A ratchet only turns one way. In economics, the ratchet effect is the tendency for levels of effort, cost, or commitment to move up and then stick. Past performance becomes the new baseline; reversals are rare. Bonuses and quotas ratchet up when times are good and do not fall when times are bad. Government spending and entitlements expand in crises and rarely contract when the crisis passes. Wages and benefits in union contracts move up in good years and are not cut in bad years. The mechanism is asymmetric: the system responds to positive shocks by adjusting the baseline upward and fails to respond symmetrically to negative shocks. What went up does not come down.
The term gained prominence in the Soviet planning literature. Managers who exceeded targets one year received higher targets the next; they learned to avoid overperforming so they would not be punished with an unreachable baseline. The same logic appears in sales quotas, budget cycles, and compensation. If you know that this year's result will become next year's floor, you have an incentive to cap your performance — to avoid the ratchet. So the ratchet distorts behaviour in two directions: it pushes baselines ever upward when they do adjust, and it discourages people from revealing true capacity when they anticipate the ratchet.
In strategy and investing, the ratchet appears wherever commitments are sticky. Capex programmes, headcount, and dividend policies tend to ratchet up. Cutting them is politically and psychologically costly; raising them is expected when results are strong. That creates a one-way bet for costs and a vulnerability when revenue or growth stalls. The disciplined move is to treat ratchets as irreversible until proven otherwise — and to avoid building ratchets into your own incentives or commitments unless you want the one-way behaviour they induce.
In mature organisations, ratchets compound. Each round of raises, each new benefit, each expansion of headcount becomes the new floor. Reversing any of it is treated as a breach of implicit contract. The result is a cost structure that only works when growth and revenue are strong. When the cycle turns, the organisation discovers that it has few levers that can move down. The strategic discipline is to build in reversibility early — before the ratchet has locked in — and to avoid treating "we have never cut X" as a virtue. It is a vulnerability.
Section 2
How to See It
Look for quantities or standards that move up in response to good outcomes but do not move down when outcomes worsen. Budgets, quotas, wages, entitlements, and baseline expectations are common ratchet variables. The diagnostic: would this number ever be reduced? If the answer is effectively no, you are looking at a ratchet.
Business
You're seeing Ratchet Effect when sales quotas are set from last year's actuals. Top performers get higher quotas next year; when the market softens, quotas are rarely cut proportionally. Reps learn to manage the ratchet — sandbagging in strong periods so the baseline does not become unattainable. The company loses information about true demand and true capacity.
Technology
You're seeing Ratchet Effect when a platform's feature set and complexity grow with each release. Features are added; they are almost never removed. The product ratchets toward bloat. The same applies to tech stack and infrastructure: services and dependencies accumulate and become the new normal. Simplification or rollback is treated as a failure rather than a routine adjustment.
Investing
You're seeing Ratchet Effect when a company's dividend or buyback programme is raised when earnings are strong. When earnings fall, the dividend is "sacred" and is maintained or cut only as a last resort. The payout ratio ratchets up over time. The same applies to leverage: debt increases in good times and is hard to pay down in bad times. Balance sheets and payout policies are often one-way ratchets.
Markets
You're seeing Ratchet Effect when government spending or regulation expands in a crisis — stimulus, bailouts, new agencies — and does not revert when the crisis ends. The new level becomes the baseline. The same applies to central bank balance sheets and policy rates: cuts are fast, hikes are slow; the "normal" level drifts. Fiscal and monetary policy often ratchet.
Section 3
How to Use It
Decision filter
"Before setting quotas, budgets, or commitments that are tied to past performance, ask: will this ever go down? If not, you have built a ratchet. Expect gaming (people will cap performance to avoid a higher baseline) and expect cost creep when conditions worsen. Design reversibility or reset rules if you want to avoid one-way drift."
As a founder
Avoid ratcheting your own costs. Headcount, office spend, and "standard" perks tend to ratchet up in good years. When revenue flattens, cutting feels like failure. Build in reversibility: variable comp that can fall, contracts that allow downsizing, and a culture where "we did more with less" is acceptable. On the incentive side, be careful with quota systems that set next year from this year's result — you will get sandbagging. Use independent targets, rolling baselines, or explicit reset rules so that the ratchet does not distort behaviour.
As an investor
Identify ratchets in the businesses you own or evaluate. Payout ratios, labour contracts, and regulatory commitments that only move up create downside leverage when the cycle turns. Prefer companies that have room to cut costs or that structure pay and capex so that bad years can be absorbed without a crisis. When management has ratcheted dividends or buybacks to a level that is unsustainable in a downturn, the equity is more fragile than the headline yield suggests.
As a decision-maker
Use the ratchet to explain why some organisations cannot shrink. Once a programme, a headcount, or a benefit is in place, removing it is politically and psychologically costly. That makes expansion easy and contraction rare. When designing policies or contracts, build in sunset clauses, review thresholds, or symmetric adjustment rules so that the baseline can move down when conditions warrant. Otherwise you lock in one-way drift.
Common misapplication: Treating every increase as a ratchet. A ratchet is specifically asymmetric — up is easy, down is blocked. Symmetric adjustment (costs that rise and fall with conditions) is not a ratchet. The model applies when reversibility is systematically blocked.
Second misapplication: Ignoring the incentive effect. The ratchet does not only lock in higher levels; it also distorts behaviour ex ante. People underperform to avoid raising the baseline. When you design incentives, account for that. If you do not want sandbagging, do not ratchet the baseline from prior performance.
Buffett has warned about the ratchet in capital allocation and compensation. Berkshire avoids long-term employment contracts and bonus structures that ratchet up and never down. He has said that the worst thing you can do is set a compensation baseline that can only go up — it locks in cost and encourages gaming. His approach: keep variable comp truly variable, and avoid committing to spending levels that cannot be cut when earnings fall. The ratchet is a source of permanent cost and hidden risk; the fix is to design reversibility into pay and commitments.
Charlie MungerVice Chairman, Berkshire Hathaway, 1978–2023
Munger has emphasised the ratchet in organisational behaviour. Once a company adds a layer of management, a perk, or a "standard" level of spend, it becomes the new floor. The organisation ratchets toward complexity and cost. His advice: resist adding things that cannot be removed. Prefer simple, reversible structures. The ratchet is why bureaucracies grow and why cost cuts are so often delayed — the baseline has been pushed up and reversing it is politically painful.
Section 6
Visual Explanation
Ratchet effect — Baselines move up when conditions improve and do not move down when they worsen. Asymmetric adjustment creates one-way drift and distorts incentives (e.g. sandbagging to avoid a higher target).
Section 7
Connected Models
The ratchet effect connects to incentives, path dependence, and irreversible commitment. The models below either explain why ratchets form (incentives, status quo), extend to related stickiness (path dependence, sunk cost), or help design around them (reversible decisions).
Reinforces
Incentives
When next period's target or reward is set from this period's performance, the agent has an incentive to limit current performance — the ratchet. Incentive design must account for the ratchet: use independent benchmarks, rolling averages, or explicit reset rules so that high performance is not punished with an unreachable baseline.
Reinforces
Status Quo Bias
People prefer to keep the current state. The status quo becomes the reference point. Combined with the ratchet, that means any increase in spending or commitment becomes the new status quo and is then defended. The two reinforce each other: ratchets create new status quos; status quo bias prevents reversing them.
Tension
Path Dependence
Path dependence means history constrains future options. The ratchet is a form of path dependence: once a level is reached, the path back down is blocked. The tension: path dependence can be efficient (sunk investment, learning); the ratchet is often inefficient (sticky cost, distorted effort). Both imply irreversibility; the ratchet adds asymmetric adjustment.
Tension
Reversible vs Irreversible Decisions
Reversible decisions can be undone at low cost. The ratchet turns reversible situations into irreversible ones: what went up does not come down. The strategic discipline is to identify which decisions create ratchets and to either avoid them or build in explicit reversal triggers.
Section 8
One Key Quote
"When the planner uses the current output to set the next period's target, the manager has an incentive to restrict current output so as to avoid an increase in the future target."
— Martin Weitzman, 'The New Soviet Incentive Model' (1976)
Weitzman captured the incentive distortion at the heart of the ratchet. It is not only that baselines move up and stick; it is that rational agents respond by limiting performance. The ratchet punishes success with a harder future. Once you see that, you see why quota systems and target-setting that tie the future to the past produce sandbagging and lost information. The fix is to break the link — independent targets, external benchmarks, or symmetric adjustment — so that high performance is not penalised.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
Ratchets are everywhere in organisations. Quotas, budgets, headcount, and "standard" benefits tend to move up and not down. The cost shows up when revenue or growth stalls — you are left with a cost structure that was built for a better world. The discipline is to build reversibility into as many levers as possible. Variable comp that can fall. Contracts that allow downsizing. A culture where "we cut back" is acceptable. The alternative is a one-way bet that blows up in a downturn.
The incentive effect is as important as the level effect. When people know that this year's result will become next year's floor, they have an incentive to cap performance. You lose information about true capacity and true demand. Sales leaders see it in quota games; investors see it in managed earnings. If you want honest performance, do not ratchet the baseline from prior results. Use independent targets, market benchmarks, or rolling windows that allow the baseline to reset.
Government and policy ratchet too. Spending and regulation expand in crises and rarely contract. The new level becomes normal. That is a ratchet. When evaluating policy risk or fiscal sustainability, assume that new programmes and new spending are permanent until proven otherwise. The same applies to central bank balance sheets and rate floors — the "normal" level drifts over time.
Design against the ratchet when you can. Sunset clauses, review thresholds, and symmetric adjustment rules (e.g. pay that rises and falls with profit) reduce one-way drift. When you cannot avoid a ratchet — e.g. union contracts, political commitments — factor in the downside: when conditions worsen, adjustment will be delayed and painful. The ratchet is a source of hidden leverage.
Watch for ratchets in deals and contracts. Multi-year commitments, step-ups in minimum guarantees, and "we never cut" cultural norms are ratchets. They lock in a baseline that may have made sense when signed but becomes a burden when the environment changes. The best contracts build in flexibility — volume-based pricing, force majeure, or explicit renegotiation triggers — so that both sides can adjust when reality shifts.
Section 10
Test Yourself
Is this mental model at work here?
Scenario 1
A sales team consistently hits 110% of quota. Next year, quotas are raised 15% for everyone. Reps start closing deals in Q1 of the following year instead of Q4 to avoid exceeding quota and raising the bar again.
Scenario 2
A company raises its dividend every year for a decade. In a recession, earnings fall 30% but the board maintains the dividend, drawing down cash reserves.
Scenario 3
A startup ties engineer bonuses to a percentage of salary. When revenue grows, bonuses go up. When revenue falls, bonuses are cut proportionally.
Scenario 4
A government launches a large stimulus programme during a recession. When the recession ends, the programme is extended and made permanent with a new name.
Section 11
Summary & Further Reading
Summary: The ratchet effect is asymmetric adjustment: baselines (quotas, wages, spending, commitments) move up when conditions improve and do not move down when they worsen. That creates one-way drift and distorts incentives — people underperform to avoid a higher future baseline. Use the model to spot sticky costs and gaming, and to design reversibility into targets and commitments. When you cannot avoid a ratchet, factor in the downside: when conditions worsen, adjustment will be delayed and costly.
Weitzman formalises the ratchet in planning: when the principal sets next period's target from current output, the agent restricts output to avoid a punishing baseline. The foundational incentive-distortion result. The paper shows that the distortion is rational for the agent and costly for the principal.
Kahneman's treatment of reference dependence and loss aversion explains why cutting feels worse than not raising — the psychological basis for asymmetric adjustment in wages and benefits. The current level becomes the reference point; losses loom larger than gains.
Harford discusses ratchet effects in compensation and organisational behaviour in accessible form. Useful for seeing the model in everyday settings — salaries, budgets, and organisational complexity.
Buffett and Munger have repeatedly warned about ratchets in compensation and cost structure. The letters illustrate how to design pay and commitments with reversibility in mind. Search for "compensation," "ratchet," and "cost" to find the relevant passages.
Holmström reviews incentive theory and the ratchet problem in dynamic contracts. Technical but definitive on why baselines that depend on past performance distort effort. Covers the trade-off between insurance and incentives in multi-period settings.
Leads-to
Moral Hazard
When baselines ratchet up, agents may game the system — underperforming to avoid a higher target. That is a form of moral hazard: the structure of the contract induces hidden behaviour that harms the principal. Designing to avoid the ratchet reduces the incentive to game.
Leads-to
Sunk Cost Fallacy
Once a cost or commitment has been incurred, people are reluctant to reverse it. The ratchet formalises that: the "sunk" level becomes the new floor. Distinguishing between genuine sunk costs (irreversible by nature) and ratchet-induced stickiness (reversible but politically blocked) helps identify where adjustment is possible.
