Professor Nelson Chow's report on possible pension plans for Hong Kong has been submitted to the government. It is as yet unclear when these will be released to the public. The SAR government has to come up with a viable proposal that will address the city's many requirements for a public pension plan.

A viable proposal will have to be fiscally sustainable. It has to provide an adequate income to meet basic needs. It certainly should be free from "moral hazards" - or abuses. This means it should not encourage workers to work less, or to squander or hide their assets in order to qualify for pension payouts. It should also not present too onerous an administrative burden. It should offer support for those on incomes so low as to preclude adequate pension contributions.

All pay-as-you-go pension schemes - using today's tax contributions to pay today's pensions - risk being unsustainable. Demographic vagaries make pay-as-you-go plans a gamble. Even if actuaries say the schemes are sustainable - population projections for 50 years hence are anything but reassuring, they may actually prove to be unsustainable in the long-term. This is why the last governor Chris Patten's pension proposals were rejected by a number of economists. Because of these risks, we should never decide on such a scheme.

But individualized saving accounts - such as the MPF scheme - do not work either. Individuals face two types of uncertainty: longevity and the rate of return. The concern about living long enough to use up one's savings, and the fear of one's investments performing poorly, create considerable anxiety. This is precisely the reason why a public pension plan is needed. We should address these uncertainties.

The only way to deal with potential risks is by "risk pooling". A risk pool is one type of risk management commonly used by insurance companies. These companies work together to develop a pool of funds, which can provide protection against serious risks such as floods, earthquakes and so forth.

Now in order to reduce the risks associated with superannuation schemes, I recommend a cohort-based pension plan. Members of each cohort contribute to a pool from which funds are drawn to support members' pension withdrawals. There is no inter-generational transfer, which is both haphazard and unfair. So a cohort-based plan is like a "shared MPF" - shared by members of the same cohort. Members contribute the same amount each month and they will eventually withdraw the same pension each month. The pension has to be adequate, in order to determine the level of contribution required - given the life expectancy of the cohort. The great thing about a "shared MPF" is that while the longevity of any member of the cohort is uncertain and unknown, the life expectancy of a cohort is relatively stable and is known.

In my column last week I recommended a rate-of-return insurance mechanism for funds deposited in the MPF. The same should apply to funds invested in pension plans. With a guaranteed minimum rate of return, it is easy to calculate the level of pension that can be fully funded by contributions and investment returns. In the event that actual returns exceed those projected by the original guarantee, bonus pensions will become available. These bonus funds can be distributed based on excess returns. Shortfalls in returns are subsidized from the insurance pool; returns beyond the guaranteed 2 percent (after inflation) would be taxed to contribute to the pool.

I would also recommend subsidizing the contributions of the poor - those whose incomes preclude contributions at the stipulated amount. This could be funded from general government revenue as part of its income redistribution strategy.

Subsidizing contributions is far better than limiting pensions to those elderly who are poor. Pension means testing may appear to make sense, because the wealthy, such as Li Ka-shing, evidently don't need it. Disqualifying the rich from the pension plan would seem to make the pension scheme more affordable. However, I do not recommend this. A means test is administratively cumbersome. It would be subject to error and could stigmatize recipients.

Administrative costs aside, there is the risk that some deserving elderly poor people may fail to apply, in part because they may not know how to, or be physically incapable of applying, and in part because they shun the stigma of "old and poor". There is also the risk that some not-so-poor may spend all their money or hide their wealth in order to qualify for the pension. This is why I would recommend universality. My proposal would even cover housewives and require the breadwinners to pay their pension contributions. If breadwinners' incomes are too low I believe subsidies from the government would be a good idea.

The author is director of the Center for Public Policy Studies at Lingnan University.

(HK Edition 08/05/2014 page9)