Wednesday, September 24, 2014

Liza Featherstone on Focus Groups

Yesterday, we at the John Jay Economics Department hosted my friend Liza Featherstone, author of Students Against Sweatshops and Selling Women Short: The Landmark Battle for Workers Rights at Wal Mart, for a talk about her new book, on the history and politics of focus groups. (That's me introducing her.)

The whole thing is worth watching. As with so many institutions, the history of the focus group has interesting twists and turns that you wouldn't guess just from its current state. I hadn't realized, for example, that the focus group originated -- like so many technologies -- in the US war effort of the 1940s. The first focus groups, apparently, were convened to improve the quality of radio propaganda. Only later was the technique adopted by commercial advertising, before migrating into the electoral arena in the 1980s and 1990s.

The most interesting part, though -- for met at least -- is the end, where Liza talks about the funny parallels between focus groups and the kind of consensus decision making practiced by mass movements like Occupy. (And by popular movements at least back to the 18th century, for that matter.) In contrast to surveys, polls and elections, focus groups and assemblies do not assume that people enter the process with well-formed views that just have to be registered and tallied. Instead, they assume that people's true views only emerge in a process of active exchange with others.

The classic example in the marketing context is New Coke. In blind tests, clear majorities preferred the new flavor, but in a setting where the two cokes were discussed, people were somehow convinced that they preferred the old flavor after all. Apparently focus groups sponsors often complain in cases like this that one strong personality bullies everyone else. But isn't that how people's choices get shaped in the rest of life too?

Is it better to keep our private views intact? There's a view that instability in asset markets arises precisely because people allow their views to be shaped by interaction with others -- herd behavior, the madness of crowds. To have efficient asset markets, it's important that people trade on their private information. We use secret ballots in elections and some people even consider it rude to ask who you are voting for. Too much discussion corrupts the process of aggregating up private beliefs. More generally, there's a sense that the way our ideas change in contact with others can be a problem, and that we may need to protect our authentic selves from social pressures.
To love you must have someone else,
Giving requires a legatee,
Good neighbours need whole parishfuls
Of folk to do it on -- in short,
Our virtues are all social; if,
Deprived of solitude, you chafe,
It's clear you're not the virtuous sort. 
Viciously, then, I lock my door.
The gas-fire breathes. The wind outside
Ushers in evening rain. Once more
Uncontradicting solitude
Supports me on its giant palm;
And like a sea-anemone
Or simple snail, there cautiously
Unfolds, emerges, what I am.
That's Larkin, who would not have had much time for either focus groups or general assemblies.

Meanwhile, on the other side, along with the focus groups we have 12-step groups, psychoanalysis, the self-criticism practiced in some revolutionary groups. All of these elevate the process of communication itself as the source of beliefs and desires, as opposed to the liberal idea that we first come into possession of these individually and then act on them or communicate them.

What do we make of this similarity? The negative answer is that focus-group politics have displaced more effective forms of political organization on the left as well as in the mainstream. People have come to feel that communication is an end in itself -- that the important thing is to have your voice heard, and only incidentally to exercise power. (Liza shared a rather depressing anecdote after the talk, about a union staffer who said that it was impossible to get members to come to meetings by saying there would be an important vote, but they would come if you told them they'd be taking part in a focus group.) Of course even this way of looking at things isn't entirely negative -- people do need to get their voices heard, especially people without the privileges that make it easy to be listened to.

But there's another way to look at these parallels. Maybe the marketers, in their desperation to "catapult the propaganda" (in the words of one of our most focus-grouped, and focus group-deriding, politicians) have stumbled on a truth about human nature that the left has always known. We are not monads, with a fixed set of preferences. As Liza says in the talk, human beings are profoundly social creatures -- our selves don't exist in isolation from others. (This is why solitary confinement is a form of torture.) Capitalism is intolerable but it has, historically, produced genuine progress in science and technology, and there's a sense in which focus groups could be an example. It's grotesque that this insight -- that people's beliefs and desires only emerge in exchange with others -- has been mainly used to sell soft drinks and candidates. But it's a real insight nonetheless.


  1. Or, as Andre Orlean puts it, "the reduction of human desires to a purely utilitarian calculus has little or no basis in fact. After all, if the demand for goods were a function of material needs alone, one would expect it to decline markedly over time. But the opposite is true. Marketing and publicity, which deliberately seek to stimulate demand by appealing to the mimetic urge, provide incontrovertible evidence of the importance of this aspect of economic behavior."

    The "fixed set of preferences" is one of the postulates required to make the Walrasian model produce its cherished equilibrium. Taken together, the constraints of the model are not merely "unrealistic" but bizarre and contradictory (e.g., every market actor is a solipsistic monad EXCEPT the auctioneer who is a consumate altruist). So when the auctioneer dies, how do they find a replacement?

    I think the prototype of the "sovereign individual" of neoclassical analysis can be found in the Spanish Baroque how-to manual for courtiers by Baltasar Gracian. If we assume that every representative agent is a courtier wannabe, engulfed by a multitude of 'necios' (fools), then we have a picture of the "society of the market." This has about the same probability as everyone scoring above average on an exam...

  2. Yes indeed. I've just ordered that book.

    1. Excellent! I'm about a third of the way through. It is riveting, brilliant, far-reaching and persuasive.

  3. Is it against the spirit of the blog to ask if it comes in an ebook for iPhone or Kindle?

    1. It doesn't exist yet, she's still writing it. I should have made that clearer.

  4. Thanks for sharing this. Very enjoyable thinking. I suspect that people's beliefs and desires emerge more through attempts to present them to others than hearing others' presented.

    First, the human brain will always believe that its conclusions are logical and well substantiated. It is only by trying to express the internal logic that we can perceive any flaws. (Yes, there are narrow exceptions to this rule.) In a non-threatening conversation we may actually recognize some flaws and change our preferences, especially if there are people that we like who have different ones. At least we have to recognize that a monad will have many preferences that are both strongly stated and critically unexamined.

    Second, we (most of us) desperately want to have an impact on other people. (I do recognize the humor of me making this argument in a blog comment.) I think the whole phenomenon of anonymous blog ranting is evidence of a fundamental desire to have an impact on other people, even if it is just to make people angry. We can't resist an opportunity to influence the preferences of other people.

    Focus groups are selected for people who most want the opportunity to influence others. While solitude holds great attractions for people who do not enjoy the pain of trying to rationally express ideas that seem clear internally. Thus, groups are subjected to the vagaries of persuasive interactions and polls are full of unexamined preference.

    For a market to function well you need the intermediate and unstable situation where actors argue decisions in small independent groups before acting individually. It is less important that individuals hear the arguments of others but rather are forced to present their own.