Bertold Brecht and the Paradox of Institutions

After the uprising of the 17th of June
The Secretary of the Writers’ Union
Had leaflets distributed in the Stalinallee
Stating that the people
Had forfeited the confidence of the government
And could win it back only
By redoubled efforts. Would it not be easier
In that case for the government
To dissolve the people
And elect another?

-Bertold Brecht

When Brecht wrote his “Solution Poem” in 1953, he must have imagined himself mocking the authoritarian government of East Germany.  Little did he realize, I would imagine, that he was mocking every institution in politics and economics.  Only 9 years after Brecht’s poem, The Theory of Political Coalitions by political scientist William Riker was published–and the idea goes back a while longer.  Riker’s argument was that every member of a coalition is, in effect a cost center, while the benefits conferred by adding a coalition member stops when the last coalition member needed to capture power is recruited.  In other words, no coalition larger than the absolute minimum needed to “win” is necessary.  To the degree that the winning coalition controls power, the strategy of coalition building is effectively the process through which the government elects people with whom it seeks to associate with.

This holds true not only for political coalitions but also economic coalitions.  At the time of Adam Smith’s writing, the idea that an additional customer might be a cost center would have been unthinkable:  if a customer is willing to pay real money greater than the cost for a good, why should he be a loss?  The short answer is monopolistic competition coupled with customer discrimination, which, to be fair, did not get seriously appreciated until the work by Avinash Dixit and others in 1970s.  The somewhat longer answer to the Smithian question is that it is more profitable for the seller to sell a single item with bells and whistles to one consumer rather than two items without bells and whistles to two consumers.

To illustrate via a simple example, consider an industry characterized by increasing returns to scale.  The break-even price for the producer is P* = AC(Q*), where AC(Q) = average cost at quantity Q.  Assume that there is enough competition that, for an undifferentiated product, the firm makes zero profit.  However, suppose that the firm can become a partial monopolist by an alternate version of the good with bells and whistles to the same good at a nominally higher cost and that some fraction K<1 of the consumers are willing to pay a price P+ >> P* for this version.  Further suppose that the cost of producing the alternate version is relatively inexpensive.  These imply that, if Z(Q) is the average cost function of the alternate version such that Z(Q) > AC(Q) for all Q, it is more profitable for the firm to produce the alternate version since:

(P+)KQ* – Z(KQ*) >P*Q* – AC(Q*)Q*> P*KQ* – AC(KQ*)KQ*.

(I am cheating a bit, since I did assume zero profit in the competititve equilibrium, but the intuition holds regardless–I will confess to being a bit sloppy with the specifics.  If you are more economically inclined, you would notice that since P* = AC(Q*) and KQ* < Q*, the firm will be actually losing money at quantity KQ*, which would be the better comparison)

What this implies is that, for a segment of the market that corresponds to the fraction K that would pay a lot more for the bells and whistles, the firm is best off shifting the production to the alternate version from the old one.  Now, the demand for the version without bells and whistles has shrunk to only (1-K)Q*.  Given the increasing returns to scale assumption, AC((1-K)Q*) > AC(Q*).  So the new break even price P’ > P* and the new equilibrium quantity demanded at P’, Q’ < (1-K)Q*.  Thus, KQ*+ Q’ < Q*.  In other words, by allowing the firm to specialize and meet a subset of the consumers that want bells and whistles, we have effectively eliminated some of the demand, creating a market failure where some of the consumers who were being served are no longer being served.

The situation is more complicated for those who are still being served:  those who are getting the alternate version of the good with bells and whistles are probably better off:  they did, after all, choose the more expensive version at a higher price.  Those who are still consuming the plain version are, however, paying a higher price without an improvement in quality.  The firm is making profit off of the first group, but none off of the second, despite the fact that the latter have been made worse off by the utilization of the technology.  Indeed, if the firm sought to increase its profits by better discriminating against its consumers, it would seek to find some other set of bells and whistles that the remaining consumers might pay a premium for and abandon those who do not.

This underscores the dynamic that Smith did not understand yet.  The assumption of constant returns to scale implies that, even if the demand for a certain variety of good shrinks by availability of an alternate version, all incumbent customers would be retained as they will continue to face the same price.  The effects of increasing returns to scale on the market was recognized by Marx in late 19th century, although it did not come to be fully appreciated by economics profession until 1970s.  While the tools developed for understanding of monopolistic competition did imply the effect just demonstrated roughly the same time, the practical impact has not, it seems.  The practical impact of utilizing information technology is that it vastly enhances the producer’s bargaining power vis-a-vis consumers:  it allows them to extract all surplus in form of increased profits from those who want something extra for a price, while ruthlessly abandon those who do not want anything fancy since the latter cannot be milked for profit.  The better the technology, the smaller the deadweight loss–the standard example of the cost imposed by monopolies:  it comes only from the relatively small number of consumers who do not wish to pay more for extra goodies.  But there is no consumer surplus left:  it is entirely appropriated by the producers.  This goes beyond the shares of capital and labor incomes that obsessed Marx and, more recently Piketty.  This is the microfoundation through which the producers exploit the consumers.  To flip the old Marxian logic, it is not so much that consumers are exploited because most of them happen to be labor, as per the old argument, but that labor is exploited by firms because they happen to be consumers.  It is in the consumption, not production, that the technological exploitation takes place.

The political version is even more extreme:  while every additional consumer brings potential profit for the producer, as long as they are willing to pay extra for bells and whistles, there is no political profit for the politicians beyond the minimum winning coalition.  The old superfluous coalitions existed not so much because of a rational need for oversized coalitions, but because entire demographic segments had to be appealed to largely as if a monolith because the ability to microtarget was limited.  Thus, the voters were farmers, African-Americans, Evangelicals, Southerners, etc:  huge blocks with some shared characteristics whose may have been diverse and heterogeneous internally but could not be easily disentangled for particular messages.  In today’s environment, however, each segment CAN be targeted separately.  There are separate messages that can be so precisely constructed that a relatively small group with specific wishes can be recruited to pay the price, in the form of political support for the coalition.  The same technology that permits producers to discriminate among potential consumers, the parties can use to discriminate among potential supporters.  Much the way producers ditch the troublesome consumers who do not yield them profit, parties ditch the supporters in whom they have little confidence, that they would not reliably support them at minimal cost to the parties themselves.  They can keep ditching them until they have absolutely the minimum necessary to maintain power.

I think this is where something paralleling the Marxian crisis of capitalism emerges:  without resorting to the antiquated and rather wrongheaded Marxian economics, one could effectively sum up the crisis of the Marxian capitalism as emerging from the destruction of demand.  By exploiting the labor and keeping all the fruits of technology for themselves, the capitalists ensure that there is no one to buy their product, however cheap they get.  This is really the argument by Piketty in (only slightly) different words.  The rate at which technology can lower prices is slower than the rate at which the capital expropriates consumers who also happens to be the labor.  In the political variant, there is another problem:  how stable is a minimalist winning coalition?  The requirements for what a coalition needs to “win” is defined by institutions, but in the end, all institutions are merely scraps of paper, sustained by the tacit (or explicit) acceptance by all members of society of their legitimacy.  To the degree that the institutions ensure systematic transfer of surpluses, whether in terms of enjoyment greater than or equal to willingness to pay or in terms of profits, to the barest majority at the expensive of the barest minority, why would the exploited half put up with the institutions?

Observe that there is little room for new entry in this market, although the political market is more restricted than the economic.  The number of consumers who want specific kinds of bells and whistles is going to be limited.  The easier point of entry for new firms is to raid the consumers of the generic, undifferentiated good, by identifying the specific kinds of bells and whistles that they are willing to pay extra for.  The consequence is still that the consumers get what they want, in terms of the goods, but the firms get all the surplus, while trying to outcompete those who are already delivering specific bells and whistles by developing better bells and whistles is not nearly as profitable–for now, as long as the market for generic products exist in large enough a size.

A useful analogue is the beer market:  some time in 1990s, technological change allowed for small quantities of beer catering to specific tastes to be produced efficiently, starting the craft brew revolution.  All manner of tastes are being served in small quantities, at a high profit margin.  While the market for cheap, relatively tasteless beer remains large–Bud Light remains the best selling beer in the market, despite being the prime example of cheap and tasteless brew–it is both shrinking and receives little attention from the breweries, who are eagerly acquiring small craft breweries that can be exploited for profit.  The competition between craft brews is not especially intense: beer production is still characterized by a reasonably increasing returns to scale and the market for specific taste is small enough that entry is not very profitable–which, of course, is why craft brew is more profitable than the mass produced beer.  The rate of technological change is not enough to make all beer cheap:  it is merely enough to make possible a sufficiently cheap expensive beer that generates profit for those who control that technology.

This is an awkward argument:  craft brew aficionados may be “poorer,” in that they pay more for beer, but at least it is the better beer for their taste.  Are they being exploited?  It is difficult to say.  Anheuser-Busch (or Interbrew) may not be selling as much Bud (or Bud Light) as before, but they are making more profits by selling a broad portfolio of craft brew.  The only “exploited” group in the mixture are the Bud Light drinkers, who are not being given as much attention as before while their beer stagnates in quality and getting relatively more expensive, if only to reflect the change in average cost in face of decreasing demand.  Changing technology will introduce more craft brew, drawing away the incumbent Bud Light drinkers who will pay extra for even slightly better beer, but as those with “obviously” distinct tastes are drawn away, the beer drinkers with “no taste” are left to overpay for ever decreasing quality, without any incentive for the sellers to care.

If this is a difficult argument to sustain for markets, it is a much more dangerous argument for politics.  The same process creates an underclass of political consumers who are both underserved and forced to pay an ever higher price for lousy goods.  Unlike beer drinkers, they face little opportunity to simply go sober.  This is the heart of the problem.

Advertisements

4 thoughts on “Bertold Brecht and the Paradox of Institutions

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s