invest. At 3x leverage, running the level of volatility we were experiencing in the semiconductor space as of April, May, you would need a return in excess of 170% a year in order to simply break even on the volatility harvesting.

To dollar cost average into something that has a 170% break-even is absolutely absurd. But again, a byproduct of the lack of education and candidly the tools that we have put out into the marketplace with an objective to attract people to shiny objects as compared to thoughtful investment vehicles.

Well, again, it depends on how you use them, right? So, a hammer is a very dangerous tool if used improperly. It is a very productive tool if used properly as volatility harvesting regimes and volatility harvesting tools. These can be used by professional investors to effectively short a realized volatility framework and create conditions under which profits can be generated by providing effectively the financing for those vehicles. South Korea has already banned the levered ETFs. They've now been forced to close their market multiple times over the past several weeks, literally doubling the number of times it had been closed in a three-week period over its entire history since roughly 1990. They've recognized that these products create almost no social utility. In the United States, we are still trapped by market fundamentalism. And we see that in everything from Kevin Warsh's recent testimony at the Fed to the general view on regulatory frameworks within the United States. Just let the market decide. There's a very good reason we don't do that. There's a reason we now have labeling on drugs that tell us what the addictive contents of them are. We now have labeling on food that tells us what the ingredient list is. And the reason why is because we used to have the same general view.

Well, that's extremely difficult for an illiterate immigrant to figure out if the sausage that they are eating is filled with potato flour or if it is filled with meat. We recognize that. We took steps to address it. And in many ways, I think we often go too far in this. We do need to recognize that there is a role for experimentation and the utility of tools as I was describing. We could ban hammers because somebody hurt somebody with a hammer. That would be a mistake I think under most economic framings. But in this case, we have created effectively a gambling environment in which people are increasingly nihilistic in their interpretation of prices. Effectively assuming in many ways that governments are stepping in to support these prices or simply print money to create wealth to paper over the many problems that we see in our society. You and I have discussed some of those. That's simply untrue. I just would emphasize for the younger audience, never substitute conspiracy when incompetence will suffice. We have regulators who have largely abandoned their role. And as a result, we are left with a series of products that are being created that I would describe as half-boiled spaghetti being thrown at the wall in an attempt to see what sticks and attracts investor dollars.

Well, unfortunately, as you know, this is one of the key concerns and it's tied to my work around market structure, the growth of passive and price insensitive. And in this case, we're referring to leverage vehicles that actually do not consider what they're buying with leverage a good thing or a bad thing. They're simply fulfilling an investment mandate, what I call a systematic portfolio rebalancing. Those create conditions under which these types of feedback loops can play out. And I would highlight to South Korea that actually much of the problem was not generated in South Korea. We imposed these conditions in South Korea through the introduction of an unlevered memory-centric ETF, DRAM, in the United States which exploded in size to almost as large as Leo Aschenbrenner's portfolio and was sending roughly half of its dollars in Korean won hedged terms. So they were selling the currency, buying the stock in Korea. It brings to mind the 1971 experience from US Treasury Under Secretary John Connally saying to the emerging markets, 'It's our currency, but your problem.' This was our ETF and their problem. And I think unfortunately you're going to see the regulatory environment begin to recognize that and it very well may be forced to change.