If we are not careful, changes in incentives to save could sharply reduce aggregate savings and investment.

In 1986, more than half a million firemen, soldiers, workers, medicos and volunteers entered the exploded Chernobyl nuclear plant. Despite being exposed to all manner of radiation, they stayed until the fire was put off, and the area was made safe behind the concrete structure that encloses the reactor today. At least half of all those who entered are dead, chronically ill or disabled today.

Elsewhere, financial markets have slumped significantly in the last 18 months. Prudent individuals across the world, and particularly in developed economies, have found their savings greatly eroded by stock and bond market collapses. And as reckless individuals get increasingly bailed out by governments, savers must wonder if it was really better to have saved and lost.

What do the two situations have in common? A recent piece* by three economists offers some insight, especially into behavioural dynamics. A peculiarity about humans, the authors say, is how they sometimes sacrifice themselves "for non-relatives who are unlikely to return the favour."

Today, the worry is that a change in incentives against saving may tip the balance in favour of those who don't save. The concept of moral hazard is that rewarding an individual who acts selfishly and causes public harm, sets a bad example for others. Otherwise cooperative individuals might then find it worthwhile to behave selfishly. If their resulting behaviour changes, this could lead to the worst outcome for the public. That is, if enough people agree, the resultant drop in aggregate savings will have poor implications for business investment and economic growth.

This is a stark and somewhat frightening prediction. When translated, it would mean that after recent events people might not continue to invest their savings publicly. If profligacy pays, what price prudence? The knock-on effects of this would then be felt through a drop in retail deposits, which shall be transmitted by the banking sector to the real economy.

Of course, people don't always act to maximise their interests alone. Members of groups often forego individually better options for collective ones. People do follow laws and norms even against their individual interests. This paper looks at such situations.

It might have been in the interests of the fire-fighters to leave, but they stayed. This presents what is called a cooperative dilemma, which the authors define as a "situation in which a person must decide whether or not to subordinate his own interests to those of the group."

Chernobyl is an example of gratuitous cooperation, an extreme case in which the beneficiaries to the action are unknown, and the favour is one-sided. The authors show that some aspects of cooperative behaviour � conformism, altruistic punishment and gratuitous cooperation -- evolve together. Conformism is the tendency to imitate the most frequent behaviour, while altruistic punishment is the tendency to punish free-riders at a cost to oneself. Moreover, contrary to the notion that this constitutes cognitive impairment (i.e., a "mistake"), they suggest that such cooperation is often a well-thought-out strategy.

Conformism, for example, is a powerful binding force. In a cooperative dilemma, conformism is the equivalent of the instruction "cooperate if the majority cooperate, do not cooperate if the majority do not cooperate." This is a stable strategy. A small number of defections will never induce the majority to defect. What's more, they say, we are strongly encouraged by our upbringing to behave this way.

Of course, the context of individual savings is somewhat different. Individuals don't save in the interests of the public. But those without incomes, such as the retired, will not be able to compensate for the loss in savings they have suffered. On the other hand, for those who earn, the economic situation demands more saving. But the rewards of good behaviour are now uncertain, and the potential risk magnified.

In actual life, people may continue to save, as they have always done. But the authors warn that while such a cooperative equilibrium may be stable, it is not inevitable. Both "save" and "do not save" are stable majority outcomes, and either can be strongly self-reinforcing.

Moreover, a sudden change to the system from one state to the other can happen by concerted leadership or shock. Today, individual incentives to save are being affected by both. Whether aggregate saving falls remains to be seen.

The author is a doctoral student of economics at the Indian Statistical Institute, New Delhi madhav8r@isid.ac.in

*Rowthorn, Robert E, Ricardo Andres Guzman, and Carlos Rodr�guez-Sickert, "Theories of the evolution of cooperative behaviour: A critical survey plus some new results", January 4 2009, Munich Personal RePEc Archive, http://mpra.ub.uni-muenchen.de/12574/