23 October 2016
How much a generation can save for its own retirement may hinge on how much it can earn. Photo: HKEJ
How much a generation can save for its own retirement may hinge on how much it can earn. Photo: HKEJ

How universal insurance can level intergenerational bias

In the local discourse on retirement protection, views about intergenerational justice have narrowly focused on the distribution of financial burdens of a pension system across generations.

It is considered unfair or unjust if a pension system imposes heavier financial burdens on certain generations.

Following this line of reasoning, a pension system is fair only when each generation bears a similar level of financial burdens for a given level of pension benefits.

This understanding of intergenerational justice has led one commentator to suggest that in order to avoid some generations imposing heavier burdens on others, pension systems should be founded on the principle of “each generation supporting itself” (自己一代養自己), that is, each generation should save for its own retirement needs.

In terms of pension system design, according to this view, a pay-as-you-go (PAYG) system should be avoided not only because the system is not founded on the principle expounded above but also because it will easily give rise to questions about intergenerational justice.

In a PAYG system, contributions made to the pension system by active workers will be used to pay for the pensions of retirees.

In order words, payers and payees of the system are of different generations.

It is contended that if the paying generation is fewer in number than the receiving generation, then the heavier financial burdens on the paying generation constitutes injustice to the payers.

This “each generation supporting itself” principle, however, overlooks the other side of the saving equation, that is, how much a generation can save for its own retirement may hinge on how much it can earn.

If different generations face different economic prospects for earnings and savings, then how should the issue of intergenerational justice be understood?

Put it another way, should not considerations of intergenerational justice also take into account the distribution of prospects for earnings and savings across generations?

An examination of factors limiting people’s ability to accumulate adequate savings for a reasonable standard of living in old age in the absence of state intervention may shed some light on the questions above.

First, consider economic fluctuations.

The economy goes up and down. Businesses succeed and fail. The economic structure shifts over time.

All these economic fluctuations affect workers’ earnings positively or negatively. Most importantly, while economic fluctuations impact on every generation, their impacts vary across generations.

The duration of economic fluctuations varies. Economic fluctuations may be less frequent or more frequent.

It is therefore not unlikely that while one generation experiences a long recession or frequent economic fluctuations, another generation experiences a long boom or less frequent economic fluctuations.

Thus, for generations that experience long recessions, it is very likely that a substantial number of workers may find themselves on low incomes, become unemployed for extended portions of their careers and have limited time working in the next economic boom to make up for losses in earnings.

As a result, workers of these generations may enter retirement with savings less than originally anticipated or even with very little savings.

Next, consider inflation.

Inflation erodes the purchasing power of savings. Its impacts on people’s earnings and savings are obvious. What is less obvious, however, is that retirees are particularly vulnerable to inflation.

Different reasons account for this.

For retirees, the losing purchasing power of savings owing to inflation is almost permanent because compared to active workers, retirees have fewer opportunities to earn additional earnings or to increase savings to make up for the lost ground.

Another reason is that retirees nowadays tend to live longer than those in previous generations, therefore, inflation’s accumulative adverse impact on retirees’ standard of living can be huge, even at modest levels.

For example, with 3 percent annual inflation, the real value of one’s savings after 10 years will be only 74 percent of its original value, and only about one-half after 20 years.

In the case of bouts of high inflation, the corresponding impact on retirees can be catastrophic.

Thus, even if workers of different generations retire with a similar level of savings, due to different levels of inflation encountered, other things being equal, their standards of living in retirement can be very different.

Furthermore, consider investment risks.

What average wage earners can do to protect themselves against inflation and hence the purchasing power of their savings in the absence of state intervention is to save more.

But there is a limit to how much more average wage earners can save. To save more, they must earn more. But for them, there is also a limit to how much more they can earn.

There are few available options for average wage earners to secure additional earnings.

One possible option is to invest in the financial market through, for example, purchasing investment products or participate in private personal pension plans.

Yet, the financial market can be very volatile. Investment is in fact a risky business.

Take the example of participating in private pension plans.

Owing to financial market fluctuations, holders of similar private pension plans with similar life-long earnings and contributions may end up with very different pensions if they retire at different times.

One study compares the hypothetical pensions US workers would have obtained between 1911 and 1999 if they held private pension plans.

The hypothetical workers are assumed to be identical in all important aspects, such as career paths, earnings and contributions and differing only in respect of the year in which they retired.

The study finds that “workers who follow an identical investment strategy but who retire a few years apart can receive pensions that are startlingly unequal”.

Overall, the study shows that the difference in pensions among workers can be as large as five times.

The essential point here is that there is a lottery element in how much pension workers can get.

Economic fluctuations, inflation and investment risks are basic facts of economic life and beyond the control of any single individual.

They are unavoidable and can only be managed and adjusted to. But, there is a limit to what can be done in this regard, especially by the majority of average wage earners.

Without denying individual striving, a person’s prospect for earnings and savings does inevitably contain a certain degree of fortune and misfortune.

Some people and some generations are more fortunate than others. Is this fair or just?

Given differential income gains and losses resulting from different patterns of fortune and misfortune experienced by different generations, cannot a case be made on the ground of justice that such differentials should be redistributed and leveled out?

If one finds the notion of redistribution too welfarist, one might approach the issue of unequal distribution of fortune and misfortune in association with income losses and gains across generations from the perspective of insurance.

Insurance is about risk pooling and sharing. It is welfare-improving for all if everyone’s risks are pooled and shared.

The more widely the risks are shared, the more welfare-improving it will be. Thus, it will be in the interests of all if the economic risks faced by different generations can be shared.

In this connection, it is noteworthy that there has been at least one study that shows that the return to people’s individual pension savings will be substantially higher if a pension system can spread investment risks across generations.

The consideration of income redistribution and risk sharing across generations provides some direction for pension system design.

Since only the state can organize pensions across generations and redistribute income and risks accordingly, the system should be a public system.

Since everyone will face uncertain prospects for earnings and savings, the system should be “universal”, that is, covering and protecting all.

To redistribute and share risks across generations, the system should also contain a PAYG element.

Indeed, the welfare gains of a pension system from sharing risks across generations shown by the study mentioned above are in fact done through a PAYG element in the system.

Focusing merely on pension system financing has blinded the local discussion to many of the limitations that people face in ensuring economic security in old age, to what justice actually demands, to what retirement protection people may need, and to what other social functions pension systems can serve, in particular the insurance function.

The insurance function must not be underestimated. Pension systems are more than poverty relief or income distribution.

Even if the entire population were middle class and presumably self-supporting, there would still be need for a system to enable them to cope with economic risks of various kinds.

From the perspective of insurance, regardless of one’s position on the value of income redistribution, there is a strong case for instituting a PAYG universal public pension system.

[Chinese version中文版]

– Contact us at [email protected]


Member of Hong Kong Democratic Foundation

EJI Weekly Newsletter