The last couple of weeks have been a rollercoaster for markets, particularly technology stocks, where excessive risk-taking has compounded an already eventful earnings season. The volatility appeared to accelerate with Alphabet and Meta’s second-quarter results, as investors questioned whether the company’s enormous capital expenditure programme would generate sufficient returns after free cash flow turned negative. Combined with stretched valuations across much of the technology supply chain, the result was a sharp sell-off.
Since then, sentiment has improved, with the Nasdaq recovering most of its losses and trading within a couple of percentage points of the all-time highs reached in early June. However, as we will hopefully demonstrate, the recovery was far from straightforward and did not benefit all market participants equally.
This episode reinforces our view that markets are becoming increasingly detached from fundamentals, with short-term price movements driven more by emotion and momentum than underlying business performance. Warren Buffett made a similar observation at Berkshire Hathaway’s shareholder meeting in May, describing today’s stock market as “a church with a casino attached” and remarking that “we’ve never had people in a more gambling mood than now”.
The common thread running through each of the following examples is not simply speculation, but what happens when investors take so much risk that they are unable to stay invested when markets inevitably become volatile.
Roundhill Memory ETF (DRAM):
One of the clearest examples of this increasingly casino-like market behaviour has been the meteoric rise of the Roundhill Memory ETF (DRAM). The fund is an actively managed thematic ETF that follows a rules-based methodology, rather than traditional discretionary stock selection, and does not seek to replicate a benchmark.
Launched on 1 April 2026, the ETF amassed more than $20 billion of assets under management (AUM) by the end of June, just three months after launch. While some of this growth reflected exceptionally strong investment performance, attracting that level of capital from a standing start is extraordinary. Assets have since peaked at approximately $27 billion.
As Ranmore Global Equity fund manager Sean Peche observed, most fund managers spend an entire career trying to raise this much capital (and very few ever do). The speed and scale of DRAM’s asset gathering is therefore remarkable.
For further context, DRAM is now around six times larger than the Roundhill Magnificent Seven ETF (MAGS), which has approximately $4 billion of assets under management despite launching three years earlier, on 10 April 2023. While a single fund is not representative of the market as a whole, we believe it is another piece of evidence suggesting investors are becoming increasingly willing to take concentrated risks.
The portfolio itself reinforces that view. Despite holding just 13 stocks, the top five positions account for 82% of assets, including weightings of approximately 24% in Samsung Electronics and 21% in SK Hynix. We will return to these holdings later.
Investors are also paying 0.65% per annum for this exposure, more akin to the fees charged by traditional active managers than a rules-based ETF, and more than double the 0.30% charged for Roundhill’s own Magnificent Seven ETF.
Source: Morningstar, Cumulative Return in GBP, 01/04/2026-05/08/2026.
A review of Reddit discussion boards (not a source we routinely cite!) suggests that, for some investors, simply owning the ETF was not enough. Encouraged by its rapid price appreciation, many were instead discussing ways to amplify returns through options, introducing an additional layer of leverage.
This anecdotal evidence is supported by Koen Hoorelbeke, Investment and Options Strategist at Saxo, who highlighted that DRAM had already accumulated millions of options contracts in open interest despite being less than four months old.
In just four months, around $25 billion has flowed into a highly concentrated portfolio focused on a single segment of the market, with at least some investors then choosing to apply further leverage through derivatives. Taken together, it is difficult to argue that today’s market is not exhibiting increasingly speculative behaviour.
KOSPI Crash – Korean Single-Stock Leveraged ETFs
The second piece of the puzzle comes from the experience of South Korean retail investors. In April 2026, South Korean regulators approved the launch of domestic single-stock leveraged ETFs, partly to discourage retail investors from accessing similar products overseas.
The first two products, based on Samsung Electronics and SK Hynix, the two largest companies in the Korean market, began trading in May. Retail participation quickly rose, increasing exposure to the risks inherent in leveraged investing. These companies also represent the two largest holdings in Roundhill’s Memory ETF (DRAM).
Source: Morningstar, Cumulative Return in GBP, 26/05/2026-05/08/2026.
By June, the KOSPI had reached record highs, fuelled by enthusiasm for the AI memory chip theme. Samsung Electronics and SK Hynix together accounted for almost half of the index, a reminder that market concentration is not solely a US phenomenon. However, exceptionally high expectations, particularly for SK Hynix, combined with weaker-than-expected earnings and more broadly concerns over hidden debt, record capital expenditure, circular financing and increasingly stretched valuations, triggered a sharp reversal.
For many leveraged retail investors, the consequences were severe. As prices fell, margin calls forced positions to be liquidated, locking in losses. It is estimated that around 3.4% of South Korea’s adult population received a margin call, with retail investor losses estimated at up to US$39 billion.
Perhaps the cruellest aspect was what happened next. Following the wave of forced selling, markets recovered strongly, meaning only those who remained invested participated in the rebound. While we cannot know the extent of leverage being employed globally, the episode serves as a powerful reminder of how concentrated positions and leverage can combine to produce permanently impaired capital, even when the underlying investment thesis ultimately proves correct.
Situational Awareness
The highest-profile example of excessive risk-taking during this period has been the collapse of Leopold Aschenbrenner’s Situational Awareness hedge fund.
Aschenbrenner published his now well-known Situational Awareness essay in late 2024. Its central thesis was that Artificial General Intelligence (AGI) would emerge by around 2027, driving unprecedented demand for computing infrastructure, power generation and semiconductor chips.
Less than two years later, he launched a hedge fund with approximately US$225 million of seed capital to invest directly in that thesis. Initially, the results were extraordinary. His views on AI infrastructure proved correct and, amplified by gross leverage of around 400%, meaning every dollar of investor capital controlled approximately four dollars of market exposure, the fund’s assets under management grew to almost US$45 billion by 1 July 2026.
However, being right about the long-term trend was not enough. As markets corrected, the leverage that had magnified gains worked equally powerfully in reverse. The fund suffered a drawdown of almost 80% before margin calls forced the liquidation of its public equity portfolio, which was subsequently acquired by Ken Griffin’s Citadel in a discounted block trade.
The parallels with the experience of South Korean retail investors are striking. In both cases, the underlying investment thesis may ultimately prove correct, but excessive leverage prevented investors from remaining invested long enough to benefit from the subsequent recovery. While early investors still generated attractive returns, they fell well short of what would have been achieved had the positions not been forced into liquidation.
What this means for Tideway investors:
Markets are increasingly driven by sentiment. Charlie Munger famously warned that “There are only three ways a smart person can go broke: liquor, ladies, and leverage.” Recent events have provided a timely reminder. While leveraged investors, including South Korean retail investors and the Situational Awareness strategy, were forced out of the market, those who remained invested have participated in the recovery.
The ability to stay invested is invaluable. Tideway portfolios do not employ leverage, meaning our investors are not exposed to forced selling through margin calls. The greater risk for long-term investors is selling at the wrong time, whether because withdrawals coincide with market weakness (sequencing risk) or because volatility becomes too uncomfortable to endure. This is why we combine robust portfolio construction with proactive financial planning to help clients remain invested through market cycles.
Diversification is a form of risk management. No one knows how markets will behave over the short term, particularly in today’s increasingly speculative environment. By constructing genuinely diversified portfolios, we seek to minimise concentrated risks and give investors the best chance of remaining invested through periods of volatility, avoiding difficult decisions to sell assets at depressed prices to meet spending needs.
Tideway Equity Positioning
We have no doubt about the transformative potential of artificial intelligence. Where we differ from the wider market is in our confidence around identifying the long-term winners. By avoiding the most crowded and highly valued areas of the market, we believe we can reduce the risk of an extended drawdown should expectations become overly optimistic.
Source: Morningstar, Tideway, Data range 30 April 2026-31 July 2026.
We believe global equity indices currently contain a significant exposure to these risks. However, we are not seeking to avoid the AI theme entirely. Through exposure to both the “picks and shovels” companies enabling AI development and businesses benefiting from the integration of AI into everyday processes, we believe our portfolios have sufficient exposure to participate in the upside while avoiding a level of concentration that could materially impair long-term returns during periods of volatility. We believe this represents an attractive balance.
Source: Morningstar, Cumulative Return in GBP, 30/06/2026-05/08/2026.
Importantly, opportunities exist beyond the most popular areas of the market. Tideway’s approach of allocating to fund managers focused on valuations, fundamentals and a margin of safety continues to deliver. Since the end of the second quarter, our core deep value managers, Schroder Global Recovery and Redwheel Global Intrinsic Value, have performed particularly strongly. James also highlighted the longer-term performance of both strategies in our previous market update and more recently on LinkedIn.
