Nov 9, 2020 · 1h 7m · capital-allocators
Karyn Williams – Measuring Risk Practically at Hightree Advisors (Capital Allocators, EP.163)
0:00 / 0:00
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of Capital Allocators, host Ted Seides interviews Dr. Karyn Williams, founder of Hightree Advisors, exploring her journey across quantitative finance, the limitations of traditional mean-variance optimization, and the development of intuitive factor-based metrics designed to empower fiduciary decision-making.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Ted holds 17.3% of the talking time here. How this is scored →
Ted as informed peer 3.8 Guest teaching 5.2 Guest disagreement 0.5 Ted pushing back 0.5
speaking balance: gold is Ted, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 16:10 Dismantling mean-variance optimization orthodoxy
Karyn forcefully dismantles conventional industry reliance on standard deviation, pointing out that variance is almost completely disconnected from real-world balance sheet risk.
Hardest push from Ted ▶ 35:35 Questioning factor analysis in venture capitalTed directly challenges Karyn's quantitative framework by questioning how factor exposures can be mapped to venture capital funds characterized by power-law return distributions.
Biggest teaching moment ▶ 43:00 Educating on probability-weighted metrics Pi and EtaKaryn educates the audience and host on formulating Pi and Eta, explaining how to replace confusing statistical point estimates with intuitive probability weighting and dollar-denominated portfolio efficiency.
Ted holds their own ▶ 35:45 Drilling down on non-normal venture distribution realityTed leverages his deep private equity and asset allocation knowledge to highlight the mathematical mismatch between public factor modeling and private VC outcome dynamics.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Ted as informed peer | Guest teaching | Guest disagreement | Ted pushing back | Why |
|---|---|---|---|---|---|---|
| Early Career: Academic Roots and Financial Engineering Research | 3 | 5 | 0 | 0 | Ted prompts Karyn on her early academic research at Claremont Graduate University. Karyn explains her work analyzing high-frequency quote data and price discovery dynamics across large and small market participants. | |
| Wilshire Associates and Early Realizations on Risk Modeling | 4 | 6 | 1 | 0 | Ted asks Karyn what obvious flaws she observed upon transitioning from academia to Wilshire Associates. Karyn explains the disconnect between simplistic mean-variance optimization theory and institutional practice. | |
| The Flaws and Misapplication of Mean-Variance Optimization | 4 | 8 | 2 | 0 | Karyn delivers a comprehensive critique of standard mean-variance optimization, explaining why standard deviation fails as a real-world risk metric for balance sheets and boards. Ted listens as she explains sequence risk and fat tails. | |
| Unpacking Common Risk Factors Across Asset Classes | 3 | 6 | 1 | 0 | Ted asks about Wilshire's multi-asset risk models, and Karyn explains how heavy point-solution models for asset managers fail to serve allocators because they do not reflect true institutional purpose. | |
| Restructuring the Portfolio and Governance at Farmers Insurance | 4 | 5 | 0 | 0 | Ted inquires about Karyn's transition to Farmers Insurance and how she handled restructuring 50 balance sheets. Karyn details managing operational complexity and moving from single-manager concentration to a factor-based framework. | |
| Designing Factor-Based Portfolios and Commitment Matching | 4 | 6 | 0 | 0 | Karyn outlines the core set of macroeconomic and market factors used at Farmers to balance portfolio risk against insurance liabilities. She explains that factors remain uncorrelated during crises whereas traditional asset classes converge. | |
| Sponsor: Ridgeline | 6 | 6 | 1 | 4 | Ted pushes back on mapping venture capital portfolios into quantitative factor models given binary return distributions. Karyn concedes VC requires qualitative evaluation and fund-by-fund structural analysis. | |
| Time at Two Sigma and the Genesis of Hightree Advisors | 3 | 4 | 0 | 0 | Ted asks about Karyn's tenure at Two Sigma and the decision to launch Hightree Advisors. Karyn emphasizes the necessity of fiduciary independence when offering risk technology to asset allocators. | |
| Practical Risk Metrics: Formulating Pi and Eta | 4 | 8 | 1 | 0 | Karyn details the creation of Pi and Eta with Caltech researchers, translating abstract statistical risk metrics into intuitive probability-of-success metrics and dollar-value efficiency measurements. | |
| Regime-Switching Modeling and Practical Risk Diagnostics | 5 | 6 | 0 | 2 | Ted asks about the empirical reliability of the underlying mathematical models. Karyn explains their regime-switching methodology that explicitly accounts for normal versus crisis market states. | |
| Productizing Hightree: Board Governance and Industry Solutions | 4 | 5 | 0 | 0 | Ted asks about productizing Hightree and the competitive landscape with OCIOs and traditional consultants. Karyn describes enabling existing practitioners with diagnostic tools rather than building a direct competitor. | |
| Concluding Questions: Daily Habits, Gravel Biking, and Life Lessons | 2 | 1 | 0 | 0 | Ted and Karyn move through concluding standard closing questions, discussing gravel biking in Ojai, daily cappuccino routines, and lessons from her father about taking risks. | |
| Premium Subscriber Questions: Pet Peeves, Career Advice, and Reading | 3 | 2 | 0 | 1 | Ted conducts premium subscriber questions on career advice, pet peeves regarding backward-looking board discussions, physics reading, and potential future podcast guests like Richard Ennis. |
Statements from this episode (15)
Karyn Williams: Smaller banks drive FX price discovery, not large banks
“The analysis that I ran suggested that the prices were being updated by the smaller players, not the larger players. And that made sense to me because if you're going to put information in the market, move that market, you had a trade to make and somebody on t…”
Williams: Mean-variance optimization is a conceptual framework, not a direct tool
“Really surprised that the mean variance optimization framework was applied directly and simply. Like, even from early days, it didn't make sense to me that these concepts, really, the point was to say something about how risk should be perceived, that it's sys…”
Williams: Standard deviation rarely reflects the actual risks institutional investors face
“So the expected return being a point estimate May connect well into, I want to cover expenses, and I want to grow, and I want to pay fees, but the risk, meaning of variance and their standard deviation, only by some strange coincidence would be connected to th…”
Williams: Mean-variance optimization ignores fat-tail market drops and sequence risk
“For most mean variance optimization problems, you're throwing away the most important information. So think about making a decision about a loss limit over an investment horizon, and you have a real balance sheet to protect, but you're not actually taking into…”
Williams: Granular trading risk models are a bad fit for institutional allocators
“Well, that's helpful to someone who's trading A portfolio, say, of a number of stocks and bonds, multi-asset class, we brought those models to institutional investors, but that's not at all the right way to approach the problem. They have different purposes. S…”
Williams: Farmers Insurance had nearly 50 balance sheets and $1B in cash
“Balance sheet that had grown by acquisition. There was something like almost 50 balance sheets. And, you know, there's a billion dollars in cash.”
Williams: Institutions unable to measure risk default to taking less risk
“So, so if you can't really measure where you are in terms of risk, And that means you can't really manage it. What you would tend to do, and a lot of organizations tend to do, is err on the side of less risk.”
Williams: A single asset manager held 80% of Farmers Insurance's assets
“When I arrived and the team was much smaller than nine people managing the process and this manager Had 80% of the assets.”
Williams: Equity growth risk permeates high yield, real estate, and private equity
“The major driver of portfolio risk is growth and equity growth. But that exists in equities, and that exists in high yield, potentially, real estate, certainly private equity, and maybe even other investments. And so having a very clear picture of where that e…”
Williams: Farmers Insurance used six core risk factors for portfolio allocation
“So those included an equity factor, a duration factor, a credit factor. We had term, which was really representing inflation risk, commodity, and illiquidity. So very, very simple set.”
Williams: Asset classes correlate during a crisis, while risk factors do not
“If you shifted from an asset allocation approach to a factor approach, And you really wanted to get very clear on what the unique risks are, that if you ended up in a crisis, would tend to not be correlated. Asset classes are. Factors by construction are not.”
Williams: Institutional asset allocation is shifting toward factor-based risk models
“And I think that's largely where we're headed in, in the market for allocation and thinking about how the world of investment management evolves with technology and better information. That's one of the things that I would see coming is a refinement of the pro…”
Williams: Uncontrolled Macro Factors Frequently Swamp Active Managers' Bottom-Up Stock Selection
“And what we found generally was that a lot of times the factors would swamp the good work that they were doing on the individual positions in the portfolio. So all this great analysis and research Being put into individual decisions, but macroeconomic effects …”
Williams: Capturing VC Power-Law Home Runs Reliably Requires Roughly 500 Holdings
“You really have to hold an index, something like 500 securities in order to know that you're going to actually hit one of those .5 home runs.”
Williams: Investment committees waste time scrutinizing minor, backward-looking line items
“It really bothers me when the conversation veers toward the small and insignificant and backward looking. So think, Wow, this portfolio has, you know, these 15 different investments. Like, you delegated that decision, and that decision's been made.”