Lesson 3 of 7
Explanation that starts from people
Methodological individualism is a rule about where an explanation may begin. It buys a specific kind of detail and it costs a specific kind of measurement.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- Methodological individualism is a rule about where an explanation may start, not a claim that groups do not exist or do not matter.
- An aggregate such as total investment is treated as something to be explained, never as the cause of what the people in it did.
- The cost is real: an explanation built this way is harder to measure, harder to test against published data, and easier to argue about.
Every economic statistic is a sum of things people did. Methodological individualism is the rule that an explanation has to reach down to those people, and that stopping at the sum leaves the work unfinished.
Two places an explanation can stop
- 1AggregateTotal investment fell 8 percent
- 2Aggregate causeBecause demand fell
- 3Stops hereNo person has appeared yet
What the rule does and does not say
It does not say that only individuals exist. Firms, banks and governments are all real, and they all act. It says that when you explain what one of them did, the explanation has to end in people making choices, because a company does not have preferences and a person inside it does.
It does not say that aggregates are useless. It says they are results. Total investment falling is the thing to be explained; it is not an explanation of why any particular plant was not built.
Mises states the position for the whole tradition in the opening part of Human Action, where economics is built up from the fact that people act toward ends they have chosen.
What it buys
It buys a specific kind of question. Instead of asking what happened to a total, you ask who changed a plan, what they were reading when they changed it, and what they had already committed and could not undo. Those questions have answers a reader can check against a filing, a contract or a decision that was recorded somewhere.
It also rules out a common shortcut. If a total is a cause, then two years with the same total look the same. If the total is a result, they do not: the same number can be produced by a very different set of decisions, and the difference is exactly what the next year depends on.
What it costs, and this is not small
Individual decisions are not published. The tradition therefore explains a good deal that it cannot directly measure, and its accounts are harder to test against the series everyone can see. Caplan's 1999 paper in the Southern Economic Journal pushes on exactly this joint, arguing that the claim that only choices actually made can count as evidence is too strong, since a person can know their own preferences by introspection without ever acting on them. The dedicated critique course later in this track takes that argument at full strength.
Check yourself
What does methodological individualism claim?
- Groups and firms are not real
- An explanation must end in individual people choosing
- Aggregate statistics should not be published
Show the answer
Answer: B.
It is a rule about where an explanation may stop. Firms and groups are real and they act; the explanation of what they did has to reach the people who chose.On this rule, what is an aggregate such as total investment?
- The cause of the decisions inside it
- A result to be explained
- A measurement error
Show the answer
Answer: B.
The sum is produced by the decisions. Treating it as the cause of those decisions reverses the order of explanation the rule insists on.What is the honest cost of this commitment?
- It cannot describe firms
- Individual decisions are largely unobserved, so the accounts are harder to test against published data
- It requires more mathematics
Show the answer
Answer: B.
The rule pushes explanation toward things that are not published. That is a real cost, and Caplan's 1999 paper argues the tradition's standard of evidence here is too strong.Sources
- Mises, Human Action, 1949, part one, on building economics up from individual action.
- Menger, Principles of Economics, 1871, English translation 1950, chapter I, where goods are defined by a person's needs rather than by their physical properties.
- Caplan, 'The Austrian Search for Realistic Foundations', Southern Economic Journal, 1999, on preferences that are known by introspection but never revealed in action.
What this leaves out
- This is a heterodox tradition. The mainstream account of the same phenomenon differs, and nothing here is a forecast or a view on any security.
- Both examples are invented illustrations of a style of explanation, not descriptions of any company or year.
- Methodological individualism is a rule about method. It implies nothing about politics, and this lesson takes no position on any.