Notes — Gaurav Kapadia on New York City, Investing, and Contemporary Art

Notes on Gaurav Kapadia in conversation with Tyler Cowen — Conversations with Tyler (https://conversationswithtyler.com/episodes/gaurav-kapadia/), 10 December 2025.

Four questions [Adler frame]

Q1 — What is it about as a whole? A portrait of a single sensibility — judgment of quality — applied across four domains that rarely share a practitioner: the governance and geography of New York City, concentrated long-horizon investing, contemporary American art collecting, and craft. Kapadia, founder of the investment firm XN and a deliberately low-profile figure, treats all four as variations on one skill: developing your own taste rather than borrowing someone else’s, and holding a small number of high-conviction positions. The conversation moves from Queens housing density through XN’s ten-to-fifteen-position portfolio to why he collects living American artists his own age, closing on Totei, a new magazine celebrating craftsmanship.

Q2 — How is it argued? By reasoning from concrete cases rather than doctrine. Kapadia works from specifics — his parents ‘house-hacking’ a two-family home into a four-family one, XN’s Figma investment after the blocked Adobe acquisition, David Chang’s ‘good enough plus something special’ hiring test, a particular Kara Walker on the wall behind him — and generalises upward to claims about density, asymmetry, culture, and taste. Cowen presses binaries (Robert Moses hero or villain; is art an asset class); Kapadia repeatedly refuses the frame or reframes it around process and judgment.

Q3 — Is it true, in whole or part? Much of it is judgment and taste rather than checkable fact, and is offered as such. The investing claims (concentration, sector-exclusion, asymmetry) are internally coherent and match XN’s reported practice, though the conversation offers no performance data. The NYC-governance claims (that a few hundred taxpayers drive city revenue; that off-cycle June primaries produce low-turnout mayoral outcomes) are broadly consistent with common analyses but are asserted, not sourced here. [?] The art-canon predictions (Dana Schutz, Rudolf Stingel, Christopher Wool likely to endure) are explicitly flagged by Kapadia as educated guesses.

Q4 — What of it? A transferable model of connoisseurship: the same discipline — narrow the funnel by criteria, then ask ‘do I get it, do I love it, can I live with it’ — governs his art buying, his hiring, and his portfolio. It supplies the wiki with a concrete, practitioner account of concentrated investing that rhymes with Value Investing but is argued from asymmetry and sector-transparency rather than margin-of-safety alone, and a rare first-person view of how a major collector and museum trustee thinks about art as intellectual exercise rather than asset class.

Glossary

House hacking — buying a small multi-unit home, living in one unit and renting the others so tenants cover the mortgage; Kapadia’s immigrant parents did this in Flushing, converting a two-family home into a four-family one. [§ On Queens and NYC’s geography]

Concentrated investing — holding very few positions (XN runs 10–15 public holdings) because genuinely good ideas are rare, so capital is concentrated on the best rather than diversified across many. [§ On XN’s investment philosophy]

Asymmetry (risk-reward) — an investment where the downside is relatively bounded but the upside is large or open-ended and forecastable; Kapadia’s core test, illustrated by Figma bought at roughly half Adobe’s blocked offer price with a large net-cash balance sheet. [§ On XN’s investment philosophy]

Founder mode — the entrepreneurial, reinvention-driven energy of an owner-operator; Kapadia argues the best investment firms have it and most, facing low barriers to entry and industry pressure not to innovate, do not. [§ On founder energy in investment firms]

Deaccessioning — a museum selling works out of its own collection to raise funds; one of the fundraising pressures Kapadia names as museums run at operating deficits. [§ On how museums are run]

Deadweight of the résumé — Kapadia’s borrowed test (from chef David Chang): a candidate must be ‘good enough’ on credentials and then have ‘something special’ — an extra gear of curiosity or ingenuity — applied to every role at XN, non-investment staff included. [§ On founder energy in investment firms]

Totei — Kapadia’s new venture: a digital-and-physical magazine celebrating craft and craftsmanship broadly conceived (from quilt-making to stand-up comedy to design), on the premise that almost everyone respects dedication even if not everyone respects contemporary art. [§ On Totei, a new venture]

Key claims by section

On Queens and NYC’s geography [§ On Queens and NYC’s geography]

Queens prospered without much new infrastructure because New York overall did well and residents moved to the outer boroughs; existing transit (the 7 train) plus proximity to Manhattan drove development around transit hubs.
The single most valuable NYC intervention almost everywhere is building more housing; Flushing grew fastest when it added density by replacing past-their-prime single-family homes, and that has stalled.
His family’s two-to-four-family ‘house hack’ let ~15 people live in a home zoned for two — density experienced firsthand.
On borders, the Long Island–Queens line is unnatural, but NYC’s rivers and five-borough identity make it more coherent than sprawling cities.
New York is proportionally small relative to the US compared with, say, Vienna to Austria; Kapadia thinks the best outcome for America is a smaller-proportion New York achieved by strengthening the middle of the country, not by shrinking NYC.
Robert Moses: both hero and villain, but Kapadia would nudge people to like him less — his lack of due process is now showing up as an inability to build; the ideal is a balance, some planning for parks and highways, less one-man taste for a whole city. [?]
The claim that a few hundred taxpayers drive a large share of city revenue is broadly true, but NYC is more stable than cities reliant on one or two payers (New Jersey after David Tepper left; Bentonville without Walmart), and its strength is being no single group’s ‘company town’. [?]

On New York City mayors and electoral politics [§ On New York City mayors and electoral politics]

Most underrated mayor: Mike Bloomberg ‘by a country mile’ — low popular ratings, very high long-term impact that people wrongly assume would have happened anyway.
Talent avoids running because candidates assume they cannot win in so large a city; recent presidential and mayoral results have shown anyone can, which should draw more entrants (a ‘new Bayesian analysis’).
NYC’s peculiar mechanics suppress competition: an off-year election with a low-turnout third-week-of-June primary lets a candidate win with tens of thousands of votes; fixes are an on-cycle even-year election, a more normal primary date, and possibly an open primary.
The non-obvious quality he screens mayoral candidates for is optimism — extraordinarily scarce against pervasive doomerism, and the thing that inspires.
Most underrated part of NYC: the Bronx (cultural identity, cuisine, music).

On building a career in investing [§ On building a career in investing]

He thought he was far better at 23 than he was, but had an investor’s ‘DNA’ from childhood as de facto landlord of the family’s rental units — collecting rent, negotiating late payments — which taught ‘a great sense of humanity’.
Good mentors were less picked than attracted: genuine curiosity and visible passion draw people who like engaging with it. ‘I was very lucky that the mentors picked me.’
He chose BCG over Goldman and Blackstone at one-third the pay — a choice almost no Wharton student would make — to learn how corporations and leadership actually work, because investors wrongly assume ‘what’s on a spreadsheet is how organizations work’. A Goldman contact told him it was ‘the worst mistake of your career’.
He met Rishi Sunak around 2005–06 as an analyst covering the same railroad-renaissance companies at a parallel long-term-focused firm.

On XN’s investment philosophy [§ On XN’s investment philosophy]

XN does two things: concentrated public-markets investing (rarely more than 10–15 positions, ‘rhymes with what Buffett has done’) and opportunistic best-in-class private investing (about a third of capital).
It excludes whole sectors it cannot be best at — healthcare above all, because of the regulatory-plus-scientific-research combination — since the goal in everything is to be the best at what they do.
A sector is ‘transparent’ when business analysis, logic, and valuation can be tied together to reach differentiated, high-fidelity conclusions; the best investments are ‘obvious in retrospect’ but hard at the time — you can ‘write the narrative in advance’ (more power for data centres; short housing stock). Industrials and media lend themselves to this.
Figma was attractive not as an early bet but as an asymmetry: after antitrust blocked the Adobe deal, roughly half the intended price plus a break fee, a net-cash balance sheet, a founder-mode leader compounding earnings fast, and high confidence in growth and leadership.
Investing has more mistakes than wins; you must be comfortable with that. He loses no sleep over deals that soared for reasons he could not have predicted, only over passive misses — being close but not engaging through distraction or busyness. With so few positions, XN adds only one to three ideas a year, so the bar is very high and FOMO is fatal.

On founder energy in investment firms [§ On founder energy in investment firms]

Founder mode applies to investing firms: the best (Sequoia, Andreessen Horowitz, Blackstone) are highly entrepreneurial; low barriers to entry mean many mediocre firms lack it.
Complacency starts at the top — ‘the top gets lazy, so that culture seeps all the way through’ — so the antidote is to keep the rope tight, live a mantra of ‘rigor and kindness’, and surround yourself with people who will tell you the truth if you slow down.
Hiring: too many firms select purely on résumé, GPA, and modelling. Using David Chang’s test, a hire must be ‘good enough’ and then have ‘something special’ — extra curiosity, ingenuity, a spark — and XN applies this to every role, keeping the bar high for its ~40 non-investment staff too (the firm is under 50 people, 10–12 investors).
Open-plan office is deliberately good now because software, power, and semiconductors must be analysed together; the layout enables a Socratic, cross-pollinating method against a more connected world.
He did not ‘retire’ from investing but from managing outside capital: after Soroban he ran XN as a family office on the same philosophy, which restored ‘the spark and the purity of the intellectual challenge’ he had lost late at his prior firm.

On the sociology of finance in NYC, London, and the UAE [§ On the sociology of finance in NYC, London, and the UAE]

London finance is more buttoned-up and banker-like; New York (and California) carry more ‘swashbuckling investor energy’, a culture that seeps in even for those, like Kapadia, who do nothing like Paul Tudor Jones or Henry Kravis.
UAE finance is evolving fast: begun with sovereign wealth funds, now amplified by a very favourable tax regime (no income tax, no capital gains) drawing non-US talent, plus large regional wealth creation — high optimism and entrepreneurial energy.

On how AI reshapes investing [§ On how AI reshapes investing]

Cowen has been his ‘coach’ in embracing the maximal case for AI; realising it needs entrepreneurial energy to force through the ‘detritus’ of an SEC-regulated firm.
Two proximate effects: on investing, faster and deeper initial company analysis (much analyst time is spent there), freeing partners to apply judgment and taste to the facts; on running the organisation, big gains in legal, compliance, operations, and tax.
Custom tooling will change the analyst’s desktop, today just ‘Bloomberg, Excel’ that Julian Robertson would still recognise, dramatically within a year.
Most firms lack the wherewithal to change; founder-mode firms will thrive because industry pressures otherwise discourage innovation.
On J.P. Morgan reportedly spending ‘$2 billion to get $2 billion in savings’, his reaction is ‘nothing’ — many are in the ‘headline-generating business’ while XN is in the ‘results-generating business’; it is very early and he doubts institutions are yet spending what they claim.
Personal use: AI as a ‘knowledge augmenter’ and first port of call, replacing the tutors he used to hire (emailing a university art-history department for a masters student to build him a custom curriculum).

On how museums are run [§ On how museums are run]

Museums are ‘a public good that’s largely financed by private capital’; almost all run at an operating deficit, admissions are a small share of budget, and fundraising pressure has intensified (inflation, the recent administration).
The biggest mistake is letting fundraising override mission — letting donors (himself included) over-influence programming or community; institutions always regret it. Pressures point toward more deaccessioning and reliance on large donors.
Roughly 1 percent of a museum’s archive is ever on display — a tragedy; the opportunity is to bring collections out digitally and physically (he praises the V&A’s public storage site as ‘a museum unto itself’).
AI use case: personalised, any-language, in-depth tours via AR-plus-AI, giving context that crowds, small captions, and language barriers otherwise deny.
The Whitney’s underappreciated strength is its principled commitment to community and artistic freedom (free entry under 25) and tight adherence to its mission as a museum of American, largely contemporary, art. [He is a Whitney trustee and declines to be pressed on it.]

On favorite artists [§ On favorite artists]

What makes a specific Kara Walker stand out amid work that can feel generic: meticulous craft, clarity of point, and visual arresting-ness on a complicated subject — the ‘je ne sais quoi’ he keeps trying to refine.
He fell in love with art before investing and has collected since he was ‘broke’; getting serious at 29, he made a deliberate rule — collect American art by artists of his own generation (his age plus or minus 20 years) — and rejects the cliché of the finance guy who gets into art.
Rashid Johnson: visually arresting and intellectually complex work that ‘stays with you’; admired for refusing to coast (directing a play, Dutchman, in a sauna; making installations and films) and taking real risk.
Buying and evaluating art makes him a better investor ‘100 percent’ — not as an asset class (he rejects that outright) but as right-brain intellectual exercise that forces him to develop his own taste and judgment rather than borrow another’s; those soft skills transfer to all domains.
Salman Toor: technically among the most talented painters ever, painting entirely from memory with no reference image; personally meaningful to Kapadia as an Indian American — Toor’s Traveler series depicted post-9/11 TSA experiences of people who look like him.
His mother, not Jainism directly, is the source of his multi-domain curiosity: a data scientist who started a jewellery company, then at ~68 moved alone to a city in China to attend art school.
The ‘trade not made’ in art: Mark Grosjean fits his criteria and others love the work, but it never stuck — criteria only narrow the funnel; then ‘do I get it, do I love it, can I live with it’ decides.

On tastes in art and how the canon will evolve [§ On tastes in art and how the canon will evolve]

Every period looks unclear from inside; there is always a ‘Cambrian explosion’ that winnows to the important few, and the commercial and critical judgments are now coalescing around a smaller set of names than 10–15 years ago.
His educated guesses for the future canon: Dana Schutz (underrated today), Rudolf Stingel, Christopher Wool, and likely Rashid Johnson and Salman Toor.
On his favourite institutions: the (now closed) de la Cruz Collection in Miami over the new Rubell; ‘there’s a lot of bad art’ at any institution that shows a lot.
On art-market ‘arbitrage’ (textiles, drawings, Old Masters, ancient bronzes cheaper than contemporary paintings): the market is in a ‘mega reset’ shaking out mediocre work and galleries; but if art is an intellectual pursuit not an asset class, best-value questions matter less — it is fine to appreciate works you cannot afford. [?] Controversially, he thinks a group of overlooked white male artists holds significant undervalued importance.
Silicon Valley futurists who scorn contemporary art (and sometimes institutions and cities) are missing that not everyone must like everything; still, many dot-com founders (Bezos among them) have moved from low to very high art-market participation as they evolved.

On Totei, a new venture [§ On Totei, a new venture]

Totei is a magazine celebrating craft and craftsmanship broadly — quilt-making, stand-up comedy, design, anything demanding sustained effort to make something excellent — launched in digital form late 2025 and physical form early 2026 (physical two-to-three times a year).
The premise: almost everyone respects dedication and craft even if not everyone respects contemporary art, and there are few resources exploring ‘the art of getting really good at something’. His own craft is investing.
On the general public: a ‘positive inflection point’, not decline — pessimism is high (COVID aftereffects, economic haves-and-have-nots, turbulence) but there are large reasons for optimism, and ways to build rather than complain are entering the culture. AI’s proximate effects — more knowledge, fewer joyless tasks — he reads as optimistic against conference-circuit doomerism.
What he wants to learn next: the practical ‘lost art of making things work’ in local, state, and federal government — how the guts of policy get done — so a concerned citizen can push the ball forward (a callback to the Robert Moses conversation).