The Risk-Compensation Check
Assume every safety tool will be spent on taking bigger risks
- Difficulty
- Easy
- Time to result
- ~ongoing to results
- Steps
- 6
- Confidence
- 75%
Schrager's model is that risk reduction and risk expansion are the same lever pulled in opposite directions. In big-wave surfing, every innovation replays the argument: leashes meant a wipeout no longer cost you a twenty-mile swim, so mediocre swimmers could surf; jet skis meant tow-in access to eighty-foot waves, and someone who belonged on a five-foot wave could now attempt a twenty-foot one and endanger rescuers. The current fight is over inflatable vests, which do not always inflate but give people the sense that they can do anything. The same shape appears in finance, where a derivative introduced as insurance is used to lever up. The check is not to reject the tool — imperfect protection beats none — but to forecast the behaviour change and be explicit that no device makes an eighty-foot wave safe.
Origin
Extracted from Masters in Business — Allison Schrager's reporting from a big-wave surfers' risk conference, where she found the debate over inflatable vests as rigorous as any finance conference on systemic risk.
Core principles
- 01Anything that reduces risk can be flipped around to exacerbate it.
- 02Safety gear does not only protect the same behaviour — it licenses new behaviour.
- 03Financial tools designed as insurance get used to lever up.
- 04Nothing makes the world truly risk free; risk stays an estimate.
- 05Imperfect protection is still worth having, provided its limits are named.
How to run it
- 1
State the protective claim
Write down exactly which failure the tool prevents and in which conditions it has been shown to work.
Watch out Devices that work most of the time get treated as devices that work every time.
- 2
Forecast the new entrants
Ask who is now able or willing to attempt the activity who previously was not, and what their actual competence is.
Pro tip Look for the skill the tool quietly replaced — swimming, in the case of surf leashes.
- 3
Estimate the escalation, not just the substitution
Users typically do not keep the old behaviour with more margin; they spend the margin on a bigger attempt.
Watch out Levered use of a protective instrument can leave net exposure higher than before it existed.
- 4
Check the spillover onto others
Ask whose burden it becomes when the newly emboldened user gets into trouble — rescuers, counterparties, the wider system.
- 5
Choose the gate, not just the warning
Decide whether access should depend on demonstrated competence, or whether a life-saving tool should be available to all and the risk accepted.
Pro tip There is rarely a clean answer here — make the trade-off explicit rather than pretending it away.
Watch out Restricting a genuinely life-saving device carries its own cost.
- 6
Publish the limits
Say clearly what the tool does not make safe, so confidence is anchored to the real protection rather than the presence of gear.
In the wild
Schrager found the big-wave surf community re-running the same argument with every innovation. Before leashes, a wipeout meant losing your board and swimming as far as twenty miles, so only exceptional swimmers surfed; leashes let mediocre swimmers into the water. Jet skis, introduced by Brian Keaulana, could tow riders onto eighty-foot waves, but also let someone who belonged on a five-foot wave attempt a twenty-foot one and create rescue risk for others. Inflatable vests are the current battleground: they do not always inflate, yet they give people a sense that they are fine and can do anything.
→ A community that keeps its safety innovations but argues openly about who should be allowed to use them.
Ritholtz raises the plateau and even rise in annual automobile fatality rates and asks how much of it comes from confidence in airbags, crumple zones and anti-lock brakes leading people to drive faster and more dangerously than they would without those provisions. Schrager treats it as the case any book on risk has to include: once you hand people tools that reduce risk, you have to warn them not to feel so safe that they go and take whatever risks they want.
→ Safety equipment that improves survivability per crash while partly funding riskier driving.
Common mistakes
Scoring the tool in isolation
Measuring only the danger removed, with no adjustment for the behaviour it unlocks, systematically overstates the benefit.
Treating protection as elimination
Risk remains an estimate of something immeasurable, so no device converts a genuinely dangerous activity into a safe one.
Rejecting imperfect protection outright
Because a safeguard is not perfect does not mean you should skip it — the answer is awareness of its limits, not abstention.
Is it for you?
Best for
Anyone introducing a protective tool, product or policy into a system where users choose their own exposure.
Not ideal for
Passive protections in settings where users cannot vary their behaviour in response.
From the transcript
“Because anything that can reduce risk can also be flipped around to exacerbate risks.”
“you have to be mindful of not feeling so safe that you can then go and take whatever risks you want, because nothing makes the…”
“even if you have every safety device in the world and you're going to surf an eighty foot wave, it's still not going to be…”
From the episode
Allison Schrager Discusses Monetary Policy
Allison Schrager