Conventional wisdom is that governments in the developing world lack the budgets to fund and train officials to stop street vending. Yet many cities with the resources to enforce, like Santiago, Chile, still have rampant street vending problems in some neighborhoods.
From a political perspective this is not surprising, however, because enforcement has important distributive consequences. Street vendors don’t pay business taxes and rents to sell goods in busy areas. Governments provide a substantial—although informal—form of welfare when they allow street vendors to work unchecked. In Lima, Peru, for example, vendors say that the ability to sell on central city streets is worth about $1,560 per year to them. In contrast, the largest formal welfare program in the city transfers only around $115 to poor families.
In a forthcoming article in the American Journal of Political Science, I show that politicians in Latin American cities make choices about informal welfare benefits, such as the tolerance of street vending, in much the same way that they do about formal welfare provision in more advanced democracies. Where mayors need the support of poor voters to win elections, they are more likely to let vendors stay on the streets, even if they anger middle-class groups.
I use three types of evidence to show that politicians deliberately forgo enforcement against street vending to provide informal benefits and win elections. First, I expect that politicians who draw their main support from poor voters are less likely to enforce street vending regulations. To see this, take the case of Bogotá, Colombia. The bars in the figure below show the number of operations against vendors by mayoral administration. Starting in 2003, mayors have drawn their strongest electoral support from poor voters. Enforcement has plummeted.
Of course, it’s possible that the conventional wisdom is right and that these days the Bogotá city government can’t control street vending due to resource constraints. However, the budget allocated to the police has increased in recent years (the solid line). Even with greater funds to put toward enforcement, mayors have been explicit in their rejection of operations. As former mayor Luis Garzón put it, he was “not going to tell the police to repress.”
Second, politicians who run for office in districts with large concentrations of poor voters should be more sensitive to the welfare effects—and thus enforce less. For instance, in Lima, Peru, each district in the city elects a mayor who directs operations against street vendors. Lima also is a socioeconomically segregated city. Districts on the urban periphery have heavy concentrations of poor voters, shown in dark blue on the map on the left. These districts do fewer operations against street vendors (the light blue on the middle map).
This spatial pattern is surprising when we take into account the number of street vendors that work in poor areas of the city. Just as the police might do more patrols where more crime occurs, we expect to see more operations where there are more street vendors. Against such natural expectations, poor districts do less enforcement even though they have large—in fact, often larger—concentrations of street vendors than wealthier areas (the right map). Admittedly, resources can be scarce in poor districts. But, I see the same trends even in cities like Santiago, which rely on a single well-trained police force citywide.
Lastly, if non-enforcement functions as a type of informal welfare system, then politicians and citizens should think of it as they do other social welfare programs. I interviewed dozens of local politicians and found that more than two-thirds avoid enforcement due to its “social costs.” They also worry about appearing insensitive to the poor’s needs. Chilean politicians, for example, feared that enforcement would make them “worse than Pinochet.”
Politicians’ concerns are real. As a part of an ongoing book project on informal welfare policies, I ran a public opinion survey and found that candidates who propose to enforce against street vendors are seen as unlikely to favor the poor’s interests in office. Poor voters also say that they are less likely to support politicians who enforce than those who let vendors work unchecked. The takeaway is that enforcement has political costs. Not all politicians are willing to take this hit.
Although it may come as no surprise that politicians have a hand in how street vending regulations are enforced, my theory of “forbearance,” or deliberate non-enforcement of the law, has different policy implications from the conventional view. Policymakers often stress the need to strengthen the police or streamline bureaucracy to limit activities like street vending. But, if politicians don’t want to enforce, then strong, well-trained bureaucracies might not get very far. My research suggests that the removal of political discretion, or increases in in-kind welfare provision, is more likely to boost enforcement.
Attention to the welfare consequences does raise the question of whether greater enforcement should be the goal in many developing countries. Street vendors from São Paulo to Tunis argue that until governments can offer real employment alternatives, they should be allowed to work. Looking the other way may be the least that local politicians can do for the poor.
Alisha Holland is a junior fellow at the Harvard Society of Fellows. She studies the comparative political economy of development with a focus on urban politics and Latin America.