Photo by Artyom Kabajev on Unsplash

Fiscal sustainability and optimal fiscal policy are central issues to demographic changes.

Without reform, studies suggest that the current United States social security system will not be able to sustain a balanced budget. With social security that is largely a pay-as-you-go program, social security benefits payout to beneficiaries will gradually exceed payroll tax revenue from today’s workers when there is a shift in the population share towards older cohorts.

To address this problem, many studies have evaluated and compared commonly available fiscal reform options in terms of their policy outcomes on macroeconomic performance and social welfare. Among the widely used policy options are to increase the payroll tax rate, to reduce social security benefits, and to extend the retirement age.

However, my study suggests that the conventional evaluation approach may omit some important aspects which, when incorporated, will yield new insights and can alter the ranking of policy recommendations based on improvements to welfare.

How fiscal policy alternatives are usually evaluated?

The standard approach in macroeconomics models is to use consumption and leisure as determinants of wellbeing to evaluate welfare during demographic transitions. Reforms that allow households to consume more and enjoy more leisure will be more desirable in terms of welfare.

In this regard, the existing literature finds that reducing social security benefits or raising the retirement age are more welfare improving for future generations compared to increasing the payroll tax rates. This is because, even at the cost of less leisure time, higher labour supply, resulting from either being more independent in retirement financing or being eligible for social security benefits at an older age, gradually enables more consumption over time.

What are missing in the conventional framework?

My study argues that the conventional welfare assessment omits at least two important aspects: age-dependent risk aversion that plays a crucial role in economic decisions, and changes in future uncertainties that may complicate individuals’ lifetime planning. But why are these two aspects necessary?

First, for most people, welfare improves not only with consumption and leisure but also with how well individuals can follow through their life plans with certainty. In other words, most people are risk averse, preferring a certain stream of consumption and leisure over an uncertain one with the same expected value. Therefore, a spike in uncertainties for the same expected value will make planning more difficult and hurt welfare, as there will be a higher chance that available resources for future consumption and leisure may deviate from people’s original plan. Accounting for such high uncertainties requires people to save more as a precaution, an amount which could have been used for instantaneous consumption and leisure if future uncertainties are low.

Consider a case of population ageing. People’s exposure to uncertainties will change as a result of policy reforms as well as the longer life expectancy that expands the planning horizon. In such a case, they may prefer a policy alternative that let them follow through their life plans with more certainty even at the cost of slightly sacrificing a part of their lifetime consumption and leisure.

Secondly, an aspect of age-specific behaviour plays an important role. The effects of incorporating uncertainty is amplified when people are increasingly risk averse with age. Intuitively, as people get older, their ability to earn extra income diminishes, making them less resilient to income shocks compared to younger adults. However, not only will this affect older adults’ decision, but younger adults will also increase precautionary savings expecting themselves to become more risk averse as they get old.

Any changes in policy-induced uncertainties will therefore affect welfare across all generations. The effect is direct, via changes in uncertainties; and indirect, via forgoing lifetime consumption and leisure to increase precautionary savings and keep the risk exposure under control.

What are the main findings?

My study develops a new framework that takes into account aspects of uncertainties and age-dependent risk aversion to revisit the welfare evaluation of fiscal policy alternatives. Based on the new framework, my findings suggest a different welfare ranking to prior studies that assume constant risk aversion across all ages.

Under the projected demographic transition, reducing social security benefits and extending the retirement age result in higher future uncertainties and make an individual’s retirement planning more difficult compared to the case of increasing payroll taxes. This therefore makes the first two options not as strongly preferred as previous studies have suggested.

Such higher uncertainties are caused by an increase in the wage rate for the reason that a higher capital to labour ratio is expected to be the result from the first two policy options, which will lead to higher income uncertainties as people progress in age.

For the case of reducing social security benefits, people will build up more savings to compensate for lower social security benefits payout, leading to higher capital intensity. Whereas for the case of extending the retirement age, people are required to spend more time in the labour force, for they will not receive benefits until the age of 75. This leads to a greater accumulation of assets that peaks closer to the new retirement age before running down at a more gradual pace due to higher benefit payout in the years post-retirement.

Together with the pattern of risk aversion as well as earning ability that depends on age, an increase in wage uncertainties will amplify adverse effects on welfare.

Future research

To summarize, my study suggests that in order to evaluate welfare impacts of demographic shocks comprehensively, it is crucial to factor in aspects of policy-induced uncertainties and age-dependent risk aversion, which together result in significant differences in policy implications compared to the more limited traditional approach. The new framework developed in my study serves as an initial step to do precisely this.

Future research may extend the framework to be more realistic by taking into account a risky rate of return on capital and intentional bequest motives, for example.


The full study can be found at

Leave a comment

Your email address will not be published. Required fields are marked *