In 1951, economist Kenneth Arrow published a book titled Social Choice and Individual Values in which he identified a list of five criteria that any fair voting system would have to satisfy. Arrow then went on to demonstrate that it is logically impossible for any voting system to satisfy all five of these criteria simultaneously.
The result, which has come to be known as the “Impossibility Theorem,” applies even to exotic voting systems that are radically different from those familiar to most of us. For example, one could have a kind of weighted voting system in which, rather than casting a single vote for a single candidate, each voter assigns preference rankings to a list of candidates. The candidate that receives the highest aggregate preference wins. One could also have proportional representation systems in which groups of candidates win and thus share offices in proportion to the number of votes they receive.
Countless other variations are possible, yet they are all subject to the Arrow Impossibility Theorem. Against this backdrop, I have several questions:
Would it be accurate to interpret a free market as functioning in essence as a kind of voting system, with consumers “voting” with their purchases on which goods and services are to be provided?
If the answer to the first question is yes, then what implications, if any, does the Arrow Impossibility Theorem have for evaluating the market’s ability to produce a mix of goods and services that accurately reflect consumers’ actual preferences?
If the answer to the first question is no, then what are the crucial differences between voting systems and a free market that prevent the Impossibility Theorem from applying to markets?
It is worth noting, incidentally, that Arrow himself favors subsuming both voting systems and markets under a more general and inclusive category of collective social choice based on ordinal rankings of individual choice. Thus, I believe he would answer the first question in the affirmative. It also seems to me that Arrow’s view of collective social choice has much in common with the Austrian notion of subjective value. In any event, I would be grateful for any thoughtful responses to my questions.
Each dollar spent is a vote towards a specific good or service on the market. You show preference by choosing one good or service over another, and by not choosing.
Yeah, the immediate objection I can come up with is that the market is not a social choice system, but an individual choice system. Our buying decisions don’t determine whether all of us should have either vanilla or chocolate ice cream, we can each have what we want.
Although ultimately the direction of economic development is determined by what the majority of customers want. So the market could be considered a social choice system. I don’t know how that works out with the impossibility theorem.
Edit: I suppose technically speaking there can be a dictator in the free market. Say that for some reason society can only produce one brand of ice cream this year, and that society consists of 100 customers. If 33 want vanilla, 33 want chocolate and 33 want strawberry, then that last customer determines what the most profitable flavor will be. He is a dictator.
Good ole wikipedia says we are talking about an ordering of preference, without a number assigned to it. So that one can say “I like TV sets more than hot dogs, and hot dogs more than peanuts”, but not “I like TV sets 5 units of liking, and hot dogs 3 units, and peanuts 2 units”.
Which fits like a glove with the Austrian understanding subjective value.
Now Arrow is talking about taking everyones preferences, mixing them together in some kind of scheme, and coming up with a grand mega preference list that somehow reflects everyones individual preferences. He says it is impossible. Which seems to me to say that central planning of an economy is impossible.
At any rate, my question is, if we assume for a moment the impossible, that such a list could exist, of what use would it be? If we are talking about elections, then we can say “the people prefer Smith to Jones”. But what does an entrepeneur learn when he finds out that people will spend their money on a TV set first, any left over money going to hot dogs, and what’s left over going to peanuts?
If he has a peanut factory, he knows or guesses what and how much to produce based on an understanding of the market for peanuts, and past history of the peanut market. But he has no need or use for the knowledge of people’s preferences for peanuts compared to hot dogs.
The voting metaphor is useful for conveying the idea of consumer sovereignty, but should not be taken any further. Exchange is individual choice, whereas voting is a mechanism of collective choice. The aggregation of individual choices/exchanges should not be considered a “collective choice” which market participants “voted for” through their trading. Using the metaphor in that way loses a lot of information.
That’s not how it works. Relative prevalence of preference determines the flow of investment, not what is to be produced. Minority preferences are, ceteris paribus, more expensive to fulfill as they receive investment last. As prices are driven downward in a majority sector, minority-desired goods become relatively more profitable to produce. Eventually this drives investment toward these goods as well. Minorities aren’t denied their preferences, they are just more expensive to obtain. What you describe in your example is a situation where a one-person difference makes the other options too expensive to undertake at all, in which case the dictator situation is at best temporary.