Servaas Storm, who’s always worth reading, has posted on the INET website a summary of a new working paper he coauthored. This issue goes way back with me—I first started looking into and writing about the labor rights/wage/trade/development nexus back in the 1980s. Working on my own, I had a lot of false starts, and I’m happy to see others digging much more deeply today.
I won’t comment on the substance of this paper, but I think an important piece is missing: how dual economies articulate, and in particular the role of clientelism.
Countries in which formal sector jobs are highly valuable but scarce, in a sea of abundant but unremunerative informal employment, have to have some mechanism for allocating them. Some classic economic models to the contrary, it never happens through lotteries. My hypothesis, based on what I’ve seen and read, is that the predominant mechanism is clientelism.
A brief digression: Most of the literature on clientelism appears in political science, where it refers to the exchange of votes for personally targeted services or transfers by politicians. I use the term to refer to a much broader phenomenon, the exchange of personally targeted benefits in return for the performance of loyalty between patrons and clients. Patrons have access to resources from which they can supply benefits to clients, while the extent of client loyalty is a determinant (but not necessarily the only one) of how many resources a patron can command. Conceptually, the client-patron relationship is a dyad, although clientelist systems are constellations of such exchange relations across whole populations: many dyads, multiple levels (patrons are clients of higher-level patrons), competing networks.
A large gap between formal and informal employment increases the tendency for clientelism to expand as an allocative system. Clientelism is not all bad—it can moderate frictions that market or formal administrative processes generate—but to the extent it replaces these “modern” alternatives it reduces social efficiency. For instance, allocating scarce formal sector jobs through client-patron exchanges is relatively harmless if the people getting the jobs are no less qualified than those left out of the system, stuck in the informal sector. If clientelist networks override formal qualification (administrative) or competitive performance (market) criteria, however, they degrade outcomes. It’s a matter of degree.
From this perspective, the most important point about labor regulation in a developing country is that it should not exacerbate imbalance, increasing the gap between the formal and informal sectors and loading more weight on clientelist mechanisms. The best forms of regulation are either universal or written to apply at least as strongly at the bottom of the labor market as at the top. I’ve written a bit about how that can be done in the realm of health and safety, and there’s no reason it can’t also guide policy in wage regulation, union rights and all other aspects of labor policy.
For instance, take minimum wages. By definition, these apply only in the formal sector, and if the effect of raising them is to intensify the formal-informal gap, that can be a problem. But there’s a way to avoid this: for every increase in the minimum wage, pair it with an increase in income transfers or similar social protection measures for those outside minimum wage coverage. Hold the gap constant or reduce it. It’s not impossible once you know what you’re aiming at.
My personal experience is that, once you’ve trained your eyes to see clientelism, you notice it everywhere. It’s not confined to low income countries or economic goods. It isn’t necessarily harmful, although, when it metastasises and displaces other social and economic arrangements, it can be deadly. I suspect it is the main factor in differences in x-efficiency, and if true, this makes it one of the main determinants of the wealth of nations. I find it incredible that organizations like the World Bank could dispense reams of development advice without considering how its proposals will pan out in a clientelist world.