Uncapped with Jack AltmanInside the Mind of a University Endowment Manager | Dan Feder, University of Michigan | Ep. 14
CHAPTERS
- 0:00 – 0:55
Rumsfeld’s “unknown unknowns” and why venture thrives on uncertainty
Dan opens with a framework separating what’s measurable (risk) from what’s fundamentally not knowable (uncertainty). He argues venture capital’s core edge is in navigating uncertainty, not optimizing known distributions. This sets up the episode’s central lens for evaluating venture within an endowment.
- •Distinguishes risk (knowable) vs uncertainty (not knowable)
- •Uses “known knowns/unknown unknowns” to frame venture’s domain
- •Claims venture’s durable advantage comes from uncertainty, not risk premia
- 0:55 – 3:34
From lawyer to endowment investor: Swensen, Princeton, and hiring for “fast processors”
Dan recounts how he stumbled into endowment management from a legal career through a chain of introductions (including Dave Swensen). At Princeton, he was hired to lead venture and private equity despite lacking the typical background. The experience shaped his views on hiring, judgment, and investing craft.
- •Career entry via chance lunch and a referral network
- •Introductions: Paul Levy → Swensen → Princeton’s Andy Golden
- •Key hiring heuristic: “fast processor vs full hard drive”
- •Early role leading venture/private equity informed his later philosophy
- 3:34 – 4:59
What an endowment is (and isn’t): pooled structure, spending needs, and perpetual objectives
Dan clarifies that a university endowment is a pool comprising thousands of separate endowments, akin to a mutual fund. The portfolio must support annual spending, preserve purchasing power against higher-ed inflation, and ideally grow beyond that—while smoothing volatility for intergenerational equity. These requirements drive an equity-oriented baseline.
- •Endowment pool made of many distinct endowments (UMich: ~13,000)
- •Core goals: 4–5% spending, inflation protection, and real growth
- •Higher-ed inflation exceeds CPI, raising the required return
- •Intergenerational equity implies dampening volatility over time
- 4:59 – 6:57
Zero-based endowment portfolio construction: public equities first, then diversification into illiquids
Starting from an all-liquid baseline, Dan explains why public equities are the logical foundation for a perpetual ~8% nominal target. Diversification (including semi/illiquid assets) is added to improve the return/risk profile using portfolio theory. Endowments’ modest liquidity demands enable them to hold illiquid exposures.
- •Fully liquid public equities as the starting point for return targets
- •Use of diversification to reduce volatility per unit of return
- •Mean-variance concepts guide high-level mix decisions
- •Low annual liquidity need (spending rate) supports illiquid allocations
- 6:57 – 10:16
Allocation vs investing: why asset allocation is risk-based—and behaviorally protective
Dan separates asset allocation (a largely risk/volatility/correlation-driven exercise) from investing (selecting specific opportunities). Allocation, while imperfect, serves as a behavioral guardrail against overconfidence and trend-chasing. It helps constrain how much an institution can overcommit to an area like venture.
- •Allocation relies on historical return streams, volatilities, correlations
- •Investing is selection within buckets; allocation is portfolio-level design
- •Risk-based frameworks can prevent behavioral overreach
- •Model portfolio as an “outer space” baseline for decision discipline
- 10:16 – 16:42
Risk vs uncertainty in venture: Frank Knight, “adventure capital,” and durable economic profit
Dan argues venture selection should not be driven by risk models, because venture’s advantage sits in uncertainty where information is asymmetric and outcomes are not knowable. He proposes reframing venture into “adventure capital” (true uncertainty/ambition) versus “capital for ventures” (more conventional financing). This also explains why stage labels (seed/late) can miss the real point: adventurousness can exist at any stage.
- •Frank Knight’s distinction: measurable risk vs true uncertainty
- •Unknown-unknown opportunities can create durable alpha
- •Venture’s power comes from backing founders with unique information/insight
- •Reframe: “adventure capital” vs routine venture financing
- •Stage is less important than whether the work is genuinely adventurous
- 16:42 – 22:26
LP incentives and career paths: outcome metrics, narrative gravity, and long-horizon “input” work
Dan describes how endowment management evolved from a mission-aligned niche into a more standardized career path—bringing pressure to show near-term results. That shift can move people from input-quality (learning, networks, patient theses) to outcome-chasing (visible marks and quick validation). He explains how his approach emphasizes slow compounding of relationships and ideas over years.
- •Endowment roles used to be mission-driven; now there’s a clearer career ladder
- •Career incentives increase focus on short-term visible outcomes
- •Long-horizon edge comes from “inputs”: listening, learning, building networks
- •VC narratives can encourage outcome/mark-driven behavior
- •Metaphor: playing “cards” (build a hand) vs a “craps table” dynamic
- 22:26 – 26:05
Backing conflicting strategies on purpose: concentrated relationships and complementary return drivers
Jack probes the cognitive dissonance of funding managers with opposing worldviews. Dan explains the portfolio logic: you don’t need to be right about everything, but you must be right enough in the areas that matter. Michigan keeps a deliberately concentrated manager roster, aiming to pick a distinct “best” exposure per niche rather than owning many similar firms.
- •LPs must hold multiple mutually inconsistent manager worldviews
- •Portfolio resilience can require complementary, not identical, strategies
- •Michigan approach: intentionally constrained manager count
- •Goal: avoid owning many lookalike funds; try to pick “the one” per area
- •Concentration forces harder choices and clearer differentiation
- 26:05 – 28:03
Why LPs invest in venture at all: obligation, overconfidence, and the lure of the few companies that matter
Dan bluntly notes many institutions invest in venture partly because they feel they’re “supposed to.” He adds that most people believe they’re above average at selection, despite wide dispersion of outcomes and shrinking edge. The asset class is only great if you access the small set of exceptional underlying companies.
- •Cultural/competitive pressure: endowments feel they must allocate to venture
- •Above-average bias is common among LPs and GPs
- •Venture is unattractive unless you reach top companies/drivers of returns
- •GP storytelling and LP appetite for narratives reinforce flows
- 28:03 – 32:29
Michigan’s endowment framework: five buckets—and a shift from “allocator comfort” to institution-specific advantage
Dan outlines Michigan’s conventional top-level structure (cash/fixed income, public equities, absolute return, real assets, VC/PE). He argues the old playbook—being a pure allocator with privileged access—won’t keep working as markets and LP offices professionalize. Michigan therefore looks for investments that align with its specific advantages: access, time horizon, and occasional ability to influence outcomes.
- •Five endowment buckets, with VC/PE as a distinct sleeve
- •Alternatives are no longer “alternative”: capital and sophistication increased
- •LP offices are better staffed; privileged access is harder to sustain
- •Michigan’s three edges: access, long time horizon, selective influence
- •Conclusion: no universal recipe; each institution must play to its positioning
- 32:29 – 36:39
People vs platforms: getting fidelity to the best investors (and reconsidering firm-level exposure)
Dan frames manager investing as a means to an end: efficient exposure to great underlying companies. He discusses the difficulty of disentangling an individual’s edge from a firm’s brand, resources, and deal flow. He also shares an evolving view: sometimes it’s better to back smaller teams or solo GPs to increase “fidelity” to the specific investors who drive a firm’s performance.
- •True objective: exposure to underlying companies, not “owning” manager brands
- •Firm context can amplify (or mask) individual skill
- •Challenge: where the “seat” ends and the person begins
- •Evolving view: consider more small teams to avoid paying for non-core partner productivity
- •Tradeoff: firm breadth vs concentrated exposure to standout individuals
- 36:39 – 40:48
Big vs small funds: size as symptom, strategy fit, and the value of being “in the conversation”
Dan rejects simplistic takes that fund size is always bad, arguing size reflects strategy and can be too small or too big. Undersized funds may be unable to lead or maintain exposure; oversized funds can dilute quality and drift from mandate. He emphasizes the LP’s role as an engaged partner—supporting uncomfortable but correct strategic choices, not dictating them.
- •Fund size can be problematic at both extremes
- •Size is an outcome of strategy; sometimes ambition requires more capital
- •Too big can dilute deal quality/fidelity; too small can limit meaningful exposure
- •Ability to lead rounds can matter for both founders and returns
- •LP value-add: engaged dialogue and support for hard-but-right decisions
- 40:48 – 48:25
How Michigan picks managers—and how herd dynamics distort LP decisions
Dan explains manager selection as a portfolio-fit problem first: even great opportunities may be redundant. Sourcing is primarily via trusted introductions, and diligence centers on whether a manager can source, execute, own, exit, and run the firm without distracting from performance. He critiques “taste” as an overused shortcut, then closes with thoughts on herd behavior in LP land—where stability hides weak decision-making until stress exposes it.
- •Selling to LPs requires understanding what the “buyer” (portfolio) needs
- •Sourcing via qualified referrals; don’t try to “see the whole market”
- •Diligence framework: source, transact, own, exit, and firm operations
- •Skepticism of ‘taste’ as a substitute for slow, deliberate thinking
- •Herd mentality: can work in calm periods but breaks under stress; aim for independence
