The Korean Levered ETFs Shaking Markets All Around the World
Alex Altman from Barclays discusses the explosive growth of leveraged single-stock ETFs, particularly in Korea, which have ballooned from $12-13 billion to $50-55 billion in Asia-Pacific over months. These products create mechanical rebalancing flows that function as negative gamma in markets, potentially amplifying volatility, while the underlying stocks—particularly semiconductor companies like Micron and SK Hynix—have attracted retail investors seeking outsized returns.
Summary
The podcast explores the rapid proliferation of leveraged ETFs, products that use derivatives and swaps to provide 2x or 3x leverage on underlying stocks. Altman explains that leveraged ETFs grew from approximately $120 billion in US AUM in April 2024 to over $200 billion by July, with Asia-Pacific growing threefold from $12-13 billion to $50-55 billion. The key distinction is that US growth came primarily from price appreciation of existing holdings, while Korean growth came from aggressive new share creation alongside price increases.
Altman details the mechanical rebalancing mechanics: when a stock falls 10%, a 3x leveraged ETF loses 30% of its AUM, requiring it to sell 10+ billion dollars in notional exposure to maintain the promised 3x leverage ratio. This creates persistent selling pressure on down days and buying pressure on up days—effectively introducing negative gamma to markets. This contrasts with traditional long-only ETFs which provide positive gamma.
The leverage itself comes from prime brokerage swap agreements with banks, creating significant notional exposure on dealer balance sheets. While the absolute AUM ($250-270 billion globally) may not seem enormous, the rapid growth rate and the multiplicative effect through leverage create concerns about financial stability. Altman notes that 93% of Korean leveraged ETF ownership is retail investors compared to 75% in the US.
Altman contextualizes this within broader retail equity exposure: 34% of US household wealth is now in equities (a record high), exceeding real estate holdings (26%) by an eight-percentage-point gap—also a record. A 20% equity market decline would destroy approximately $16 trillion in wealth and likely trigger recession through reduced consumption.
The discussion covers Barclays' BETTY (Barclays Equity Timing Indicator), a quantitative model with 19 inputs tracking discretionary flows, non-discretionary flows, volatility positioning, and real yields. BETTY has been in record warning territory due to momentum crowding, suggesting asymmetric forward returns over two-month horizons—a 42% chance of positive returns versus the 73% baseline.
On valuations, Altman notes the S&P trades at 20.2-20.3x earnings while real yields are at the 95th percentile (~2.3%). Historically, when real yields were this high, S&P multiples averaged 14-15x post-GFC and 18.5x post-COVID, suggesting current valuations are 10% elevated. He attributes this to margin expansion and argues higher-quality businesses deserve higher multiples, though the fiscal deficit and government incentive to prevent equity market disruption support current prices.
Altman discusses how AI and quantitative analysis have become central to modern equity investing, shifting from fundamental analysis to understanding flows, positioning, derivatives mechanics, and factor dynamics. He describes his career pivot from fundamental analyst to quant-derivatives specialist, and how sophisticated investors now require deep forensic modeling rather than generalist coverage. He notes that AI is effective at identifying component ingredients for baskets or hedges but still requires human judgment on optimization, liquidity, and implementation—expertise remains valuable despite AI proliferation.
About this episode
<p>Retail participation in the stock market is booming. And of course the biggest story in markets is the AI trade, which has created an incredible amount of demand for chips and memory. These two broad themes have come together in the form of leveraged, single-stock ETFs. And while these products are popular in the US, the scale coming out of Korea is enormous. It's a good week to talk about this intersection, because some of the biggest stories of the week include Samsung's earnings and SK Hynix's new US listing. Barclays's Global Head of Equities Tactical Strategies Alexander Altmann has used the word “terrifying” to describe the amount of notional exposure coming from these levered ETFs. He explains to us why that is and we talk to him about why, in such a short period of time, the world of levered ETFs has gotten to be so large, with AUM increasing threefold in Asia alone. He also us gets into how he is thinking through risk management and how we as society — and retail investors in particular — got to be overexposed on equities and why that keeps him up at night.</p> <p>Read: <a href="https://bloom.bg/4f68KA5">SK Hynix’s $26.5 Billion Listing Reopens Asia Route to US Market</a><br /><br />Only Bloomberg.com subscribers can get the Odd Lots newsletter in their inbox, plus unlimited access to the site and app. Sign up: <a href="https://www.bloomberg.com/subscriptions/oddlots?in_source=oddlotspodcast">bloomberg.com/subscriptions/oddlots</a></p><p>See <a href="https://omnystudio.com/listener">omnystudio.com/listener</a> for privacy information.</p>
Key Insights
- Leveraged ETFs function as short gamma instruments, requiring mechanical selling on down days and buying on up days to maintain leverage ratios, creating a multiplicative amplification of market moves that didn't exist at scale a few years ago.
- Korean leveraged ETF growth has been driven primarily by new retail investor inflows combined with price appreciation, whereas US growth came almost entirely from price appreciation of existing positions, indicating different investor behavior patterns across markets.
- The absolute AUM of leveraged ETFs ($250-270 billion) is less concerning than the threefold growth rate in Asia-Pacific and the multiplicative notional exposure these products create on dealer balance sheets through swap agreements.
- Barclays' BETTY model indicates record warning conditions for equity market returns due to momentum crowding, with only 42% probability of positive two-month S&P returns—though it explicitly measures forward asymmetry, not crash probability.
- Current S&P valuations at 20.2-20.3x earnings are approximately 10% higher than historical levels when real yields were similarly elevated (95th percentile at 2.3%), though margin expansion and business quality improvements partially justify current levels.
- The shift from 26% household real estate wealth to 34% equity wealth—an eight-percentage-point gap at record highs—creates structural government incentive to prevent equity market disruption, as a 20% decline would trigger recession-level consumption impairment.
- Retail investors own 93% of Korean leveraged ETFs versus 75% in the US, concentrating leverage exposure in a cohort with potentially less sophisticated risk management and higher sensitivity to momentum-driven drawdowns.
- AI models excel at identifying component universe for baskets and hedges but remain weak at optimization decisions involving liquidity, implementation costs, and practical constraints, preserving value for human expertise despite access to infinite data.
Topics
Transcript
Now a message from Meta. Meta is launching America's Workforce Academy. The program offers paid training, a job, and a path to America's future. Because the future is for everyone. Learn more at meta.com slash America's Workforce Academy. The thing about AI for business, it may not automatically fit the way your business works. At IBM, we've seen this firsthand. But by embedding AI across HR, IT, and procurement processes, we've reduced costs by millions, slashed repetitive tasks, and freed thousands of hours for strategic work. Now we're helping companies get smarter by putting AI where it actually pays off, deep in the work that moves the business. Let's create smarter business. IBM. When you own your own business, you…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Odd Lots
One of the World's Largest Hedge Funds on Its 86x Growth in Token Spending
Man Group executives discuss their implementation of generative AI across asset management operations, revealing an 86x increase in token consumption since January and demonstrating how AI agents are systematizing quant research, augmenting PM workflows, and identifying novel trading signals across multiple data modalities.
These Are the Sharps Actually Making Money on Prediction Markets
This OddLots episode features journalist Adam and prediction market traders Brian and Daniel discussing how skilled traders consistently profit from prediction markets through rigorous research, data analysis, and ground-level reporting rather than luck or AI. The conversation explores market efficiency, insider trading concerns, the zero-sum nature of prediction markets, and how professional traders systematically beat retail investors.
How a Major Grocery Store Chain Can Dramatically Lower the Cost of Food
Scott Patton, Chief Commercial Officer of Aldi U.S., discusses how the grocery chain dramatically reduces costs through operational efficiency, including limited SKUs, private label dominance, direct farmer partnerships, and technological innovations like digital price labels and optimized barcode placement. The conversation covers Aldi's new Times Square location, supply chain strategies, and how consumer trends like GLP-1 adoption are reshaping grocery shopping behavior.
What Dan Wang Saw on His Last Trip to China
Dan Wang, author of "Breakneck" and China expert, discusses his recent month-long trip to China with podcast hosts Joe Weisenthal and Tracy Alloway, covering observations about phone culture, influencer culture, demographic challenges, and the tension between material abundance and widespread discontent among young people.
Carmen Li's Plan to Build a Futures Market for Compute
Carmen Lee, CEO of Compute Exchange and Silicon Data, discusses her efforts to build futures markets for GPU compute on the CME, drawing parallels to oil commodity markets. She explains how GPU price indices are constructed, who the natural buyers and sellers of compute futures would be, and addresses challenges like GPU performance variance and fungibility. The conversation covers spot markets, refurbished chips, residual value calculations, and the broader question of whether AI represents a bubble.