What do governments do? What do we want them to do? On the heels of last month’s state assembly elections and ahead of a national election in a few months time, it’s worth considering these questions seriously. The public mood in India is notoriously anti-politician, admittedly with excellent reason given our candidates’ length and vivid criminal records and shameless displays of corruption. But our disdain for the individuals – inadequate though they might be – who occupy the ahlls of power ought not to make us naïve about the importance of government to the lives of hundreds of millions of ordinary citizens. It might be fashionable in some circles to quip that ‘India grows at night’, or to argue that governments, rather than being the solution is the problem, but such clichés ignore two immutable truths of politics: 1) citizens desire a vast array of public services that governments are best suited to provide on a national scale; 2) the provision of such services has distributional consequences as those benefitting most from government are often the least well-off, while the relatively privileged, conveniently oblivious of the myriad ways they benefit (and have benefited) from government policy, complain about high taxes and the burden of corruption.

Middle-class concerns about taxation and corruption have spawned the twin agendas of neoliberalism and good governance. The first seeks to limit the scope and extent of state activity by draining the treasury. The second offers technocratic solutions to policy problems in a quest to maximize efficiency, Both have their virtues, of course, but neither is an appropriate solution to the problems facing developing countries. Indeed, I would argue they are counterproductive in most cases.

Pradeep Chhibber (University of California, Berkeley) and I, analyzing electoral returns from state assembly elections across the 15 largest Indian states from 1967 onwards, found that states characterized by limited ‘fiscal space’ were more likely to experience higher levels of electoral volatility and greater anti-incumbency swings. What is fiscal space? Simply put, it is the share of the state budget available to governments to spend on public policy initiatives after accounting for their fixed commitments, such as salaries for the civil administration and policy. What affects a state’s fiscal space? Well, obviously, overspending on bureaucrats raises expenditures, but, just as acutely, limited tax capacity imposes a constraint on the spending side. To improve fiscal space, the Indian states collectively must improve their woeful tax extraction rates, but such an emphasis lies at odds with a neoliberal agenda that stresses the shrinking of the state budget as a primary goal. Good governance advocates attack the state from a different angle, advocating the delegation of many of the tasks primarily performed by states to private agencies. These agencies are thought to be more efficient and less prone to the rent-seeking that undermines our public institutions. I do not doubt that the private sector can do things more efficiently than state governments, but that misses the point. What are the consequences of reducing the role of the state in citizens’ lives? One plausible effect is that it hampers the states’ ability to build performance-based legitimacy amongst its citizens. When citizens associate high-performing services with the outsourcing of those activities from the government, they will also allocate any credit for that performance to the private-sector actor with which they interacted and not to the state.

Our understanding of the consequences of such dynamics is still limited. In a project co-authored with Thomas Edward Flores (School of Conflict Analysis and Resolution at George Mason University), I argue that the neoliberal-governance agenda has reduced the democratic dividend of national elections worldwide. We assemble a global data base of national-level elections across all nations, and analyze whether holding an election helps build democracy. The data are sobering. Across all elections in the developing world, the democratic dividend of elections, which we define as the gain in democracy following an election, is virtually zero, and, in some decades and regions of the world, negative. Simply put, elections are more often than not counterproductive from the perspective of deepening democracy.

The goal of our project is to explain variation in the democratic dividend of elections. We have identified several factors, including the country’s democratic institutional stock and the absence of violent civil conflict, that boost this dividend. But the primary factor that reduces the size of the democratic dividend is a limited fiscal space. For our cross-national analysis we define fiscal space simply as the country’s per capita tax revenues. The results are striking. Even when we control for a host of plausible theoretical correlates and allow for country fixed effects which should account for any unmeasured factors, countries with limited fiscal space have negative or null democratic dividends, while countries with greater fiscal space have positive dividends. Having money to spend on public policy changes the terms of the debate on which elections are contested. The 2012 US presidential election illustrates my point. In that election, contested by the incumbent Barack Obama and the challenger Mitt Romney, a major issue was the Affordable Care Act (aka. Obamacare), which promised a wholesale reform of the US health care system. The debates were not about ethnicity-based reservations, about communal favoritism or discrimination, or, indeed, about any identity-based issues. They centered on a major policy initiative possible only in a country that has invested significantly in its state capacity and that boasts high tax extractive capacity. But, in the majority of the developing world, where building a health care system is inconceivable given low state capacity, elections are contested instead on the basis of caste, clan, religious, and tribal loyalties. The consequences for democracy over the long-run of these divergent strategies are profound.

In making the arguments above, I do not seek to “throw the baby out with the bath water.” Reform movements to reduce corruption and increase efficiency of the state are important, but they cannot be the only goal. We must first build a state before we can perfect it.

*Irfan Nooruddin teaches political economy at The Ohio State University.