Vineer Bhansali warns of market fragility from dispersion trading and bimodal AI valuations
Summary
Vineer Bhansali argues that current equity markets exhibit fragility similar to the 2008 financial crisis, driven by dispersion trading and low implied correlation that can lead to cascading deleveraging. He suggests that high implied volatility in AI stocks reflects a bimodal outcome—either massive success or bust—rather than standard uncertainty. Bhansali compares the current environment to the dot-com bubble, noting that private-market valuation multiples have been applied to public companies with restricted floats. He identifies an 'unwritten third mandate' for the Federal Reserve to ensure financial stability, implying a likely pivot if markets crash. Finally, he recommends using two-year Treasuries to finance convex optionality trades to capture returns in volatile conditions.
Key takeaways
- 1 Hedge fund 'pod shops' operate like call options for their sponsors: they have limited downside (stop-losses) but unlimited upside potential. This structure incentivizes maximum leverage and correlated trading, leading to synchronized deleveraging when markets turn.
Unidentified speaker (speaker_0) · 11:55
Episode sections
- 0:00 – 1:31
Market Volatility and Dispersion Trading
Speaker_0 notes that long-dated volatility for stocks like Micron traded at 80-100%, leading to distorted delta values in deep out-of-the-money options. Speaker_2 introduces the topic with the VIX above 30, setting the stage for a discussion on market fragility and dispersion strategies.
- 1:31 – 4:37
Macro Parallels and Private Market Influence
Speaker_0 compares current speculative fervor and valuations to the 1999-2000 dot-com bubble, noting that private market-style valuations are now applied to public companies with restricted float. He attributes this to tax policies and a risk-on mentality.
- 4:37 – 10:42
Market Mechanics and Liquidity Fragility
Speaker_2 asks for an attribution of the November 5th market drop. Speaker_0 explains that dispersion trading and low implied correlation create fragility, where market makers rely on e-mini futures liquidity that can evaporate during stress, leading to cascading deleveraging.
- 10:42 – 15:25
Crowding in Hedge Funds
Speaker_2 asks about crowding. Speaker_0 describes the hedge fund ecosystem as a 'zoo' where pod shops are sponsored to take specific, correlated risks (like dispersion) with tight stop-losses, leading to synchronized exits and market-wide cascades when trades go against them.
- 15:25 – 22:43
Bimodality Hypothesis and Option Pricing
Speaker_2 introduces the concept of bimodality to explain extreme option pricing. Speaker_0 explains that high implied volatility for far out-of-the-money calls reflects a binary outcome (AI success vs. bust) rather than standard distributional volatility, making options act like prediction markets.
- 22:43 – 29:05
Anthropic IPO and Hedging Challenges
Speaker_2 discusses the upcoming Anthropic IPO and its potential volatility surface. Speaker_0 predicts high volatility and limited hedging options due to low float and the risk of short squeezes, comparing it to the SpaceX IPO dynamics.
- 29:05 – 32:33
Public vs. Private Asset Classification
Speaker_0 debates whether these new public companies should be viewed as financial assets or commodities. He suggests that if they are commodities with limited supply, high volatility is justified; if they are financial assets with infinite supply potential, volatility should collapse.
- 32:33 – 36:05
Credit Correlation Dynamics and Market Fragility
Speaker_0 draws parallels between current equity dispersion strategies and the synthetic credit markets during the Global Financial Crisis (GFC). He explains how implied correlation in credit remained low until deleveraging forced unwinds, causing a cascade where realized correlation spiked. This historical frame suggests that current equity market stress could trigger similar cascading unwinds in correlation-based trades.
- 36:05 – 38:56
Fed Policy, Financial Stability, and the 'Third Mandate'
Speaker_2 highlights the upcoming December 9th Fed meeting and notes the VIX recently closed at 28 and hit 30, comparing it to June 2022 when the Fed hiked despite high volatility. Speaker_0 argues the Fed now has an 'unwritten third mandate' of financial stability, making them relevant to market liquidity. He suggests the Fed is likely to pivot or signal accommodation if asset prices crash, even if inflation remains high.
- 38:56 – 41:03
Asymmetric Trade Ideas in Short-Dated Treasuries
Speaker_0 proposes the two-year Treasury as an asymmetric trade, citing a 4.75% yield. He argues that if rates rise 300bps, the trade is roughly breakeven due to roll-down and carry, but if the Fed pivots due to a crisis (rates drop 300bps), the trade yields 12-15%. He advises focusing on the front end of the curve rather than long duration, noting the Fed funds rate is in a 'zone of danger' where bad things historically happen.
- 41:03 – 42:49
Convexity and Optionality Framework
Speaker_2 asks if option structures are preferable to holding two-year bonds directly. Speaker_0 suggests using the yield from the two-year Treasury to finance optionality, creating a convex trade that benefits from volatility in either direction. He emphasizes that in an environment of uncertainty and potential bimodality, holding yield plus explosive convexity is a robust design element for capital allocation.
Assets discussed
| Asset | Kind | Stance | Context |
|---|---|---|---|
| NVIDIANVDA | equity | neutral | People talk about Cisco, Nvidia, talk about unprofitable companies that are getting valuations that are just unbelievable, right? |
| MicrosoftMSFT | equity | neutral | I was just looking, so it's at 50 now. It's a variant swap weighted on NVIDIA and Microsoft and so forth, and comparing that to the VIX. So you get VIX EQ of 50, VIX of 30. Go back to July, and this was sort of peak micron and some of what you were writing about in terms of bimodality. You had a |
“Stance” is how the speakers talked about the asset, not a recommendation.
Claims worth verifying
Long-dated volatility for Micron traded at 80-100% levels, leading to distorted delta values.
Unidentified speaker (speaker_0) ·0:00
How to verify: Check options chain data for Micron (MU) long-dated options during the period discussed (late 2026) to verify implied volatility and delta values. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
The VIX peaked above 30 during the recording period.
Unidentified speaker (speaker_2) ·1:31
How to verify: Check historical VIX data for November 2026 to confirm the peak level. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
The market fell 4.4% on November 5th, while the QQQ fell 6.5%.
Unidentified speaker (UNKNOWN) ·5:02
How to verify: Check historical price data for the S&P 500 (SPX) and NASDAQ 100 (QQQ) for November 5, 2026. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
Implied correlations were destroyed due to dispersion trading.
Unidentified speaker (speaker_0) ·6:44
How to verify: Analyze historical implied correlation indices for the S&P 500 and NASDAQ 100 to observe the drop in correlation prior to the market move. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
The VIX EQ was at 50 while the VIX was at 30 at the time of the conversation.
Unidentified speaker (speaker_2) ·8:22
How to verify: Check the price of the VIX EQ ETF or index data for the specific date of the podcast. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
SpaceX IPOed at a trillion-dollar valuation.
Unidentified speaker (speaker_2) ·23:15
How to verify: Check press releases or financial news reports regarding the SpaceX IPO valuation in 2026. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
Anthropic IPOed at a trillion-six valuation.
Unidentified speaker (UNKNOWN) ·22:48
How to verify: Check press releases or financial news reports regarding the Anthropic IPO valuation in 2026. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
Delta of a 150% out-of-the-money option can be over 0.5 in Black-Scholes world when vol is very high.
Unidentified speaker (speaker_0) ·0:20
How to verify: Calculate delta using the Black-Scholes formula with high implied volatility (e.g., 150%) and a 150% OTM strike to verify the mathematical result. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
VIX tracking strategies, including dispersion strategies, require buying VIX back or index volatility to exit.
Unidentified speaker (speaker_0) ·7:24
How to verify: Review the mechanics of dispersion trading and VIX futures/ETFs to confirm the hedging requirements for exiting these positions. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
Market makers do not make markets when liquidity evaporates, leading to fragility.
Unidentified speaker (speaker_0) ·9:13
How to verify: Observe order book depth and trade execution quality in e-mini futures during periods of high volatility to assess market maker participation. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
Synthetic credit tranches, specifically super-seniors, were considered default-proof and kept implied correlation in credit low until the GFC.
Unidentified speaker (UNKNOWN) ·32:51
How to verify: Review historical credit index tranche pricing data (e.g., CDX or iTraxx) from 2026-2026 to observe low implied correlation levels, and cross-reference with GFC timeline and deleveraging events. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
The CDX index consists of 125 equally weighted names and can be tranched into equity, junior mezz, senior mezz, and super senior layers.
Unidentified speaker (speaker_0) ·33:11
How to verify: Check the official documentation for the CDX NA IG Index from Markit (now S&P Global) which specifies the constituent count and tranche structure. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
The VIX closed at 28 and hit 30 recently, and the Fed has only hiked once when the VIX was above 30 the day before, in June 2022.
Unidentified speaker (speaker_2) ·36:44
How to verify: Check historical VIX closing prices for the date prior to the podcast release (Nov 2026) and compare with Federal Reserve FOMC meeting dates and policy rate decisions in June 2026. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
In June 2022, the Fed policy rate was 87 basis points and year-over-year inflation was 8% when the VIX was above 30.
Unidentified speaker (speaker_2) ·37:16
How to verify: Verify the Federal Funds Effective Rate on June 2022 and the CPI YoY inflation rate for May/June 2022 from the Bureau of Labor Statistics. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
The two-year Treasury is yielding 4.75%.
Unidentified speaker (speaker_0) ·39:16
How to verify: Check the current yield on the 2-Year US Treasury Note from the US Department of the Treasury or major financial data providers. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
Historically, when the Fed funds rate was within +/- 40-50 basis points of the current level, bad things happened.
Unidentified speaker (speaker_0) ·40:35
How to verify: Analyze historical Fed funds rate data to identify periods where the rate was near the current level (as of Nov 2026) and correlate with subsequent market crises or significant downturns. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
Speaker_0 bought protection on credit tranches at PIMCO during the GFC and it worked out well.
Unidentified speaker (speaker_0) ·33:51
How to verify: Review PIMCO's annual reports or investor presentations from the 2026-2026 period detailing their credit trading strategies and performance attribution. For recent events, verify 2026-08-21 through 2026-11-21. Use another period only when the speaker explicitly names it.
Do your own research
- Verify the current VIX level and historical Fed hiking patterns when VIX > 30 using Federal Reserve Economic Data (FRED) or CME Group data.
- Check the implied correlation levels for S&P 500 single-name options versus index options via Bloomberg or CBOE data to assess dispersion trading activity.
- Review the float and short interest of companies like Anthropic or other low-float IPOs to evaluate hedging difficulties and short squeeze risks.
- Analyze the yield curve, specifically the 2-year Treasury yield, and calculate roll-down and carry effects for the proposed asymmetric trade.
- Examine hedge fund filings (13F) and industry reports to assess crowding in dispersion and volatility strategies.
- Compare current public company valuations (EV/EBITDA, P/S) to private market comparables to validate the claim of 'private-market multiples' in public equities.
- Research the Federal Reserve's dual mandate and any recent speeches or minutes referencing financial stability as a policy consideration.
- Assess the historical performance of VIX futures and options during periods of high correlation spikes, such as March 2020 or August 2007.
Full transcriptexpand
0:00Unidentified speaker (speaker_0): cost of risk should be lower in time. What happened that both you and I observed, and I think you did point out in one of your notes, which actually really triggered my brain, was the fact that long dated volatility for micron and for many of these stocks was actually trading at that 100%, 80% type of level. And I started thinking about what that does, right? So the one of the observations that it leads to is when vol is very high like that for a deeply out of the money option, not at the money, deeply out of the money, 200% two hundred percent or 150% a hundred and fifty percent out of the money option. It's delta mathematically in Black Scholes world actually is very, very high. So you can have a hundred and fifty percent out of the money option with a delta over point five, which makes no sense because delta is supposed to be the probability of you ending up in the money. Typically the delta of a at the money option is point five or higher, not something that's a hundred 150% and fifty percent out of the money.
0:48Unidentified speaker (speaker_1): Hello, this is Dean Kernutt, Curnutt, and welcome. to the Alpha Exchange, where we explore topics in financial markets associated with managing risk, generating return, and the deployment of capital in the alternative investment industry.
1:06Unidentified speaker (speaker_2): I'm joined here today with Veneer Bonsali. He's the founder and the CIO of Long Tail Alpha, and a man who, like me, lives in the tales of the distribution. Vhereer, what a great time for us to catch up and have this call. Yeah,
1:19Unidentified speaker (speaker_0): likewise, Gene. Good to reconnect. Thank you for putting me on on your podcast multiple times, inviting me to your fantastic annual conference. It's always great to connect and talk vol and talk tales. And today, definitely the tail day.
1:31Unidentified speaker (speaker_2): I'm a big fan of your framework, your thought process for assessing optionality and across assets, really. And so we're having this conversation. The VIX just peaked its head above 30. We've been waiting for this to happen. When you look at the current backdrop, whether it's the rate side, the geopolitical side, we'll talk about market concentration. Give us your rendering of what you see now in risk markets and if it reminds you of anything at all.
2:00Unidentified speaker (speaker_0): That's like an a easy one to, you know, try to hit out of the book. ballpark here because that having been in the industry now close to thirty five years and having been through 2000 and 2008 and 2013 and 2018, 2020, right? So every anecdote you could kind of look and try to start painting rhymes and similarities and so on. So there's a lot of similarities and I know you and I have talked about this and also written about this, but there's obvious similarities to nineteen 1999 ninety nine and two thousand 2000. People talk about Cisco, Nvidia, talk about unprofitable companies that are getting valuations that are just unbelievable, right? So in the old days in 2008, 2010, if we saw a unprofitable company come in with a market cap of let's call it 50 billion or 25 billion or 100 billion, that was incredible. It was like, wow, what's going on here? Now it's like 2 two trillion is nothing. It's like, you know, bar for the course, and the next one is like three trillion. So I think there's a lot of similarities. And I think the fact that you have very restricted stock companies, and we'll
3:00talk a lot more about this in a minute here, trading at such incredible nominal valuations tells you that what has happened is something pretty im important, which is, in my view, is that the private markets have found their way into the public markets. Think about SpaceX, think about anthropic, again, which we'll talk about a lot. People are willing today to pay private market type evaluation without control, without voting power for these so-called public companies with almost no float out there. So there's a lot of similarity. Now there's been articles written in the last few days. You've got to have it. And again, I don't know which way the world ends up. And that's my bimodality hypothesis that we can talk about here. in a minute. But I don't know which way this turns out, but there are a lot of parallels. And I think right now the speculative fervor is at absolute peak. The macro backdrop is that it is very much risk-gone by the dip mentality, by assets, equity assets, especially. And then we can also talk about the other things that have happened since the early 2000, two thousand having to do with tax policy, which makes it suboptimal for people. to sell their stock out because you don't want to pay taxes. So I think the system is set up in a way macroeconomically right now that's very conducive to not only owning stock, but buying more stocks. And so it has follow on effects. And we can talk about the Fed and so on as well if you're interested.
4:17Unidentified speaker (speaker_2): We're going to talk about some of your work on buy modality and that episode, especially over the summer when the bid to very far out of the money calls on stocks like Micron, which was a trillion dollar company for a period of time, was just outrageous. It was just so high. But let's just start at the macro level with the VIX itself. So as we have this conversation, it's at 30, no idea where it'll wind up. But we know that something between 30 and 40 is pretty unsustainable. Sustainable. In other words, it's either going down or it's going up. It can't really stay in. the mid-30s for any extended period of time. We know that the VIX has an acceleration component to it, a quality to it. Things, de-risking events tend to take on a life of their own.
5:02Unidentified speaker (UNKNOWN): So
5:02Unidentified speaker (speaker_2): let's go through November 5th first. The market falls four point four percent. The triple Q falls, I think, six point five percent. What's your attribution on that? How do you get to something so significant where you know the backdrop was we're realizing fifteen to twenty and suddenly we have a four point four percent down move in the S P?
5:24Unidentified speaker (UNKNOWN): Yeah,
5:24Unidentified speaker (speaker_0): probably just mathematically, right? So again, you and I both know this and everybody who's watches this type of podcast knows this that when you think about a fifteen annualized volatility, you're capturing standard deviation. But the narrower you make the distribution, the fatter the tails get because the total area under it, the integration under the probability density function, if you want to get wacky about this, has to be one. So the fact that you can get a 15-vol but still get a five sigma, 10 sigma event in financial markets is nothing that one should be particularly surprised about. Now, so it comes back to again behavior, right? So if you're anticipating that you're just gonna get a 15-vol, 15 vol, you're going to get 1%. move at most with one sigma, and a 3% move is like a three sigma move, you're basically saying the normal distribution rules, which we know it doesn't. And you're in the space right now in the marketplace where absolutely the normal distribution should not be respected, right? So think about the concentration of stocks. Things are not normally distributed in the index. So SP, 30, 40%, driven at least in
6:25terms of risk by seven or eight stocks. QQQs maybe 70 or 80% written by these stocks. And this comes back to the theme of implied correlation, which you pointed out, and also Micron, to a certain degree. The implied correlations got completely destroyed in the process of the last year and because of dispersion trading and so on, which basically requires, as you know, it requires um buying single name volatility and selling index volatility. So one of the things that I've been talking about for about three months with our clients is the fact the single name large gap wall is inverted, right? Call skew is the wrong direction. Index volatility is traditional, right? It's over calls. When you combine these two things along with the very large weight of these large gaps on the call side, implied call side dispersion is at its theoretical minimum. or was, because correlation has really popped in the last few days. So part of the VIX shock that you were highlighting, the fact that the market moved a lot, has to do with the fact that there are a lot of VIX tracking strategies, including dispersion strategies, where the only way out of these strategies is to actually buy the VIX back or buy index volatility back. And correlation and index volatility is basically a one-to-one trade right now. So one of the things that we had been talking about to our clients is one of the simplest hedges today is
7:45think about buying, for instance, the VIX futures or call options on the VIX. In my view, that trade has not quite fully worked out yet. So I would not be surprised if before this whole thing is over, you actually get a spike up to the recent highs forty 40 or fifty 50 in the VIX.
8:02Unidentified speaker (UNKNOWN): Yeah.
8:02Unidentified speaker (speaker_2): And to get there, you really, you know, if you say 40 VIX and we do our little square root of 252, you're moving two and a half, maybe even three percent a day to get into the 40s. And you certainly need a lot of correlation among stocks to get there. The SIBO has these two, obviously it has the VIX. It's also got this VIX EQ, right? The single stock VIX. And I was just looking, so it's at 50 now. It's a variant swap weighted on NVIDIA and Microsoft and so forth, and comparing that to the VIX. So you get VIX EQ of 50, VIX of 30. Go back to July, and this was sort of peak micron and some of what you were writing about in terms of bimodality. You had a VIX EQ of 50 and the VIX of 17. So
8:45Unidentified speaker (UNKNOWN): it's
8:46Unidentified speaker (speaker_2): just incredible how much pickup in the VIX without any commencement. commensurate movement in single stock vol.
8:53Unidentified speaker (speaker_0): Exactly. And I mean the other thing you'd see, I mean, I've almost religiously look at the depth of the e-mini futures market, right? Which to me is like the ultimate lightning draw for everything. Because at the end of the day, if you have to hedge yourself, you have to trade something like the e mini futures if you're a market maker. And even yesterday, when the market was, you know, obviously going leg down, even after the amazing meeting anthropic IPO, you could see. liquidity basically evaporated until the very last minute, until the close. And it's because market makers are not going to make markets. So it is this latent illiquidity, it's this fragility that's hidden. I would not look at implied volatility. I would not look at VIX and say, well, that should be our risk. Unfortunately, there are so many strategies in the world today: dispersion and risk parity, and you name it, even trend following. to a certain degree, that are indexed off that one number. You're running this risk that happens every five to seven years macroeconomically, where everybody in a very self-preservation, rational way, is trying to control their own risk and they're connecting the dots and saying, if wall goes up, I'm gonna do this. They're watching the same number, which is not an accurate reflection of reality and risk. And they all do the same thing together, and then they find out. that the e-mini futures contract or the depth of the market is not large enough for them to all do the same thing at the same time. It keeps happening over and over again.
10:12People keep misinterpreting the liquidity when you don't need it for the liquidity available when you do need it. It does not work that way, in my view.
10:22Unidentified speaker (speaker_2): Yeah, I think your old colleague from Pimco, Mohammed El Iron, has said liquidity is a morale. mirage when you need it the most. So some part of what you're saying there is we're sort of linking strategies or behaviors, investment decisions, to a statistic or a metric that is too common, right? This sort of is leading me to a little bit of some thinking I'm on crowding. And if you go back to uh November 5th, there certainly has been a lot of focus on pod shops, those types of long short strategies, and whether that de-risking, which shows up in the main prime brokers as a huge deleveraging event, long short portfolios, that the portfolios are leveraged, large, and sane. I'd love for you to just help us think about how you think about crowding from a risk standpoint, from a behavioral standpoint.
11:15Unidentified speaker (UNKNOWN): Yeah,
11:15Unidentified speaker (speaker_0): exactly. So what happens, I mean, uh, you know, this is an ecosystem, this is a zoo, this is a forest. I've lived in this jungle for 35 years, so have you. And in this jungle, as the market has evolved, what has happened is many of the shops, pod shops, and so on, they are their mandate when one of the large pod shop sponsors by the new fund. Of course, they've got a lot of capital coming in, they'll tell the pod, you can only do this. This is your channel. You can only do dispersion. You can only do distrust credit. You can only do whatever, right? You can't go beyond it. Now, if you have to make returns, which is how you get paid, and you have a very tight stop loss, because this is from the perspective of the sponsor of the pod shops, they're running a portfolio of call options, right? They've got hundred 100 pods, everything is a call option. If it goes down 3%, you're out, you're fired. If you go up 20, great, you get a nice chunk of that profit. So think about it from the psychology of a pod shop trader. They know I'm running a call option in which my stop loss is very well known because my risk the my sponsor's risk system will shut me down. On the other hand, I need to maximize the upside optionality, which means I should go max leverage and hope that when the market starts going against me. I can get out before all the other similar pod shops do. So they're the same trade because what has been working is what is sponsored. What
12:35is not working, i.e., buying tail protection, is not sponsored because of the negative expected return trade by itself. So the pod shops have the same trade on because that's where the money that's looking for returns is going. And everybody is optimally. levering it up to the maximum amount. And when they all find out, oops, we're doing the same exact thing, maybe not exactly the same thing, but similar things in terms of the underlying index, they all exit at the same time. And that just generates basically a cascade at the macro level. So that's the way the system is set up. It works really well because it's basically taking the venture capital model for making a lot of investments where you hope that two or three will do really well, four or five will do okay, and maybe three or four will get bust. If you can sustain that kind of business model very tightly controlled and regulated and risk managed, it is a positive expected return for the pod sponsor. But it does create this externality that the overall market level, in my view.
13:33Unidentified speaker (speaker_2): Yeah, it's almost as if the pod sponsor has gifted the pods with some. version of optionality. It's kind of like an outperformance option. There's a lot of leverage, but the hedging of an outperformance option is going to mute correlation. That's kind of how I see it. So when they de-risk, they in some ways are adjusting the delta hedge of an a outperformance option. In other words, it goes against them, they've got to sell their winner, and they've got to buy back. their loser. That's going to mute correlation. And I think that's some part of the explainer of just how incredibly low this correlation is. And it does remind me a little bit from a Vol standpoint of 2017, right? That giant gamma well where the market found itself just stuck. And of course, once the dam breaks, you can get something significant a la Volmageddon in Feb 18.
14:29Unidentified speaker (speaker_0): Exactly. And I think the recurring phenomenon, right? So I don't get surprised with 20 sigma moves anymore. To me, they mean nothing. I've seen so many 20 sigma moves in my life. Again, you can never forecast them, obviously, because otherwise we'd be very rich doing just that. But you could always anticipate that another 20 sigma move is coming. And a lot of what's going on in the markets right now looks like that to me. The fact that many diversifying assets are being sold at the same time as equities and long short funds are having a really hard time and so on and so forth. Feels to me that there is mass-scale liquidation going on on top of the equity market sell-off.
15:04Unidentified speaker (UNKNOWN): You
15:04Unidentified speaker (speaker_2): mentioned the 20 Sigma move. There's this great quote, maybe four or five days into the quantquake of 2007, where I think it was the CFO of Goldman, John Winkle Reed. He says, We were seeing 10. standard deviation moves three or four days in a row. And I was like, it's hard to make those 10 standard deviation moves if you take the normal distribution literally.
15:24Unidentified speaker (speaker_0): Exactly.
15:25Unidentified speaker (speaker_2): Let's talk about this bimodality. So I'll just set it up. And we were talking about this a lot over the summer. You had this just epic spot up vol Vala up period where assets were going up a lot in price, and the vol was rich. richening quite a bit. I mean, it was a la 99 in 2000, but I think even more significant. And I'll just remember back in late June of this year with Micron as an example where you could have done a zero cost collar on this thing with a 10% down put strike out two years and sold a 170% call. You know, the vol skew and the vol itself was so distorted. You and I talked about it, and you have this mathy, so you're gonna have to dumb it down. But this explanation for how the market can take a one trillion dollar company and assign a hundred plus vol to a two-year far out-of-the-money call. How do we get to that pricing?
16:23Unidentified speaker (UNKNOWN): I
16:23Unidentified speaker (speaker_0): won't dumb it down, but I will just explain it in the way that I think can be very easily understood by anybody. Right. So we start always when we look at option prices, we look at volatility. Then the general thinking and understanding is that if a new company comes in, a brand new IPO, in the short run, its volatility and credit spreads will be high because we don't know if it's going to survive or not. But in the long run, if it's survival, conditional on and surviving, its implied volatility will fall, its credit spreads will fall. That's the inverse. inverted curve because if it survived, its cost of funding and cost of risk should be lower in time. What happened that both you and I observed, and I think you did point out in one of your notes, which actually really triggered my brain, was the fact that long-dated volatility for micron and for many of these stocks was actually trading at that 100%, 80% type of level. And I started thinking about what that does, right? So the one of the observations that it leads. to is when Wall is very high like that for deeply out of the money option, not at the money, deeply out of the money, 200% or 150% out of the money option. It's delta mathematically in Black Shoals world actually is very, very high. So you can have a 150% out of the money option with a delta over 0.5, which makes no sense because delta is supposed to be the probability of you ending up in the money. So typically the delta of a at the money option is 0.5 or higher, not something
17:43that's 150% out of the money. So that triggered my brain. I said, why is this happening? And the reason it's happening is because that implied wall number that you have to plug in is very, very high because that's where market or buying, you know, people are buying and selling options. Now we can get into the mechanics of why people are willing to buy that option. One potential explanation is that there are a lot of levels. levered ETFs. And those levered ETFs, the market maker who sells those ETFs, has to effectively hedge himself against a large move up or a large move down because they have to basically buy and sell the stock in a levered form. In order to do that, they have to buy ratchets. A ratchet option basically strikes every month, it has a new at-the-money at the money strike and a new expiration. So you have a sequence of ratchets which basically moves up. the whole world term structure. So that's one explanation from a demand and supply mismatch. But then you also say, well, okay, fine, that's true. But then why is the retail market or why is the institutional market not taking advantage of it? Why are they willing to buy that option at a very high volatility? So this is where the bimodality hypothesis comes in. So let me try to explain that in a very clean and simple way here. So uh for a second, abandon black shoals. Okay, don't think it's a unimodal fat distribution. But think about two possible outcomes. A very good outcome in which AI rules the world. Microns chips are
19:02sold now forever. You're not gonna make any one of them, any more of them, and they will trade like commodity. They're they'll trade like oil or nuclear fuel or something like that. So that's a good world. So you have a distribution where the stock could go up 200% and stay there. That's probability potential number one. The second one is complete bust, and we find out that the whole thing was a complete mirage, and we go on the left side, and Micron and all these other AI companies turn out to be just another rerun of 2000 or whatever you want to call it. Well, that's your second outcome. Very bad. outcome. And very simply, let's just assign those two returns some number, plus 50 minus 50, or plus 50, minus 25, and give them a certain volatility, which is not 100%, but something small, like 20% or 30%, which is effectively par for the course for tech stocks. Well, what happens is when you calculate this and calculate the probability-weighted standard deviation of this combination distribution, left and right or up and down, the bimodal, you find that the variance or the volatility is naturally a very high number, not because of the volatilities of the individual distributions being high, but because of the difference in the expected returns. That's a very
20:21simple fact, which says that when you have very disparate outcomes like a coin flip, your variance is actually a function of the difference in the expected return squared. That's what this whole thing is about. So it's basically saying, hang on, let's step back for a second. And let's say the market is actually not a black shoals market, but the market is actually a really good, really bad Dr. Jekyll, Mr. Hyde market. What is the rational price of uncertainty? It's the difference. of the mean of those two outcomes. And that's how you should rationally think. And anecdotally, just to put a little bit of meat beyond the quantitative mathy stuff on it. I have actually spoken with investors who a few months ago just threw the towel in and said, We can't fight this AI rally anymore. If we don't participate, we're going gonna to lag so badly, we just have to get exposure either outright or by buying options. We can't miss it anymore. So that kind of behavior results in this bimodality.
21:21Bimodality is very rational when you're in the zone of maximum confusion. And if you got to have anthropic and got to have SpaceX, got to have micron at that price, but you're not willing to pay the price, your only choice is you have to buy options on it. So that's my little rationale for why you can. have a longer dated $5 trillion, $3 trillion company trading at implied volatilities of close to 80 or 90%.
21:46Unidentified speaker (speaker_2): It's almost like the option montage is a prediction market. It's a yes or no.
21:51Unidentified speaker (speaker_0): Exactly. That's exactly what it is. It's in the zone of maximum confusion, it is exactly what you just said. It is a binary prediction market. And you can easily justify plus minus 80 ball based on that.
22:03Unidentified speaker (speaker_2): And I think some part of your thinking and mine too is a tail wagging the dog. Our institutional money management business, there's lots of risks. One is the risk of missing out. And when the benchmark is so populated with these high flyers, your capacity to deviate from that benchmark, unless you're a very, very rare manager where you're your investors just give you sway to lag, that benchmark really matters. And so you get forced in, and that becomes a little bit of a pylon effect. It's so interesting. Let's talk about anthropic. So the stock debuts yesterday on a 2% down move in the SP. I don't know if it's broken price yet, 83, but it's close.
22:48Unidentified speaker (UNKNOWN): You
22:48Unidentified speaker (speaker_2): know, God bless America. They start a company six years ago. and uh it IPOs for a trillion six. The old days of starting a business and hoping to run it to uh $50 million are long behind us. So, in some sense, that's good. I think it speaks to how pathbreaking this technology is and how your mind is forced to do incredible what-ifs what ifs on something like this. We saw SpaceX IPO months ago. a trillion five. And so what I wanted to do was talk to you about the vol surface of anthropics. So I think the options are coming Monday, and I'm very sure they are going to be very liquid, very fast. That's the setup for something like SpaceX when it went public. The options took very, very quickly. People want to trade them across the entire vol vault surface. What would you say? Just big picture the characteristics of a giant IPO with a brand new technology with $230 billion in the bank. What do you think it's going to look like?
23:51Unidentified speaker (UNKNOWN): Yeah,
23:51Unidentified speaker (speaker_0): that's great. And again, you know, I can't talk about specific trade strategies and all that, but I can tell conceptually what we can talk about here is what I expect the vault surface to look like. And to your point, SpaceX is a great example. So let me give you a little anecdote that'll tell you what I think will happen with the anthropic vault surface. So right after SpaceX went public in June and options started trading, we started getting a lot of inquiries. And being a tail specialist, one of the questions that came from a lot of investors was: how can we hedge this? And this comes back to what you were saying, Dean earlier. Various ways of hedging. One way to hedge it is to buy a put option, another way is to make it less negative cost, which is selling out of the money call to finance a put, right? Volatility is high in both sides because of the bimodality thing that we just talked about. Now, why doesn't that work? The reason people wanted to hedge was because a lot of investors had stock that was locked up with their managers. They don't own the stock. They don't even know when they're actually going to get the stock. but they want to hedge the price risk of it. Now you can buy a put option, but buying a put option will cost you a lot of money. I mean, the ball was some 60%, right? So you've got to pay 30% or so premium to buy a one-year option. So that's obviously out of the picture because you're going to give up most of your windfall profits that you made on this IPO. Well, what
25:11about selling a call option to buy a put option? That looks like a great trade on paper. The risk is if you sell a call option, which is American style on single stocks, and there is suddenly a big squeeze on the stock and somebody exercises that out of the money option, you have to go buy the stock to deliver. You cannot just say, I'm gonna exit it. So you could end up actually selling a quote unquote covered call, but if you don't own the stock, you really cannot deliver on this. So I think the situation will be very similar here. Coming in, only a few percent is actually floating. You can trade this thing if you want. And we have some strategies that we're probably going to do just on the margin. But you really cannot do the big trade, which you and I have talked about in the past, is the Mark Cuban trade, right? The one where when Yahoo came in 1999, he did that big risk reversal that is. now epic and famous and is written about in legendary places where he sold the call to buy bought a put, it effectively locked in that market crash. That's the trade you want to do, but you can't really do that. So what that results in potentially is this biomodality hypothesis playing out all over again. There will be a lot of forced buyers of the ball because they need to own both tails on it. But there won't be very much forced
26:26Unidentified speaker (UNKNOWN): by the ball.
26:27Unidentified speaker (speaker_0): hedging, and there probably won't be very much selling because people are afraid that they're going to get squeezed. Like you, I'm very curious where it ends up. I expect it to be extremely high in terms of employees and probably similar to where we just talked about long dated ball was trading for Micron and SpaceX and all that. Because if you don't own the stock, and your only way out to keep up with your benchmarks and have this exposure of something that potentially could be the biggest technology to be ever seen, you're gonna buy that option. You don't have a choice of not buying that option if you don't own the stock. So I don't know if that answers your question or not.
27:05Unidentified speaker (speaker_2): I look at SpaceX and I had made some forecasts, just guessing, as to what the vol surface would look like. So I think the first characteristic is almost surely it's going to be downward sloping. A business is expected to mature over time and the market kind of gets comfortable with it. And really it's just about the feedback of realized vol. SpaceX comes out, you get a bunch of eight to 10% moves. That's going to allow one month to probably clear in the 80s or 90s.
27:35Unidentified speaker (UNKNOWN): And
27:35Unidentified speaker (speaker_2): then you kind of walk it down. Maybe three months month is 70 and six month is 60. And that's kind of what's happened. Over time, realized on SpaceX has gone basically from 100 to 50 and implied is travel down with it. Of course, not perfectly, but it's traveled down with it. So that's the first one is I think an inverted term structure. The other one's the SKU, The skew, which I think is going to be super interesting, especially with anthropic, and I would say, in some ways, more. than SpaceX, the put skew in anthropic, I think is going to be a really important indicator. Here's a company that's made a lot of commitments. It's first and foremost, the information that we're going to get through both the options volvice and just the stock itself is going to be so valuable. Right now, we get a report from the company in its pre-IPO period where they're giving us some metrics. No one really knows exactly what they're counting and what they're not counting. And so price is going to be an incredibly valuable barometer for just how the AI ecosystem and how the frontier labs are doing. So my guess is there's going to be more put skew than in something like SpaceX. And I think they'll certainly, as you've been talking about your bimodality work, there'll be a lot of call skew. for sure. And that'll set up some interesting, interesting trades to do as
28:55well.
28:55Unidentified speaker (speaker_0): What it reminds me a lot is this is not in a way, it's obviously new because we've never seen this type of dynamic and structure and single name equity options ever, right? In my 35 years of doing it. But at the same time, I have seen this in commodities. So if I connect them and you know make a little leap of faith or, you know, maybe connection, can we think of these stocks as commodities? They're in limited supply, only a little bit of float. You got to have it if you're going to participate. But it can be manufactured in infinite capacity in the future. As a matter of fact, if we stay on the spending plan for compute and power and all that, you're probably gonna have to issue a lot more stock just to make sure that they have the funding to do it. And by the way, the technology is unbelievably good, right? So the question here is. is this really a commodity? Because the market is pricing it like it was pricing micron, like it was pricing SpaceX, as a commodity where limited manufacturing capacity for the world. This is not like oil. This is not like gold or whatever else where you might get skews on both sides because the thing is in limited supply. So I think that's the thing that I'm trying to dispel and figure out is should I really change my frame of reference and say when a large private market comes to public market, and now we have kind of hybridized private and public markets, should we be thinking of private markets
30:15in the public market trading environment as a precious limited supply commodity? Or should we be thinking of this as a public commodity? The results are pretty stunning, right? Depending on which way. you end up. If it really is a public asset that will trade like a stock, implied volatility will come basically collapsing down over the next few years. And if one was a ball trader, one would be selling that ball, saying, this doesn't make any sense. This is not a commodity. This is a financial asset that can be manufactured in infinite supply. So you should be a ball seller. On the other hand, if you say no. you're not gonna get this stock. This is really like oil or natural gas or whatever, you can't make any more of this stuff. Then it will trade like a commodity and maybe implied go up to 200 because you have to own the the convexity on this stock. So it's an open question. I don't have an answer to it. But I think commodity traders probably should be trading AT A and B stock options more than us, the financial market guys.
31:13Unidentified speaker (speaker_2): I wonder at some point. we don't know what the cash burn will be for anthropic, but boy, a convert would satisfy a lot of constituencies, right? They'd be monetizing outrageous vol, and they would provide a very interesting opportunity for the buyer if the VOL was priced interestingly. That seems to me like a great risk transfer. And a lot of these high flyers have done converts, maybe the Bitcoin miners and and so forth. I wanted to just circle back to correlation as well. So in the old days, you could do the dispersion trade. People would wait and say, Oh, if it gets to 60 bid, I'll do some 65. And you know, in crisis periods, we've seen it bid 90, not just GFC, but Eurozone sovereign crisis has got 90 bid. And now we've gone from in insanely low to just very low, right? From maybe 10 to 20. And I keep asking myself, what is it that's gonna really create a tremendous amount of correlation in the markets? And I'd love for you to tell me you're thinking on this because I want to also finish with your view on the rate side. Is it that the market risk side of things, the de-risking, does it need to start to flow through to the real economy? Where maybe it's the wealth effect or the CapEx pullback has started to hurt the
32:33real economy, which outside of CapEx doesn't seem great.
32:37Unidentified speaker (UNKNOWN): So
32:37Unidentified speaker (speaker_2): I'm just wondering like what the pathway is to like a bona fide 50 realized correlation environment and not just for a week around the tariff tantrum. How do you think about something that gets considerably higher from here? What would need to happen?
32:51Unidentified speaker (UNKNOWN): Yeah,
32:51Unidentified speaker (speaker_0): so that's a great question. So again, I'll bring my second parallel, having again pivot from 2000 to the GFC, 2008 to 2010. Parallels, not exactly the same thing. So if you remember, synthetic credit was a big thing. Synthetic credit tranches, super seniors were the thing where you would sell implied correlation in credit, right? So CDX index, if people don't know this, 125 equally weighted names, CDX index itself. can be tranched into four different layers today: equity, junior mes, senior mes, and super senior. Super senior isn't a correlation trade in a very similar form, like the correlation trade in equity, but it's in credit. So what happened then? So let's take some learnings from that environment. Correlation did not rise for a very long time in credit because that structure, the super senior selling protection on it. was considered to be default-proof as it is. You're not going to get a 15% economy by default. So what happened during the GFC was as spread started to migrate and delevering started to happen, many of those trades that were actually pretty good trades started to widen out because people had to exit them. That drove implied correlation in credit from a very low number to a very, very high number. in the index tranche market, which I've written a number of papers on. I was part of buying protection on it at the time at PIMCO, and it worked out really well. So I bring
34:11that frame of reference now here. So it's less about realizing correlation and implying correlation at the 70-80 level. I think the first move will happen, not because of a realization of correlation like that. Of course, the thing that would drive it would be what you just mentioned. The proximate cause would be an economy-wide unwinding of the AI trade. But the perception in people's mind that that could happen. And the capital structure in equities, if you want to take an analogy, starting to filter up the scale where people say these very nice, persistent strategies, think about dispersion, think about other correlation-based trades, diversification. internally inside of large cap, small cap, et cetera, et cetera, the many guises that that trade is on a mark-to-market basis under risk. That results in an unwind. And I think we're beginning to see a little bit of that. And then it becomes a cascading effect as many of those participants start to effectively roll over. And then, because implied correlations get to that point, realized realize start. to do the same thing. So the same thing happened in the GFC. Once implied tranche correlations got really high, credits actually started to behave like they were correlated. And at that point, and maybe this is kind of like the final comment on how this whole thing ends, at that point, somebody who's not
35:31very smart like us, but manages liquidity very well, steps in and says, Okay, this is a very one-sided situation now. I'll take implied correlation in 99.99. Right. I think that's what we're talking about here, right? That's why we do what we do is you maintain liquidity, you maintain potency, you maintain the ability to step in when all the smart money is just running over each other, right? That's when you step in and you say, Well, I got liquidity to buy something at these dirt cheap prices. So that's kind of my story, the arc that I have. observed at least five times in my own career.
36:05Unidentified speaker (speaker_2): Yeah, it's interesting. Like if you look at implied correlation, again, it got down to 10, some just incredibly low levels because I was realizing two or three. So there's still a tiny bit of carry there, but it's not as if someone's going to look at it, put their Graham and Dodd hat on and say, wow, at 30, there's a lot of value here, right? So it's 10 to something potentially. much higher if you get the levy break. Here's where I want to finish. So we've got our Fed meeting coming up on December 9th. In some ways, it feels to me that they really are trying, they're on a credibility regaining tour in some ways. You know, inflation is high, but it's really the persistence of it that they're really trying to fight back or feeling like they have to fight back. on. I want to show this chart for you and get you to talk through it with us. So
36:56Unidentified speaker (UNKNOWN): I
36:56Unidentified speaker (speaker_2): just started looking at this. You know, the VIX closed at 28 yesterday. It was at 30 this morning. And I just I went on Bloomberg. They've got that little AI thing they call Ask B. And I asked it, I said, how many times has the Fed hiked with the VIX above 30 the day before? The answer is once. It was June of 2022. The VIX closed at 3260 the day before. But of course, you'll remember the policy rate was 87 basis points, and year over year inflation was 8%. They were way, way behind. This is not we're behind. It's the persistence, and we've got to do something at some point. But I just wonder, and this is where I'd love for you to just talk to us. and help us close this out is at what point does financial stability, that unwritten third mandate, creep into the equation for Warsh and his team? So, how much asset price damage or turmoil can occur and the Fed still go through with their hike?
37:56Unidentified speaker (speaker_0): Very little. I think you'll you've hit the nail on the head. So earlier this year, I wrote my piece in January saying the year of the Fed irrelevant. irrelevance, the year of the horse, the fire horse, which happens every 60 years. I was, in a sense, okay, but also in a sense, I was pretty wrong because the Fed has become very relevant again now from the point of view of what you just talked about. They are tightening because they're behind in inflation. They are relevant from the point of view of financial markets again, because they could create the thing that nobody wants, which is a massive liquidity crisis and a massive. unwind, which they don't want to do to solve the inflation issue. So, yes, the third mandate is very much alive. I think the likelihood of them pivoting if they see a big crash is very, very high. But they're not going to pivot by saying something like, oh, we're not going to tighten. They're going to say something like, We are watching financial conditions at this point. The liquidity of treasury markets is extremely important, the liquidity of financial markets. is very important. That signal, that code will be enough to tell the market, okay, game on. Now, what do you do if you can't bet on it, right? What you do is you find places of asymmetry where you sort of win in both sides. And I'm going to be very simplistic here, but I always go back to things like the two-year treasury. If I look at the two-year two year treasury for my own capital, I'm not recommending any security,
39:16but I'm just telling you what the two year treasury is. Two-year treasury is giving you a 475 yield. If you hold it for two years, you're going to make 475 a year, guaranteed, nominally at least. If rates go up 300 basis points, which probably they might have to do to quell inflation, you barely lose any money on a one-year your horizon because you're getting your roll down, you're getting your carry, you're losing on the duration of one point. 1.8. Net net, okay, fine. You're slightly negative, maybe zero, slightly positive. Well, what about the thing that you just mentioned? What about if we actually have a financial crisis and they actually pivot? A negative 300 basis point in the two-year treasury makes you almost 12 or 15%, right? Your yield plus take 1.8, 1.86 six times whatever your yield fall is, you know, about 10, 12 percent. That's a pretty darn asymmetric trade right now. So yes, we've been bond haters not loving long duration for a long time. But at this point of the cycle, you have to start thinking that the Fed pivot in a financial crisis is a possibility. Pay attention to the front end. Back end, I'm not so sure because inflation is still a problem. But that's the thing, the framework in which I'm going in is basically like that framework. And by the way, there's one other chart you should put up on your Bloomberg is just the level of Fed funds, right? Last time we've gotten to this type of level, plus or
40:35minus 40, 50 basis points, bad stuff happens. So this is a zone where you are now in the zone of danger. So I don't know which way the world turns out, but in terms of contrary thinking, in terms of thinking what's unexpected, I do think what you said, Dean, is exactly right, is the Fed pivot is not priced in the market, but you should be paying attention to a possibility of it happening.
41:03Unidentified speaker (UNKNOWN): And
41:03Unidentified speaker (speaker_2): the way you described the two-year is it's sort of long optionality. And again, you've described because you're getting that baked-in And yield, there's a convexity to the up for down move. Are there option trades to do? What would you say? about just being long twos versus having some option structure to be along them? Is there value and optionality? How do you think about that?
41:22Unidentified speaker (speaker_0): Absolutely. So you look back two weeks, you should have been holding two years, make a 475, take a percent of that 475, throw it at some optionality, right? I mean, it doesn't matter which direction. I mean, that's a pretty convex trade. I wouldn't be very cute. I mean, there are a lot of cute trades you can. could do in single stock and forward ball and all kinds of stuff. But I don't think you have to be particularly sophisticated right now. The dynamics of the world in the longer macro context is one of uncertainty. And if this thing escalates in either direction by modality, I think having that explosive convexity plus yield is kind of your design element. And then you can fill it with your favorite trades in it, but I think yield plus optionality is the framework within what one should be thinking about.
42:13Unidentified speaker (speaker_2): Yeah,
42:13Unidentified speaker (speaker_0): we'll both be looking at the listing on Monday, right? So we'll see if we were right or we were totally wrong.
42:17Unidentified speaker (UNKNOWN): Exactly.
42:21Unidentified speaker (speaker_1): Alpha Exchange. If you've enjoyed the show, please do tell a friend. And before we leave, I wanted to invite you to drop us some feedback. As we aim to utilize these conversations to contribute to the investment community's understanding of risk, your input is valuable and provides direction on where we should focus. Please email us at feedback at alpha exchange podcast.com. Thanks again and catch you next
42:45Unidentified speaker (UNKNOWN): time.