A recent Carnegie Endowment for International Peace report, authored by Minxin Pei, claims that using a conservative estimate, kickbacks and theft account for about 10 percent of government spending and transactions. This suggests, according to the report, that corruption and bribery costs China at least $86 billion a year.
I assume that Pei’s figures do not include corruption involving the private sector. This can be pretty large. One of my Peking University students told me last year that his job (he was very apologetic in describing it) consists of making friends with the heads of computer services at the hundreds of banks in China so that the small company he works for, which creates ATM-related software, can sell their products.
Why is this friendship important? Because the only way to get one’s products purchased, according to him, is by paying off the head of computer services, and since the person receiving the bribe needs assurances that the bribe-offerer is not working for the police, he will only accept bribes from people he knows well. My student assured me that all his competitors have someone filling a similar function.
According to the Carnegie Endowment report corruption is concentrated in areas with extensive state involvement, such as infrastructure projects, real estate, government procurement and financial services. I don’t know how good Pei’s figures are but if they are reasonable, they probably understate the cost of corruption significantly since they focus on public sector corruption, and as my student’s story suggests, corruption is widespread in the private sector too. $86 billion is a high number – about 3-4% of GDP. China, by the way, was ranked 72 on the Corruption Index among the 180 countries ranked in 2007 by Transparency International.
Although corruption is always harmful to an economy, when thinking about its economic impact we should distinguish between different kinds of corruption. If corruption is stable and predictable it can nonetheless allow businessmen to continue to develop their businesses, employ workers, and create value. In that case corruption is merely a kind of tax – inefficient in that it distorts business activity, like all taxes do, and of course even less likely than normal fiscal expenditures to be put to socially beneficial uses, but it need not significantly harm economic growth prospects. This is not to say anything about the political and social impact of corruption, which of course may be highly destabilizing.
However there are at least two cases in which corruption can be much more harmful to a country’s economy. If corruption is random and unpredictable, it can raise business uncertainty enough to cause a significant drop in investment. As a Brazilian friend of mine told me, before the presidency of Collor de Melho in the late 1980s he used to know who he needed to bribe in order to get approval for projects, access to water and electricity, roads maintained, and peace from the unions. He hated to make the payments, but at least the rules were clear and he could plan his expenditures with reasonable certainty.
Under Collor, however, things became so random and government officials so rapacious that he was never sure who he had to pay next, and since once he had made a significant investment the corruption escalated (because he had already made a big enough commitment that it made sense for officials to bleed him dry), he decided to stop investing altogether and to take his money out of the country. When corruption significantly increases the uncertainty under which businessmen operate by disrupting the rules of the game, they disinvest.
The second case is when corrupt officials discount the future at very high rates (perhaps because of uncertainty about their period in office, or factional fighting, or political instability). This changes their attitude towards long-term development. A “rational” corrupt official wants to maximize his economic rent (I love this value-neutral term for corruption) over the long term, so it is in his interest that local businessmen over whom he has power are successful and their profits grow over the long term. The bigger and more profitable they are, and the longer the bribe-giving relationship, the more money the official receives. In these cases long-term growth is very compatible with official corruption, and in fact it is pretty easy to think of cases that exhibit this kind of “far-sighed” corruption.
In cases where the corrupt official has reason to question the stability of his stay in power, however, what matters is to maximize rent in the short term, not in the long term. In that case he is more interested in looting the company of whatever assets it has that can be easily liquidated, even at the risk of killing the company. These are particularly harmful cases because the bribe-offering managers also know that the risk to the company, so presumably they also become much more interested in looting assets than in building a profitable business. This kind of corruption (along with the former) was characteristic of the last few years of Nationalist rule in China.
I have no idea what the mix of corruption in China is information celebrex, but I suspect that we are largely experiencing the “good” kind of corruption (blog readers please disagree if you think otherwise). It is not just the current mix that is important, however, but also how things can change. For example, if factional fighting becomes bitter and so forces officials to discount the future at higher rates (because it is harder to predict how long they will retain power), I would imagine that we would see more of the latter kinds of corruption develop. In a way the fight against corruption, if it ever becomes effective (which I doubt), might even have the perverse consequence of increasing the looting mentality because corrupt officials will be concerned about getting enough money out of the country as quickly as possible before they get noticed and caught.
Most importantly we should understand how a great deal of corruption, even of the “good” kind, can add instability by creating a self-reinforcing mechanism that amplifies decline. By this I mean that in case where there is any sort of political or economic crisis, of the threat thereof, it may force officials to change the type of corruption – as Color de Melho’s presidency in Brazil in the late 1980s did – so that it amplifies the decline. In that case the danger of corruption is that, probabilistically speaking, it fattens the tails of the distribution of potential outcomes and, as every finance geek knows, fatter tails raise financial distress costs. In fact maybe we can think of corruption is a type of financial distress cost – the economic cost of corruption rises with the probability of crisis.
; information celebrex