Tax and non-tax financial incentives can promote saving for retirement by providing an overall tax advantage to individuals through a reduction in total lifetime tax paid, although this has a fiscal cost. Incentive design should account for retirement saving needs and capacities for all income groups and, as a minimum requirement, provide for tax neutrality between consumption and savings.

Tax rules should be straightforward, stable and consistent across all retirement savings plans. Tax credits, fixed-rate tax deductions or matching contributions can be used to provide an equivalent tax advantage across income groups. Countries using tax credits may consider making them refundable and paying them into pension accounts. Non-tax incentives, in particular fixed nominal subsidies, help to boost low-income earners’ savings. Countries with an “EET” tax regime should maintain the deferred taxation structure and all countries considering the introduction of financial incentives should examine their fiscal capacity and demographic trends.

Providing pension services involves costs such as administration and investment activities which are paid for by members and employers. These costs can greatly affect the ultimate value of accumulated retirement savings. Some pension arrangements can be also more expensive, such as those providing more choice.

Market mechanisms have often been insufficient to align charges with the actual cost to providers due to market failures, such as asymmetric information or behavioural biases.

Measures to improve transparency are essential, but are not enough to align costs and charges. They work best when supported by pricing regulations and structural solutions. To maximise net returns, policy makers and regulators can also use measures such as benchmarking and tying investment expenses more closely to portfolio performance.

Several nationally significant investment institutions have common features and provide evidence of good governance and investment approaches. They have regulatory and legal frameworks at arm’s length from government; clearly stated missions to guide investment policy; an oversight board that is accountable to the competent authorities and to members; and transparency about their governance arrangements and their investment and risk management to keep them accountable to different stakeholders.

These institutions express their performance objectives in terms of their mission and monitor performance against this long-term goal rather than against a market benchmark. Target date and lifecycle funds are the preferred strategy for institutions with individual accounts. Long-term return strategies may offer better returns, but at a higher risk that insufficient funds will be available to members at retirement.

Low levels of financial knowledge and behavioural biases can lead people to make unsuitable decisions for retirement.

Mechanisms such as automatic enrolment and escalation of contributions can harness inertia to make pension systems more inclusive and help increase contribution levels. People unable, or unwilling, to choose a contribution rate, a pension provider, an investment strategy or a post-retirement product, may benefit from default options.

There are also other tools to help with decision making, including: web applications, limiting options and making comparisons easier, and financial incentives. Pension statements can convey key information simply, while financial education seminars and financial advice can help people understand the information.

Individuals in low socioeconomic groups have a lower life expectancy than high socioeconomic groups. They may be financially disadvantaged if they spend a shorter time in retirement relative to their working life, receiving a lower “return” on contributions made towards their funded pension source. Public pensions and tax rules can help to offset some of this disadvantage.

Policies to improve the sustainability of pension systems in light of increases in life expectancy will need to consider how those in different socioeconomic and gender groups may be impacted. Generally, working for longer will be required, but not all groups will necessarily be able to. Increased flexibility around retirement age is key to improving pension outcomes for all groups and ensuring that lower socioeconomic groups are not penalised in retirement for having shorter life expectancies.

Survivor pensions are still needed to smooth living standards after a partner’s death. However, recipients should not be eligible for a permanent survivor pension before retirement age. Instead, temporary benefits should be available to help adapt to the new situation.

The cost of survivor pensions should be internalised within each couple or, as a minimum, among all couples. In a budget-neutral reform, this means that the pension level of singles will be higher than someone living in a couple and benefiting from survivor pensions.

Partners from former unions should not be eligible as they have no current consumption to smooth. Splitting pension rights offers some advantages, although some countries favour the individual treatment of partners, including to promote gender equality.