Not only has it scored at or near the bottom of Transparency International's corruption perception index, but Bangladesh also scores in the bottom quartile of many of the Worldwide Governance Indicators. Yet this same country has seen its per capita income rise by one percentage point every decade (it is now close to 5 percent). It has already achieved universal primary enrolment, and an equal number of girls and boys in secondary school. It is on track to reducing child mortality by two-thirds (relative to 1990 levels) by 2015. Finally, Bangladesh reduced poverty by eight percentage points in five years—a rate of poverty reduction that is twice as fast as India’s.

What can explain this apparent paradox? One response is that the data about Bangladesh’s development successes are inaccurate. This reminds me of a poster a colleague had in his office: “If the facts don’t fit the theory, change the facts”. But anyone who has studied or visited Bangladesh knows that the growth and dynamism of the place is real. Another response is that the governance indicators are mis-measured. True, alongside these indicators of high corruption is a vibrant free press, an active NGO community, and since 1991, democratic elections.

A third way of resolving the paradox is to recognize that the relationship between governance and development is complex. The same level of corruption could do a lot of damage in one country, and less in another. Even in the same country, corruption could be more or less harmful at different times. In the case of Bangladesh, the main sources of growth and human development arose by going around the government. The garment export industry grew thanks to duty-free textiles, which bypassed the customs department, where there was considerable corruption. The improvements in health have a lot to do with the work of NGOs in delivering contraceptive and other services that helped reduce Bangladesh’s fertility rate below 2. Girls’ enrolment in secondary education is helped by a scholarship scheme that also gives a stipend to the school—be it public, private, NGO-run or religious (madrassa)—depending on the number of girls enrolled. Finally, the reduction in poverty is associated with the widespread use of microfinance and remittances from Bangladeshi workers in the Middle East.

Of course these explanations beg the question: how did Bangladesh manage to introduce these programs that got around government? Why haven’t other countries been able to do so? These questions are hard to answer, but let me offer some conjectures (developed in conjunction with my colleague Binayak Sen). First, when Bangladesh was formed in the aftermath of the 1971 liberation war, there was hardly a government. The international and domestic NGOs filled the vacuum in delivering services. Second, Bangladesh is a densely populated, homogeneous society. In such a configuration, innovations spread like wildfire. So family planning, microfinance and other programs took off in Bangladesh more easily than they might have elsewhere. Finally, by the time the government got strong enough to control the NGOs and others, it was too late: microfinance, family planning and private schooling had already become commonplace. To its credit, the government recognized this and proceeded to support the providers of essential services with finance. The net result is a country with significant levels of corruption but where development has progressed at an impressive rate.

Can it last? Many observers felt that weak governance would constrain Bangladesh in the future, even if it hadn’t done so in a major way in the past. Evidence of this is the acute power shortage that the country faces. Until very recently, Bangladesh has not had an independent power producer investing in the country since 1999 or so. One of the obstacles was the high level of corruption. Furthermore, the current caretaker government is pursuing an active anti-corruption campaign, including arresting some high-profile individuals on charges of corruption. How these developments will play out in the future is unclear, but one thing is certain: Bangladesh’s unique development path has forced us to rethink any simple-minded notions of the relationship between governance and development.