The Hindu Business Line : On corruption and its `sister activity'

DUKE Vincentio says in Measure For Measure, "I have seen corruption boil and bubble till it o'er-run the stew." Which is what we saw over the week, in fair measure, as stories about how corruption helped many babus make so much money that the dough simply boiled and bubbled till it overran the till and took the form of hotels and land.

The accounting field is no exception . There is this latest accounting scandal in the US about one of two government-chartered mortgage finance companies, Fannie Mae, as AccountingWeb (www.accountingweb.co.uk) informs. The site mentions a Dow Jones Newswires story that spoke of `new and pervasive accounting violations', allegedly involving acts such as "embellishing the company's earnings by overvaluing assets, under-reporting credit losses and misusing tax credits".

Meanwhile, trouble is brewing also at the postal and logistics company TNT, informs AccountingWeb, referring to the ongoing tax probe.

Deciding to catch up later with the `twists and turns' of bad news that seems to come in pairs, I turn to a September 2005 research paper from NBER (National Bureau of Economic Research, www.nber.org) titled `Bribery: Who Pays, Who Refuses, What Are the Payoffs?' by Jennifer Hunt and Sonia Laszlo. "Many development economists fear that corruption reduces equity as well as efficiency, constituting a regressive tax, causing the poor to be excluded from public services, and skewing growth in favour of the rich," notes the intro.

"Most empirical research on the causes and consequences of corruption is not only conducted at the macro (country) level, it is almost exclusively based on perceptions of corruption, rather than actual, measured corruption," observe the authors, before plunging into `newly available data from Peru' to analyse `bribery-related interactions between public officials and households', and more importantly `to measure the burden of bribery across income classes'.

Where did the data come from? Hold your breath, the data came from the Peruvian statistical agency. It had included `a module on bribery in the 2002 and 2003 national household surveys', after petty corruption (meaning `bribes paid in the course of daily life') became a major worry, and the country was a middle ranking on country in Transparency International's Corruption Perceptions Index (CPI).

This is how the module worked. "One respondent per household (half are the household head) is asked numerous questions pertaining to the household's use of 21 different types of officials," as the paper informs. "If a particular type of official was used in the last twelve months, then respondents are asked a series of questions in connection with use of this official type in this time - frame, and possible bribery: whether the official asked for a bribe, gift, tip or `coima' (slang for bribe), whether the respondent felt obliged to bribe, bribed voluntarily, or refused to bribe, and the amount of the bribe if she bribed."

There's more in the module that Peru tried out. It asks respondents about the quality of the services received from the official type: "whether they saw an official immediately, the number of visits to the official, whether they concluded their business with the official, whether they consider the services received to be `good', `regular' or `bad' and whether they wasted significant time or money in connection with using the official (for example, on transportation)." Please note that Peru scored 67 in the 2004 CPI, and India, 90. Wonder if we too should have collected bribery data as part of census.

There are alarming charts at the end of the paper depicting bizarre findings. "The police account for 35 per cent of bribery episodes and the city (municipal) government for 21 per cent, with the judiciary in third rank with 12 per cent. These three institutions account for 68 per cent of bribery episodes," write the authors. In terms of total bribe payments, the judiciary was at the top with a 42 per cent share of the money paid in bribes, followed by the police with 27 per cent and city government with 11 per cent, making a total of 80 per cent. A case of inverse proportion, between the number of episodes and payment.

Theoretical model in the paper speaks of `two agents, the public official and the client,' interacting in a two-stage game. "The official has a monopoly on the service he provides. The official plays first, and decides whether to angle for a bribe or not. If the official does not angle for a bribe, he carries out his job `honestly' in both stages (which means not shirking, and following required procedures, including possibly unnecessary red tape)."

What about the other, the not-so-honest type? He `angles for a bribe' by shirking in the first stage, and the client gets the message! The official sets the amount of the bribe. The client bribes, or doesn't, though in the latter case, "the official punishes her by shirking in the second stage as well".

There is usually honour among thieves too, it is said. So, there is service delivery, though not an improved one, for the extra money received, lest the greased official falls foul of the Bard's line in Twelfth Night: "I hate ingratitude more in a man than lying, vainness, babbling, drunkenness, or any taint of vice whose strong corruption inhabits our frail blood."

Towards the end of the paper, Hunt and Laszlo explore whether bribes are a regressive tax, that is, whether the vice constitutes a larger burden on the poor than the rich. "Among users of officials, the rich bribe more frequently, and pay higher bribes, and the distribution of types of bribery episode is the same for rich and poor. The burden of bribery can therefore only be higher for the poor when considered relative to income," observe tha authors.

However, what the authors find in Peru's data is that among users of public officials, bribery is at worst a flat tax. Because `the prospect of having to bribe' isn't discouraging the poor more, nor giving the rich a better service. "We therefore consider it unlikely that the burden of bribery is regressive for the whole population... The main distributional consequence of bribery appears to be a transfer from clients to officials in return for no net improvement in service," they conclude.

Seen thus, all action against corruption, which we know touches but a tip of a massive iceberg, isn't going to make lives less expensive for the majority. Perhaps, it again achieves a miniscule redistribution — from a few who received the bribes to the government, in the form of penalty.

Among the references at the end of the paper is a research by Marianne Bertrand, Simeon Djankov, Rema Henna and Sendhil Mullainathan, titled `Obtaining a Driving License in India: An Experimental Approach to Studying Corruption' (MIT Working Paper, 2005). An abstract of this on www.econ.yale.edu informs that the work analyses data from an experiment on bureaucratic processes conducted by the World Bank in 2003-2005. The researchers selected about 800 individuals applying for a driving license in New Delhi and `carefully tracked' them. "All procedures and related payments from the time of application for a temporary license to the time that a permanent license is granted (or not)" got documented. "Some individuals were randomly assigned to a condition that either offered them free training or increased their ability to pay for a license. At the end of the study, all subjects were asked to take a surprise driving test."

Here comes more surprise: The authors' findings reject the `grease-the-wheels' view of corruption, where bribes enhance efficiency! "Among those that obtained a license, individuals who paid more were 20 per cent more likely to fail the surprise driving test than a control group. Also, individuals that were better drivers were only 2 per cent more likely to get their license than a control group." Catch a `preliminary and incomplete' draft of this paper on http://163.152.84.199, the site of Asian Institute of Corporate Governance. Moral: Corruption doesn't pay.

Another research of interest is a 1993 NBER paper titled, simply, `Corruption', by Andrei Shleifer and Robert W. Vishny. The authors say two things: One, that "weak governments which don't control their agencies lead to ultra-high corruption levels". Well, that's not too surprising. But the second point they mention is insightful: "The illegality of corruption and the need for secrecy make it much more distortionary and costly than its sister activity, taxation."

AccountSpeak@TheHindu.co.in

D. Murali

Article E-Mail :: Comment :: Syndication :: Printer Friendly Page

Stories in this Section
India's vote on Iran at IAEA — National interest versus foreign policy

The value of share buybacks

A base for rural connectivity

Too important to be left to political compulsions

Presentation of financial instruments

On corruption and its `sister activity' — taxation

Are tax lures to woo investment passé?

Making sense of the Sensex

Rural lending