The economic problem of watching the watchmen

The perennial problem in both the private sector and government is the problem of “watching the watchmen”. In the language of economics, it is a principal-agent problem; how to hold top management (agents) accountable to shareholders (principal). This is captured in the famous Latin phrase: ‘Quis custodiet ipsos custodies?’ Translated as: “Who will watch the watchmen” or “Who will monitor the monitors.”

When outcomes are not measured merely in terms of financial profit, the problem becomes even more complex. This is because there are no simple indicators or metrics. This is the case with public sector institutions, where the shareholders (principals) are the people of the country and their elected representatives are the top line of management (agents).

COPE: Trying to hold public sector accountable

In Sri Lanka, every two to three years a parliamentary committee called COPE (Committee on Public Enterprises) reports on the management due diligence of government-run institutions. The latest COPE report was released this month, in July 2013. COPE reports routinely find huge departures from proper accounting and operating procedures and request rectification, but does this mean violations are diminishing? One thing is clear; the shareholders – the people of the country – don’t seem to think so.

People perceive a high level of corruption

Of the 174 countries surveyed by Transparency Inter national’s 2013 Corruption Perceptions Index, Sri Lanka came in at 79. That is, the perception of corruption by Sri Lankans was greater than the perception of corruption by people in 78 other countries. On this account, Sri Lanka is doing better than most African and South Asian economies but also significantly worse than other countries in these regions: Such as Ghana, Rwanda and Bhutan.

Limits of comparison

It is important to understand that this is mainly a corruption PERCEPTION index. And this has limitations for cross country comparison. This is because social and cultural history can significantly affect perceptions and therefore the magnitude of perception is not likely to be linked to the magnitude of corruption in the same way from one society to another.

For instance, in Colombo, the invitation to pay a municipal parking attendant without receiving a ticket does not cause outrage – perhaps due to sympathy for these beleaguered attendants. But in London, such a proposition would seem terribly outrageous.

Actual bribe payments

The latest survey however went beyond perception to ask about actual payments of bribes by those who came into contact with government institutions. The results are captured in Figure 1, which shows the figures for Sri Lanka and Malaysia.

The problem seems huge. 43 percent of Sri Lankan households that came in contact with the police paid a bribe in 2012; likewise, 22 percent paid a bribe to the judiciary and 20 percent to the tax authorities. Malaysia, considered to have some similar political and social dynamics to Sri Lanka, has starkly less bribe payments.

Perception is that it is getting worse

The challenge in Sri Lanka is not just that corruption is both perceived and experienced to be very high but that it is also perceived to be getting worse.

In 2010-11 49 percent of Sri Lankans thought that corruption had increased and 19 percent thought it had stayed the same. In 2012-13, 64 percent think that corruption has increased and 19 percent continued to think it had stayed the same (see Figure 2). This means that most of those who thought it was staying the same or decreasing a little bit in 2010-11 felt gravely mistaken by 2012-13.

Business is probably more affected than consumers

Businesses are particularly affected by corruption in Sri Lanka. Consumer-related services such as education, healthcare and utilities are relatively less corrupt compared to services used by business such as paying taxes, registrations/permits and the legal system. This again is what game theory analysis in economics predicts: Businesses as repeat customers are more vulnerable because they can be “repeatedly punished” with delays and denials of service if they refuse to pay and their capacity to pay is also higher.

But this does not mean that consumers escape. The corruption cost of business is passed on to consumers through higher pricing and weaker competition. Given that the survey is of households and not businesses, it shows that the corruption habit of institutions that deal with businesses is spilling over in a big way to the consumer experience as well.

Chicken or egg? In the second generation it is both!

A previous Verité Insight found that Sri Lanka’s poor score on the Doing

A single party dominated parliament means that COPE is essentially the fox protecting the hen coop. ‘Quis custodiet ipsos custodies?’ If citizens and the private sector want to see a reduction in corruption, they have to find other means that go much further than routine COPE reports, to let the foxes know that they are accountable

Business Index was largely caused by three factors: Difficulties in paying taxes, enforcing contracts (police and judiciary) and registering property. Figure 1 corroborates that finding. It shows that institutions which are responsible for these three areas are precisely those that elicited the most bribes by those who dealt with them.

There is a chicken and egg dilemma here. Is there more bribing to overcome the inefficiency of these institutions, or are they kept inefficient to facilitate more bribing? Whatever came first, economic research is clear that the perpetuation involves both. The Verité Insight on the justice system showed that it was plagued with needless delays. The results of the corruption survey suggest that this may be also by design: Keeping it in “slow motion” is creating lucrative bribe opportunities.

Watching the watchmen

The data shows that Sri Lanka’s problem is particularly acute because it is the “watchmen” that are perceived to be the most corrupt. As Figure 3 shows, 64 percent thought the police corrupt, 51 percent said it of political parties and 39 percent said it of parliament and the legislature. These are the primary “watchmen” institutions and also the three that are perceived to be “most corrupt”. This is much worse than if the same level of corruption was perceived to be in ‘religious bod- ies’, ‘medical services’ and ‘NGOs’.

Initiatives such as the COPE report of parliament don’t adequately deal with the Principal-Agent problem highlighted in economics. A single party dominated parliament means that COPE is essentially the fox protecting the hen coop. ‘Quis custodiet ipsos custodies?’ If citizens and the private sector want to see a reduction in corruption, they have to find other means that go much further than routine COPE reports, to let the foxes know that they are accountable.

(Verité Research provides strategic analysis and advice to governments and the private sector in Asia)