Comparative Statics¶
Core Idea¶
Comparative statics compares two equilibrium states of a system — the resting state before a parameter changes and the one after — while deliberately declining to model the trajectory between them: perturb one exogenous parameter, re-solve for the new equilibrium, and report only the before-after difference.
How would you explain it like I'm…
Two Photos
Imagine taking one photo of where all your toys settled, then changing one thing and taking another photo after they settle again. You only compare the two photos — you skip watching the toys roll around in between. You just ask: did this toy end up higher or lower than before?
Before and After, Skip the Middle
Comparative Statics is a way of answering 'what changes if I tweak one thing?' by comparing two settled states — the one before the change and the one after — without tracking the journey between them. You assume the system comes to rest, you change one input while holding the rest still, you solve for the new resting state, and you look at the DIFFERENCE: which direction did it move, and roughly how far? You deliberately ignore how fast it got there or what wobbles it went through. This saves a huge amount of work whenever the only question you care about is 'where does it end up after I do X?' — not 'what's the path?'
Compare the Resting Points
Comparative Statics is the move of comparing two equilibrium (resting) states of a system — the one before a parameter changes and the one after — while refusing to model the path between them. You assume the system rests at an equilibrium that depends on a set of parameters; you perturb one or a few while holding the rest fixed; you re-solve for the new equilibrium; and the answer of interest is the difference between the two — its sign, size, or pattern. The transient dynamics (how fast, along what path, with what overshoots) are explicitly out of scope. It rests on three ingredients: equilibrium-first modelling (a well-defined resting state per setting), parameter perturbation (isolate what's exogenous, hold the rest fixed), and differentiating the equilibrium with respect to the parameter. The price is precise: it says nothing about the adjustment itself, and silently assumes the new equilibrium is actually reached and is the right one.
Comparative Statics is the structural move of comparing two equilibrium states of a system — the state it settles into before a parameter changes and the state it settles into after — while deliberately declining to model the trajectory between them. The system is taken to rest at an equilibrium that depends on a vector of parameters; one or a few are perturbed while the rest are held fixed; the model is re-solved for the new equilibrium; and the answer of interest is the difference between the two equilibria — its sign, its magnitude, or its qualitative pattern. The transient dynamics — how fast, along what path, with what overshoots the system moves from one resting state to the other — are explicitly out of scope. What survives is an answer to a single question: if this parameter changes, in which direction and roughly how far does the equilibrium move? The move rests on three ingredients. First, equilibrium-first modelling: a well-defined resting state for each parameter setting, normally characterised by first-order conditions, a market-clearing or balance condition, or a fixed point. Second, parameter perturbation: isolate the exogenous variables, hold the rest fixed, and ask the system to absorb the change. Third, differentiation of the equilibrium with respect to the parameter — via the implicit-function theorem, monotone-comparative-statics machinery, or numerical re-solving — to read off the comparative answer without simulating the path. The whole apparatus is a static reasoning device that buys an order-of-magnitude reduction in modelling effort wherever the only policy-relevant question is 'where does it settle after I do X?' The price is precise: the analysis can say nothing about the adjustment itself, and it silently assumes the new equilibrium is in fact reached and is the right one.
Broad Use¶
- Microeconomic theory: tax incidence — raise a tax and ask who bears the new equilibrium burden.
- Engineering trade studies: solve an equilibrium at a design point, perturb a component spec, re-solve for the new operating point.
- Operations research: increase lead time by a day, hold demand fixed, re-compute the optimal base-stock level.
- Policy analysis: raise a minimum wage, assume the market re-clears, read off projected employment and price changes.
- Ecology and epidemiology: compare two stable states of a predator-prey or SIR system as carrying capacity or vaccination coverage shifts.
- Political science: median-voter models predict policy shifts by comparing equilibria as the median position moves.
Clarity¶
Makes explicit which question is answered and which is refused — the long-run resting state, not the transition — and keeps the robust sign result categorically separate from the fragile magnitude result.
Manages Complexity¶
Compresses an unbounded dynamic problem into a finite static one: solve the equilibrium symbolically, differentiate with respect to the parameter, and never carry the trajectory.
Abstract Reasoning¶
Supplies portable moves — implicit-function reasoning, envelope and Le Chatelier shortcuts — that read off the direction of an equilibrium's response from the structure of the problem without simulating the path.
Knowledge Transfer¶
- Climate adaptation / organizational change: the refusal to model dynamics travels as a discipline of honesty — the answer is conditional on the new equilibrium actually being reached.
- Systems engineering: implicit-function reasoning ports to feedback-loop steady states and control-system set-points.
- Policy with tipping points: the multiple-equilibria caution guards against confident prediction in path-dependent systems.
Example¶
The textbook tax-incidence result — that the more inelastic side of a market bears more of a tax — is derived by implicitly differentiating the market-clearing condition, yielding a robust sign while staying silent on the weeks of adjustment.
Relationships to Other Primes¶
Parents (1) — more general patterns this builds on
- Comparative Statics presupposes Equilibrium — The file: 'Equilibrium is the resting-state object comparative statics operates on; comparative statics is the second-order move of comparing two such states... the comparison operator that sits one level above the equilibrium noun.' It presupposes equilibrium.
Path to root: Comparative Statics → Equilibrium
Not to Be Confused With¶
- Comparative Statics is not Perturbation because comparative statics perturbs an exogenous parameter and reads off only the new equilibrium, whereas a perturbation studies the system's response to a disturbance including the transient.
- Comparative Statics is not Sensitivity Analysis in Operations Research because comparative statics re-solves a behavioral equilibrium and reports the direction of its shift, whereas sensitivity analysis probes how robust a fixed optimal plan is as inputs vary.
- Comparative Statics is not Equilibrium itself because equilibrium is the resting-state object, whereas comparative statics is the second-order comparison of two such states across a parameter change.