Media Briefings

INADEQUATE SAVING: Age dependence of people’s ability to impose ‘self-control’ implies that the young need more government incentives to save for retirement

  • Published Date: June 2017

People’s ability to focus on long-term goals is stronger in later adulthood than earlier in their lives. Moreover, this improvement in ‘self-control’ is particularly fast at younger generations. This lifecycle pattern of self-control implies that the government should subsidise savings but that subsidies to encourage savings should decline with age.

That is the central message of research by Nicola Pavoni and Hakki Yazici, published in the June 2017 issue of the Economic Journal. Their study calculates optimal age-dependent saving subsidies for the US economy, and compares them with the subsidies implied by the typical 401(k) retirement plan. The results indicate that while optimal subsidies decrease over the lifecycle, the 401(k) plan implies an increasing pattern of subsidies as people get closer to retirement.

Economists traditionally assume that people have what they call ‘time-consistent intertemporal preferences’: when people make plans, they stick to them. But there is an abundance of experimental and anecdotal evidence from our everyday lives that points to the opposite: people face problems of ‘self-control’.

For example, we often make plans about how we will start a healthy diet or stop smoking or start studying hard for the upcoming exam at the beginning of next week. But when the actual day of the unpleasant activity (eating healthy, quitting smoking, etc.) arrives, we fail to follow through on the promises we made to ourselves.

One such notoriously unpleasant activity is saving for the future. Many people fail to save enough for their old age even though they have planned to do so. There is in fact a large body of empirical evidence from laboratory and field studies that document that people face ‘self-control problems’ regarding their saving behaviour.

The new study asks the question: can governments do something to help people get over the problem of inadequate saving? The authors find that government should motivate people to save enough by subsidising market returns to savings.

Most importantly, these subsidies should be age-dependent and, in particular, decline with age. This is at odds with the current practice in the United States where the 401(k) programme implies age-dependent saving subsidies that increase with age.

The finding that subsidising returns to savings can help people increase their savings is intuitive. This study first shows that people do indeed save too little compared with the efficient level. Moreover, while the presence of inadequate savings has been suggested by previous research, that body of work has been silent about the age-dependence of such subsidies.

This is mainly because of a common assumption in research on self-control that the degree of self-control is constant over people’s lifetimes. That assumption contrasts with a significant body of empirical research indicating that like many other personality traits, people’s ability to exhibit self-control changes as they get older.

Both personality psychologists and experimental economists provide strong evidence that people find it more difficult to direct their choices toward long-term goals (they have less self-control) in early adulthood but, as they grow older, their ability to control themselves increases. Moreover, this increase in one’s ability to impose self-control slows down as people age.

The new study takes account of this empirical regularity and shows that under such assumptions about the evolution of self-control, subsidies should decline with age.

The authors also calculate actual optimal age-dependent saving subsidies for the US economy. As Figure 1 shows, optimal subsidies are quite sensitive to age. This figure also compares optimal subsidies with those implied by the 401(k) plan.

If the very last periods before retirement are excluded – during which the subsidy rate in the 401(k) essentially mimics the employer’s matching rate – the subsidy levels in the two cases are of comparable magnitudes. A marked difference emerges, however: the 401(k) plan implies an increasing pattern of subsidies while the optimal subsidies decrease over the lifecycle.


Notes for editors: ‘Optimal Life-Cycle Capital Taxation Under Self-Control Problems’ by Nicola Pavoni and Hakki Yazici is published in the June 2017 issue of the Economic Journal.

Nicola Pavoni is at Bocconi University, Milan. Hakki Yazici is at Sabanci University.

For further information: contact Nicola Pavoni on +39-02-58-36-33-73 (email: website:; or Romesh Vaitilingam on +44-7768-661095 (email:; Twitter: @econromesh).