A precedent is a past decision or outcome that is used to guide or justify a current one. In law, precedents (case law) bind or persuade future courts: what was decided before shapes what can be decided now. The logic extends beyond courts. Organisations have precedents in how they have handled similar situations. Negotiations set precedents for what is acceptable. Personal and institutional behaviour creates patterns that others — and future selves — will cite. Precedents reduce the need to re-decide from scratch; they also constrain deviation. Every decision you make is a potential precedent for the next.
The force of precedent comes from consistency and predictability. When past behaviour is known, it creates expectations. Breaking precedent without explanation or justification costs trust and invites the question: why was that case different? So precedents are both a resource (you can point to the past to support your position) and a constraint (you must either follow or distinguish). Distinguishing — arguing that this case is different in a principled way — is the art of managing precedent without being trapped by it. Bad use of precedent is cargo-culting: doing something because it was done before, without asking whether the prior context still applies. Good use is pattern-matching with a check: does this situation match the prior one in the ways that justified the prior outcome?
Strategic implications: (1) the first time you do something, you are setting a precedent — choose with that in mind; (2) when you want to change course, you must either distinguish the new case or explicitly overrule the old precedent and explain why; (3) in negotiations and policy, the precedent you set in one deal or ruling will be cited in the next. Precedents compound. Design them deliberately.
Section 2
How to See It
Precedents are at work when someone says "we did X last time" or "that would set a bad precedent" or "there's no precedent for that." Look for references to past decisions in meetings, contracts, and strategy. The absence of precedent is also a signal: when something is truly novel, the lack of prior case can be an argument for caution or for freedom.
Business
You're seeing Precedents when a company refuses a customer exception ("if we do it for you, we have to do it for everyone") or grants one and then faces repeated requests citing "last time." The first decision created a precedent. Pricing, refunds, and contract terms all set precedents that shape future negotiations. Smart operators know which precedents they are setting and which they are willing to break — and how to distinguish.
Law & Policy
You're seeing Precedents when a court cites prior cases to support or limit a ruling. Stare decisis — standing by what was decided — is the doctrine that gives precedents binding or persuasive force. Lawyers argue that the facts fit or do not fit the prior pattern; courts either follow, distinguish, or overrule. The body of precedent defines the playing field for future disputes.
Investing
You're seeing Precedents when a valuation or term sheet is justified by "comparable" deals. Those comparables are precedents. They anchor expectations and constrain deviation. When you set a new precedent — a new high or low for a sector — you shift the reference class for the next round. Precedent is why first deals in a new category matter so much.
Organisations
You're seeing Precedents when how the company handled one layoff, one promotion, or one conflict is cited in the next. "We promoted X without a formal process" becomes a precedent that weakens the next "we need a process" argument. Leaders set precedents with every visible decision. The question is whether they are intentional.
Section 3
How to Use It
Decision filter
"Before deciding, ask: what precedent does this set? Is that the precedent I want for future cases? If not, either choose differently or be ready to distinguish or overrule later with a clear rationale. When citing precedent, check that the prior case is analogous in the ways that matter — not just superficially similar."
As a founder
Every first — first hire, first fire, first customer exception, first investor term — sets a precedent. Grant a one-off discount and you have set a precedent that discounts are negotiable. Enforce a boundary once and you have set a precedent that the boundary is real. Choose early decisions with the next ten in mind. When you need to break precedent, do it explicitly: "we did X last time; this time we're doing Y because [reason]." That preserves the norm while allowing change.
As an investor
Term sheets and side letters set precedents for the fund and the market. The first time you accept a term, you make it harder to refuse next time. The first time you refuse, you set a ceiling. Portfolio companies and co-investors will cite your past behaviour. Be consistent or be clear when you are changing course — and why. Precedent is part of your reputation.
As a decision-maker
Use precedent to justify and to constrain. When you want to say no, "we don't do that; we never have" is a reason — if true and if you want to keep the precedent. When you want to say yes in a novel case, consider whether you are creating a precedent you can live with. Reference class forecasting uses past cases to calibrate; precedent is the legal and organisational version. Match the precedent to the situation; do not confuse correlation with relevance.
Common misapplication: Treating precedent as destiny. "We always have" is a reason to continue only if the reasons for the prior choice still hold. Contexts change. Precedents can be distinguished (this case is different) or overruled (that precedent was wrong or obsolete). The mistake is following precedent without re-checking fit.
Second misapplication: Ignoring the precedent you set. One-off decisions are rarely one-off in effect. The exception you grant will be cited. The boundary you relax will be tested again. If you are not ready for the precedent, do not set it. If you set it by accident, correct it explicitly before it compounds.
Buffett is known for consistency in valuation, governance, and communication. He sets precedents deliberately: Berkshire does not do certain things (hostile takeovers, earnings guidance) and does others (transparent letters, long-term holding). When he breaks precedent — as with certain investments or governance choices — he explains why. The precedent structure is part of his reputation: counterparties know what to expect. That reduces transaction costs and builds trust.
Charlie MungerVice Chairman, Berkshire Hathaway; 1924–2023
Munger emphasised learning from the best prior thinking — "elementary, worldly wisdom" from multiple disciplines. His use of mental models is a form of precedent: apply the same reasoning patterns that worked in the past to new situations. He also warned against slavish precedent: "We have a bias toward consistency. But we're willing to change when the facts change." Precedent as guide, not prison.
Section 6
Visual Explanation
Precedent is a chain. Each decision is a link. Future decisions can (1) follow the chain — same outcome, same reasoning; (2) branch — "this case is different" (distinguish); or (3) break the chain — "we were wrong then" (overrule). The chain gives structure and predictability. The ability to distinguish or overrule gives flexibility. The art is knowing when to follow and when to break — and always having a reason.
Section 7
Connected Models
Precedents sit at the intersection of path dependence, consistency, and learning from history. The models below either explain how past decisions shape future ones (path dependence), how to use past data for calibration (reference class forecasting), or how to avoid over-weighting the past (status quo bias).
Reinforces
Path Dependence
Path dependence means current options are shaped by prior choices. Precedent is path dependence in decision-making: what you did before narrows or expands what you can do next. Both suggest that early decisions have outsized impact — set precedents and paths deliberately.
Reinforces
Lindy Effect
The Lindy effect: the longer something has lasted, the longer it is likely to last. Old precedents that have survived many challenges may be more robust than new ones. When considering whether to follow or overrule, the age and durability of the precedent is information.
Leads-to
Historic Recurrence
History does not repeat exactly, but patterns recur. Precedents are the institutional form: we store past decisions and apply them to new situations that match the pattern. Historic recurrence warns that the match may be imperfect; precedent requires checking that the analogy holds.
Reinforces
Reference Class Forecasting
Reference class forecasting uses outcomes from similar past cases to calibrate predictions. Precedent is the same idea in law and policy: find comparable cases, see how they were decided, apply or distinguish. Both require choosing the right reference class — the right set of precedents.
Section 8
One Key Quote
"Stare decisis is usually the wise policy, because in most matters it is more important that the applicable rule of law be settled than that it be settled right."
— Justice Louis Brandeis, Burnet v. Coronado Oil & Gas Co. (1932)
Brandeis captures the trade-off: precedent favours settlement and predictability over perfect correctness in each case. Sometimes the cost of changing a rule is higher than the cost of living with an imperfect one. The implication is that overruling precedent should be rare and justified — but when the rule is wrong enough, or the world has changed enough, overruling is the right move. Precedent is the default; it is not absolute.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
You are always setting a precedent. The first time you make a decision in a category, you create the reference for the next. Treat first decisions as precedent-setting: hiring, firing, pricing, terms, exceptions. Ask what the next ten cases will look like if you do this. If you do not like that future, choose differently now.
Cite precedent when you want to constrain or justify. "We have never done X" is a reason — if it is true and if you want to keep that norm. "We did Y in situation Z" is a justification when the situations are analogous. The key is analogy: the prior case must be similar in the ways that justified the prior outcome. Lazy precedent-citing is cargo-culting.
Break precedent explicitly. When you need to change course, do not pretend nothing happened. Say "we did X before; we are doing Y now because [reason]." That preserves the norm of consistency while allowing change. Silent deviation erodes trust and creates confusion about what the rule actually is.
Precedents compound in negotiations. The first deal in a relationship sets the baseline for the next. Concessions you make once will be expected again. Boundaries you hold once will be tested again. Negotiate with the precedent in mind — and know which terms you are willing to set as the new normal.
Distinguish or overrule; do not ignore. When precedent points one way and you want to go another, you have two moves: distinguish (this case is different in a principled way) or overrule (that precedent was wrong or no longer applies). Ignoring precedent without explanation invites the charge that you are arbitrary. Explaining your move preserves the system.
Section 10
Summary
Precedents are past decisions used to guide or justify current ones. They provide consistency and reduce re-decision cost; they also constrain deviation. Use them by asking what precedent each decision sets and whether prior cases are truly analogous. When you need to change course, distinguish or overrule with a clear rationale. In law, business, and negotiation, precedents compound — design them deliberately.
Kahneman on reference class forecasting and using past cases to calibrate predictions. Application of precedent-like reasoning to judgment under uncertainty.
Buffett's annual letters as examples of consistent precedent-setting in communication, valuation, and governance — and explicit explanation when course changes.
Reinforces
Common Law
Common law is the body of judge-made law built from precedent. Courts follow, distinguish, or overrule prior cases. The system evolves through accretion and occasional overruling. Understanding precedent is understanding how common law works — and how any precedent-based system works.
Tension
Status Quo Bias
Status quo bias is the preference for the current state simply because it is current. Precedent can reinforce that bias: "we have always done X" may be inertia, not reason. The tension: precedent supports consistency and learning; status quo bias can trap you in suboptimal patterns. Distinguish following precedent because the reasons still hold from following it because it is there.
