Economics, Business & Finance: Ceteris paribus and the assumptions underneath it

What you'll be able to do: Find what a model is holding constant that you never said out loud, which is where the marks go, and where a correct model produces a wrong answer.

Why this is the first method for economics

Every result in the subject arrives with two words attached: ceteris paribus: other things equal. Demand slopes down as price rises, other things equal. A minimum wage above the market rate reduces employment, other things equal. A tax cut increases disposable income and spending, other things equal.

Students learn to say the phrase. Almost none of them learn to list what "other things" means for the case in front of them, which is a problem, because the list is exactly what determines whether the model's conclusion still holds.

For the downward-sloping demand curve, "other things" includes income, the price of substitutes and complements, tastes, and expectations about future prices. Change the price of a good and raise incomes in the same period, and you cannot read the quantity change off the demand curve as if price were the only thing that moved, you've confounded a price effect with an income effect, and the diagram you were taught doesn't warn you.

Where the unstated assumptions actually hide

Market structure. "A price floor above equilibrium reduces employment" is the standard prediction in a competitive labour market. It is not the prediction in a market with a dominant local employer (a monopsony) where a wage floor can raise both wages and employment simultaneously, because the employer was suppressing both below the competitive level to begin with, Whether your local labour market looks more like the first case or the second is an empirical question the diagram is silent on.

What the policy is being compared to. "This fiscal stimulus will raise output" assumes the central bank doesn't raise interest rates to offset it, and assumes the economy has spare capacity to expand into. Neither is stated, and both can flip the sign of the effect, not just its size.

Time horizon. A tariff might raise domestic producers' output in the short run before input costs and retaliation catch up. A result true in one horizon and false in another is not a contradiction; it's an unstated assumption about which horizon you meant.

What "the market" includes. GDP's own definition. C + I + G + net exports, silently assumes that what's produced equals what's sold, which is only true once you count inventory changes as a form of investment. The identity looks like arithmetic and is actually a modelling choice about how to treat unsold goods.

Whose behaviour is fixed. A merger analysis that says "prices will rise" usually assumes competitors don't respond. If a rival can enter or expand, the prediction weakens or reverses.

How to run it

Here's my analysis: [paste your reasoning, model or conclusion]. Don't tell me if the conclusion is right. List what I'm holding constant that I haven't stated. For each one: if it were false, would my answer just change in size, or would the whole conclusion flip, could the sign reverse?

That second question is the one that matters. Some unstated assumptions move a number by a few percent. Others, market power instead of competition, offsetting policy instead of none, a short horizon instead of a long one, mean the model was answering a different question than the one you asked it.

Across the disciplines

Microeconomics. Elasticity assumes a fixed set of substitutes; introduce a new one and the same price change produces a different quantity response.

Macroeconomics. "Money supply growth causes inflation" assumes stable money demand and full resource utilisation. Both are testable, neither is automatic, and the relationship has visibly weakened and strengthened across different decades as those conditions changed.

Accounting. Depreciation assumes a useful-life estimate and a residual value; change either and reported profit changes with no cash ever moving. The number looks objective and is a modelling choice wearing a suit.

Finance. A discounted cash flow assumes a discount rate, a growth rate after the forecast period, and that the forecast cash flows themselves are right. Most of the disagreement over whether a company is "worth" a given price is disagreement about these three unstated numbers, not about the formula.

Management and marketing. A pricing recommendation assumes competitors don't retaliate and that the customer segment you measured demand on is representative of the one you're about to sell to.

Entrepreneurship. A unit-economics model assumes customer acquisition cost and retention stay where they were in the pilot once you scale past the group of early adopters who found you yourselves.

The exam version

Here's an exam or case question and my answer. Which assumptions did I state explicitly, and which ones did I rely on without saying? Which of the unstated ones, if the grader assumed the opposite, would cost me the most marks?

Marking schemes in this subject routinely allocate credit for stating assumptions separately from the calculation. Students treat that line as preamble and lose it as volume, not as a single dramatic error.

Pitfalls

1. Confusing an assumption with a mistake. Most are legitimate simplifications. The goal is knowing you made them, not eliminating them.
2. Listing the trivial ones. "Consumers are rational" is not usually the interesting assumption in a specific problem; the one about market structure or time horizon usually is.
3. Stopping at the list. The what-changes-if-it's-false question is the method; the list alone is just an inventory.
4. Only checking assumptions when the answer looks wrong. Run this on analyses you're confident in, an unstated assumption sits most comfortably in a conclusion nobody has doubted yet.
5. The tell: every assumption surfaced is one you'd already have named if asked. Give it your actual reasoning, not a tidied-up version of it.

Try this today

Take the last piece of economic or business reasoning you did, a homework answer, a case analysis, a claim you made in an argument. Paste it and ask only: what am I holding equal here that I haven't said?

Then, for the one that matters most: does your conclusion just move in size if it's false, or does it flip sign?