The two books under review focus on the politics of the subaltern, but in different ways. Rina Agarwala seeks to understand how informal workers, generally viewed in political economy as hapless victims of capitalism, organize themselves. Tariq Thachil attempts to explain how a political party catering to elite interests can also capture the vote of the subaltern. Both books concentrate on India but frame their inquiry comparatively and theoretically.

Informal workers comprise 93% of India’s labor force. They “construct buildings, build roads, grow and sell fruits and vegetables, clean homes and streets, sew clothes, weld car parts and make shoes—not to mention the boxes they come in” (Informal Labor, Formal Politics, and Dignified Discontent in India, p. 2). They are informal because their work conditions, benefits, and wages are not contractually protected, as is true of the “formal working class.” Because of the nature of their work, their dispersion, and their widespread illiteracy or low literacy, they are viewed in the larger literature as “commodified victims, shorn of agency” (p. 3).

Yet in 2008, with little resistance from parliament, the government of India passed a bill that offered them “social security,” especially health insurance. The overall budgetary commitment was not astronomical, but the symbolic significance of government recognition and absence of parliamentary opposition was substantial. More robust benefits were gained at the state level. According to India’s constitution, labor welfare is primarily a subject concerning the state, not the central government.

How does one explain this nonintuitive development? Agarwala’s explanation leans heavily on the voting significance of this group. These workers are “using their power as voters to demand state responsibility for their social consumption or reproductive needs, such as education, housing, and health care” (p. 15). The author develops this argument on the basis of a four-state comparison: Tamil Nadu, Maharashtra, West Bengal, and Kerala. Her focus, moreover, is on two sectors marked by the predominance of informal workers: construction and tobacco. The latter is also female dominated, which means that its marginality is doubly pronounced: by class and by gender.

Agarwala’s contribution is to show that the welfare of informal workers cannot simply be read out of the market-based economic logic of capitalism. A great deal depends on whether the polity in which capitalism operates is democratic or not, and how party politics are organized. If political parties must compete for the vote of the poor in order to come to power, as is likely in a democracy where a significant proportion of the population is poor, informal workers can exercise their vote to acquire gains that they would not otherwise obtain. These gains may not be large in fiscal terms, but for the poor, each addition in welfare, even if small, is highly valuable. Discussions of worker welfare in the developing world should take note of the nature of the polity and party competition.

Thachil’s ambition is also theoretically and comparatively anchored. Why poor people support political parties that represent the economic interests of the wealthier communities is a puzzle amply discussed in the literature on the developed world. But it remains a black box in the developing world. Combining ethnography, large-n statistical work, and good old “soaking and poking,” as Richard Fenno put it, Thachil enters that black box.

The author’s focus is on the Bharatiya Janata Party (BJP), historically viewed as a party that draws most of its support from the Hindu upper caste. The upper castes are an estimated 16%–18% of India’s population. Clearly, the BJP cannot come to power by courting the privileged, and their affection, alone. It must reach downward, but if it does so, will it not lose the upper caste vote?

Separating Hindu nationalism into its formal political unit and its nonparty organizations turns out to be the key to the resolution of the puzzle. The party concentrated on its elite base, but its affiliated organizations, not contesting elections, provided social services, primarily health care but also education, to the subaltern. India’s government machinery was not doing a good job of reaching the social peripheries in several states. The Hindu nationalist nonelectoral organizations saw an opportunity and seized it.

Thachil also argues—and herein lies his distinctive contribution—that the relationship between these organizations and the subaltern is not marked by clientelism, as the larger literature, especially that emerging from Latin America, would normally put it. There is no quid pro quo involved: “Unlike vote buying . . . such services are not distributed under quid pro quo protocols” (p. 15). Rather, ties of “goodwill, friendship, suggestion and rumor” (p. 27) are built. The expectation is that the subaltern would reward such painstaking work, but the links are not episodic and election specific, nor is the service withdrawn if the voters do not vote as expected. India has for years practiced the secret ballot, and it is impossible to find out who voted which way. We only know which electoral booth favored what party, not how each individual voted. As a result, the possibility of detecting individual defection and inflicting individual punishment is minimal. At best, collective punishments can be meted out. These BJP-related organizations, however, do not even do that. They, as Thachil argues, just keep working. Unrelenting social service, not clientelistic exchange, produces votes.

It is noteworthy that as Elite Parties, Poor Voters was about to be published, something eerily confirming its basic arguments happened. In India’s 2014 national elections, the BJP polled more votes than the Congress Party among three of the biggest subaltern electoral categories: socially, the Scheduled Tribes (Adivasis) and the Scheduled Castes (Dalits), and economically, the poor. The BJP was expected to win a higher proportion of the votes of the upper castes and the rich and of the middle classes, which it did. But it also left the Congress Party far behind at the lower end of the social and economic scale, which was unprecedented. Thachil had already worked for years on how this denouement might come about. He was able to identify the deeper forces and strategies at work.

These strong books are not without problems. But the problems should be put in perspective. No first books, nor for that matter books written at the height of intellectual maturity, are faultless. The best way to understand, and present the problems, is to say that these books resolve some puzzles but leave others unresolved, leaving them for the next round of research, or the next set of researchers. Research is a cumulative process of inquiry.

Agarwala’s big unresolved puzzle is this: If informal workers use their vote to influence politicians, why are they able to do it only in some states, not in others? After all, in an economically poor democracy like India, in which the poor were historically a majority and are still a plurality, most political parties should be gravitating toward the poor. Agarwala’s answer is that the informal workers become electorally meaningful only when political parties compete for the vote of the poor. But that pushes causality one step further. Why would any political party, seeking to capture power in a low-income democracy, not court the poor? We need to ask whether the poor voters always see themselves as part of a poor economic class, or also as members of caste or religious communities that are not defined by economics.

Some of Thachil’s arguments have an eerie fit, as I said earlier, with the 2014 election results in India. But the 2014 elections also pose a huge problem for him. One of the states he studied, Uttar Pradesh (UP), India’s biggest, having a population of more than 180 million, was, according to Thachil, a state where the services of nonelectoral organizations, affiliated with the BJP, failed to produce votes. The workers of these organizations “openly admitted that they had been unable to translate the goodwill they accrued through service work into electoral results” (p. 222). He adds that for the lower castes, these workers were always suspect because they came from the upper castes. What happened in central India, therefore, could not be repeated in UP.

It turns out that in the 2014 elections, the BJP’s biggest success was precisely in UP, where it won 73 of 80 parliamentary seats. The BJP would not have come to power in Delhi without its UP performance. Realizing the gravity of the question, Thachil wrestles with it in the “afterword” of the book. In general, he rightly says, the national results tallied with his argument, but “the most significant discontinuity came in Uttar Pradesh” (p. 283). He hypothetically attributes the results in part to Narendra Modi’s personal popularity and in part to the Hindu–Muslim polarization that the BJP campaign engineered.

Social science books should not be judged entirely on predictive criteria. We are not a deterministic, but a probabilistic, science. Our predictive powers can be overestimated. So UP alone cannot be held against Thachil’s impressive research and arguments. But it does pose a larger intellectual question. Voting is a multidimensional exercise. Social services alone, or, for that matter, economic or cultural issues alone, cannot be expected to decide elections once and for all. Why exactly one set of issues triumphs over the other remains a puzzle to be resolved, not simply in India but elsewhere, too.

This problem has two sides. First, voters must be viewed as members of an economic class, as well as members of social or cultural communities, constituted by race, religion, language, caste, tribe, and so on. As human beings, we have multiple selves. Our electoral behavior cannot be reduced to our economic interests alone. Identities can, and do, matter. Second, a trickier problem is trying to sort out when we vote our economic interests and when our cultural/social/religious identities triumph over them. Political leaders repeatedly tell you about the multidimensionality of the voting public. Political science is yet to respond adequately to the analytic complexity posed by this multidimensionality. (For a larger theoretical statement of this problem, see Ashutosh Varshney. “Nationalism, Ethnic Conflict and Rationality,” Perspectives on Politics 1 [no, 1, 2003]: 85–99.) As far as Indian voting per se is concerned, religion and caste are recognized to have been the “master narratives” of politics for over half a century, not class. (For the impact of religion on politics and voting, see Pradeep Chhibber, Religious Practice and Democracy in India, 2014.) More carefully grounded empirical studies of the kind reviewed here will provide building blocks of a better understanding.