Dan Loeb: The Lost Art of Short Selling, and Why Stock Picking is Back
Dan Loeb, CEO and CIO of Third Point, discusses his evolution from anonymous internet troll and event-driven investor to running a multi-strategy $30B AUM hedge fund. He covers the resurgence of short selling, the importance of management quality, AI's impact on investing, and his philanthropic work in criminal justice reform including his role in securing Ross Ulbricht's pardon.
Summary
Dan Loeb opens by reflecting on his early days as an anonymous participant on internet chat boards like Yahoo Finance and Silicon Investor, describing himself as the 'original troll' who enjoyed uncovering fraudulent companies in the 1990s, including a company called Actrade that was repackaging factoring securities under a fake technology brand called TADS. He frames this period as genuinely formative for his short-selling instincts.
Loeb traces his investing education through a series of formative experiences: starting as a child investor at Payne Weber, learning enterprise valuation at Warburg Pincus, working in risk arbitrage, and then finding his real education at Jefferies on the distressed debt desk. He credits not just senior mentors but peers and clients — including Eric Mindich and David Tepper — as key influences, describing himself as a 'Chinese corporation reverse-engineering' the best practices of everyone around him.
Third Point began as a classic event-driven fund focused on complex transactions — spinoffs, bankruptcies, demutualizations, and risk arbitrage — where management sandbagging and information opacity created reliable alpha. Over time, as technology became more central to the economy, Loeb shifted toward a greater emphasis on business quality, moats, thematic investing, and macro awareness. He argues that being technologically or economically illiterate is no longer viable for any serious investor.
Loeb describes Third Point's current structure as a multi-strategy platform encompassing a long/short equity and credit hedge fund, a CLO business, private credit and direct lending, a credit solutions arm for workouts, and a wholly-owned insurance company that captures the investment-grade portion of their strategy. He emphasizes that all these businesses are interconnected.
On the topic of short selling, Loeb says he avoids purely valuation-based shorts due to the risk of meme-stock squeezes. He highlights a recent thesis on homebuilders as an example of a multi-factor short — combining structural issues (fake asset-light models with real land commitments) with post-COVID pricing dislocations and affordability pressures.
In a panel discussion with David Sacks and others, Loeb argues that NVIDIA is undervalued on a forward earnings basis and that the market is struggling to process the concept of a multi-trillion dollar dominant technology company, drawing parallels to how investors misjudged Google and Amazon as safe shorts in earlier eras.
Loeb also discusses the human element in investing, arguing that pattern recognition in assessing management teams remains entirely qualitative and subjective even after 30 years, and that human networks and interpersonal trust will remain irreplaceable even as AI agents proliferate.
On philanthropy, Loeb describes a framework centered on addressing income inequality through education reform — particularly his work with Success Academy charter schools — and criminal justice reform. He explains the three categories of unjust incarceration: false conviction, disproportionate sentencing, and cases warranting clemency due to rehabilitation. He details his multi-year effort, working with Charlie Kirk and attorney David Warrington, to secure the presidential pardon of Ross Ulbricht, founder of Silk Road, who had been sentenced to double life plus 40 years. Loeb is careful to share credit and acknowledges the complexity of Ulbricht's case while framing it as a clear example of disproportionate sentencing.
Key Insights
- Loeb argues that NVIDIA is undervalued on forward earnings and that the market's inability to conceptualize multi-trillion dollar companies leads to systematic undervaluation of dominant technology platforms, just as it did with Google and Amazon.
- Loeb claims that being technologically or economically illiterate was survivable as an investor before the GFC, but is now disqualifying given how deeply technology and macro factors permeate every asset class.
- Loeb describes his homebuilder short thesis as rooted not just in rates but in the structural deception of builders pretending to be asset-light like NVR while carrying deep commitments to land pools, compounded by unsustainable post-COVID cost inflation.
- Loeb argues that purely valuation-based shorts are dangerous because stocks with no rational valuation can still be squeezed by Reddit communities or speculative momentum, making catalyst and structural analysis essential to short selection.
- Loeb contends that assessing management quality remains entirely qualitative and subjective even after 30 years, relying on pattern recognition rather than any formalized rubric or quantitative framework.
- Loeb describes his philanthropic framework as targeting income inequality not by focusing on wealth concentration at the top, but by arguing the real crisis is failure to equip vulnerable children with intellectual tools — which he blames on union-protected adult interests overriding accountability and merit in public education.
- Loeb characterizes the Ross Ulbricht pardon effort as requiring a presidential pardon specifically because the justice system offered no internal recourse for someone with a life sentence, making political mobilization through figures like Charlie Kirk the only viable path.
- Loeb describes his early investing career as deliberately reverse-engineering the thought processes of clients and peers like David Tepper and Eric Mindich, framing mentorship as horizontal and peer-driven rather than exclusively hierarchical.
Topics
Full transcript available for MurmurCast members
Sign Up to Access