Broad-based public support for the desired outcome and system design is critical, especially when the program experiences temporary setbacks. 

In defining the desired outcome and the role of contributory components, policymakers may wish to take into account any retirement-related government benefits (public pension system, national health care coverage) and relevant cultural norms. 

Consider how reliable those benefits and norms will be over the long term. For example, will a societal norm that children financially support their parents in retirement work as well in the future as it has in the past? Demographic trends — increased longevity, a working-age population that is shrinking relative to the population of retirees — should be incorporated into a long-term outlook. Finally, it is important to establish a plan for measuring progress toward the desired retirement outcome, as well as the overall success of the retirement system.

Savings
Mandatory, voluntary and in-between

Mandating retirement saving is a straightforward way to try to ensure adequate funding. Most Asian retirement systems currently incorporate some level of required retirement savings, often with either the employer or the employee, or both, making a specified contribution. For other systems, a mandate may not be possible if citizens prioritize personal choice or resist a directive as a form of government or employer “overreach.” 

Whenever a mandate is deployed, transparency is key: Be straightforward about what the required savings rate is meant to cover. People may assume that it will be sufficient to fully fund a desired retirement outcome when, in fact, it was never intended to do so.

Complementing mandatory saving programs with voluntary saving components can strengthen a retirement system. For the voluntary saving component, we advise systems to use behavioral finance techniques — they’re often called nudges — to boost savings through the power of human inertia, or the tendency to do nothing. Individuals can be automatically defaulted into saving at a particular rate and investing according to a predetermined approach — unless, that is, they take action and opt out. But often inertia kicks in and most people don’t opt out. Using this kind of a nudge is a quiet way to encourage greater personal responsibility. Systems can also automatically increase employee savings rates on an annual basis until a target is reached — and again, people can opt out. 

In U.S. defined contribution (DC) plans, these plan design features are known as automatic enrollment and automatic contribution escalation, and the evidence shows they can be highly effective. In our 2018 survey of U.S. defined contribution plan participants,¹ 33% said they were unlikely to have enrolled in their DC plan without automatic enrollment. In the UK, the introduction of automatic enrollment helped spur a significant rise in participation in private sector workplace pensions, from 20% of eligible employees in 2012 to 63% in 2016.²

Anchor saving at appropriate levels

Employees tend to see the default savings rate and imposed annual dollar amount caps as recommendations, or as anchors that affect their decision-making process. Make the rate or cap too low and savings will be inadequate to fund a successful retirement outcome. Make them too high and they will scare people off. But set them at an appropriate level and employees are nudged to better behaviors. The same principle applies to the mandated savings rate: It should be set at a level suitable to the employee population.

Investments
Limit core/default investment options

When it comes to investing, we recommend a professionally managed, well-diversified approach. In some models, governments or institutions make investment choices on behalf of individuals. In other models, individuals make their own investment decisions. Most people don’t have the needed expertise or time to invest on their own. Emotion comes into play, too. People often underestimate how much risk they can take on when they are investing over a very long time horizon — and they may take on too much risk when time is short.

In designing a retirement system, an important consideration is the level of financial advice that is available to individuals. This can help identify which overarching investment strategy could be most effective. When advice is not readily available, we recommend target date strategies, which gradually shift the allocation away from stocks and toward conservative investments as a participant nears retirement age. When advice is more accessible, target risk strategies can work well. In these strategies, asset allocation is aligned with an individual’s stated risk tolerance, typically categorized as conservative, moderate or aggressive. For those employees who decide to make their own investment decisions, we suggest a limited menu of diversified investment options. An open menu that provides a large number of individual choices may be overwhelming to many employees and result in misuse or inaction.

Determine the appropriate roles of insurance and guarantees

A guarantee, whether it refers to guaranteed principal or guaranteed lifetime income, is an appealing notion. But the security it provides can be expensive and may lead to unintended consequences. Investment strategies designed to guarantee principal will need to overweight fixed income, an inappropriate approach for a younger investor with a multi-decade horizon to retirement. Strategies to provide guaranteed lifetime income, which necessarily involve an insurance component, may limit the potential for higher return and generally require some sacrifice of liquidity. That said, the security of a minimal level of income is an attractive feature. We recommend that guarantees be considered as part of a long-term plan but not the sole component.

Access
Minimize plan leakage

The potential for plan leakage — giving individuals the ability to withdraw retirement-designated savings and use them for another purpose — can pose a risk to the long-term sustainability of even the best-designed plan. Set a clear policy explaining when and how savings can be accessed. Again, there are tradeoffs. Restricting access may prevent leakage, but if the rules seem too onerous they may inhibit people from joining a plan and/or committing additional savings to retirement. You may wish to encourage a reserve fund for emergencies — it helps to segregate short-term liquidity from long-term investments. Treating withdrawals as interest-paying loans may motivate people to be more responsible and nudge them to keep saving. Tax disincentives can also discourage plan withdrawals. Making retirement plans portable, allowing employees to consolidate savings when they move from one employer to another, can help minimize leakage as well.

Automatic programs

In addition to automatic enrollment and automatic contribution escalation, periodic re-enrollment of these automatic features may be beneficial, making opting out of any choice an ongoing or periodic requirement rather than a one-time decision.

Tax incentives

Tax incentives, if they are meaningful, can change behaviors and spur additional saving. Experience shows that tax incentives do not encourage saving among lower income households because they typically pay very little, if any, tax. However, tax incentives do lead higher income households to dedicate more of their savings to retirement. For that reason, we think tax incentives may be best deployed to encourage usage of fully voluntary personal retirement savings vehicles, which higher income households will need to meet their lifestyle requirements in retirement.

Tax policy that encourages more systematic withdrawals and penalizes lump-sum withdrawals may encourage better behaviors, keeping savings invested and compounding returns vs. withdrawing the entire balance in a single lump sum. For example, in Korea’s corporate DC pension system, the individual retirement pension and personal pension taxes are deferred until retirement assets are withdrawn from the plan. At that critical juncture, an employee pays less in tax when assets are systematically withdrawn vs. taking a lump-sum payment. 

Finally, some countries give employers a tax incentive to contribute to an employee’s retirement saving plan, on either a mandatory or a voluntary basis (including through a matching contribution, as we discuss in the following section).

Matching contributions

In a matching contribution program, the employer matches the participant’s contribution by a certain percentage, up to a specified limit. This nudge really works, because an employer’s matching contribution often feels like “free money” to the employee receiving it and incents the employee to save as well.

The employer’s role in a retirement system

As many Asian policymakers have recognized, employers can play an important role in helping to build and sustain a successful retirement system.

EMPLOYERS CAN:

Through employers, one of the most effective behavioral tools — matching contributions — can help bolster retirement security. In a matching contribution program, the employer matches the participant’s contribution by a certain percentage, up to a specified limit.

Employer matches are typically found in voluntary retirement saving programs. But when a mandated employer contribution is structured as a match, it can persuade more employees to save. (As we have noted, an employer match feels like “free money” to an employee receiving it.)

Employers can choose among a range of approaches to matching employee contributions. In Taiwan, for example, employers are currently required to contribute 6% of an employee’s salary, there is no match, and employee contributions are voluntary. If, instead of mandating employers to contribute 6%, the mandate were structured as a one-to-one match (in which employers are mandated to match employee savings, up to a cap of 6%), more employees would be encouraged to save. That is because they would only receive the employer match if they themselves were saving for retirement. 

But, as is often the case, there are trade-offs to be made. If policymakers require employers to structure their mandated contribution as a one-to-one match, it could help one segment of the population — employees who are encouraged to save more of their own salary — while simultaneously hurting another segment: the employees who can’t or won’t save to receive the employer match and thus receive no employer contribution at all.

We advise policymakers to carefully consider all the relevant factors (including the ability and willingness of employees to save) to ensure that the right level of mandated contribution and contribution match is in place to incent greater contribution. Finally, policymakers should make sure employees understand that the matching rate may not be meant as recommended savings rate. 

We all know the power of compounding in capital accumulation. But making an early start to saving is about human nature as much as mathematical equations. One of the biggest motivators of continued saving, we have found, is having successfully saved … something. Positive reinforcement works.

Provide personalized account statements that clearly show progress toward a goal

• Present accumulated wealth as the level of income it will provide in retirement. (If retirement savings look like a bank account balance, they may seem easier to spend.)

• Make sure individuals have the information they need, when they need it. 

• Create tools to help people take specific actions that will boost their odds of a better retirement outcome

• Test whether individuals understand what is being communicated to them.

Beware of unintended consequences

Try to anticipate unintended consequences before they happen — and move quickly to remedy them when they occur. In the U.S., for example, the 2006 Pension Protection Act did a lot of good, including establishing a strong legal framework for automatic enrollment. But one unintended consequence has been unfortunate: Because an Internal Revenue Service ruling on automatic enrollment provided an example that used a 3% rate, many employers interpreted the ruling as guidance. They concluded that the 3% rate would be seen as the “safe” — the legally protected — option. In fact, 3% is far too low if it is sustained as a lifetime savings rate.³

Plan for the full life cycle

When designing or enhancing a retirement system, we suggest that you plan for the full life cycle, not just where the majority of the population is today. (Many Asian populations are still relatively young, but not for much longer.) Address both phases of the cycle — accumulation of assets during working years and decumulation in retirement — and the transition between the two. For example, most Asian pension systems pay lump-sum benefits and retirees are left to their own devices to make complex decisions about investment products and withdrawal timing.

A comprehensive retirement system should consider how retirement savings can be converted into income streams in retirement. The system should also provide guidance on how to leverage the benefits of both asset management and insurance products to solve an individual’s retirement income needs. Annuities are a well-established approach, of course, and some governments have taken steps to introduce public annuity programs. For example, Singaporeans are mandated to convert part of their retirement savings into a monthly income stream to support a basic standard of living during retirement via CPF LIFE (Central Provident Fund Lifelong Income for the Elderly). The Hong Kong government launched a voluntary life annuity scheme in July 2018. Taking a different tack, the Australian government has been exploring a framework for retirement income products that offer greater flexibility.

We recommend a well-defined review process to re-evaluate how well the system is doing. Use an agreed-upon methodology so that discussion and debate are fact-based. (Facts will be your friend when political pressures come to bear.) Flexibility helps, too. Retirement systems should evolve as demographics and societies change and new knowledge and experience are acquired.