Positive and Normative Economics

The distinctions table

A worked example: a minimum wage in a small town

Sridhar runs a workshop with eleven workers in Bhiwandi. A notification raises the minimum wage for his class of establishment.

Four statements are made in the debate. Sort them.

1. "After the notification, Sridhar's wage bill rose by eighteen per cent." Positive. Verifiable from his books.
2. "He responded by not replacing two workers who left." Positive. Verifiable, and it is a claim about behaviour.
3. "A higher minimum wage reduces employment among the least skilled." Positive, and much harder. It is a general claim about cause and effect, and economists genuinely disagree about its size because the evidence is mixed. That the answer is contested does not make the statement normative; it makes it a difficult positive question.
4. "Nobody should have to work for less than a living wage." Normative. No study settles it.

Why the sorting matters to a lawyer. If the notification is challenged, statements 1 to 3 belong to evidence and statement 4 belongs to the object of the statute and to the constitutional standard. A petitioner who leads only statement 4 has produced no evidence. A State that answers only with statement 4 has not met the evidence. Courts routinely have to do exactly this sorting, and doing it badly is how a hearing turns into an exchange of opinions.

Where the distinction gets blurred, and an examiner will test you on it

An answer that says only "positive is what is and normative is what ought to be" is worth half marks. The rest of the marks are here.

A positive statement can carry a hidden value judgment in its choice of subject. An economist who studies the effect of a subsidy on the fiscal deficit and never studies its effect on child nutrition has made a judgment about what is worth measuring. Selection is not neutral even where measurement is.

The line between the two is not always clean. Saying "this policy is efficient" sounds positive and is usually said as praise. Efficiency is a criterion with a value built into it, namely that a larger total is better, which is precisely the point [Why a Law Student Studies Economics] makes about efficiency and equity.

A normative conclusion needs a positive premise. "The tax should be raised" is worthless unless we know what raising it does. Bad normative economics is nearly always bad positive economics wearing a moral tone.

Most real policy statements mix the two in one sentence, and unpicking them is the skill. "The fuel subsidy is a wasteful giveaway that costs the exchequer a lakh crore" contains a positive claim about a number and a normative claim in the word wasteful, and the two must be answered separately.

What this does NOT mean

It does not mean normative economics is unscientific rubbish. Every policy decision is normative. Somebody has to decide what the objective is. The point is to know which kind of statement you are making.

It does not mean an economist should have no opinions. It means the opinion should be labelled as one.

It does not mean positive statements are always true. They are testable, which is a different thing. A positive statement can be confidently made and completely wrong.

Quick revision

1. Positive economics describes and predicts. It is testable in principle and free of value judgments. Form: if A then B.
2. Normative economics prescribes. It rests on value judgments and cannot be settled by evidence. Form: A ought to be done.
3. The test: could evidence in principle show this to be false? If yes, positive.
4. Marker words for normative: should, ought, fair, just, desirable, too high. The marker is a clue, not the test.
5. Origin: Nassau Senior, then John Neville Keynes in 1890 who separated a positive science, a normative science and an art. Robbins pressed the separation; Friedman argued in 1953 that many apparently normative disputes are really positive ones in disguise.
6. The blurring points, worth marks: the choice of what to study is itself a judgment; efficiency is a criterion with a value in it; every normative conclusion needs a positive premise; and real statements mix the two.
7. For a lawyer: evidence answers positive questions, and the object of the statute and the constitutional standard answer normative ones.

Test yourself

1. Define positive and normative economics and give one example of each. Positive economics deals with statements of fact about economic behaviour which can in principle be verified or falsified by evidence, for example that an increase in the price of a good reduces the quantity of it demanded. Normative economics deals with statements about what ought to be, resting on value judgments, for example that essential medicines ought to be exempt from tax. The first can be tested; the second is a question of what we value.

2. Classify each of these: (a) India's fiscal deficit was higher last year than the year before; (b) the fiscal deficit is too high; (c) a larger fiscal deficit raises interest rates; (d) the government ought to spend more on primary schools. (a) Positive, a statement of fact checkable in the Budget documents. (b) Normative, because "too high" appeals to a standard of what is desirable. (c) Positive, a causal claim that evidence can test even though economists dispute its size. (d) Normative.

3. "A disagreement about policy is always a disagreement about values." Discuss. It is often not. Friedman's argument in 1953 was that many policy disputes between people who share the same objective turn on a positive question about what a measure actually does. Two people who both want less poverty may disagree about a minimum wage because they disagree about its effect on employment, which is a factual question capable of investigation. Some disputes are genuinely about values, for example how much present consumption should be given up for future generations, and those cannot be settled by data. The useful discipline is to separate the two before arguing.

4. Can positive economics be entirely free of value judgments? Give your reasons. Not entirely. The measurement of a relationship can be neutral, but the choice of which relationships to measure, which variables to include and which effects to report is a judgment about what matters. Terms that appear technical, such as efficiency, also carry a criterion of what is better. The honest position is that positive economics can be much freer of value judgments than normative economics, and that the judgments it does contain should be stated rather than hidden.

5. Why is the distinction useful to a lawyer? Because a court hears both kinds of statement in a single argument and must treat them differently. Claims about what a measure does are matters of evidence, to be proved by data and expert testimony. Claims about what ought to be done belong to the object of the statute, to the directive principles and to the constitutional standard of reasonableness. Separating them shows which parts of a case need proof and which need argument, and it exposes a submission that offers a value judgment where evidence was required.

6. Restate this sentence, separating its positive and normative parts: "The petrol subsidy is a wasteful giveaway costing the exchequer a lakh crore." The positive part is the claim that the subsidy costs the exchequer approximately one lakh crore rupees, which can be checked against the Budget's subsidy statement. The normative parts are "wasteful" and "giveaway", which assert that the money would be better spent otherwise and that the recipients do not deserve it. The positive claim is answered with figures; the normative claim is answered by arguing about who benefits and what else the money would buy.