The SEC's own definition puts both tests in one phrase: risk tolerance is your ability and willingness to lose some or all of your original investment in exchange for potentially greater returns. Willingness is a feeling measured by a questionnaire. Ability is arithmetic: up to six months of income already in cash, the date the money is needed, and a $3,000 annual cap on deducting a realized loss — and the arithmetic wins every time the two disagree.

Ability has published reference points

The reference points sit on the same investor education site. On cash: many savers keep enough to cover an emergency such as sudden unemployment, and some hold up to six months of their income so the money is there when it is needed. On time: money for a goal five years away or less belongs in something that does not have to be sold at a loss on a fixed date, while a 35-year horizon changes which products fit.

On taxes the constraint is asymmetric. Under IRS Topic 409 a realized net capital loss can be deducted against other income only up to $3,000 a year, or $1,500 for married filing separately, with the excess carried forward. A $9,000 realized net loss with no offsetting gains therefore takes three years to deduct in full, while a $9,000 gain is taxed in the year it is realized.

What a broker is required to collect

FINRA Rule 2111 tells a firm what to know before recommending anything. Paragraph (a) requires reasonable diligence on the customer's investment profile, and lists nine items: age, other investments, financial situation and needs, tax status, investment objectives, investment experience, investment time horizon, liquidity needs and risk tolerance. Risk tolerance is item nine of nine. Eight of the other items describe capacity.

The rule has been in force since 2012 and was last amended 30 June 2020; its Supplementary Material .08 states that it does not apply to recommendations covered by the SEC's Regulation Best Interest. Either way the list is the same list, and it is a workable self-audit even with no recommendation involved.

Worked example: a $30,000 account

Example figures, published rules. An account holds $30,000, with $10,000 in one position. The position falls 20 percent, a $2,000 loss, leaving $8,000. Getting back to $10,000 from $8,000 takes a 25 percent gain, not a 20 percent one, because the base changed; the guide on why a 20 percent loss needs a 25 percent recovery works the general case.

Now add capacity. If monthly income is $3,000, the six-month cash reference point is $18,000. An investor holding $30,000 in the market and $4,000 in cash fails that test before any position moves, so a 20 percent drawdown is not the risk — the risk is having to sell into it to cover a month of expenses. An investor with the $18,000 already set aside meets the same drawdown with a choice instead of a deadline.

The two questions separate cleanly

Willingness asks how a 20 percent fall feels. Ability asks three things with numbers in them: how many months of income are already in cash, what date the money is spoken for, and what a realized loss does to this year's tax return. A questionnaire answers the first. The second set is answered with a bank balance and a calendar.

Where the answer shows up in the portfolio

Capacity sets position size more often than it sets product choice. Concentration is where the two meet, because a share of the account large enough to matter is exactly the share that makes a sector move a household event; the guide on five holdings that are still one bet covers how to measure that share.

Drift moves the answer without a decision. The SEC's allocation page gives the case of a portfolio that starts at 60 percent stocks and reaches 80 percent on gains alone, with rebalancing at six or 12 month intervals or at a pre-set percentage. A 60-percent portfolio that has become an 80-percent portfolio has raised its risk without anyone's tolerance changing.

Recheck dates, not feelings

The inputs that change are the cash balance, the date the money is needed and the tax position; all three are checkable on a statement. Simulated results say nothing about any of them, which is the point of the guide on what a paper-trading result cannot prove.

Source references

Primary sources used for this guide. Follow the links for the original scope, definitions and conditions.