There are many good reasons for unemployment insurance (UI). It provides a safety net for the unemployed, enabling them to 'smooth down', rather than dramatically slash, their consumption - which also makes their job search more efficient.

UI obviates the need for 'surrogate' safety nets or safety net substitutes - like ad-hoc handouts to the jobless, an artificial expansion of government-sector employment, or worse, a high tolerance for heavy loss-making state enterprises (as in China and India, for example) on the grounds that they provide jobs.

UI also serves wider economic goals. It injects purchasing power into the economy when this is most required - that is, during a downturn - and it does so automatically, without the need for any official spending package - which is why it is called an 'automatic economic stabiliser'.

And - this may also be relevant in Singapore's case - UI enables economic restructuring (which can involve layoffs) to proceed smoothly; without UI, such restructuring might have to be minimised or delayed.

But designing unemployment insurance needs special care. While the system should serve all the goals indicated above, it should not do so at huge cost nor kill incentives for people to find jobs, or to work.

Fortunately, as a latecomer in adopting unemployment insurance, Singapore can learn from the experiences of other countries - as well as the now-considerable research on the subject - to build a state-of-the-art safety net.

One of the findings of the research is clear. Badly designed UI schemes do indeed reduce incentives to work - thus perpetuating unemployment - and can be hugely expensive. That is why, during the 1990s, a number of countries were forced to reduce their unemployment benefits and tighten eligibility rules: benefits are now no longer paid unconditionally.

Another finding is that the more people pay for their own unemployment benefits, the greater their incentives to work and to find jobs (as you would expect), as well as the less the public costs of providing those benefits.

According to recent research, schemes that are largely savings-funded are indeed feasible: economists calculate that, during a working life of 45 years, a person who is unemployed three times for six months at a time would lose only around 4 per cent of lifetime earnings - and this amount can be covered by individual savings, or by loans against future earnings.

Thus, economists Martin Feldstein and Daniel Altman of the National Bureau of Economic Research in the US propose interest-bearing 'unemployment insurance savings accounts' in which individuals put aside 4 per cent of their salaries, and which can be drawn upon during periods of unemployment. If accounts are exhausted, individuals can borrow from government and repay, with interest, from their future earnings.

A variant of this scheme - as suggested in a paper co-authored by Nobel Prize winning economist Joseph Stiglitz - involves the integration of unemployment insurance with a pension scheme. Here too, people have individual accounts, with part of the contributions earmarked for unemployment insurance.

As Singapore already has an individual account mechanism in the form of the CPF, a savings-funded scheme of this sort could be one way to go - although Mr Stiglitz does suggest that it could be supplemented by a tax-funded scheme as well.

Many do's and dont's

When it comes to tax-funded schemes, there are a number of do's and don'ts. Both experience and research show that tax-funded schemes work best when: