Some ideas are both radical and conservative at the same time. This is a thought triggered by reading Robert Frank's latest book, Under the Influence.

"Context shapes our choices to a far greater extent than many people realize" he writes. "In virtually every domain, evidence suggests that our consumption patterns tend to mirror those of others in our social circle." He shows that behaviours such as smoking, becoming obese, committing crime (pdf) or even how many children we have are all heavily shaped by what our peers do.

This, he says, means that much of what we do imposes externalities upon others. The biggest externality of smoking, he says, is not so much the passive smoke that smokers impose upon others but the fact that their smoking encourages some others to take up the habit. This, he says, means there is an orthodox Pigovian case for taxing things like cigarettes or sugary drinks.

He goes further. There are, he says, expenditure cascades: others' spending makes us spend more:

Higher spending at the top has shifted the frames of reference that define adequate from those just below the top, and so on, all the way down.

This, he believes, leads to a case for a progressive consumption tax.

What's more, he thinks that opposition to higher taxes on the rich is motivated by "the mother of all cognitive illusions". A lot of what we buy are positional goods, such as big houses or big cars. Taxing these more heavily won't affect their positionality, any more than would shortening a ladder change the fact that some people are on a higher rung than others.

We should, I suspect, distinguish here between behavioural externalities and informational externalities. For example, if I buy an SUV because others' huge cars make me feel unsafe on the road in a small car, it is a behavioural externality. If, however, I buy a big car because I believe doing so is what somebody in my position should do, we have an informational externality. (The latter, I think, contribute to asset price bubbles, among other things.)

Informational externalities, says Frank, influence our political views. He cites rising support for gay marriage or the legalization of cannabis as examples: the more that others' hold such positions, we more we are likely to see them as mainstream and so give them credence.

From one perspective, all this is unexceptional. Frank is just building on a point made by Adam Smith:

A linen shirt, for example, is, strictly speaking, not a necessary of life. The Greeks and Romans lived, I suppose, very comfortably though they had no linen. But in the present times, through the greater part of Europe, a creditable day-labourer would be ashamed to appear in public without a linen shirt, the want of which would be supposed to denote that disgraceful degree of poverty which, it is presumed, nobody can well fall into without extreme bad conduct. Custom, in the same manner, has rendered leather shoes a necessary of life in England. The poorest creditable person of either sex would be ashamed to appear in public without them.

Frank and Smith are expressing a form of cultural determinism - which has been a mainstream, if controversial, position in anthropology for years.

From an economist's point of view, however, it is radical. It is yet another challenge to the fiction of homo economicus, which envisages us as rational agents with exogenous preferences and beliefs. We should perhaps read Frank alongside Shiller's Narrative Economics, which shows how our beliefs are shaped not just by facts but by stories of sometimes dubious truthfulness.

Indeed, we can push Frank's theory a little further into much more radical territory than he explores. If our behaviour and ideas are influenced by context, why shouldn't our identities be also? Although he never mentions it, there's nothing in Under the Influence that contradicts Simone de Beauvoir's claim that "one is not born, but rather becomes, a woman" - that femininity is a social construct.

And if our preferences are so endogenous, why should policy-makers attach weight to them? Pushed only slightly further, Frank's ideas fit with Jason Brennan's Against Democracy. And they certainly pose Daniel Hausman's question: why satisfy preferences? (a question he answered mostly in the negative.)

We can go even further. Although Frank tries to reconcile his views on endogenous preferences with liberalism by pointing out that taxes are less restrictive than regulations, his work poses a challenge to liberalism. If our beliefs, wants and even identities are heavily socially influenced, then we are what Michael Sandel called in Liberalism and the Limits of Justice "radically situated subjects", and we cannot slough off our identities to go behind Rawls' veil of ignorance. And that brings into question a foundation of liberalism.

Many of us will find Frank's points empirically unexceptional. And yet they raise some big questions. Which reminds us that economics cannot be a merely technocratic discipline.