Nigeria’s finance minister, Kemi Adeosun, speaks at a news conference in Lagos on April 9. (Akintunde Akinleye/Reuters)

Like most of Nigeria’s states, tiny Osun in southwestern Nigeria relies a lot on transfers of oil revenues from the federal government. In 2014, these transfers accounted for more than 85 percent of the state’s annual budget. While signing into law a new bill that expanded the state’s collection of property taxes earlier this month, Osun Gov. Rauf Aregbesola said that the fall in oil revenues showed that “living on other people’s bakery is never reliable. It is high time we started to bake our own cake on our own, if we don’t want to starve.”

Osun is not alone. Across sub-Saharan Africa, national and local governments are more interested in tax collection now than at any point since independence. Yet there are very real obstacles to building domestic tax bases in Africa. For decades, African governments have relied on natural resource rents, aid, and easy-collect customs and excise taxes. They rarely have the bureaucratic capacity to assess tax liability and enforce compliance.

Tax-seeking governments must also bargain with citizens who distrust state institutions and have spent decades evading “capture.” More than 80 percent of Africa’s workforce is employed in the informal or “off-the-books” economy, which accounts for 55 percent of the region’s gross domestic product.

In Nigeria, tax evasion also extends far into the official economy. In 2013, Nigeria’s minister for economy and finance estimated that 75 percent of officially registered firms did not pay corporate taxes.

How can a government build a tax base in this environment? If the government is not strong enough to punish those who evade tax payment, is the collection of direct taxes a lost cause?

In a new article in the British Journal of Political Science, my co-author Cristina Bodea and I draw on public opinion surveys from urban Nigeria to look at when and why citizens become willing to pay taxes. We argue that citizens’ beliefs about taxation are especially important in countries where social contracts between governments and citizens are still in flux. Encouraging pro-tax attitudes — or what might be termed “voluntary compliance” — is likely to be a quicker and less costly route to tax payment than investing in stronger bureaucracies and enforcement.

Even in a place like Nigeria, where tax evasion carries little real stigma or expected penalty, we found evidence of the beginnings of voluntary compliance. Only 42 percent of those we polled believed in a citizen obligation to pay tax, and those respondents were 26 percent more likely to have actually paid taxes. Intriguingly, neither tax attitudes nor tax payment is associated with class, formal sector employment, ruling party membership, ethnicity or any of the other factors that we might expect.

So why do citizens adopt pro-compliance beliefs? We identify two plausible explanations. First, delivery of public services — e.g., water, roads, clinics, etc. — matters. In our survey, those who felt the government was using its revenue wisely and those who had concrete experience of services delivery were more likely to feel obligated to pay taxes. This is consistent with a vast literature in political science and economics that views enduring social contracts as evolving out of this kind of taxes-for-services exchange.

But social capital matters as well … and perhaps not in the way we would expect. Robert Putnam has famously argued that a robust civil society can boost government performance and help economic development. Is it possible that this relationship requires a level of state capacity and reach that does not yet exist in the developing world? Might robust social capital sometimes prevent the formation of strong states that can deliver?

Residents of Nigeria’s crowded and fast-growing cities have very different levels of access to social capital and community-provided services. In some places, communities have been able to overcome their differences and figure out ways of providing for security, contract enforcement, and access to credit on their own. We found that residents who had access to these kinds of non-state services were far less likely to believe in a citizen obligation to pay tax. In contrast, those who did not have access to this kind of social capital — including those living in violent and riot-prone neighborhoods — were far more likely to believe in paying taxes.

This is intriguing evidence that tax-seeking governments must do more than provide services if they want citizens to pay taxes. They must also grapple with existing non-state actors who already provide some services to citizens. Does this mean that states must crush their rivals to ensure societal compliance? Recent tax collection success in Lagos, Nigeria, might suggest an alternative pathway.

The Lagos state government has increased its tax revenue more than eightfold since 2003. In stark contrast to other Nigerian states, internally generated revenue accounts for more than 75 percent of the state’s budget. In order to accomplish this task, the state government has entered into arrangements that might be termed “tax farming,” or the use of non-state actors to collect tax on behalf of the state. Given the vast size of Lagos’s informal sector, reliance on intermediaries — notably market associations — has allowed Lagos state to “borrow” the stronger societal roots of these organizations to expand tax collection. The state government has also started to use billboards, the radio and other public campaigns to build stronger beliefs of citizen obligation to pay tax. By mixing organizational innovation and public campaigns that tie tax payment to civic virtue, Lagos’s tax model is a powerful one for other states in the developing world to consider.

Adrienne LeBas is an associate professor of government at American University and a residential fellow of the Woodrow Wilson International Center for Scholars in Washington, D.C. Follow her on Twitter at @amlebas.