The Three Risks

A risk questionnaire measures one thing and gets treated as though it measured three. Here is the distinction, named so it can be argued with.

The claim

Tolerance, capacity, and behavior are three different things. The industry measures the first, calls the result a risk score, and treats it as though it answered all three.

It does not, and the gap between them is where plans actually go wrong.

One: tolerance is a disposition

Risk tolerance is how much volatility someone is willing to accept. It is captured by asking, in a calm room, with no money moving.

That is real information. It is also the easiest of the three to collect, which is why the industry standardized on it.

Two: capacity is arithmetic

Risk capacity is how much loss the plan can absorb before the plan stops working — time horizon, required withdrawals, fixed obligations, and how much the goal can move.

It has nothing to do with how anyone feels. Two people with identical questionnaire answers routinely have completely different capacity, and the questionnaire cannot see it.

Three: behavior is what actually happens

Behavior is what someone does in the decline rather than what they predicted they would do. It is the hardest of the three to measure and the one a questionnaire cannot see, because a strategy abandoned partway through has delivered nothing.

Tolerance is stated. Capacity is calculated. Behavior is revealed. A single score implies all three agree.

Why collapsing them is not a rounding error

The three can point in opposite directions, and the dangerous combination is specific: high tolerance, low capacity. You have clients like this: comfortable with a level of risk their plan cannot survive, and nothing in the conversation feels wrong until it is.

One number cannot express that, because one number cannot express a disagreement.

Why the category does it anyway

A single number is easy to show. It fits in a proposal, it produces an artifact for the file, and it makes the client conversation feel settled.

Those are real benefits. The mistake is treating the artifact as the analysis.

What to do instead

Keep the questionnaire, and stop asking it to do three jobs. Calculate capacity from the plan separately. Ask directly about conduct in previous declines, which is revealed rather than stated. Then reconcile the three in writing wherever they disagree.

The disagreement is where the advice is.

What this is not

The name is not a product; use it freely. riskDNA, which I co-founded, builds analytics on the premise. Planners have separated tolerance from capacity for years; what is usually missing is behavior, and a written reconciliation when the three disagree.

It is also not a prediction. Nothing here forecasts what any investor will do.

If you think this is wrong, the contact page is the route. I answer.