The Market Disregards Correlation

[00:00:01] Hello, this is Dean 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. Right, Alpha Exchangers, it's been a busy year indeed in markets and also on the podcast front.

[00:00:26] We put out 25 pods thus far this year and I expect we'll get to 45 in 2026. We've got an exciting fall schedule and I look forward to bringing you some extraordinary guests. As I look back on our 265 episodes over nearly eight years, I have been the quote guest on roughly 45 of them.

[00:00:47] That includes retrospective episodes that review events like the 87 crash and LTCM and more recently the 2021 frenzy in GameStop. I love financial market history with the view that in the words of George Santayana, those who fail to remember history are doomed to repeat it. Show me a time when things went horribly wrong and I will show you an experience to learn from in this business.

[00:01:14] My solo appearances on the Alpha Exchange also include work I've done to highlight women on the podcast and efforts in the industry to advance the cause of career growth for females in markets. It's a topic I'm passionate about and at my recent Macro Minds conference, I was really excited to support 100 Women in Finance as one of the event's three beneficiaries.

[00:01:38] In addition to being a podcast host, I'm also a market practitioner who spends a fair amount of time thinking about risk and optionality and trying to look around corners to spot sources of vulnerability in asset markets. My process begins and ends with market prices because they are what is actionable. You can't trade GDP futures, as they say, although Calci and Poly are probably going to change that sooner or later.

[00:02:07] But prices drive P&L and that's of course what matters for investors. What I'd like to do over the next half hour is share with you what's on my mind with respect to the prices we all stare at and engage with for hours daily. It's been not just a busy year for the pod, but also one for students of market risk and specifically those card-carrying vol nerds like me who consume the vast information coming from the derivatives markets.

[00:02:37] I'll take you through the cross-asset picture as I see it right now and tie together three themes that look separate on the surface, but are really one story. Let's get dialed in, as they say. Here's what I want to cover. First, correlation. Realized and implied correlation among stocks have fallen to levels never seen before, and this is doing something specific to index volatility.

[00:03:02] Second, a large part of this market is in what I call a spot-up-volup dynamic, where a stock rallies sharply and its options get more expensive at the same time. And third, across asset classes, the cost of insurance is quite inexpensive and I believe too low relative to the forces of uncertainty imposing themselves on markets.

[00:03:26] Let's start with correlation or, in today's market regime, the utter lack of it. If I've been fascinated with one thing in markets over the last 18 months, it has been the record low level of correlation among stocks and its dampening impact on volatility in indices like the S&P. If you're following me on Twitter, you may have tired of the commentary by now.

[00:03:55] And I do get that. But I wouldn't keep coming back to it unless I was so convinced this was a risk hiding in plain sight. It's true. I said the same thing last year. I'm looking back on a solo podcast I dropped in October 25 entitled, quote, Low correlation is the defining risk in markets. I'm not one prone to exaggeration and clickbait is not my forte.

[00:04:21] So the title should be considered more in the context of conviction. The correlation cometh, I say. Maybe not today or tomorrow, but it's underpriced. Let me explain. Since October of last year when I published that pod, the realized correlation of stocks in the S&P has only fallen. It's roughly 10 percent.

[00:04:43] To give you a sense as to how unusual this is, the correlation of the S&P to the TLT since that time is around positive 35 percent. What? Yep, we live in a world in which common stocks diversify themselves better than exposure to the good old risk-free bond market. These are indeed unique times. Let's step back and consider the interaction between realized and implied correlation.

[00:05:14] In a giant derivative complex like the options market on the S&P 500. To put it simply, implied is so low because realized is even lower. Let's explore the linkage. Especially when it comes to derivatives, market participants are beholden to pricing what they experience. And by experience here, I mean the daily fluctuations in the underlying asset.

[00:05:40] In options markets, the marginal price setter is the replicator. The vest-wearing, model-driven, matcha-drinking, mathy type who seeks out puts and calls that can be hedged at a cost favorable to the market price. These arbitrageurs, as they're called, are responsive almost entirely to how a trade carries. And favorable carry boils down to some comparison of realized to implied volatility.

[00:06:08] The incredibly simplified version. When you can buy options at an implied vol lower than the subsequently realized vol, your pod leader is happy. When you can sell options at an implied vol higher than subsequently realized, pod leader also happy. Carry is everything in markets. It underpins the attractiveness of curve trades and rates in various commodities.

[00:06:34] In options, it's the essence of the VRP, or vol risk premium strategy, that mostly works but of course can blow up in equities. And in the S&P, correlation carry is a thing as well. If realized correlation in the S&P is 10, you can bet that implied is just a little bit higher.

[00:06:54] It's high enough to motivate very intelligent investors to remain in the dispersion trade that combines a long position in single stock vol with a short one in index vol. That spread is the carry that has been a reasonably consistent source of P&L with a few notable hiccups like the 2025 tariff tantrum and the 2026 oil shock.

[00:07:18] Because the boss pays out on real-time profits, not those you suggest may materialize in the future, vol, correlation, and credit risk are priced based on how they carry in real time. In this way, realized vol, realized correlation, and realized defaults are the anchors for pricing. They set the conditions through which profits can be manufactured.

[00:07:42] If realized correlation is near zero, it matters not what Dean Curnutt says about forward-looking risks. You can do the dispersion trade at very low levels of implied correlation and still generate profits. But, but, but. In 2006, one could have said the same about credit risk in the U.S. banks. Morgan Stanley CDS traded in the 20s in 2006. That was simply because it made money to sell it there to earn carry.

[00:08:11] This is an incredibly cherry-picked example of a derivative price ultimately proving far too sanguine relative to the risks that would unfold. But the point is that carry matters most in pricing. And when realized vol, or in the case of the dispersion trade, realized correlation is so low, market pricing relationships are going to reflect it, irrespective of the danger that comes from such a low margin of safety.

[00:08:40] When an environment persists for long enough, when something has not happened for a good while, the market starts to price it as if it cannot happen at all. I care about these never-seen-before levels of correlation so much because we have no real experience with them. There's no playbook in such uncharted territory, and I argue that it matters not just for the hedge funders,

[00:09:04] but also for long-only investors who haven't touched a listed option or have not even a passing interest in the dispersion trade. Why do I say that? As Les Grossman said in Tropic Thunder, let's take a giant step back. Alright, I won't say the rest, but what a cameo from Tom Cruise. The volatility of an index is not just the volatility of the stocks inside it. It's very much also the degree to which they move together.

[00:09:32] You can hold a basket of individually volatile stocks and still get a quiet index as long as those stocks are not moving in the same direction at the same time. Correlation is the multiplier. Right now, and for the better part of 24 months, that multiplier is close to zero. Let me give you the numbers because this is a case where you have to be specific.

[00:09:56] On July 22nd, one month realized correlation across the S&P 500 printed 0.4%. That is basically the zeroth percentile over 15 years. Out of something like 4,000 daily observations, only three have been lower. All of the exceptionally low levels have come since 2025. This is not a fluke print. This is where we live right now.

[00:10:24] Implied correlation tells the same story going forward. One month implied correlation fell to an all-time low around 4% recently. One year implied correlation recently reached 17%. To see how low that is, go back to 2011. As you all know, I am obsessed with financial market risk anniversaries.

[00:10:47] In early August of 2011, the VIX surged to 48% as the U.S. debt ceiling crisis unfolded and correlation among stocks was incredibly high. Realized correlation in 2011 was 75%. And, shocker, one year implied correlation reached 75% as well. Carry, as they say, rules the world.

[00:11:12] Today's diminutive, skinny, negligible, paltry, piddling, and utterly pint-sized level of implied correlation is perfectly consistent with the concurrent carry, but is way too low relative to macro risk, in my view. We are pricing close to the polar opposite of a correlation event. The market, as the podcast title suggests, disregards correlation.

[00:11:38] I like buying disregarded events, especially when I can make a macro argument as to why that same event might actually materialize. Today's disappearing correlations occur in an index that has become considerably more top-heavy. As I posted on Twitter back in 2011, the top 10 names in the S&P constituted roughly 18.4% of the index market cap.

[00:12:05] Today, that same top 10 comprises 39.6%. One might be forgiven for thinking that with more than double the concentration among the top 10, and with those stocks considerably more volatile than 2011's leadership, you'd have a more volatile index. It's the exact opposite.

[00:12:26] The simple average of the 45 pairwise correlations among the top 10 was 61% for all of 2011. Over the last year, it's 22.6%. The average realized volatility of those top 10 stocks was 27% in 2011. Today, 35%. So the stocks themselves are more volatile.

[00:12:52] But the realized volatility of the S&P was 23.4% in 2011. And today, it's about half that, around 12.7%. In plain words, we have a much more concentrated index made up of considerably more volatile, but much, much less correlated stocks. A rule of thumb that travels well in audio.

[00:13:16] About five correlation points translate into one index point from a volatility standpoint. If you want a single image for how far this has gone, take two of the largest stocks in the world, NVIDIA and Google. Their one-month realized correlation recently went negative, down around minus 10%. Two of the largest, most connected names in the index moving in opposite directions.

[00:13:44] Over that same window, the correlation between the S&P and the TLT, stocks to bonds, was positive, around 71%. So your supposed traditional diversifier was moving with equities, while two mega cap tech names were moving against each other. That is the world we are pricing now as normal. For me, there is a strong reason to think that the true risk is higher than the realized numbers suggest.

[00:14:12] There's been a lot written recently about what many call the circularity question in the AI trade. When you map it all out, there is a heavily overlapping set of loans, backstops, customer financing arrangements, and equity stakes among the key players. These companies are all attached at the hip. There is a commonality of exposures that does not show up in the daily correlation of realized returns.

[00:14:37] And precisely because it does not show up, S&P volatility is a lot lower than it otherwise would be. I think that it is likely that these companies prove considerably more correlated than a market price like implied correlation currently handicaps. So the market disregards correlation. It prices a correlation event as if it cannot occur, specifically because it has not occurred for a long period of time.

[00:15:06] But these things do occur. Risk enthusiasts like me marked the 10-year anniversary of Brexit in late June. A decade ago, on that day, June 24th of 2016, the euro stocks fell by 8% and the euro fell 2% on the same day. The incoming correlations beforehand suggested to borrow a line from Ace Rothstein in Casino that it could not happen. It would not happen.

[00:15:37] It did happen. And low correlation is not a stable state. We got a live preview earlier this year when the average pairwise correlation among the biggest 10 names in the S&P increased from roughly 25% early in the year to 57% over the course of a month to as high as 78% in a single week. It can move from the low single digits to the high 70s in a hurry.

[00:16:04] When stocks become more volatile, they also become more correlated. And because index vol is so dampened by low correlation today, there is enormous scope, an enormous amount of room for it to re-rate higher if that correlation does show up. Fresh off the poorly received Kevin Warsh press conference and the hurried unwind of the Situational Awareness Fund, the events that would boost correlation are not exotic.

[00:16:32] A geopolitical development. Rising oil prices feeding back into inflation. Account. Tweet Gamma, anyone? A hyperscaler deciding to ease off on CapEx. A financial accident inside the leveraged ETF product complex. An illiquidity problem in private credit. There are many macro shocks that have historically pushed correlation up, and the market is charging almost nothing for them.

[00:17:01] Back to long-only investors for whom the dispersion trade is of no concern at all. They do have something very much in common with the MIT types who populate the QIS desks on the sell side. They both are responding to record low levels of realized correlations. The long-onlys are taking considerably more risk at the portfolio level than they may realize very much because of the unsustainably low level of correlation

[00:17:30] and the unsustainable amount of diversification they are enjoying as a result. If 5 correlation points is worth a vol point at the index level, a return to 40 on realized correlation adds 6 additional vol points of realized vol to the index. That's 50% higher than what we are realizing now and could mean that many investors are simply miss-sized, larger than they really ought to be.

[00:17:58] Did I forget to mention that realized volatility in stocks and realized correlation among them are themselves correlated? The very same macro shocks that cause stocks to become more volatile also cause them to become more correlated. This empirical fact amplifies the risk at the portfolio level. A couple of ending thoughts as we close out this portion of the discussion. First, sizing is everything in markets.

[00:18:27] Just ask situational awareness. Or, 20 years ago, Brian Hunter. Both received an incoming call with the most terrifying eight words. We're here from Citadel, and we're here to help. Both were comically too large and concentrated in their positions. Both would have done well to listen. First, to Hyman Minsky, who told us that, quote, success breeds the disregard of the potential for failure.

[00:18:55] And then to Voltaire, who said, uncertainty is uncomfortable, but certainty is absurd. Now to the second piece, which is where a lot of the single stock volatility is coming from in the first place. A big part of the market is in a spot-up, vol-up dynamic. It turns out that cratering correlation is, to an extent, the result of the unique behavior of single stock vol. Let me explain.

[00:19:21] Normally, when a stock rises, its implied volatility falls. Fear comes off. Market caps expand. Credit risk declines. And options get cheaper. Spot-up, vol-down. That's the default. But in a spot-up, vol-up regime, the stock goes up and the options get more expensive from an implied vol standpoint at the same time. Demand for call options surges.

[00:19:49] The skew to the call gets bid up. The term structure of implied vol often inverts. History is fairly clear on how these episodes, if they are protracted enough, resolve themselves. I say it often. Spot-up, vol-up most often ends in spot-down, vol-down. Think of GameStop in 2021. It peaked in late January, and three months later, the stock was down around half,

[00:20:15] with implied vol collapsing hundreds of points in the process. The move up and the move down are not symmetric, and that matters a great deal for how you position. The clearest warning I have seen on this front recently has been in the South Korean chipmaker, SK Hynix. I started waving the flag on this in late May, suggesting that buyers, especially of the toxic leveraged ETF,

[00:20:42] ticker symbol 7709, should be careful. The stock has, as expected recently, imploded. But it was up roughly 900% on a year-over-year basis recently, and its implied volatility had gone from 40 to 120 in the process. That is beyond rare. It was a sign of massive speculation, driven by a genuine shortage of memory,

[00:21:08] extraordinary profits, and the circular demand that comes from both leveraged ETFs and option products. Sitting on top of the already volatile stock is the leveraged product, again, ticker symbol 7709, traded out of Hong Kong, that peaked at roughly $14 billion in assets. I covered this carefully on Twitter and encourage you to review what happened there.

[00:21:34] The TLDR is that its sponsor was forced into the options market on SK Hynix because it could not get the leverage it needed from the swap providers. Its prospectus initially limited it to holding 25% of its exposure through options, but with permission to exceed that under exceptional circumstances. Apparently, these are exceptional times because the option exposure reached 36% in June.

[00:22:02] The fund was buying deep-in-the-money calls that expired in July, and the ratio of option open interest in calls to puts got to nearly 100 to 1, basically 1 million calls to 10,000 puts. I am an old-timer indeed, and I have never, ever seen anything like it. The ETF provider had nowhere to go but to buy incredibly expensive options from sellers who saw it coming a mile away.

[00:22:31] You can guess who got the better end of these trades. The mechanical properties of leveraged ETF products have been well covered by me and others. To rebalance each day, the fund had to buy when SK Hynix rose, and during April and May, the stock surged 5% or more on one of every three trading days. Assets chased performance, and the fund got larger and larger.

[00:22:57] At one point, it held $28 billion of exposure to SK Hynix, a stock realizing more than 100 vol. The hedging runs both ways. When SK Hynix falls, 7709, the ticker for the leveraged complex, had to rebalance to get smaller, selling into a falling volatile market, amplifying already large moves to the downside. And because it bought in-the-money calls,

[00:23:25] the dealer community was caught short gamma. From a risk standpoint, a deep-in-the-money call is the same as a deep-out-of-the-money put. So whoever sold those calls bought stock to hedge and had to dump that hedge very quickly as SK Hynix dropped. It's a toxic brew. We heard a consistent stream of chatter around OTC trades in which swap dealers sought to buy one-day crash risk

[00:23:53] in the underlying at enormous premiums. Another tell that the situation was becoming too large, too volatile, and ultimately unstable. Now, the read-through for me, and this is the part I want U.S. listeners to sit with, is that the same structure has lived right here in the U.S. semiconductor complex. The reaction function for those leveraged ETFs could not have been more specific. We know exactly what they have to do. On one recent day,

[00:24:23] the leveraged complex in Micron and SanDisk products, tickers MUU and SNXX, had to buy around $2 billion of Micron and 1.8 of SanDisk. MUU started the day long about $18 billion of Micron. SNXX started long $13 billion of SanDisk. And don't forget SOXL, the three-times levered semiconductor ETF,

[00:24:50] which peaked at just under a long position of $100 billion of exposure to the SOXX index. There's also KORU, incredibly a 3X fund on the COSP that had nearly $2 billion in AUM and $6 billion of exposure. It had a 42% down move on June 5th, a 34% up move on June 11th,

[00:25:18] and a 36% down move on June 23rd. These are casino-style swings that never end well. The mechanism is a feedback loop. The stock goes up on the shortage story. The leveraged ETFs are forced to buy more to rebalance into the move. That forced buying pushes realized vol higher. Higher vol and a rising price pull in more directional speculation through the calls, and around it goes.

[00:25:47] It is value investing flipped upside down. You are mechanically buying more of the thing precisely because it went up and got more volatile. Memory is indeed undersupplied, and I would add so are memories. People have short ones, or more bluntly, they never learn. I have been calling this the XIV of 2017, just with a different ticker. Back in 2017, as the S&P put in a 50-year low and realized vol,

[00:26:17] the inverse VIX complex got larger and larger. There's a story about a hedge fund CIO having to explain to an excited cab driver who thought he had found a great stock pick that the XIV was not actually a company. It captured how over-consumed that short vol trade had become by people who just assume that number goes up. Today's version is not built on an undersupply of realized vol.

[00:26:45] It's built on an undersupply of memory. But there's one important difference that changes how you trade it. In 2017, the problem was low vol. This time, the defining feature is high vol. That makes trade construction trickier, but not impossible. There's a piece of this that verges on the deceitful and is worth calling out. The leveraged ETF prospectuses carry a standard matrix showing how the product would perform

[00:27:15] against the underlying at various levels of future realized volatility. Several of them include a row for 10 realized volatility. You know what actually realizes 10 vol? Almost nothing. Maybe the TLT, long treasuries, moving 30 to 80 basis points a day. SanDisk regularly moves north of 10% in a single day. A realized vol table for a leveraged ETF product

[00:27:44] on SanDisk should start at 50 and run to 150 vol. When you build it that way, it looks much, much worse and much more honest. I think the leveraged ETF providers in the US are susceptible to litigation and the regulators are worth watching. The top security regulator in South Korea has already expressed regret about the Frankenstein that has been unleashed in that market. That regulatory angle cuts in a direction

[00:28:14] people may not expect. The SK Hynix sponsor is now going the other way. Instead of availing themselves more leverage through the options market and raising the notion all they could use there, it is now giving itself flexibility, not necessarily aiming for two times the returns each day. This is a de facto deleveraging event. It's actually good for financial stability, but if your long options on SK Hynix's or, heaven help you,

[00:28:44] long options on the leveraged ETF itself, it's not good news because it's vol reducing and lower vol is bad for the price of options that depend on it. When the VIX ETF complex cut leverage after the Feb 18 implosion, option prices on those instruments got hit hard. To be long an option on a leveraged ETF is to be on the wrong side of a decision the sponsor can make willingly or in response to the demands of a regulator

[00:29:14] to lower its leverage ratio. This spot-up vol-up dynamic has made its way into the pricing of long-dated options on U.S. high-flying names as well. To review some additional work I did on Twitter, the pricing of long-dated vol and skew in Micron has been fascinating and incredibly actionable for hedgers. As of June 22nd, with the stock at $1,211, the Micron December 2028 call

[00:29:43] struck at $2290 at a bid of $465. Let's put that into words. You could have sold a call and collected 38% of the stock price that breaks even on the upside at $2755, up 227% from the spot price. As Alec Baldwin said in Glenn Gary, Glenn Ross, have I got your attention now? Good.

[00:30:11] And because very high implied vol on far out-of-the-money calls on long-dated options produces almost inconceivably high deltas, this option struck at nearly double the stock price at the time, carried a delta of 63. All I can say is that Micron cannot stay a 90 vol, $1.5 trillion company forever. Timing is everything, and I just happen to have top-ticked the recommendation

[00:30:41] for a zero-cost collar on Micron. But as of June 22nd, you could have used the proceeds from the sale of that $2290 call to buy a put struck at $1080, also expiring December 2028. With the stock closing at $739 on July 29th, that put is a nice, long-dated insurance policy to own. So that's the second theme. A meaningful part of this market

[00:31:10] has been pushed higher and made more volatile by a machine that has to buy. It's also recognition that we live in a world in which innovation is happening so quickly that stocks themselves have become options. The options written on them are options on options. And the options on the levered ETFs on those stocks, well, you get the picture. I want to point you to a recent Forbes piece by Longtail Alpha founder and CIO,

[00:31:39] Veneer Bonsali. It's called, Canaries in the Coal Mine are Telling Us It's Time to Own Both Tails. Veneer outlines just how the market comes to price a company with a market cap north of a trillion dollars with long-dated volatility of 90. It's a function of bimodality in potential outcomes. I think we live in some version of a bimodal world where the status quo simply cannot and will not hold.

[00:32:08] The option pricing implications here, as Veneer outlines, are profound. To close out on the spot-up-vala portion of this discussion, I want to be clear that while I have argued strongly that history consistently shows that these dynamics are indicative of speculative excess that will not end well, this is not a call for a bear market. It has been a call that there was an accident waiting to happen here, an extreme sequence of daily moves

[00:32:37] amplified by the overlay of leveraged ETFs and the options on top of them. Move fast and things break, as situational awareness has learned. So far, we have focused on two market cross-currents that don't make their way to the front pages of the Wall Street Journal or the FT. They're pretty nuanced. I'll revisit a question that I alluded to earlier. How is ultra-low correlation at the index level related to spot-up-vala

[00:33:07] at the single-stock level? I am glad you asked. Let's hold the two ideas next to each other because this is where they connect. Single-stock volatility has been screaming. The VIX EQ, the single-stock version of the VIX, reached 51 in late July. It has been at that level before, in 2024 and 2025. But back then, the VIX itself was double or more than double the level of the VIX in 2026.

[00:33:37] At one point, the VIX EQ was more than three times the level of the VIX. For the relative pricing level to reach such an extreme, you need a degree of implied correlation we have never seen before. And because implied correlation follows realized, that means a level of realized correlation and degree of diversification we have also not experienced. How do you reconcile a single-stock vol in the low 50s

[00:34:06] with an index vol in the mid-teens? Correlation. The stocks are driving the index and investors are paying a great deal for single-stock options, but they are leaving index vol behind because the low level of correlation is expected to continue. The market is disregarding a correlation event in equities. Here's an insight I think is especially important. When we think about how that spread between single-stock and index vol will be resolved,

[00:34:35] we mostly think about a correlation event in which index vol spikes, pulling the spread in. We saw exactly that in August 24, April 25, and in March 26, though each event was relatively short-lived. But there's a second way. Instead of index vol catching up, single-stock vol catches down. Here, the speculative fever cools and the spot-up vol-up names come back to earth. As I posted on the good old Twitter,

[00:35:05] you saw this happen earlier this year in silver and you are seeing it again occur in KOSPI through the EWY. In the first case, index vol catches up. In the second, single-stock vol catches down. Either way, the spread compresses. Theme 2 is pushing the numerator up. Theme 1 is holding the index down and they are two sides of the same coin. That brings me to the third theme and it pulls the whole cross-asset

[00:35:35] picture together. I have had strong conviction that the pricing of insurance, of convexity in this market, is exceptionally low. I have taken to saying that insurance costs a ton in life, just not in markets. Health insurance, car insurance, flood insurance, there isn't a good deal to be had. This could be a good time for a short detour into my own misadventures in driving and car insurance. I'm happily headed east on Route 287 in Westchester

[00:36:04] and in less than a second's time, I'm literally driving over half a blown-out tire, likely from a truck. At 65 miles an hour, I go right over this thing. $9,000 of damage later, my insurance person is asking me if it's wise to submit a claim, lest they raise my premium going forward. Back to the market price of insurance, which I argue is considerably low on two out of three metrics that I think we should all care about.

[00:36:33] Let me make the case. First, I built an index to track this and recently shared on Twitter. It's the average of the five-year rolling percentiles of five measures of insurance costs. The VIX for equity vol, the VXTLT for treasury vol, the CVIX for currency vol, high-yield credit spreads, and the VIX HY for high-yield credit vol. Equities, rates, FX, and credit. The compensation for bearing volatility and spread risk

[00:37:03] across all of it in a single number. That number sits at an exceedingly low level. Go around the horn. The VIX was recently as low as 16. Treasury vol, the VXTLT, recently as low as 10. And in credit, one month implied vol on the HYG got down to around 4.8%. Credit spreads themselves, despite the widening seen in hyperscalers, remain in very low percentiles. These are nominally

[00:37:33] very low option prices and spread levels. That's the first point. For a given spend, you can buy a lot of insurance at current levels. The second observation is that these prices, while nominally low, are not, I repeat, not inconsistent with how they carry. Two things are true at the same time here. On a pure carry basis, you can argue insurance is not cheap because the daily moves in macro assets have been modest. Thus,

[00:38:02] the carry component that plays such an important role in pricing is keeping option prices down. But, in my view, it's the nominal price that matters more. And that brings me to the third component, which is the overall level of uncertainty in the economy, in technology, and in global affairs. Because correlations are so friendly and because the capex trade is so dominant, these factors are not contributing to volatility at the macro and index level yet.

[00:38:32] But I would argue strongly that the amount of uncertainty we are living with justifies higher prices for market-based insurance than we are currently being charged. Insurance can be had for a song, and an option gives you the right to change your mind, a right being handed out cheaply at a moment when the pace of change is already high and, if anything, accelerating. Rates deserve their own word here because I think interest rate vol has been too low

[00:39:01] and it's a defensive asset worth paying for. One of my sayings is that equities are short the straddle on rates. A large move in treasuries up or down mostly spells trouble for the stock market. And with the correlation between stock and bond prices often positive these days, it's higher rates that get the attention as the threat. But, warts and all, treasuries still probably rally if there is a big enough risk off to leave the S&P in a serious drawdown.

[00:39:31] So, rate vol protects you on both sides. The arrival of a new Fed share sharpens this. There is complaining that Kevin Warsh has withheld some of the reaction function guidance the market has grown accustomed to. Fed shares have to master what I call the sweet art of saying nothing. But there is a balance. And when the financial press and the sell side start parsing a chair's words closely as they did with early missteps from Bernanke and Yellen, that exercise alone

[00:40:00] can add volatility as we have just seen with Kevin Warsh. In the aftermath of the July presser, I'm reminded of the classic scene in Goodfellas where, with Billy Batts in the trunk, the crew stops by Tommy's mother's house for a late dinner and to get a knife. His mother says to his visibly shaken Henry, you don't talk too much. That's sort of where the market is with Kevin Warsh right now. It's worth remembering that in the first several years

[00:40:29] of Greenspan's tenure, there was no official announcement about a change in the policy directive. The money nerds had to infer it by studying the Fed's H-4-1 reserve data released on Thursdays. There's got to be some balance between spoon-feeding the market every subsequent scene in the movie and not providing any reaction function at all. The natural tension here is why we viewed long rate volatility as exceptionally cheap. The trade that captured this best for me

[00:40:59] was the largest one I recently saw, 212,000 contracts of the Jan 2028 100, 120 call spread in the TLT were bought for 62 cents. I like the time to expiry a lifetime away, 18 months, given today's incredible pace of change. I like the skew it collects and most of all, I like that you are net buying volatility in an unstable asset at extremely low levels.

[00:41:28] The 100 strike call is about 15 delta. The 120 strike call about a 5 delta. It's a solid way to part with 62 cents and protect the tail outcome. And you do not have to express any of this in a single outcome. You can build a basket across the S&P, high yield, long treasuries and currencies like the euro and the yen. Let's close out this discussion and let me pull the three threads together because they really are one.

[00:41:58] Correlation is near zero which pins index volatility to the floor even as the stocks inside get more volatile. And because the index and cross asset vol both stay quiet, the insurance that would actually pay in a shock is priced as though the shock simply cannot materialize. Each of these makes the other look more reasonable. Quiet index vol makes low correlation feel safe. Low correlation makes the single stock fireworks feel contained.

[00:42:27] And both together make cheap insurance feel like a luxury nobody needs. The market prices what it experiences. The market disregards events that are far enough in the rearview mirror to contemplate happening again. That's human nature and recency bias. But you can flip every one of those. I think about the two largest threats to the U.S. as a serious disruption to either its stock market or its bond market and a large enough disruption to one

[00:42:57] could cause a tipping point in the other. The event that matters is the one where dispersion turns into a correlation event where the stocks stop offsetting and start moving together. That's when the investment community at large is missized having badly underestimated the actual volatility risk taken. When vol risk is suddenly repriced market insurance accumulated at low prices during sunny days becomes a valuable asset in the portfolio.

[00:43:25] Well, that is it for me for now. I wish you an excellent week and appreciate you being a listener. You've been listening to the 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

[00:43:55] dot com. Thanks again and catch you next time.