Analyzing Market Volatility

Advisers – please don’t mass email all of your clients about current market volatility. Some clients have been through this before and are quite relaxed - and some don’t pay much attention to financial news. Don’t add to the ‘noise.’ Instead: Segment your clients: 1. Have they been clients for 5 years or less? If they’re relatively new to financial planning and investment, the current volatility may be concerning. 2. Have they indicated via your onboarding process, risk profiling and conversations that they feel uncomfortable with temporary portfolio declines? Clients that tick both boxes deserve your personal attention - not a generic email. 1. Call them. 2. Ask them how they’re feeling. 3. Listen to them. 4. Don’t judge - empathise. 5. Don’t send charts and graphs. Confirm to them that your own family’s life savings are invested in exactly the same way as theirs. Reassure them that we will get through this and that you’ve got their back. “This too shall pass” and in the meantime you’re with them every step of the way. Onwards…

The Volatility-of-Volatility Term Structure - This paper studies the term structure of the VVIX (volatility of volatility), a measure of expected volatility changes in the VIX (volatility index). Here are the key findings: Informational Content of VVIX Slope: The study reveals that the slope, not the level, of the VVIX term structure holds significant information about vol-of-vol risk. A steeper slope predicts positive returns on S&P 500 and VIX straddles (options that profit from price movements in either direction). Importance of Vol-of-Vol Risk: The paper highlights that VVIX slope offers unique insights beyond the VIX term structure and variance risk premium (VRP). This implies vol-of-vol risk is crucial not just for VIX options, but also for stock index options like the S&P 500. Decomposing VVIX Term Structure: The research employs a model to explore the drivers behind the VVIX slope. It identifies continuous vol-of-vol and jump risk as the main contributors, with their influence varying based on economic conditions. Economic State and VVIX Slope: During calm markets (low q/V ratio), jump risk and a constant term dominate the VVIX, leading to a flat term structure. Conversely, in turbulent markets (high q/V ratio), continuous vol-of-vol risk takes center stage, causing a steeper slope. VVIX Slope and Market Downturns: Analyzing major crises, the study shows that the VVIX slope captures a shift in the composition of vol-of-vol risk. Initially, jump risk is prominent. However, as the crisis unfolds, volatility uncertainty becomes the primary driver, suggesting market participants anticipate prolonged volatility. Overall, the paper emphasizes the significance of the VVIX slope as a predictor of returns and a valuable tool for understanding the dynamics of vol-of-vol risk in the context of stock and VIX options.

Government bonds underperformed equities, credit and commodities in this 3-year risk on market. Our analysis shows when equities sell off, Treasuries are also less diversifying compared to decades prior (chart). What’s happening? Long bond yields are made up of 2 components: ➡️ Policy path - in a world shaped by supply, central banks are more limited in their ability to come to the rescue of the economy without reigniting inflationary pressure. Hence Treasuries are less reliable when equities fall. ➡️ Term premium - it’s driven by bond volatility, inflation uncertainty, and of course fiscal dynamics. Think of it like any other type of risk premium such as equity risk premium it’s about perceived risk and additional required compensation above risk-free for holding it in portfolios. Large deficits record debt and heavy issuance mean that term premia can reprice higher, maybe especially during stress, pushing long yields up even as markets may price a lower policy path. Together, these forces weaken the traditional stock–bond hedge. I think of Treasuries now as quality income assets not the diversifiers they used to be.

This evening in ERCOT is tight. In anticipation of the inevitable finger-pointing and misrepresentations of what is causing it, here are some facts. Thermal outages at this moment are 10 GW. That's at the 'extreme' level under the SARA report. Notwithstanding that this late September, load is very, very high. Especially for a Sunday. HE19 and HE 20 (from 6 pm to 8 pm), in the chart just pulled from the ERCOT website, is showing a ~4-4.5 GW load miss in the chart below. This means that system load for today (forecast by ERCOT) is turning out to be ~4-4.5 GW HIGHER than they had forecast, a mere ~24 hours earlier. Oops? Additional load of 4.5 GW in an hour that was forecast to be 68 GW is almost 7% error. That's very poor forecasting. Why does this matter? If system conditions change this rapidly due to forecast errors of this magnitude, who fills in the gap? Some enterprising energy storage and fast-ramping gas plant operators might be lying in wait in realtime energy, having had proprietary algorithms that predicted this miss and the resulting very high prices that are materializing. But when forecast misses are this large, the ancillary markets are available at a moment's notice to be deployed if the realtime energy markets are insufficient. What we are watching today (massive forecasting misses) is literally the exact reason to have ancillary markets, and we will likely see deployment of multiple ancillary products if the load doesn't miraculously drop to what ERCOT had previously forecast yesterday. There are MANY factors and variables at play in how markets play out daily and there can be many surprises. So for today, what are the causes for the high prices and use of ancillary market products to keep the system intact? 1) Load far higher than predicted (2-4 GW higher depending on hour) 2) Thermal outages hitting 'extreme' levels for this time of year (4 GW higher than expected) 3) Wind lower than expected for these hours (1.5 GW lower) Add that up. What is the biggest contributor? Answer: It isn't the wind. Meanwhile, for those of you following along the current ancillary markets sizing process in ERCOT, there was a lot of debate last week on the sizing of ancillary products during some of the highest-risk hours, like 6-8 pm. The methodology being proposed by ERCOT would reduce volumes in those hours because those hours are now so 'predictable' based on the data from the previous two years. Today has just showed two critical points: 1) That these evening hours are not as predictable as ERCOT has suggested as recently as two days ago, and that reducing ancillary procurement in these hours will create more operating risk on the system. 2) Of the multitude of factors that lead to scarcity and high prices, high load and thermal outages are at least as large of a contributor as anything else. IMM take note. #ERCOT #energystorage #ancillarymarkets #loadgrowth

Volatility Smirk A volatility smirk, a variation of the more common volatility smile, is depicted by plotting the strike price and implied volatility of options for a specific underlying asset, such as a stock, all sharing the same expiration date. In a volatility smirk, the implied volatility (IV) of options on a given underlying security or index decreases as options move further in-the-money (ITM) or out-of-the-money (OTM). This results in a graphical representation that slopes downward, resembling a smirk. This differs from a volatility smile, which exhibits a U-shaped curve with higher IV for both deep in-the-money (ITM) and OTM options compared to at-the-money (ATM) options. In contrast, a volatility smirk's downward slant indicates that as options become deeply ITM or OTM, their implied volatility decreases. Implications of a Volatility Smirk 1. Market Expectations: Directional Bias: A volatility smirk, particularly a steep one, suggests the market has a directional bias towards potential downward price movements for the underlying asset. This is because OTM put options (offering downside protection) have higher implied volatility compared to ATM options. 2. Pricing of Options: Higher Costs for Protection: Options positioned in the "tail" of the smirk, particularly far OTM puts, tend to be pricier compared to their counterparts in a flat volatility scenario. This reflects the market's inclusion of a higher risk of substantial price drops. Potential for Skew Strategies: The smirk can create opportunities for options strategies that exploit the volatility difference between ATM and OTM options. However, transaction costs and other factors can make these strategies complex to implement. 3. Risk Assessment: Perceived Market Risk: The smirk's steepness can serve as an indicator of perceived market risk. A sharper smirk suggests a market with heightened concerns regarding potential downside risks. #options #volatility #smile #smirk #trading

Stocks AND bonds have both been moving down over the past few weeks, just like in 2022. We call this the "ziggy problem"❗ But what is the "ziggy problem" of stock-bond correlation? And how can investors avoid the negative returns trap? In 2022, a "balanced" portfolio of stocks and bonds lost value on both the stock side and the bond side. As of Q1 2026, we're seeing a similar story play out. Why does this happen? Aren't bonds supposed to zig when stocks zag? Why are bonds not protecting portfolios right now? It all comes down to INFLATION and expectations for interest rate moves. In times when inflation is a concern for investors (and for central banks), we see positive correlation. Stocks and bonds tend to move together. We've seen this before: 📌 1970s-80s (stagflation)... 📌 2022 (COVID/stimulus causing inflation) ... 📌and now again in Q1 2026 ($100+ oil causing inflation concerns). When inflation is a primary concern, central banks are reluctant to cut interest rates, because adding more liquidity could make the problem worse. So while stocks and economic data are arguing for cuts, inflation is arguing to keep rates unchanged (or even an interest rate hike). The result? We don't see bonds rally when stocks fall. ❓So which investments perform well in this environment? Uncorrelated hard assets - like infrastructure, shipping and real estate. They often have 1️⃣ built-in hedges against inflation risk 2️⃣ they pay a regular return in up and down markets 3️⃣ they are inherently uncorrelated to short-term economic factors. That's why they are among the few asset classes in the GREEN at this point in the year. Are bonds still valuable in portfolios? Of course! If high oil prices start to impact economic growth, then we're looking at recession risks. In this scenario, the Fed and central banks will look to aggressively cut rates, causing bonds to rally. We can't discount this scenario. But we're not there yet. Our outlook is still for economic growth and the bigger concern right now is inflation rather than recession. That's why the combination of hard assets AND bonds are necessary to protect against inflation as well as growth risks. 📊 This chart is on p.64 of our Guide to the Markets and p.6 of our Guide to Alternatives, available at jpmorgan.com/GTA.

The correlation between stock prices and bond yields has returned to positive territory -- hinting at a period of distress in equities and a regime shift in equity and bond markets where recession fears, rather than inflation, may be starting to drive direction of both. The correlation between the two asset classes was positive for the better part of 20 years prior to the pandemic, suggesting equities trended in the direction of yields as inflation mostly coincided with growth. Stocks held a negative correlation to yields throughout most of the 1980s and 1990s, when inflation hurt stocks -- and that phenomenon returned for the 2022-24 bear market and recovery period. Notably, major stock corrections occurred each time the correlation jumped out of its primary regime. Bloomberg Intelligence Michael Casper, CFA

Anatomy of SPX returns around the biggest 1 day losses and gains Large single-day moves in equity markets rarely occur in isolation (keyword: volatility clustering). A review of the biggest one-day losses and gains in the S&P 500 reveals a consistent pattern: both types of extremes tend to occur in already weakening markets. These events typically emerge after a period of negative momentum, suggesting they are part of a stressed market regime rather than true outliers. The largest down days often resemble a “washout.” After a sharp sell-off, the market typically delivers moderate gains in subsequent days, supported by stretched positioning and short-term mean reversion. Conversely, the biggest one-day gains usually represent reversal moves within existing downtrends. However, these sharp up days are not reliably followed by further upside — the average pullbacks that follow are mild and statistically insignificant. I also looked at the five-day volatility before and after these extreme moves. Here a notable asymmetry appears: volatility drops significantly after major reversal days, indicating that part of the market shock has been digested. After the biggest down days, however, volatility remains elevated and normalizes only slowly. This has clear implications for option strategies: 1) Delta hedging becomes challenging during volatile periods. Strong reversal moves can create abrupt swings that make it difficult to maintain stable delta exposure. In stressed markets, gamma risks rise quickly, complicating hedge management. 2) After sharp sell-offs, elevated volatility offers more time to manage long-vol positions. Because volatility does not immediately fall after large down days, traders have a longer window to take profits on long-vol strategies or rotate into volatility-selling approaches. These periods can offer both tactical opportunities and attractive entry points for vol-based strategies. Overall, the analysis shows that extreme market moves contain valuable signals for timing option strategies — and highlight the importance of disciplined risk management when volatility is elevated. #investing #options #volatility *Data from 1/1961 to 11/2025. Sample size 50.

Ever wonder why corporate bonds and equities become highly correlated when markets crash? The Merton (1974) model explains why. Merton defined a corporate bond as a combination of • a risk-free bond—in normal times, the bondholders’ upside risk is limited to the regular coupon payments and return of principal—and • a short put position on the company’s assets. If the company’s asset value depreciates below its debt, bondholders become long the company’s assets and receive what’s left through bankruptcy proceedings. (Meanwhile, as the stock price goes to zero, stockholders are wiped out.) Hence, as a company approaches default, the market starts to expect that bondholders will be left holding the bag (of the company’s remaining assets). Merton explained that “as the probability of eventual default becomes large, ... the risk characteristics of the debt approach that of (unlevered) equity." (From the book Beyond Diversification.) * Merton, Robert C. 1974. "On the Pricing of Corporate Debt: The Risk Structure of Interest Rates", The Journal of Finance, Volume 29, Number 2, pp. 449–470.

Kronos is a new AI model from Tsinghua University built to understand financial candlestick data like a language. Trained on 12 billion records from 45 markets, it forecasts prices 93% better than leading models, cuts volatility errors by 9%, and creates more realistic synthetic market data. Traders can use it for forecasting, risk management, and testing strategies. Best part: the pre-trained model is open source on GitHub. 🔹 Performance Kronos lifts price-forecast RankIC by 93 percent versus the leading TSFM and 87 percent versus the best non-pre-trained model. Volatility MAE drops 9 percent. Synthetic K-line fidelity improves 22 percent in zero-shot tests. 🔹 Data scale Pre-trained on over 12 billion K-line records from 45 exchanges and seven frequencies. Learns cross-asset, cross-timescale representations from OHLCVA that transfer across forecasting, risk, and generative tasks without fine-tuning. 🔹 Tokenizer A specialized coarse-to-fine tokenizer with Binary Spherical Quantization discretizes market moves into hierarchical tokens. Sequential subtoken prediction beats concurrent prediction and continuous baselines in ablations with matched parameters. 🔹 Investment impact Backtests on China A-shares show the top AER and IR among baselines. Test-time sampling lets desks trade accuracy for compute. Averaging multiple rollouts raises IC and RankIC without retraining. How to use it: Alpha research. Start with zero-shot price or return forecasting on your universe. Rank by Kronos signals and run quick AER and IR checks. Volatility. Plug Kronos volatility forecasts into position sizing and options surfaces. Synthetic data. Use Kronos-generated K-lines for stress tests and data augmentation. Validate with TSTR before deployment. Inference scaling. Ensemble multiple sampled trajectories for higher stability around rebalance. Credits: Authors: Yu Shi, Zongliang Fu, Shuo Chen, Bohan Zhao, Wei Xu, Changshui Zhang, Jian Li. Institute for Interdisciplinary Information Sciences and Department of Automation, Tsinghua University.