When the Rule Is Unclear: How Ambiguity Resolves Itself | HL Hunt
When the Rule Is Unclear: How Ambiguity Resolves Itself
Most financial rules are clear. The minority that are not have a property the others lack: every firm resolves them separately and nothing makes the resolutions agree. A firm facing one compares a certain cost against an expected one, and in the case below the certain cost of caution is 8.8 times the expected cost of the permissive reading. Caution only pays above a 35.4% annual enforcement probability, which no firm believes it faces. That is arithmetic rather than character, which makes it a design problem.
What you'll learn
The core thesis
Most financial rules are clear. A meaningful minority are not, and the unclear ones have a property the clear ones do not: each firm resolves them separately, and nothing forces the resolutions to agree.
A firm facing an ambiguous requirement does not interpret it in isolation. It weighs the cost of the cautious reading against the expected cost of the permissive one, and those two numbers are usually not close.
Our position is that ambiguity systematically favours the permissive reading, that this is a predictable consequence of the arithmetic rather than of anyone's character, and that it is therefore a design problem rather than a compliance problem.
This is a description of how incentives operate, not a recommendation. The firms reading a rule cautiously are frequently right on the merits and are paying for it, which is precisely what makes the pattern worth naming.
Why the cautious reading is expensive
Take a firm with $4.2 million of revenue facing a requirement that can be read two ways. The cautious reading means declining a category of business; the permissive reading means continuing and carrying a contingent exposure.
- Cautious reading: forgoes 8% of addressable volume, which is $336,000 a year, with certainty
- Permissive reading: a $950,000 potential penalty at a 4% annual probability, which is $38,000 a year in expectation
That breakeven figure is the whole mechanism. A 35.4% annual chance of an enforcement action on one ambiguous practice is far above what any firm's experience suggests, so the arithmetic points the same direction for every firm that runs it, and the firms that do not run it are deciding on something other than cost.
Per our fixed cost analysis, the figures also scale badly. A smaller firm facing the same ambiguity has less revenue to forgo and less capacity to absorb a penalty, so the ratio moves against caution for the firms least able to carry the consequence either way.
Where ambiguity comes from
It is rarely carelessness. Ambiguity is usually produced by something that was reasonable at the time.
| Source | Why it produces ambiguity |
|---|---|
| Principles-based drafting | Deliberately general so it survives changes in practice, which means it does not resolve specific cases |
| Rules written for a different product | Applied by analogy to something that did not exist when they were drafted |
| Guidance that is not binding | Indicates a view without settling the question, per our licensing analysis |
| Overlapping authorities | Two regulators, two reasonable readings, no mechanism to reconcile them |
| Enforcement as the only clarification | The meaning is established by a case, which arrives years later and applies retrospectively |
| Definitions tied to form rather than function | Per our legibility analysis, a practice can be restructured to fall outside a definition without changing what it does |
The first row deserves defending rather than criticizing. Principles-based rules exist because prescriptive ones are gamed, go stale, and fail on cases nobody anticipated. The cost of that approach is precisely the ambiguity described here, and it is a genuine trade rather than a mistake.
The fifth row is the one that produces the worst dynamics. If meaning is settled only by enforcement, then the firm that gets the clarification is the firm that is penalized, and every other firm receives the benefit of the clarity for free. That makes being first to ask for a view a cost rather than a protection.
What firms actually do
- Watch what competitors do. Per our monoculture analysis, this converges the industry on a reading without anyone having validated it, and the resulting consensus is mistaken for guidance.
- Obtain an opinion. Useful, expensive, and per our exception analysis it documents that the question was asked, which cuts both ways.
- Ask the regulator informally. Frequently answered in terms that do not bind anyone.
- Adopt the reading and document the reasoning. The defensible route, and the one that distinguishes a considered position from an unexamined one.
- Avoid the question. Restructure so the rule plainly does not apply, which per our licensing analysis is a large part of why financial products have the shapes they do.
- Resolve it conservatively and absorb the cost. Which happens, and is the option this article's arithmetic penalizes.
Item one is the most common and the least sound. An industry consensus formed by firms watching each other is not evidence about the rule. It is evidence about what other firms concluded under the same incentives, and per our monoculture analysis correlated reasoning produces correlated exposure: if the consensus is wrong, it is wrong for everyone at once.
Item four is the practically important one and is worth separating from item one. A firm that read an ambiguous rule, reasoned about it, recorded why, and applied the reading consistently is in a materially different position from one that adopted the same reading because everyone else had, even though the behaviour is identical.
A falsifiable claim
Where a rule is ambiguous, the distribution of firm behaviour should be bimodal rather than spread, clustering at the cautious and permissive readings with few firms in between, and the permissive cluster should be larger. This is observable wherever practice can be surveyed: count firms by reading on a specific contested requirement. If behaviour is spread evenly, firms are reasoning independently and the convergence mechanism described here is weak. If it is bimodal with a larger permissive cluster, the arithmetic above is driving the outcome. A finding that the cautious cluster is larger would be the strongest evidence against this article, and would suggest non-financial considerations dominate the cost calculation.
Who bears it
- Cautious firms lose business to permissive ones. The $336,000 goes somewhere, and it goes to whoever read the rule differently.
- Customers face inconsistent treatment depending on which firm they approached, with no way to know which reading applies to them.
- The eventual clarification applies retrospectively to conduct that was unsettled at the time.
- Small firms are worst placed. They can afford neither the opinion nor the forgone revenue nor the penalty, per our fixed cost analysis.
- The rule's purpose is partly defeated. A protection that half the market reads as not applying protects half the market.
The second bullet is the consumer-facing cost and it is invisible to the consumer. Per our reference point analysis, a household cannot tell whether the treatment it received reflects a settled requirement or one firm's reading of an unsettled one, which means the variation looks like differences between products rather than differences in interpretation.
What reduces it
Mostly not available to individual firms, which is why this is a design problem.
- Binding guidance on request. A mechanism to obtain an answer that holds removes the asymmetry where asking creates risk.
- Safe harbours. A stated route that is definitively compliant makes caution cheap, which is the direct remedy for the arithmetic above.
- Prospective-only clarification where the ambiguity was genuine, which removes the retrospective penalty on firms that had to choose.
- Published interpretive positions, so firms are reading a regulator rather than each other.
- Firm-level: document the reasoning. Per our audit trail analysis, the one thing a firm controls is whether its reading was considered and recorded.
- Firm-level: review the reading when conditions change, since an interpretation adopted years ago under different facts is not a current position.
Item two is the highest-leverage change and the least used. A safe harbour does not resolve the ambiguity, it prices it: it makes the cautious route cheap enough that the arithmetic stops pointing one way. Per our rationing analysis, that is the general principle that a choice nobody can afford is not really a choice.
Scenarios and what we are watching
| Scenario | What we would expect |
|---|---|
| Enforcement activity increases | The breakeven probability moves toward reality and cautious readings become rational, which is the mechanism working as intended |
| Safe harbours expand | Behaviour converges on the harbour, which is clarity purchased at the cost of flexibility |
| A clarification lands retrospectively | The firms that read it permissively absorb a cost, and the firms that read it cautiously do not recover the revenue they forwent |
| Nothing changes | The permissive reading remains the commercially indicated one and the industry consensus continues to be mistaken for guidance |
The third row is the asymmetry worth sitting with. When an ambiguity is finally resolved against the permissive reading, the permissive firms pay and the cautious firms are vindicated without compensation for having been right for years. That asymmetry is the strongest argument for the safe harbour approach: it is the only remedy that makes being careful affordable at the time rather than correct in retrospect.
Frequently asked questions
Because caution costs $336,000 with certainty against $38,000 in expectation. Caution only pays above a 35.4% annual enforcement probability.
Mostly not. Principles-based drafting is deliberately general because prescriptive rules get gamed and go stale. The ambiguity is the cost of that trade.
A consensus formed by firms watching each other reflects shared incentives, not the rule. Correlated reasoning produces correlated exposure.
Safe harbours, which make the cautious route cheap rather than resolving the question. Then binding guidance on request and prospective-only clarification.
Key takeaways
- Ambiguous rules are resolved firm by firm, with nothing forcing the resolutions to agree.
- In the worked case, caution costs $336,000 a year with certainty against $38,000 in expectation for the permissive reading.
- The breakeven enforcement probability is 35.4%, far above what any firm’s experience suggests, so the arithmetic points one way for everyone.
- Principles-based drafting produces ambiguity deliberately, because prescriptive rules get gamed and go stale; it is a trade, not carelessness.
- When meaning is settled only by enforcement, the firm that gets clarity is the one penalized and everyone else benefits for free.
- Following industry consensus reflects shared incentives rather than the rule, so correlated reasoning produces correlated exposure.
- Customers face inconsistent treatment that looks like product differences rather than interpretive differences.
- Safe harbours are the highest-leverage remedy: they do not resolve the ambiguity, they make caution affordable.
This analysis is educational and is not legal, regulatory, or compliance advice, and nothing in it recommends adopting any particular reading of any requirement. All figures are stylized illustrations using assumed revenues, volumes, penalty amounts, and probabilities, and do not describe any particular firm, rule, or enforcement environment. The existence and scope of safe harbours, the availability of binding guidance, the retrospective effect of clarifications, and penalty exposure vary by jurisdiction, by regulator, and by product. Consult qualified regulatory counsel on any specific requirement.