I just returned from India, where I was lecturing to the Indian Parliament in the same hall where U.S. President Barack Obama had recently spoken. The country was racked by scandal. A gigantic, ministerial-level scam in the cellphone sector had siphoned off billions of dollars to a corrupt politician.
But several of the MPs had also been taken aback on discovering that, when Mr. Obama spoke to them, he read from an "invisible" teleprompter. This had misled his audience into thinking he was speaking extemporaneously, a skill highly regarded in India.
Both episodes were seen as a form of corruption: One involved money, the other deception. The two transgressions are obviously not equal in moral turpitude. But the Obama episode illustrates an important cross-cultural difference in assessing how corrupt a society is.
Transparency International and, occasionally, the World Bank like to rank countries by their degree of corruption, with the media then citing where each country stands. But cultural differences between countries undermine the legitimacy of such rankings - which are, after all, based on surveys of the public. What Mr. Obama was doing was a common enough practice in the United States (although one might expect better from an orator of his ability); it was not so in India, where such a technique is regarded as reprehensible.
India certainly has corruption, like almost every other country. But India also has a culture in which people commonly assume that everyone in public life is corrupt unless they prove otherwise. Even a blind man will tell Transparency International: "I saw him take a bribe with my own eyes." Indeed, a distinguished Indian bureaucrat, a man of unimpeachable character, once told me that his mother had told him: "I believe you are not corrupt only because you are my son!"
So, if you ask Indians whether their governance is marked by widespread corruption, they will answer with gusto: Yes! But their exuberance biases India's global ranking relative to more empirically minded countries.
A similar bias arises from the occasional tendency to view political patronage elsewhere as being more corrupt than the same practices at home. For example, when the East Asian financial crisis broke out, there followed a systematic attempt to pin the blame on the affected countries: "Crony capitalism" allegedly had somehow crippled their economies. In other words, the acquaintances and benefactors of the East Asian leaders were "cronies," whereas those of U.S. leaders were "friends"?
In fact, it was clear that the culprits were the International Monetary Fund and the U.S. Treasury, which had encouraged a shift to capital-account convertibility without understanding that the case for free capital flows was not symmetrical with the case for free trade.
But where substantial corruption can unambiguously be found, as it often can, one must recognize that it's not a cultural given. On the contrary, it's often the result of policies that have fed it.
In the 1950s, India had a civil service and a political class that were the envy of the world. If that seems shocking today, the loss of virtue must be traced to the all-pervasive "permit raj," with its licensing requirements to import, produce and invest, which grew to gargantuan proportions. High-level bureaucrats quickly discovered that licences could be bartered for favours, while politicians saw in the system the means to help important financial backers.
Once the system had taken root, corruption percolated downward, from senior bureaucrats and politicians, who could be bribed to do what they weren't supposed to do, to lower-level bureaucrats, who wouldn't do what they were supposed to do unless bribed. Clerks wouldn't bring out files or get you your birth certificate or land title unless you greased their palms.
But if policies can create corruption, it's equally true that the cost of corruption will vary with the specific policies. That cost has been particularly high in India and Indonesia, where policies created monopolies that earned scarcity rents, which were then allocated to officials' family members.
Such "rent-creating" corruption is quite expensive and corrosive of growth. In China, by contrast, corruption has largely been of the "profit-sharing" variety, whereby family members are given a stake in the enterprise so their earnings increase as profits increase - a type of corruption that promotes growth.
In the long run, of course, both types of corruption are corrosive of the respect and trust that good governance requires, which can undermine economic performance in its own right. But that doesn't absolve us of the responsibility to define corruption properly - and to acknowledge obvious and important cultural differences in how it's understood.
Jagdish Bhagwati is a professor of economics and law at Columbia University and a senior fellow in international economics at the Council on Foreign Relations. He is the author of Termites in the Trading System: How Preferential Agreements Undermine Free Trade .