New working paper (in French): Social networks and socialism

This new piece of research studies the role of social networks in a “socialist” economy where access to goods is rationed (in the paper, this is because of a price cap, but the argument may apply to other forms of regulations). In the model, social networks develop as an alternative device to access a trading opportunity for those goods. As a result, people overinvest in social networks — at the margin, they make friends with people they do not really like, or spend too much time with their relatives, because such networking may provide useful consumption opportunities.
The most interesting result is that regulation may be sustained as a self-fulfilling political equilibrium. This may hold provided some people benefit from it. Here regulation maintains prices below their market clearing level. It benefits the “poor” because they have to pay less for the goods. It harms the “rich” whose income is lower in the regulated society (in the model it is because they own shares in the firms that produce the rationed goods). These “rich” include the “average” individual, who is also harmed due to the price distortion brought about by regulation. Only poor enough people benefit, and that is only because more efficient forms of redistribution are not available.
If people’s ability to invest in their social network is heterogeneous, the expectation of a regulated economy will lead to the emergence of a social group who is well equipped in social capital (and therefore expects to do well in a regulated economy), but with relatively low income. This social class benefits from regulation because they have a privileged access to the goods that are rationed, thanks to their superior social network. They benefit from low prices while not suffering too much from rationing. On the other hand, they are not poor enough to support regulation should they have not invested in their social capital. Therefore, these people validate expectations: if it is anticipated that the economy will be rigid, they invest in social networks and, because they are winners in the social networking game, end up voting in favor of regulation. If it is anticipated that markets will be left unregulated, they do not invest in social networks and vote against it. If this group of people is large enough, both rigidity and flexibilty are self-fulfilling outcomes.
The argument applies to social-democratic societies where a wide range of publicly provided goods are subjected to a price cap (or even free) and rationed. We may mention access to social housing, day care, medical care, summer camps, pension homes, music conservatories, and good schools. The model tells us that the system is supported by a social class of people who are relatively modest economically but well positioned to access those goods thanks to these connections. This article by Eric Le Boucher laments how opacity and complexity in the French higher education system ultimately favors the well-connected (in particular high school teachers) who know how to get into the right track. Indeed, high school teachers are modestly paid, but they are in an excellent situation to use their connections to get their children in the right place. And they have more free time than others that they can use to develop connections and gather information. They match the caracteristics of the social group I have just described.
Along similar lines, this article by Capital points out the pervasiveness of favoritism in the allocation of social housing in France; while sheer bribing is rare, social connections (and affinities) with the key decision makers is essential. One does not need to be miserable to be eligible for social housing. According to the article, 50 % of households in the Paris region are eligible — the proportion of social housing is less than 20 %. It is easy to figure out that among those 50%, the more connected and least poor among the eligible would be better-off in a free market system where they would not have spent as many resources to establish connections. But these resources are sunk costs and they end-up locked in in a situation where they gain from rigidity.


On the art of clientelism

Besides the train strike (which is basically over), there is another protest movement by the so-called “intermittents du spectacle” (henceforth “intermittents”). These are employees of the cultural/media industry who enjoy a specific unemployment benefit system. Essentially this system allows them to move back and forth between temporary employment and unemployment benefits, while other temporary workers (say in the tourism industry) would have to spend a larger fraction of their time in employment in any given year to be eligible for benefits. Beyond the eligibility rules, being on benefits has become a way of life for the intermittents. As a result, their specific unemployment benefit system has an annual deficit of about 1 billion euros, which is about the same, for 250.000 contributors, as that of the general system, which covers 16 million people. This is a transfer from the rest of society (including poorer people and people who have far less enjoyable jobs) to the intermittents.

As a result, the social partners, who are supposed to negotiate the financing of the welfare state, have regularly tried to cut on these conditions. In general there has been enough consensus between unions and employers on these reforms, because the intermittents are perceived by regular working class employees as privileged, while employers want to cut on labor costs and social security contributions. Thus the main private sector labor union, CFDT, has generally gone along with the employers’ proposals to cut the benefits. By contrast, the CGT labor union which represents mostly public sector employees is opposed to the cuts, and is active in organizing the intermittents. This is rational: While the CFDT median voter has to finance the intermittents scheme by paying higher contributions to the unemployment benefit system, public sector workers have job security and therefore do not contribute to the system.
In 2003 a reform was implemented and met by protests that perturbed a number of summer cultural festivals, up to the point that a sacred cow of French cultural mythology, the Avignon theater festival, had to be cancelled.

The same scenario is being played out right now as another agreement by the social partners has managed to put some mild cuts into place for the intermittents’ specific system. The movements should be less popular than ever, given the punitive tax increases that have been inflicted on most of the population since 2011. Nevertheless the Valls government has already yielded to the intermittents by transferring part of the costs to the tax payer; the agreements between the social partners is, then, purely cosmetic. The financing of the system is being put into the general government budget instead of the “social accounts”.

[Quite remarkably, exactly the same trick was played after the 2003 reforms. But you have to read the austere report of the “Cour des Comptes” (a committee of Cassandrae who open closets and lift carpets, then write reports on what they have found, and whose wise recommendations are invariably ignored by the politicians) to figure out how. The 2003 reform made eligibility rules stricter. But a special fund, financed out of the government budget, was put in place to guarantee a smooth landing for those who were no longer eligible. The fund was supposed to be temporary (like the jobs of the intermittents), but never trust a temporary measure.]

The existence of the intermittents is a remarkable example of political clientelism. Over the years, the eligible population has been multiplied by 10. This reflects the explosion of “cultural” spending by successive governments. Short-time unemployment compensation is a way of life for most people involved in “spectacle vivant” (theater, dance, clowns, happenings) and is also used by television and radio channels, including the State-owned ones, to pay their technicians.

The rise in the number of people working in the “cultural” sector, maintained by subsidies at an artificially large size, is a way for the political left to recruit its own electorate. The intermittents are a captive electorate, because the excess size of the entertainment sector puts them constantly under theat. They know that a substantial fraction of them would have to relocate to other industries (and lose part of their human capital) should the subsidies be cut. They also know that employers, and more generally union representatives for private sector employees, have targeted the intermittents benefit scheme as a priority for implementing cuts in social spending. This means that the threat they face is associated with the political right, implying that they need the left to be in power so as to protect their rents.

Distributing rents is not sufficient for successful clientelism. If the beneficiaries of the rents think that the opposite political party will treat them just as well, they have little incentives to reward your clientelism with votes. The rent must therefore be designed in such a way that your political opponents will try to eliminate it. By having the intermittents UB scheme being financed out of the UB system instead of the general budget, one makes the cost to other workers more apparent. That is, as you create the rent, you create a constituency against that rent at the same time. And that constituency is important in convincing your clientele that your opponents will attempt to eliminate their rent. The price to be paid for it is that the scheme will come regularly under attack; but these attacks secure loyalty among your supporters.

While the rent has to be fragile, it has to be resilient at the same time. You will need your clientele as an army to reconquer power if your opponents win the next election. But this army will not be around should the next government eliminate the scheme by a stroke of pen. To prevent that, make sure that the rent is not a discretionary subsidy scheme. Instead, it should be embodied in some form of entitlement.

The clientelism scheme will also work better if you can recruit more people into it. This means that the rent should be relatively small, so as to save resources for raising the number of beneficiaries. And if your clientele is an occupation or an economic sector, entry into it should be relatively easy (of course entry cannot be costless because that would simply eliminate the rent). Indeed many of intermittents are relatively mediocre artists who would have elected a different occupation in the absence of the scheme. Lowering standards and promoting amateurism goes hand in hand with the ability to enroll many people. Conversely, one would not go very far politically by promoting “excellence” in the arts instead.

Finally, make sure to “leverage” your clientelism by designing it so as to enroll indirect supporters in addition to direct ones. Suppose you are the mayor of a small city; you get more support by giving municipal jobs to members of large families than small families. By hiring one employee, you not only make him happy, but also his parents, sister, cousins, and so forth. In that respect, targeting the entertainment industry is a clever idea, since they can potentially enroll the audience. By targeting artists and intellectuals for your clientelism, you also achieve leverage. By virtue of their profession, these people influence public perceptions; and a dog does not bite the hands that feed him. Indeed each election a cohort of artists and intellectuals duly endorse the left in the media, while the right has to content themselves with a handful of athletes, movie stars and popular singers (who in fact reach out to far more people than the artists and intellectuals, but the point here is that many of the latter benefit from government subsidies).

I can only speculate about why over the years the right-wing governments have proven impotent in reducing the size of a scheme which, in addition to being considered unfair by many people, runs against them politically. Perhaps an explanation is that the educated bourgeoisie, who consume a substantial amount of “elite” cultural products, does not want their music and theater festivals to be sabotaged. This is a form of hostage taking but the hostages seem to be struck by some sort of Stockholm syndrome, perhaps rationally so. They contribute more to the system than they get, but as consumers of cultural products they do get part of their taxes back in the form of low prices and an abundant supply. They expect to lose more from a less profligate and more expensive cultural sector than they would gain from reduced social security contributions if the cuts were to be implemented.