Friday, April 25, 2008

Adverse Selection and Moral Hazard

A 2-Pager by Ajit Chaudhuri, April 2008

During my post-graduate studies, a group of us used to frequent a restaurant in Baroda that offered an inexpensive non-vegetarian buffet. As repeat customers, we should have got warmer welcomes every time we went – but no, they were increasingly icy after the first time until, fortunately towards the end of our course, the sight of us descending from our motorcycles had them scurrying for cover, desperately looking for reasons not to let us in. No, we were not the Gujarat chapter of Hell’s Angels – the reasons for their behaviour were twofold. First, we were all young men with voracious appetites (anyone familiar with Vijaynidhi, Somnath Sen, Mathew John, Dinshaw Irani, Rajvinder Kohli, et al, in those days would sympathise with any restaurant they walked into). And second, we eschewed the soups, breads, salads, etc. (all the cheap stuff put out to take up stomach space), and concentrated our efforts on the mutton and chicken. And this little example illustrates the two phenomena that are the subject of this paper.

Adverse selection occurs when policies encourage the selection of ‘bad’ customers. The restaurant featured above priced its buffet based upon the average eater, but the ‘all you can eat for one price’ policy discouraged the plate-pickers and encouraged the gluttons. As they say, ‘everyone can’t be above average’, and I suspect that that buffet scheme did not survive us. The concept of adverse selection originates from the insurance industry and is used to describe the problem of insurance being purchased only by those most in need, such as life by the old and health by the sick, and the subsequent likelihood of the insurance firm going bust. Adverse selection is prevented by various means, such as reducing the information asymmetry between buyer and seller and ensuring that those at higher risk pay higher premiums (finding out, for example, whether there is a history of heart ailments in the purchaser’s family) and laws requiring everyone to have insurance.

Adverse selection also explains why my old school’s annual alumni reunion is such good fun (for the likes of me and my friends) – it is a glitzy event in a 5-star hotel for which the alumni association charges Rs. 1,000 per person with as much booze and food as you can imbibe thrown in. And while some seedha-saadha types do land up out of patriotic feeling, most of us, men and women alike, are there to have a wild time (nobody brings their spouse unless s/he is also an alumni) and proceed to do exactly that.

Moral hazard relates to the changes in behaviour that policies, especially those that insulate a party from risk, bring about. In the introductory example, it was the sticking to mutton and chicken when we would normally have balanced our intake with vegetables, rotis and stuff. In insurance, moral hazard happens in two ways to increase costs to the insurer. The first relates to the way a person with full health insurance seeks treatment for the most trivial of ailments, or the way someone with full cover vehicle insurance is careless about locking the car. The second relates to the lack of ethics in the medical profession – the way doctors and hospitals find out how much a patient can pay and then find ways to extract precisely that much. If you have cover or can pay, you will find yourself going through meaningless tests and treatments. If you can not, you can die like a dog on the road – like the lady in Kanpur just yesterday[1]. With India being the most privatised health system in the world[2], and with health expenditure being the single most significant cause of families falling below the poverty line[3], the combination of moral hazard, medical ethics and a government health policy that mouths platitudes about universal coverage while encouraging a rapacious health system spells disaster for us all.

Moral hazard relates to finance as well – financial bailouts of lending institutions can encourage risky lending in the future if risk-takers believe that they are protected from the burden of losses – the case of Bear Stearns comes to mind. It applies to borrowers as well, as the hue and cry over the effect of the decision to forgive agricultural debt on the long term credit climate will testify to. It applies to management – when top management is shielded from the consequences of poor decisions, or when funding for projects is independent of success. It applies to sociology – the way the powerful are especially susceptible to disrespecting the law of the land, the way men behave around women in skimpy clothes on the assumption that they are ‘asking for it’, and the way private bus drivers drive when they have paid an ‘all-purpose’ hafta to the traffic police.

Are adverse selection and moral hazard applicable to the development sector as well?

By working two jobs, one with an Indian donor agency and one with a British one, I see adverse selection at its starkest every day. Care Today can fund anybody it chooses to, and we have thus worked with and funded a variety of organisations including UNDP, Actionaid (whoever heard of an Indian funding agency funding the UN or a foreign funding agency – but we have) and the Indian Army. The Paul Hamlyn Foundation, like all foreign donors in India, can fund only FCRA-registered NGOs – a much smaller universe of the good, the bad and the ugly with the one commonality that development requirements have to balance with organisational perpetuity concerns.

A recent case of moral hazard that I saw was in the Paul Hamlyn Foundation’s work with an NGO. PHF normally supports for one year and then, if suitable, another three years. In this case, we at PHF decided to approve three years support, enough for the NGO to do all that it saw as required, in one go. This would enable growth with stability, I had argued to my bosses. Monitoring systems revealed that money for salaries and infrastructure was being absorbed effectively, but none of the activities that the NGO had so effectively communicated as critical for meeting its objectives were being done. Enquiries revealed that the NGO had been utilising its time and resources in undertaking sub-contracting assignments for one-time short-term donors. The NGO, seeing assured and stable support, had downgraded the priority of the PHF-supported activities and adopted an ‘incremental’ approach to further activities and funding. Needless to add, I am rethinking the basics of the relationship.
[1] Many news channels and newspapers covered this. For the sake of convenience, I will cite just one – ‘Woman who gave birth by roadside dies, 8 doctors suspended’ in the Indian Express of 25th April 2008.
[2] According to WHO Statistical Information System: Core Health Indicators, India’s health care system is 83% privately funded in 2004.
[3] According to the WHO, 24 percent of the families of all Indians hospitalized fall below the poverty line as a direct result of hospitalization. This is from Devadasan, Van Damme, Ransom and Criel, “Community Health Insurance in India: An Overview”, Economic and Political Weekly of 10-16 July 2004.

No comments: