The Psychology of Investing w/ BlackRock’s Emily Haisley (RWH069)

ELI5 / TLDR

Emily Haisley runs the behavioral finance team at BlackRock. Her job is to sit with fund managers — privately, confidentially, with no power over their pay or promotion — and find the systematic ways they sabotage themselves. She does it two ways: by crunching their trading data to spot recurring mistakes (selling winners too soon, marrying losers, scaling into positions too timidly), and by reading their bodies (Oura rings) to catch the moment stress, not the market, starts driving their decisions. The throughline: the goal is to take yourself out of every decision and look only at the thing in front of you. Most of what wrecks an investor isn’t the market — it’s ego, pain, and the story they’re telling about themselves.

The Full Story

The job nobody else has

Haisley is a PhD psychologist who sits inside BlackRock’s Risk and Quantitative Analysis group — deliberately outside any portfolio manager’s reporting line. That structural detail is the whole point. Because she can’t affect anyone’s career, fund managers will actually tell her where it hurts. She calls it “a protected learning environment.” If she were their boss, they’d hide the very biases she’s there to find.

Her opening move with a manager is counterintuitive. If her analytics find no biases, that’s bad news.

Because if our analytics can’t find any biases, then we can’t help you. You’re kind of maybe like as good as you’re going to be in your process.

A measurable, systematic mistake is an opportunity — you can fix it. Random mistakes are just noise. So a clean bill of health means there’s no alpha left to unlock. The biases are the gold.

Loss aversion, and the three faces it wears

Most of the damage traces back to one root: loss aversion — the well-worn finding that a loss hurts roughly twice as much as the same-sized gain feels good. Haisley watches it surface in three recurring shapes.

Scaling in too slowly (myopic loss aversion). A manager forms a sound thesis but enters at a tiny size and creeps up. Sometimes that’s smart — buying into weakness. But often it’s fear: a new position feels risky simply because it’s new and unfamiliar, and change itself feels risky. There’s a subtle org-chart version too. The analyst who pitched the idea owns only a handful of positions, so any single loss is glaring and personal. The portfolio manager above them is diversified across dozens; for them the same risk is just one more sensible bet. The lower you sit in the pyramid, the more each individual risk feels like a referendum on you — so people undersize.

The CEO of a company wants everybody below them taking risk in line with their edge… But for that one individual taking the risk, they’re less diversified. Each risk is more of a reflection on them and feels riskier.

Disposition bias — selling winners, clinging to losers. Peter Lynch’s line: don’t cut the flowers and water the weeds. Haisley measures the probability a manager realizes a gain versus a loss, benchmarked. A higher propensity to bank gains and hold losers might be the disposition bias — but she only calls it a problem if it actually costs money: if the losers fail to mean-revert and the sold winners keep climbing. Her favourite tell isn’t in the data, it’s in the room:

The biggest clue also that someone’s doing this is if you start talking to them about it and you’re met with some emotion. You’re met with some sourness or some stress or some anger — that’s when, okay, we know that there’s some alpha here we can unlock.

Sunk cost (in private assets). In real estate, private equity, infrastructure — where you’ve spent months and real money on due diligence — walking away feels like a loss. You’ve grown “pregnant with the deal.” She calls this the number-one bias in private markets.

The fix is rarely “try harder”

Knowing about a bias doesn’t cure it. The host confesses he owns Alibaba — bought in 2021 because Charlie Munger and Lou Simpson loved it, now down a third, an irrelevant 1% position he can’t stop fretting over. Haisley’s diagnosis is clean: almost nothing in his reason for buying had anything to do with Alibaba. It was authority bias (smart people said so), tribal affiliation (these are my people), and ego (I have access to the smart money). The decision was about him, not the company.

Anytime you’re making a decision about markets where the reason why you’re doing it has something to do with you, then that’s when you know that there’s a problem.

So the antidotes are structural, not motivational. Defaults: the team agrees every new position starts at size X. You can deviate, but you must document why — a deliberate bit of “process sludge” that forces a real reason instead of a reflex. Kill criteria (an idea she credits to Annie Duke via the host): decide in advance what evidence would make you sell. And she flags the break-even effect — the urge to wait for a painful position to climb back to even and then dump it “to get it out of my sight.” That’s pain management masquerading as risk management, and she keeps the two separate. Are you cutting risk because the thesis broke, or because the ache got unbearable? Only the first is a real reason.

The deeper fix is having a sparring partner wired the opposite way. Haisley buys into weakness; her husband aggressively cuts losses and runs winners. Kahneman, she notes, appointed Richard Thaler as his “quitting coach” — someone licensed to tell him what he didn’t want to hear.

Teams: manufacturing disagreement on purpose

Half the conversation is about groups, because most institutional money is run by committees, and committees have their own pathologies. Groups gravitate toward shared information and things they already agree on, because agreement feels good and disagreement is genuinely unpleasant.

If a psychologist wants to do an anger manipulation, something they can do is find out what somebody believes and have them watch a video of somebody arguing the other side. Disagreement is just unpleasant. But the literature suggests that disagreement leads to objectively better decisions.

The mechanism is error cancellation: diversity and independence cancel out individual biases, which is how the wisdom of crowds actually works. So Haisley engineers disagreement back in:

Decision authority, not consensus. A lead PM decides, informed by the team. But everyone else’s explicit job is “to debias the decision maker — to attack, to challenge the decision maker.”
Quiet leaders. The best decision-makers speak last, so they don’t anchor everyone on their view. They practice “delayed judgment” — hear everyone out, then decide, then stay willing to reverse. This is Tetlock’s “active open-mindedness,” the defining trait of superforecasters: strong convictions, loosely held.
Collect votes before the meeting. Working with BlackRock real-estate head Thomas Mielke (who’d had a 100% deal-approval rate over seven years — a group of five men who never voted down a deal, never sold an asset at a loss), they introduced pre-meeting scoring. Each member rates the deal independently before discussion. The chair sees the spread, and can deliberately surface the dissenter’s view instead of letting momentum carry the room.
An outside challenger running a pre-mortem. Bring in someone not on the committee to argue the other side and imagine it’s three years later and the deal was a disaster — what went wrong? Making the attack a formal role makes it safe to challenge work the deal team has sweated over.
Blind, anonymous voting at the final stage. So nobody has to be the person who publicly “killed the deal.”
Cognitive diversity by design. The second person you add to a committee shouldn’t be the next-best expert — it should be someone who’ll make a different mistake. Two clones produce correlated errors. She extends this past gender to geography and, pointedly, politics: in a polarized age, having someone of a different political persuasion matters because politics quietly shapes what you forecast.

She also names a bias she calls the saddest one: tainted altruism — the assumption that anything doing good in the world shouldn’t or won’t make money. Look at sustainability through the lens of risk and opportunity, she argues, not tribal politics.

Reading the body, not just the brain

Here’s the genuinely unusual part. Haisley’s team links portfolio managers’ physiology — via Oura rings, voluntarily and confidentially — to what they do in their portfolios. The trigger was a study: dose people with cortisol for about a week and their risk preferences shift toward risk aversion. The worry isn’t market stress changing your risk-taking — that’s at least related to the job. The danger is your home life or office politics quietly dialing down the risk you take in markets, for reasons that have nothing to do with markets.

It’s not just their job to work hard, it’s also their job to rest hard.

Simply showing a manager that a long drawdown has kept their stress elevated — something they often swear they’d “moved past” — frequently breaks the spell and returns them to baseline. The reflex in volatile markets is to grind harder and stare at the screen longer; that’s exactly the cycle that bends judgment and ends in burnout.

On managing the stress itself she favours mindset shifts over hacks. Reframe the butterflies as fuel, not nerves (the study where girls told their exam jitters were energy did better). Remember that oxytocin is also a stress hormone — it pulls you toward people when you’re rattled, so resist the urge to withdraw and lean into the team instead; it’s also heart-protective. And treat a frightening market as a once-in-a-career learning seat rather than something to merely survive.

AI war-games and the gap between theory and the storm

A teammate built an AI simulation that loads the team’s real portfolio, throws news headlines at them (some signal, some noise — deciding which is the test), and lets them trade through invented volatility. It forces a team to get clear on its actual strategy — do we hold through volatility, provide liquidity, or manage dynamically? — and reveals that the calm “leader speaks last” model breaks under a ticking clock, where someone has to be directive. The host recalls Bill Miller discovering in 2008–09 that “almost nobody was really a value investor” when the pain got real. You can’t know what a drawdown does to your body until you’ve lived one. The simulation lets people make those mistakes — and feel the gallows humour — before real money is on the line.

Ego, the hero’s journey, and right livelihood

Underneath the technique sits a worldview. The common trait across every good investor she’s met is not being caught up in their own ego — being more interested in the market than in being right.

They want to understand what’s going to happen next in markets more than they want what they said yesterday to be true.

She frames careers through Joseph Campbell’s hero’s journey: the trials are the point, the imposter feeling means you’re still growing (“when you don’t feel like an impostor, that’s the problem”), and the climax is always a sacrifice of ego that turns out to be a false sacrifice — surrendering it is what unlocks you. Her uncle Robert, who lived his whole life expecting his weak heart to stop at any moment and so lived without fear, set her on this path and handed her Krishnamurti as a child. Asked how she squares Krishnamurti’s anti-ambition philosophy with a hyper-corporate world, she catches the host committing tainted altruism in real time: getting paid doesn’t mean you’re doing harm. Giving 35 million ordinary people risk-managed access to capital markets, she argues, is a real public good.

She closes on three New Year’s resolutions, all variations on the same move — creating distance from your own mind. Stop saying mean things to yourself (treat the inner critic as a habit: notice when and where it fires, and noticing is often enough to loosen its grip). Notice your breath, then deepen it to steer your own state. Be a better listener — to people, and to your own nervous system. The same skill she teaches investors: get a little emotional distance from the story so you can see the thing clearly.

Key Takeaways

Loss aversion — a loss hurts ~2x as much as an equal gain feels good; the root of most investing mistakes.
Myopic loss aversion — undersizing new positions because newness and change feel risky; worse for concentrated junior analysts whose every position is personal.
Disposition bias — banking gains too early, clinging to losers; only a real problem if it’s measurably costly (losers don’t recover, sold winners keep running).
Sunk cost bias — the #1 trap in private assets; due-diligence time and money make walking away feel like a loss.
Pain management vs risk management — before cutting, ask whether the thesis broke or the ache just got unbearable. Only the first is a valid reason.
Break-even effect — waiting for a painful position to reach even, then dumping it; a self-driven, not market-driven, decision.
Authority + tribal + ego biases — buying because admired people own it, because it’s “your tribe,” or because you have “access to smart money” — none of it about the asset itself.
Tainted altruism (“the saddest bias”) — assuming anything good for the world can’t make money.
Default position sizes + documented deviations — a nudge plus deliberate “process sludge” to force real reasons over reflexes.
Kill criteria — decide in advance what evidence makes you sell.
Opposite-wired sparring partner / quitting coach — pair with someone whose instincts cancel yours (Kahneman used Thaler).
Error cancellation — independence and diversity are what make the wisdom of crowds work; correlated clones add nothing.
Pre-meeting independent voting — score deals before discussion to kill anchoring and momentum.
Outside challenger + pre-mortem — a formal devil’s advocate imagining the deal already failed makes challenge safe.
Blind/anonymous final votes — remove the social cost of being the deal-killer.
Cognitive diversity by design — recruit for different mistakes (geography, politics), not the next-best expert.
Delayed judgment / active open-mindedness — leaders speak last; strong convictions, willing to update (Tetlock’s superforecasters).
Cortisol shifts risk preference — sustained stress biases toward risk aversion; dangerous when driven by life, not markets.
Rest hard, not just work hard — grinding through drawdowns deepens the bias and trends toward burnout.
Stress reframes — butterflies as fuel; oxytocin pulls you toward your team (lean in, don’t withdraw).
AI volatility war-games — rehearse decisions and mistakes before real money is at stake.
Take yourself out of the decision — the unifying rule: judge the asset, not what owning it says about you.

Claude’s Take

This is unusually good for the genre. Most “psychology of investing” content recites Kahneman and Tversky and stops. Haisley has spent a decade actually operationalizing it inside a $14T shop, so the value is in the mechanics — how you measure disposition bias against a benchmark, why you collect votes before a meeting, the documented-deviation trick — not the textbook list of biases. The cortisol-and-Oura work is the freshest thread and the most broadly applicable; the team-design material is the most actionable for anyone who decides in a group.

The honest caveats: it’s a friendly conversation between two friends, so there’s zero pushback. The Oura-ring stress-to-portfolio linkage is presented as more settled than the published evidence likely supports — n-of-one physiological correlations are noisy, and “showing someone their stress fixes it” is the kind of claim that needs a control group. There’s also unavoidable BlackRock PR varnish (the “sacred trust,” the ESG defense, the uniformly kind culture). And the host’s Alibaba confessional, while charming and genuinely illustrative, eats a chunk of runtime.

Still, the signal-to-noise is high and the central instruction — any time your reason for a decision is about you rather than the asset, stop — is worth the two hours on its own. An 8: rich, practical, rare access, lightly let down by the absence of any skeptic in the room.

The Psychology of Investing w/ BlackRock's Emily Haisley (RWH069)