Unemployment insurance premium rates should reflect market forces, argued Research Fellow Michael Beenstock at a CEPR lunchtime meeting on 21 October. Just as young Ferrari owners pay higher car insurance premiums than sedate Mini Metro drivers, so individuals who face greater unemployment risks should pay higher premiums. The Beveridge system of social insurance should be replaced by a market approach, with private insurance contracts in place of national insurance contributions. Beenstock's research indicated that the current system of unemployment insurance tended to redistribute income towards the better off. The market approach would avoid this cross- subsidization and promised to be fairer. It would also encourage a more efficient allocation of resources.

Michael Beenstock is Esmee Fairbairn Professor of Finance and Investment at the City University Business School. He is also a Director of the City Institute for Financial and Economic Research and a Research Fellow in CEPR's International Macroeconomics programme. He has written extensively on exchange rate policy as well as on the economics of unemployment in the United Kingdom.

Beenstock first outlined the current system of 'social insurance', based on Beveridge's principles. Premiums or contributions do not reflect the risks that the individual faces, but are set so as to fund the total cost of the scheme. Successive governments have let the latter principle lapse but have steadfastly adhered to the former. In the case of unemployment insurance, Beenstock noted, those who face high unemployment risks contribute at the same rate as those who face low unemployment risks (or, like tenured academics, no risk at all). If social insurance applied to cars, the Ferrari owner would pay the same premium as the Mini Metro owner.

Beenstock then discussed his proposals for a market-based approach to unemployment and other forms of social insurance. His proposals represented a radical departure from Beveridge's principles and, he conceded, were unlikely to find immediate acceptance. The 1985 White Paper on Social Security, for example, not only failed to mention the market approach, but fully endorsed Beveridge's social insurance principles.

A market approach would offer many advantages, Beenstock argued. The price mechanism is acknowledged to generate an efficient allocation of resources, he claimed, for insurance just as for product markets. Insurance premiums should therefore be priced so as to reflect the level of risk to each worker. The supply of different types of workers would then be influenced by the unemployment insurance premiums that people are required to pay. If premium rates were based on the risk of unemployment, the price mechanism would give people an incentive to choose occupations with lower unemployment risks, and this would lead to a more efficient allocation of resources.

{v20 b650}The Beveridge approach also fails to provide the same degree of protection to high and low paid workers faced with a high risk of unemployment, according to Beenstock. Whereas unemployed executives now realize that they were under-insured, people at the other end of the income distribution have been over-insured and can find unemployment more attractive than work, he claimed. Beenstock's proposals would mean lower premium rates for those on low incomes, while some of the better off might be prepared to pay more in return for improved protection if they become unemployed. In this way, Beenstock argued, the market approach will provide better service and more value for money than the Beveridge policies.

Actuaries must identify the risk classes to which individuals belong in order to calculate appropriate insurance premiums. In his research Beenstock used 1978 data on unemployment flows to identify 'unemployment risk classes' in terms of age, marital status, occupation, region and housing tenure; these variables appeared to influence rates of flow into and out of unemployment. He then presented estimates of the competitive market premiums for various risk classes which took account of the rates of flow into and out of unemployment for each class. His illustrative premiums (see Table) were for an unemployment insurance policy in which the worker can draw benefit from the moment of job loss until whenever re-employment occurs. Beenstock noted the wide variation in the estimated premium rates: it is 80 times more expensive to insure risk class 1 than to insure class 5. The actuArial,Helvetica,Sans-Serif cost of insurance is relatively high in class 1 because many members of this group become unemployed, and once unemployed are unlikely to find new employment.

Beenstock's research also indicated that there was a positive correlation between his estimated premium rates and wage levels. Groups with high estimated premiums (and therefore high risk of unemployment) tended to earn higher wages, and this implied that the present flat-rate premiums 'undercharge' those on high incomes and so redistribute income regressively. Beenstock described this as 'perverse cross-subsidization'. Implementation of his own proposals for unemployment insurance would, if anything, benefit those on low wages, he claimed.

The government could in principle administer a market-based scheme such as Beenstock advocated. He argued that in practice, however, the insurance industry would be in a better position to price unemployment risks efficiently. Beenstock therefore proposed 'denationalizing' the unemployment insurance industry. There was a precedent for this, he noted: a century ago friendly societies and trade unions provided a variety of private unemployment insurance services. Beenstock advocated a return to these arrangements during the next century, but with the insurance and reinsurance industries as the main suppliers.

Unemployment risks are not the only sources of income uncertainty, Beenstock noted. His proposals for unemployment insurance were only one example of what he called Income Maintenance Insurance Contracts (IMICs), designed to maintain income under various contingencies, whether they be unemployment, wage volatility or retirement. Pensions and unemployment insurance contracts are both examples of IMICs. Beenstock foresaw a 21st century in which there will be a range of competitive markets in IMICs, meeting the desire for economic security that Beveridge sought to satisfy through social insurance.

After he had finished his talk, members of the audience questioned several aspects of Beenstock's proposals. In particular, would not an unforeseen macroeconomic contraction bankrupt many private insurance companies if they were obliged to pay out large amounts of unemployment insurance? There were other questions about the operation of a market system of insurance. Would payment of insurance be dependent on availability for work, and how could this be ascertained by private companies? Beenstock's system of risk-related premiums would discourage people from entering occupations with high risks of unemployment, it was argued; was this desirable?