7. Communicating on investment strategies

In a defined contribution world with individual accounts, people need to choose an investment strategy for their retirement savings or to stay with the default option offered by their pension provider. Choice allows people to select an investment strategy that is adequate given their preferences, risk profile and personal circumstances. The OECD Roadmap for the Good Design of Defined Contribution Pension Plans (OECD, 2012[1]) recommends providing members of defined contribution (DC) retirement savings plans with a choice of investment strategies with different risk profiles and investment horizons, and a default option for those unwilling or unable to choose.

Individuals need to be able to understand and compare the different options available in order to choose their investment strategy. While a financial advisor may assist them in understanding the different options, not all jurisdictions require that individuals seek personal financial advice before selecting their retirement savings investment strategy. Moreover, financial advice can be costly. It is therefore important that regulators ensure that individuals receive adequate information to compare their alternatives and make a choice of investment strategy.

This chapter provides policy makers and pension plan providers with guidance on how to communicate about investment strategies to help members compare their options. It offers an overview of the practices in place in different countries to communicate about different investment strategies. It discusses the merits of different approaches and proposes policy options to ensure members can effectively use the information received to compare their investment options and choose an investment strategy that is appropriate for them given their preferences and personal circumstances.

Different investment strategies have different objectives and characteristics, which may be defined in terms of risk and return profile, of investment horizon, of costs and fees, or by reference to a benchmark. This heterogeneity requires that plan providers communicate about the different options they propose. However, pension providers are in many countries able to choose, to some extent, how to communicate about the different investment strategies they offer to members saving for retirement.

Policy makers need to require retirement savings providers to deliver information to members in a balanced, clear and not misleading manner. They also need individuals to identify the objectives for their retirement savings, based on their personal preferences, risk profile and circumstances, and to be able to assess whether a given investment strategy is suitable for them and likely to achieve these objectives. Based on the information received from regulators and retirement savings providers, individuals need to assess their needs, risk profile and preferences and select the most appropriate option, which may well be the default option.

The chapter starts by looking at the importance of communication for DC retirement savings schemes. Secondly, it explores the ways in which providers communicate on the risk profile of different investment strategies. Thirdly, it examines the communication around historical and expected performance, with a focus on the use of benchmarks to help people compare their investment options. The fourth section explores combined measures of risk and performance used to communicate on investment options. The fifth section looks at the role of regulators and supervisors in ensuring that people can identify their personal objectives and preferences with regards to retirement savings, the challenges they face to communicate about investment options, and ways to assess the effectiveness of communication methods to inform policy choices. The last section concludes with practical options to ensure members of DC retirement savings schemes receive adequate information, can compare their investment options and find an investment strategy that is appropriate given their objectives.

In a DC world, individuals are responsible for making several important decisions that will affect their future retirement income. Depending on the design and structure of the retirement savings arrangement, they may need to choose whether or not to participate in the funded retirement savings system, how much to contribute to it, which provider to select, how to invest their savings, and which pay-out product to elect upon retirement.

Given the potential impact of these choices on future retirement income, the role of communication, i.e. what information is used by people to make their decisions and how it is understood, is essential. Depending on its objective, communication about retirement savings may be broad or personalised, and may be generic or targeted at specific audiences.

OECD work on communication on retirement savings has focused on national pension communication campaigns and pension statements (OECD, 2014[2]), as well as on communicating choices (OECD, 2016[3]) and accounting for behavioural biases (OECD, 2018[4]). These studies have provided important lessons and guidance that can be useful to guide communication about investment strategies to help members compare their options.

National pension communication campaigns are useful instruments to communicate to the general population on the functioning and reforms of retirement savings arrangements. Communication campaigns should be part of an overall national strategy for financial education aiming to improve the financial awareness and literacy of the population, as recommended in the OECD/INFE High-level Principles on National Strategy for Financial Education endorsed by G20 leaders (OECD, 2012[5]) and in the OECD Recommendation of the Council on Financial Literacy (OECD, 2020[6]).

National pension communication campaigns are effective when designed according to clearly set and measurable objectives. These objectives may be defined by governments, pension supervisory authorities or other public entities, possibly in consultation with other stakeholders. Objectives may to be build consensus around the need for reform, to raise public awareness about pensions, to strengthen public trust in pension institutions, to improve people’s understanding and knowledge about pensions or to influence individual behaviours with respect to pensions. They can be linked to systemic pension reforms (e.g. the introduction of automatic enrolment) or have one-off (e.g. gaining public support for parametric changes) or ongoing (e.g. promoting personal savings) objectives. They may also cover the pension system as a whole, or only a specific component (e.g. the voluntary funded system).

Communication campaigns have a broad reach by nature but may be targeted at specific audiences. This could be the case for pension reforms affecting only a portion of the population, or could be a design feature of the communication campaign to address population sub-groups differently in order to achieve better outcomes.

A robust evaluation process should form part of the communication campaign to analyse the effectiveness (impact) of the campaign and its efficiency (cost-benefit analysis). The evaluation process should include pre-campaign research and regular monitoring of the campaign via both quantitative and qualitative tools.

Information should be disseminated in a coordinated fashion when several stakeholders are involved in a campaign, and phased campaigns may be useful to avoid the confusion created by multiple messages. Focused campaigns are more likely to achieve their goals. When private providers or employers are involved in a national campaign, public authorities need to coordinate the dissemination of information to avoid creating confusion.

Pension statements are useful tools to engage members and encourage them to take active steps to improve retirement income adequacy. Pension statements are often the most frequent form of written communication members receive about their funded DC retirement savings plan. They are used by pension providers to communicate about the features of a retirement savings arrangement (e.g. contract information, account balances, asset allocation, contributions paid, fees deducted), and may also present retirement income projections.

Pension statements can use pension projections to convey the uncertainty of future retirement income. The retirement outcome in DC arrangements will depend on several variables whose outcomes are also uncertain. The advantage of pension projections conveying uncertainty is that they can lead individuals to take action. Well-designed and appropriately communicated pension projections may affect the behaviour of individuals by steering them to take active steps such as increasing contributions, changing their investment strategy or delaying retirement.

There is a trade-off between the simplicity of a pension statement and tools such as projections to encourage proactivity. Pension projections may act as a powerful call to action for members but may also appear as complex and create confusion. Mathematical jargon and probabilities should be avoided and user-friendly terms such as “forecast” or “indication” should be adopted to facilitate comprehension and ultimately to encourage proactivity.

There is no consensus on whether nor how to communicate pension projections. The amount of information, they way to present them and the assumptions to compute such projections vary between jurisdictions.

Deterministic projections are easier to understand but may fail to convey the uncertainty around future retirement income. Deterministic projections that show the impact of member engagement may be of practical use. Scenarios illustrating the effect of investment returns and life expectancy on future retirement income require members to understand investment and longevity risks. Projections showing the effect of member engagement, such as increasing contributions for instance, may be more easily understood by members with limited financial knowledge.

Stochastic modelling provides a better picture of the probability range for future retirement income, but it is generally thought to be too complicated for members to use in a practical way. Stochastic modelling is complex to prepare and may be hard for members to interpret.

The assumptions and rules used by providers to compute pension projections should be set by pension supervisors and regulators (IOPS, 2018[7]). To facilitate comparison and ensure the usefulness of projections, the rules or ranges for the different parameters to be used by public and private pension providers should be harmonised (Stańko, 2019[8]).

One key issue when communicating about retirement savings, whether through national pension campaigns or in personalised communication such as in pension statements, is the low level of financial literacy of most individuals, in particular when managing their resources and planning for retirement (OECD, 2016[3]). In DC arrangements where people are responsible for more decisions affecting their future retirement income, the required knowledge and skills are even greater than in other types of pension arrangements (e.g. PAYG, DB) where public authorities make some or most decisions on behalf of individuals.

Financial education may address this lack of financial literacy by making sure people have sufficient financial knowledge and understanding, by promoting attitudes oriented towards the long term, and by providing people with the skills to engage about their pension arrangement. Financial education tools to address the needs of individuals in DC schemes include information and awareness campaigns, which can be general or personalised, instruction initiatives such as trainings and workshops, and the provision of advice (OECD, 2016[3]).

Behavioural biases may also affect people’s financial decisions about retirement savings, ultimately affecting their future retirement income (OECD, 2018[4]). While more individual choice may be welfare enhancing if agents are rational, behavioural economics show that this is not necessarily the case and that psychological factors affect people’s choices. Behavioural biases include preferences such as the present bias, beliefs such as overconfidence and inaccurate decision-making processes such as rules of thumb and framing.1 Box 7.1 provides an illustration of the potential impact of framing on financial decisions such as when saving for retirement.

Policy makers may use various features to improve the design of funded pension arrangements and assist people in these decisions.2 Five groups of policies improve retirement incomes while considering behavioural biases and limited financial knowledge (OECD, 2018[4]): automatic features, such as automatic enrolment; default options, such as setting a default contribution rate or a default investment strategy; simplification of information and choice, such as standardising and reducing the number of options; financial incentives, such as matching contributions and tax incentives; and financial education.

Communication is an integral part of the policies aimed to support retirement savings choices for individuals. All five groups of policies involve communication, either indirectly or directly. Communication is required indirectly for automatic features, default options and financial incentives to explain their functioning. Communication is an integral part of financial education, and of the simplification of information and choice, as both these groups of policies consist in making sure individuals have the appropriate level of knowledge and understanding to make decisions related to their retirement savings.

This body of work has provided important lessons, but the analysis of communication has to cover other areas where individuals need to make choices, such as when selecting an investment strategy. Some of the lessons learnt are useful for communicating on investment strategies to facilitate comparing options. Communication needs to address the trade-off between simplicity and accuracy. In addition, communication should focus on one issue at a time to avoid creating confusion. Moreover, appropriate framing can greatly facilitate communicating messages. Finally, it is essential to be able to evaluate the success of different communication experiences with respect to their initial objectives to prepare and implement adjustments as required.

Communicating on the different features of retirement savings schemes is essential for people to make choices. National pension communication campaigns can assist people in understanding the purpose and objectives of saving for retirement as well as how the overall retirement system works. Pension statements can encourage people to make choices towards improving their retirement outcomes. However, for this, policy makers need to understand the behavioural biases that people have when making choices and adjust communication accordingly. In particular, choosing an appropriate investment strategy is an essential element in this context. Policy makers can provide default investment strategies, following the main guidelines provided in Chapter 4 of this volume. However, people still need to decide whether to remain with the default or to select a different investment strategy, and for this they need to understand the potential risks and rewards that different investment strategies, including the default, provide them.

The communication of potential risks associated with the different investment strategies available requires choosing a metric to assess these risks. Measures of risks used in different jurisdictions include the volatility of returns, value-at-risk, expected shortfall, and the share of risky assets in a portfolio.

Volatility of returns is the most widely used indicator of risk among OECD jurisdictions. Volatility is a measure of the dispersion of investment outcomes. It is the standard deviation of historical returns, observed over a specified time horizon. It represents how much returns fluctuate around their average. Two investment strategies may achieve similar returns on average over a year or over a longer period. However, volatility captures how this average annual return is reached and how wide the daily, weekly or monthly returns fluctuate around that average.

The choice of a time horizon may have important impacts on the interpretation of volatility. For retirement income investment strategies, which are meant to deliver returns over a long period of time, it may not be appropriate to focus on short-term horizons such as one year. However, looking at very long-term returns, such as over a period of 40 years for instance, raises the issue of the availability of historical data. Even with long-term investment in mind, policy makers may still want individuals to understand that shorter-term returns may not be smooth and that fluctuations are to be expected. Short-term volatility also remains important for those at, or close to retirement, as they may be more vulnerable to short-term losses and lack the long-term investment horizon necessary to recoup them. The choice of the time horizon to examine the volatility of returns may also depend on whether the risk of retirement income investment strategies is communicated separately from that of other types of (shorter-term) investment strategies.Countries using volatility as a measure of risk consider the annualised volatility of investment strategies over periods ranging from one year to ten years, with most countries taking five years of historical data into account.3

Volatility may be based on actual historical portfolio data, or on representative assets and harmonised assumptions. In this case, the volatility of a portfolio of reference assets weighted in accordance with the considered strategy’s asset allocation is used to compute volatility. For consistency and comparability, methodologies and reference assets may be set by the regulator or an industry association (e.g. Insurance and Pension Denmark, the Dutch Banking Association).

Volatility is a widely used risk measure for pension providers, even in countries where there is no obligation to report volatility in pension fund documents (e.g. Israel, the United States).

Volatility may however not be a comprehensive representation of the risk profile of an investment strategy. Volatility is an estimate based on historical data, which may not be a reliable indicator of future investment returns. Volatility also does not take into account all investment risks, for instance credit and liquidity risks. The volatility of an individual investment portfolio may therefore not represent its complete risk profile.

Pension authorities also look at the share of equities in the investment portfolio as another metric to determine the risk of different portfolio investment strategies. Several countries classify fixed income securities and mutual funds as low risk assets and include all other assets in a “higher risk” or “growth” asset class. This higher risk class usually includes equities, alternative investments such as infrastructure, private equity and hedge funds, and corporate bonds below a certain credit rating – typically below investment grade. The share of higher risk assets in the investment strategy is used as a proxy for risk. While this type of metric is not a measure of risk in itself, it provides a basis for comparing the riskiness of different investment portfolios.

A variation of this metric is to look at the characteristics of assets in the portfolio in order to assess its risk. Looking at whether a product offers a guarantee, or at characteristics such as its legal maturity or currency risk can help measure the investment risk of a product. Italy for instance, assigns different risk metrics to products depending on whether they include a guarantee.

Stochastic modelling may also be used in order to simulate the returns of investment strategies and derive risk measures. Stochastic modelling allows generating simulations of investment returns and deriving probabilities of certain events happening. Value-at-Risk and expected shortfall are examples of metrics used by OECD countries to quantify the risk of investment strategies using stochastic modelling.

Value-at-Risk (VaR) is the pre-determined bottom percentile resulting from the stochastic distribution of investment returns. For instance, savers in a fund that has a 5% VaR of -20% over one year can expect to lose at least 20% of their savings with a 5% probability over any given year.

The expected shortfall provides a measure of the severity of the potential loss, based on the stochastic distribution of investment returns. While the VaR indicates how much savers can expect to lose at least for a given probability, the expected shortfall indicates the expected magnitude of the loss, conditional on being in the bottom percentile of returns.

Current and historical data may permit the computation of various additional descriptive risk metrics.

The risk of a portfolio may depend on the maturity profile of its fixed income components. The modified duration of a bond portfolio, i.e. the sensitivity of the bond portfolio to a change in interest rates, or the share of bonds broken down by maturity, are metrics used to reflect this risk.4

Risk may be a function of the diversification of an investment strategy. The geographical distribution of investments can provide information on the potential exposure to geographical risk.

Risk metrics may also integrate elements linked to long-term sustainability and the integration of environmental, social and governance (ESG) factors. For instance in Sweden, the Swedish Pensions Agency discloses a measure of sustainability risk for pension investment strategies, based on their holdings over one year.5 It also shows a ranking of carbon risk, based on the expected capability of a portfolio to perform in a low-carbon economy.6

Various metrics are used to assess the riskiness of different investment strategies, but many strategies are complex and require the knowledge of several financial concepts to be fully understood. However, not all individuals are able to assess the riskiness of a retirement savings portfolio based on a volatility figure, or a value-at-risk number. For instance, Chłoń-Domińczak, Kawiński and Stańko (2013[14]) found in focus studies that pension fund members in Poland had great difficulties in understanding the basic notions of risk, including the concept of standard deviation.

Jurisdictions and providers therefore often use categorisations, based on their chosen risk metric, in order to assist people in their choice of an investment strategy. Risk categorisations may be associated with a risk or investor profile, guiding individuals by providing a qualitative assessment of investment strategies. They may also be associated with age groups or investment horizons to direct people towards an investment strategy based on their age or time to retirement.

To communicate on risk categorisations, visual aids are often used to help individuals associate a risk measure to a sentiment. Colour coding risk measures, or using infographics are widespread means of communicating on the risk of different investment strategies. Although visual aids may not always improve individuals’ comprehension of complex matters such as investment risks, McGowan and Lunn (2020[15]) show that diagrams may promote the causal connections they are designed to highlight, and ultimately influence people’s decision-making related to saving for retirement.

Many OECD jurisdictions using volatility as a risk metric use categorisations or ranges to assist people in interpreting it. Providing guidance as to where on a scale a volatility measure is can be helpful for people to understand the risk profile of an investment strategy when saving for retirement. Countries or institutions choosing this approach generally define a limited number of volatility ranges (e.g. three for Insurance and Pension Denmark, five for Sweden’s Pensions Agency, seven for the European Securities and Markets Authority, the Dutch Banking Association and Turkey’s Pension Monitoring Center) as using many categories may prove confusing for individuals when assessing the risk profile of different investment strategies.

Labels may be added to risk categorisations in some jurisdictions to facilitate interpreting risk measures based on volatility. For example the Turkish Pension Monitoring Center (PMC) uses a methodology with seven risk ranges based on the five-year volatility of returns, similar to the one used by the European Securities and Markets Authority (ESMA). However, it also defines a labelling of risk ranges: a strategy may be labelled conservative or cautious with a ranking of 1 to 2, balanced with a ranking of 3 to 4, dynamic or growth with a ranking of 5 and aggressive with a ranking of 6 or 7, whereas this qualitative classification does not exist in the ESMA categorisation.7

Visual techniques may help interpreting the volatility of investment strategies. The use of a colour code or a graphic representation may help members form a view on the risk level of different investment strategies. Many OECD countries combine the use of a risk categorisation with a visual element in order to make risk more palatable for pension scheme members. Some regulators however, such as the French Autorité des Marchés Financiers (AMF), prohibit the use of a colour coding when using the ESMA scale of risk.8

Sorting risk categories from green to red is a commonly used visual aid. The traffic light colour coding is an easy way to understand scale for most individuals and is widely used by pension authorities and pension providers. The Swedish Pensions Agency, for instance, classifies investment strategies in five risk categories based on historical volatility over three years. Each category is assigned a colour from green for the lowest risk class, through orange for the average risk class, to red for the highest risk class.

Another frequently used visual to represent risk is the speedometer, although it may not always be efficient in conveying information about risk. As an example, in May 2016 the Dutch Banking Association (NVB) introduced the risk meter, a voluntary standardised investment risk indicator to facilitate comparison between investment strategies. This indicator has values ranging from 1 to 7 and its graphical representation is that of a speedometer.9 For comparability purposes, it is based on the volatility of returns of a portfolio of reference assets representing the investment strategy under consideration.10 In Chile however, focus groups showed that the use of a speedometer visual did not always help people understand the information shown, but rather frequently led them to associate the option displaying the highest “speed” with the best outcome, regardless of the explanations provided (Antolin and Fuentes, 2012[16]).

Volatility and its impact on long-term returns may be communicated via graphs. In a series of articles published on their website, a Canadian pension provider illustrates the meaning of volatility by showing: 1) the historical odds of returns being positive over various holding periods for two of their retirement strategies, represented by histograms increasing with the holding period; and 2) the effect of volatility on returns over different time periods (one year, three years, five years and ten years), represented by graphs showing the rolling returns over these periods and the diminution of negative rolling returns as the holding period increases.11

The most common approach to communicating risk based on the share of higher risk assets in a portfolio is to determine risk profiles with asset allocation limits. Jurisdictions using this approach use between three (in Colombia, France, Latvia, and Slovenia) and five (in Chile, Hong Kong (China), Italy and New Zealand) categories. Other asset allocation criteria may be used in addition to the share of equities in a portfolio. For instance, the risk categories of different investment strategies also depend on the share of investments in local fixed income assets in Colombia, or on the presence of a guarantee in Italy and Slovenia.

Associating a qualitative factor with the risk class is a common communication technique. In France, the AMF recommends the use of three risk profiles – Prudent, Balanced and Dynamic – and limits the share of higher risk assets to 30% of the portfolio for strategies to be labelled “Prudent”. In New Zealand, there are five risk categorisations for KiwiSaver pension funds, from Defensive to Conservative, Balanced, Growth and Aggressive, according to their holding of growth assets.

Retirement income providers may also use their own risk categorisation based on the share of growth assets in the investment portfolio, if the categorisation is not prescribed by law or by regulation. This is the case in the Netherlands, in Iceland and in Korea for instance.12

Assigning risk profiles to age groups may help to communicate about the risk of different investment strategies in a way that is perceived as more personalised by individuals. Chile, Colombia, Luxembourg, New Zealand and Slovenia associate risk profiles to age groups or investment horizons, and restrict investment strategies based on age. Individuals may also be defaulted into a certain risk profile based on their time to retirement (e.g. Chile, Colombia).

Age groups may be defined by law or regulation, or be left for providers to determine. In Slovenia for instance, retirement income providers must offer three life-cycle investment strategies for different target age groups, including a guaranteed option for those closer to retirement. Different supplementary pension providers have designed different target age groups and consequently different rules for their risk profiles based on asset allocation.

Like volatility or asset allocation categorisations, probabilities are often used to delimit risk categories, which are also defined qualitatively. In Australia, the standard risk measure provides retirement income scheme members with the frequency of a negative annual return to be expected for any investment strategy.13 The measure is based on the probability distribution of each investment strategy, based on return, volatility and correlation assumptions by asset class. However, the measure is communicated to savers via seven risk bands, with each band corresponding to a number of negative years to expect over a saving period of 20 years, as shown in Table 7.1. The Australian standard risk measure, with risk levels from very low to very high, helps individuals assess the risk profile of different strategies. Additionally, investment strategies labelled “Conservative” should have a risk level of very low to medium.14

Pension authorities and providers often avoid mentioning probabilities in their communications and translate probabilities into more basic concepts. Probabilities and stochastic simulations are complex. Many individuals do not clearly understand probabilities and their implications. In the Australian example, the standard risk measure is communicated as the “frequency of a negative annual return to be expected” for any investment strategy. In Denmark, Insurance and Pension Denmark’s volatility based investment risk measure is also mapped to provide individuals with the maximum decrease in accumulated assets to be expected over one year with a 95% probability, i.e. the 5% VaR of the return distribution. However, the association’s guidelines advise on communicating this figure as “the highest drop in the value of savings over one year” and adding an asterisk indicating “with 95 percent certainty”.15

This section looks at ways in which different OECD jurisdictions communicate on the potential rewards of investment strategies.

Communicating about the rewards of different investment strategies implies first deciding on how to measure the relative performance of different investment strategies and then choosing a metric to represent that performance. OECD jurisdictions use several types of performance measures to assess the rewards of investment strategies. Some countries look at the performance of investment returns, while others focus on the probability of reaching certain objectives, or on retirement income projections.

Absolute and relative performance measures may be used to communicate on the rewards of investment strategies. Absolute performance measures include measures of returns on retirement savings, savings accumulated at retirement, or projections of retirement income. Such measures may also integrate the cost and fee structure of different investment strategies in order to assess their net performance. Relative performance measures may consider similar outcomes and are based on the comparison with a benchmark, peer group or with an objective.

Many jurisdictions use both the absolute and the relative performance of investment strategies. Whether it concerns absolute or relative performance, communication may focus on actual historical performance or on potential rewards of retirement income investment strategies. While historical data only focuses on past performance of similar strategies, potential performance can be based on deterministic or stochastic projections of returns on savings or retirement income.

Most jurisdictions communicate on potential rewards mainly by requiring retirement income providers to show the historical returns of investment strategies. All communications include a mention that historical returns are not a guarantee of future returns.

Historical returns are shown using the annual or annualised return over periods of time ranging from a few months (e.g. in Finland, Israel, Poland, Portugal, Sweden and the United Kingdom) to up to 15 years or since the launch of the fund (e.g. in Finland, Hong Kong (China), Hungary, Poland, the Slovak Republic and the United States). Many pension supervisors require providers to show historical performance over different time horizons.

The most common way of presenting historical returns is to use tables or graphs. Several OECD jurisdictions use a combination of both graphs and tables in order to cater for different individual preferences.

Historical returns can be presented in the form of simulated monetary amounts. The Chilean Superintendencia de Pensiones (SP) website includes an online tool where individuals can enter their pension account balance and view the return (positive or negative) they would have received net of inflation for each of the AFPs and each of the investment strategies, if they had saved this balance in a fund for the past five years.16 Five tabs representing the five risk categories allow a comparison of the returns across the different providers and investment options, both in monetary amounts and in accumulated and annualised return terms. The tool also summarises the monetary difference between each provider and the lowest performing one.

Many countries also integrate fees and costs in their review of investment strategies’ performance. Fees and costs directly affect the potential performance of investment strategies and jurisdictions therefore often include fee indicators in the potential performance disclosure requirements.

  • Pension authorities may publish a ranking of retirement savings funds based on the fees they charge to members. Rankings may use fee levels in percentage points (e.g. Poland) or provide the estimated cost of saving a regular amount with different providers (e.g. Sweden).17

  • A cost metric may also be designed to make all costs comparable across funds and help individuals understand the impact of costs and fees on retirement savings. It may provide the annual equivalent cost over one year (e.g. Australia, Germany, Hong Kong (China), Mexico, the United States), or over different accumulation periods (e.g. Italy).18

Services offered by pension providers and linked to different investment strategies can play a role in the costs and fees charged, and hence may also be a factor individuals need to take into consideration when comparing different investment options. Box 7.2 illustrates how communicating on the level of service may be useful for individuals to compare different retirement savings providers and their investment options.

Many OECD jurisdictions use benchmarks to assist DC scheme members in comparing investment strategies. For performance measurement, a benchmark is a standard or reference point to which any investment strategy can be compared in terms of relevance as a means of investing for retirement, based on historical performance or target expected return.

Benchmarks may be set as investment objectives to reach. They can be defined across all investment strategies, or differentiated based on a categorisation of funds by risk-return profile.

  • In Colombia for instance, the Superintendecia Financiera (SFC) establishes a reference portfolio of traded indices for the three risk categories of pension funds. This reference portfolio is used to determine the minimum return attached to each risk category.19

  • Inflation is a widely used objective to reach for retirement income investment strategies. In Australia for example, APRA defines return targets for retirement savings in reference to the consumer price index (CPI) growth over 10 years and any investment option should be evaluated in its ability to achieve an inflation-related objective over 10 years for the purposes of MySuper product dashboard requirements.20 Pension providers in the Slovak Republic are also required to communicate on the historical returns of their retirement investment strategies compared to inflation over different periods of time.

Benchmarks may also be used as mere performance comparators.

  • Communication can focus on the relative performance of pension funds compared to that of a peer group. In Turkey, the PMC has put in place a fund performance assessment system in order to assess the relative performance of pension funds to that of comparable funds, as described in the law and in the Pension Mutual Fund guidelines.21 The performance of pension funds is evaluated on a yearly basis by comparing each fund’s performance to that of its comparison group, based on fund type and risk level.22 For each comparison group, the average annual performance of each fund is compared to that of the average of all funds in the group. Funds generating a return above (below) the group average plus (minus) a group standard deviation are considered relatively successful (poor).23

  • In Australia, APRA has developed a heat map which shows, among other things, a retirement product’s investment performance over different time horizons against various reference portfolios and peer products.

Benchmarks may be mandated by regulators (e.g. Australia, Colombia, the Slovak Republic, Turkey) or chosen by pension providers (e.g. the United Kingdom, the United States).

A common benchmark or a limited set of benchmarks with harmonised characteristics may be more helpful for individuals to compare options across providers, than allowing each provider to choose one or several benchmarks.

The lack of standardisation may make it hard for individuals to compare across pension providers and then among investment strategies and may increase the influence of factors outside of the risk and return profile, such as marketing and brand name, in people’s choice of an investment strategy. For example, in Mexico, confusion and the lack of transparent information can lead individuals to inefficient choices with respect to their pension and to a lack of competitive pressure among pension managers (Calderón-Colín, Domínguez and Schwartz, 2010[17]). Pension members may seem more influenced by marketing efforts, the size of the pension manager and whether it belongs to a recognisable financial group rather than by the risk and return profile of pension funds.

Allowing pension providers to use different benchmarks for different funds may allow more granularity and representativeness. Many OECD countries, such as the United Kingdom, encourage pension providers to choose a benchmark for marketing and reporting purposes. The benchmark is often related to the asset universe or asset allocation of the fund under consideration, in order to facilitate comparisons of performance between the fund and its benchmark. In the United Kingdom, pension fund managers may, however, use their discretionary power to invest in significantly different asset classes than their reference benchmark, as long as the degree of freedom available is indicated in the fund documents.24

When pension funds use their own benchmark, comparability across investment strategies and providers is impeded. In OECD countries where pension funds use their own benchmark, it is more difficult to compare pension providers and investment strategies. Using a single dedicated pension benchmark, or a limited number of benchmarks, linked to risk profiles, member ages or other factors individuals could easily identify, may allow new and existing members to have a clear view of the different options and their respective merits and shortcomings.

Such benchmark should also be set by an independent body in order to avoid any potential conflict of interest. If all retirement income investment strategies are compared to a single benchmark, it may be advisable that individual pension providers cannot influence or decide what their reference point should be.25

Using defaults as pension fund benchmarks can be a solution to assist individuals in their choice of pension provider and investment strategy, while making sure benchmarks remain aligned with pension objectives. Comparing pension funds and investment strategies to defaults could be a means to assist people in understanding their options and making a choice of pension provider and investment strategy (Gokçen and Yalçin, 2015[18]).

Regulators define default options in many countries in line with the long-term nature of retirement savings and with the majority of DC members’ pension objectives to make sure that people unable or unwilling to make choices are protected. So using readily available defaults as benchmarks could be a useful tool for individuals to compare their different retirement saving investment options.

Several countries, in line with the OECD Roadmap for the Good Design of Defined Contribution Pension Plans (OECD, 2012[1]), have established life-cycle default investment strategies, to protect people close to retirement from negative shocks. Life-cycle default strategies offer differentiated investment allocations according to pension fund members’ age, in order to adapt to the change in pension funds’ objectives and risk appetite as members approach retirement. They can therefore be suitable reference points for individuals to compare different investment strategies in terms of their risk and reward profile.

This section examines the communication about risks and rewards jointly and the ways in which both can be combined for the purpose of assisting members in their choice of an investment strategy.

Linking investment risks and potential returns may assist people in understanding the relationship between the two aspects and the trade-off between different investment strategies. In their study about the Chilean pension simulator, Antolín and Fuentes (2012[16]) find that users tend to associate risk with negative outcomes, to ignore the upside potential and to believe there was nothing they could do to reduce or act upon risk.

Classifications combining risk and rewards may be useful to communicate pension choices to individuals. In Germany, the use of a chance-risk classification allows forming an overall opinion on the risk and return profile of different pension investment strategies, based on an independent provider (Box 7.3). Similarly, in the United States or in Mexico, providers can rely on risk and return ratings provided by the independent consultant Morningstar.26

Using similar metrics to put in perspective potential risks and rewards may be useful in conveying the risk-return profile of different investment strategies.

  • Expected risks and rewards may mirror each other and be communicated in a table format. Table 7.2 shows the table included in a Dutch pension provider’s information brochure, detailing measures of expected risk and rewards for four investor profiles. The metrics used are the expected return, the downside risk, the upside risk and the investment horizon. The expected return is the expected average return over a period of 10 years. The downside and upside risks are the 5% and 95% value-at-risk over 10 years. In Iceland, another provider’s website shows the average, lowest and highest nominal annual return obtained by each investment strategy over the past five years.27

  • Monetary amounts may also be easier to understand than a percentage and hence help communicate expected risks and rewards to the general population. The Chilean SP publishes a monthly report in which it details monetary ranges for the possible monthly gains and losses of each of the five categories of funds for a given retirement pot of CLP 9 000 000, based on historical returns.28 The risk bands are presented in an infographic, with possible losses shown in dark blue and possible gains in light blue.29

  • The use of scenarios may also allow for the communication of both the risks and rewards of investment strategies. Product information sheets may include performance scenarios presenting the potential risk and rewards of any considered investment strategy, by showing the possible net outcomes both in monetary value and in yearly return equivalent for a given investment over different time horizons.30

  • Simple graphs and tables may be more effective to communicate the potential risks and rewards to individuals than more elaborate graphical presentations including probabilities. Consumer testing on approaches to show the performance of different investment strategies done by the European Commission in 2015 shows that individuals understand simple tables or line graphs better than graphs with scenario ranges (funnel of doubt) or histograms with probabilities of occurrence (London Economics, 2015[19]).

  • Cartoon figures or representative characters may be used to personify the different risk and return profiles. The Hong Kong, China MPFA defines five major types of funds, based on their risk and return profile: the conservative fund, the guaranteed fund, the bond fund, the mixed assets fund and the equity fund. Each fund type is represented by a cartoon character.31 The five fund characters are detailed over two pages in a booklet, which also contains a recap table for members to understand the main characteristics, associated risks and objectives.

Metrics combining risks and rewards can also be used to compare the profile of different strategies, if appropriately communicated. The Sharpe ratio for instance, provides a measure of risk-adjusted returns by dividing the performance of an investment strategy by its volatility over a chosen period, and is used by regulators and pension authorities to help compare different investment strategies (e.g. Israel, Sweden).32 Given the complexity of such measures, visual aids or categorisations may be required for individuals to understand them and use them when deciding on their investment strategy.

Providing the portfolio composition can also be a way to communicate on potential risks and rewards. Whether or not a benchmark is used, communicating on the asset allocation of an investment strategy can convey a sense of the risk and return expectations to members, as some individuals will infer different performance prospects from different asset classes.

Individuals need to be guided to evaluate their personal objectives and risk appetite when saving for retirement. This section covers some of the measures in place in various countries to assist individuals in assessing their retirement savings investment profile and selecting the appropriate investment strategy, and the challenges linked to communicating about different investment options to individuals. It concludes with an overview of some of the methods used to assess the effectiveness of communication choices to inform policy decisions.

People need to identify the objectives for their retirement savings based on their preferences and circumstances. While certain jurisdictions recommend that individuals saving in a retirement plan consult a financial advisor, others offer individuals or providers means to assess their objectives in terms of investment horizon and risk appetite.

Surveys and studies show that people tend to overlook general financial advice such as messages encouraging them to save more for their retirement, and to prefer more personalised advice (Box 7.4).

Financial advice may be recommended or required in order to evaluate the risk appetite of individuals willing to save in a pension plan. The French AMF requires the provision of personalised financial advice to savers in an individual pension plan in order to verify the adequacy of their investment strategy choice with their actual investor risk profile. In Colombia, personal financial advice is required for individuals considering a switch from the public DB to a private DC scheme.33 Similarly in the United Kingdom, professional advice must be sought before switching from a guaranteed pension or from a DB to a DC pension plan if the plan transfer value exceeds GBP 30 000.34

Pension regulators often use questionnaires to provide individuals with guidance on the choice of risk categorisation corresponding to their preferences. These questionnaires may be posted on regulators or other independent entities’ websites, or integrated to the subscription form of retirement savings products.

  • In Italy for instance, the Pension Fund Supervisory Commission (COVIP) requires all pension providers to include a self-evaluation questionnaire in their subscription documents to assess the risk profile of individuals willing to join a pension plan.35 This questionnaire includes qualitative items related to financial literacy and awareness, but also ranked items related to savings horizon, risk appetite and to the private savings capacity. The results of the questionnaire define which type of investment strategies may be suitable for the evaluated individual. People scoring 4 or less are directed towards investment strategies with guarantees or investing exclusively or mostly into fixed income. People scoring between 5 and 7 are directed towards balanced investment strategies or those investing mostly into fixed income. People scoring 8 or more are directed towards balanced or equity investment strategies.

  • In New Zealand, nine questions allow individuals to determine their investor type and the associated recommended category of investment strategies through the Sorted website.36

  • The Irish Pensions Authority offers an investment risk profiler consisting of five questions, which provides individuals with a measure of their personal risk appetite. Results are shown based on the speedometer visual, on a scale of 0 to 10 with 0 being the most risk averse profile. This investment risk profiler is part of a section called “Understanding your pension” on the Pensions Authority website, where individuals can find information on investment asset classes, the link between risks and potential returns, and the difference between active and passive asset management.

In order to assist members in understanding the different types of funds and how their risk appetite may change throughout their professional career according to their personal circumstances, the MPFA in Hong Kong, China set up an investment education thematic website. The website aims to educate scheme members on MPFA investment choices, including the relative risk-reward profile of the five major types of funds. The website includes online tools, such as the MPFA calculator and many infographics and visual aids, such as the risk-return profile cartoon characters.

Standardisation improves comparability between providers. Before the Insurance and Pension Denmark risk label came into force in 2018, different providers in Denmark labelled very different investment strategies as low risk. Similarly in the Netherlands, the risk meter has been endorsed by the Dutch Financial Markets Authority (AFM) as a consistent and useful tool to help consumers objectively assess the risk profile of different investment strategies across different providers.

Standardising the communication on risks may be necessary for comparison purposes, but should also be accompanied by visual aids for people to interpret risk measures. Providers using standardised risk indicators may mix them with other commonly used visuals or colours in order to ease communication. For instance, mixing a standardised volatility ranking, the speedometer and the traffic light approaches may help make the risk classification visual and accessible to individuals.

The definition of a risky portfolio needs to account for differences across countries as the choice of terminology can have an impact on people’s perception of risk. Norway’s DNB offers four risk profiles for members of its retirement savings plans: the most conservative option includes 30% equities, while the riskiest profile invests 100% into equities. In countries like Chile, an equity allocation of 30% classifies as moderate to high risk and would fall into the third out of five risk categories. The prudent profile is the less risky option in France, whereas it is the middle profile in Slovenia.37

The use of adjectives and qualitative elements by providers may blur the message of standardised risk indicators. In the Slovak Republic for instance, a provider communicates on the four different risk profiles offered in its supplementary pension plans by associating each one with an adjective and a character.38 The conservative option is represented by Mr. Cautious, the balanced option by Mr. Balanced, the growth option by Mr. Brave and the index fund by Mr. Action. These visual and qualitative elements may be useful to help people form a view on the overall risk profile of each investment strategy. However, the two riskiest strategies, growth and index, are represented by the same picture and very similar profile descriptions, although the ESMA risk ranking of the two strategies differ, with the growth fund ranked 4/7 and the index fund ranked 6/7.

Using different risk indicators may bring additional transparency and information to individuals, but can also lead to confusion for members, especially if they carry inconsistent risk rankings. In the Irish Pensions Authority example, a person with a medium risk appetite of 6 out of 10 may be confused when selecting a pension investment strategy from a provider that uses the ESMA ranking from 1 to 7.

Generic risk rankings may not be appropriate for long-term investment such as when saving for retirement. The ESMA risk categorisation applies to any investment fund in Europe, whatever its recommended holding period. It is based on the 5-year volatility of returns, is not meant to reflect long-term risk and would probably assign a high-risk ranking to balanced 30-year investment strategies. Similarly, a traffic light approach assigning a high-risk ranking to all long-term products, such as the one used in Spain, is not appropriate to classify the risk of retirement savings strategies.

Getting people’s attention in the first place may also be an important communication challenge. This is because people have a present bias and tend not to want to think about retirement. They tend to focus on what is urgent and “tunnel” down on those tasks at the expense of other priorities. Proper timing for pension communication and careful design of such communication to arouse people’s curiosity may help to overcome this challenge (Box 7.5).

Disclosing information on the risk of different retirement savings strategies is important, but should be done in a way that is understandable to the target audience. Although they might not always be fully transparent, risk categorisations help individuals interpret risk measures and better understand the risk profile of investment strategies. Some countries require a lot of transparency on risk, but it is not clear that this actually helps people get a feel for the risk of the different retirement savings options. Consumer testing by the European Commission when designing the Key Information Document of Packaged Retail and Insurance-Based Investment Products (PRIIPS) in 2015 shows that simple scales are more effective to communicate on the overall risk level of a product than multi-dimensional graphics (London Economics, 2015, p. 45[19]).39 Only 58.3% of respondents correctly classified investment strategies as high, medium or low risk when provided with a multi-dimensional graphic showing overall risk as well as market risk, credit risk and liquidity risk, versus 74.2% of those shown a horizontal scale of 1 to 7, and 73.4% of those shown a simple vertical graphic resembling an energy-efficiency scale from A to E.40

The most appealing risk indicators may not communicate risks in the clearest way. Individuals taking part in the 2015 consumer testing by the European Commission found risk rankings such as the energy efficiency scale easier to understand than horizontal scales of 1 to 7 (London Economics, 2015, p. 51 and 61[19]).41 However, when asked questions about the relative risk of different investment strategies, individuals answered correctly more often with the simple horizontal scale. The horizontal scale was especially effective in conveying the link between risks and potential returns, with over 10% more individuals correctly answering the questions about the risk-return profile when shown a horizontal scale of risk from 1 to 7 compared to those shown an energy-efficiency scale of five categories.42

Perceived complexity may deter individuals from choosing a retirement savings product. Based on a consumer testing exercise carried out in Italy, Gentile et al. (2015[23]) find that synthetic risk indicators are perceived as less complex by individuals than unbundled risk measures, “what if” scenarios and probabilities.43 The study also shows that perceived complexity is the main driver of the willingness of individuals to invest in a given option, as it always contributes to reducing the propensity of individuals to invest. Similarly, two consumer testing exercises commissioned by the ASIC and performed ahead of the introduction of a mandatory product dashboard for defaults (MySuper dashboard) and other investment strategies (Choice dashboard) by APRA in Australia, found that individuals perceiving pension choices as complex were more likely to disengage or to focus only on the elements they find simpler, ignoring what was deemed complex (ASIC and Latitude Insights, 2013[24]; ASIC and Latitude Insights, 2015[25]).44

Presenting information in a relatively simple manner helps people make decisions. The way information is presented may affect not only comprehension, but also decision-making ease and confidence, and the willingness to choose one product over another (Box 7.6).

Different elements are understood differently by DC scheme members. The Australian consumer testing found that risk bands or categorisations are better understood than probabilities, which often create confusion, while historical returns appear clearer when shown in graphs rather than tables or text (ASIC and Latitude Insights, 2013[24]; ASIC and Latitude Insights, 2015[25]). Individuals taking part in the consumer testing exercises also preferred to see asset allocations shown in pie charts, return target figures mentioned in plain language (such as “inflation + 3%”), and to see costs and fees shown in monetary amounts, with an industry average also shown for comparison purposes.

Communication about investment strategies and their associated risks, rewards and costs needs to be adapted to the target audience. Jargon and complex metrics should be avoided when communicating to individuals about their investment options.

Standardising the communication on potential risks, rewards and costs helps individuals understand and compare the different risk, return and cost profiles of retirement income investment strategies. Whether risks, rewards and costs are presented separately or together, using categorisations or straight measures, different providers should be encouraged to present risk, reward and cost metrics in a similar way so that individuals can compare these characteristics.

The use of several risk indicators can create confusion for individuals. Mixing different types of risk indicators may create confusion rather than increase transparency for individuals, unless there is a consistent mapping of the different risk indicators with risk categories or risk profiles, which people can understand and interpret.

Visual aids are effective ways of communicating on the risk and return profile of investment strategies. Using familiar colour codes or visuals associated to risk such as traffic light colours or the speedometer visual may help people interpret risk indicators or categories. Representative characters or cartoon figures with different personalities or ages can help convey information on the risk and return profile of different investment strategies. Such visual aids should however be tested to ensure they are well interpreted by individuals and do not create misleading perceptions.

Associating qualitative characteristics to investment strategies may help individuals appreciate their risk and reward profile, but may also leave room for interpretation. Regulators and supervisors should provide a framework for providers to associate a qualitative assessment to the risk and return profile of investment strategies, based on the chosen indicators.

Policy makers encouraging the use of benchmarks in communication to DC plan members should clearly define the purpose of such benchmarks. A single or limited set of benchmarks, defined according to characteristics that people can easily identify, such as age or risk profile, is more effective to help individuals compare different investment strategies. Using different benchmarks chosen by pension providers may be suitable when used only as performance targets.

Using defaults as benchmarks may be useful to help individuals compare their options. In countries where default investment strategies comply with criteria defined by regulators, using the default as the comparator could help individuals assess the performance, costs and risk profile of different investment strategies in order to choose the strategy that is best aligned with their personal preferences and circumstances.

Policy makers should consider designing tools to assist people in determining their risk appetite. When professional financial advice is not required, regulators and supervisors may be better placed than retirement income providers to help individuals get an unbiased assessment of their personal risk tolerance and investment horizon.

Supervisory authorities should provide guidance on the mapping of risk and return categorisations to that of individuals’ personal risk appetite. People should be able to understand whether an investment strategy’s risk and return profile is appropriate, given their personal circumstances and preferences.

Policy makers should assess the effectiveness of their approach to communicate risks and rewards and adjust policy accordingly. The impact of different communication approaches on individuals’ understanding of retirement savings investment strategies and on their actions, should be evaluated through, for example, consumer testing. Policies should be adjusted to reflect the findings from these evaluations.

References

[27] Acharya, S. et al. (2015), Asset Price Effects of Peer Benchmarking: Evidence from a Natural Experiment.

[32] Allport, W. et al. (2019), How the UK Saves: Effects of gender on retirement savings behaviour Member experience from the National Employment Savings Trust (NEST), Vanguard and NEST Insight.

[16] Antolin, P. and O. Fuentes (2012), “Communicating Pension Risk to DC Plan Members: The Chilean Case of a Pension Risk Simulator”, OECD Working Papers on Finance, Insurance and Private Pensions, No. 28, OECD Publishing, Paris, https://dx.doi.org/10.1787/5k9181hxzmlr-en.

[25] ASIC and Latitude Insights (2015), Report 455 - Consumer testing of the Choice product dashboard.

[24] ASIC and Latitude Insights (2013), Report 378 - Consumer testing of the MySuper product dashboard.

[13] Behavioural Insights Team (2018), Encouraging Retirement Planning through Behavioural Insights, https://www.osc.gov.on.ca/documents/en/Investors/inv_research_20180727_encouraging-retirement-planning.pdf (accessed on 6 November 2020).

[31] Bernartzi, S. and R. Thaler (2007), Heuristics and Biases in Retirement Savings Behavior.

[22] Blakstad, M., E. Brüggen and T. Post (2018), Life Events and Participant Engagement in Pension Plans, Maastricht University - School of Business and Economics - Department of Finance; Netspar.

[17] Calderón-Colín, R., E. Domínguez and M. Schwartz (2010), “Consumer confusion: The choice of pension fund manager in Mexico”, Journal of Pension Economics and Finance, Vol. 9/1, pp. 43-74, http://dx.doi.org/10.1017/S1474747209004004.

[14] Chłoń-Domińczak, A., M. Kawiński and D. Stańko (2013), System oceny i prezentacji wyników inwestycyjnych kapitałowych systemów emerytalnych (Evaluation and presentation of investment results in funded pension systems).

[21] Fuentes, O. et al. (2017), Personalized Information as a Tool to Improve Pension Savings: Results from a Randomized Control Trial in Chile, http://www.economia.puc.cl.

[23] Gentile, M. et al. (2015), Financial disclosure, risk perception and investment choices. Evidence from a consumer testing exercise, CONSOB, http://www.consob.it/documents/11973/204072/qdf82.pdf/58dc22f8-504b-4bad-9679-610306359dfc.

[18] Gokçen, U. and A. Yalçin (2015), “The case against active pension funds - Evidence from the Turkish private pension system”, Elsevier, Vol. 63/Emerging Markets Review, pp. 46-67.

[26] Hiscox, M. et al. (2017), Supporting Retirees in Retirement Income Planning, Commonwealth of Australia.

[7] IOPS (2018), Good Practices on the Role of Pension Supervisory Authorities in Consumer Protection Related to Private Pension Systems.

[12] Kahneman, D. and A. Tversky (1979), “Prospect Theory: An Analysis of Decision under Risk”, Econometrica, Vol. 47/2, http://dx.doi.org/10.2307/1914185.

[30] Koposko, J. et al. (2015), “Perceptions of Retirement Savings Relative to Peers”, Work, Aging and Retirement, pp. -, http://dx.doi.org/10.1093/workar/wav019.

[19] London Economics (2015), Consumer testing study of the possible new format and content for retail disclosures of packaged retail and insurance-based investment products, European Commission, http://dx.doi.org/10.2780/71652.

[15] McGowan, F. and P. Lunn (2020), “Supporting decision-making in retirement planning: Do diagrams on Pension Benefit Statements help”, Journal of Pension Economics and Finance, Vol. 19/3, pp. 323-343, http://dx.doi.org/10.1017/S1474747219000015.

[6] OECD (2020), Recommendation of the Council on Financial Literacy.

[4] OECD (2018), “Improving retirement incomes considering behavioural biases and limited financial knowledge”, in OECD Pensions Outlook 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/pens_outlook-2018-8-en.

[29] OECD (2016), “The Mexican pension system today”, in OECD Reviews of Pension Systems: Mexico, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264245938-5-en.

[3] OECD (2016), “The role of financial education in supporting decision-making for retirement”, in OECD Pensions Outlook 2016, OECD Publishing, Paris, https://dx.doi.org/10.1787/pens_outlook-2016-8-en.

[2] OECD (2014), “Pension communication: Pension statements and national campaigns”, in OECD Pensions Outlook 2014, OECD Publishing, Paris, https://dx.doi.org/10.1787/pens_outlook-2014-8-en.

[5] OECD (2012), OECD/INFE High-level Principles on National Strategies for Financial Education.

[1] OECD (2012), The OECD Roadmap for the Good Design of Defined Contribution Pension Plans, https://www.oecd.org/finance/private-pensions/50582753.pdf.

[11] Reinhart, A. et al. (2007), “The persuasive effects of message framing in organ donation: The mediating role of psychological reactance”, Communication Monographs, Vol. 74/2, pp. 229-255, http://dx.doi.org/10.1080/03637750701397098.

[8] Stańko, D. (2019), Design and supervision of pension projections in 26 jurisdictions, IOPS, http://www.iopsweb.org.

[28] Stańko, D. (2003), Polish Pension Funds, Does The System Work? Cost, Efficiency and Performance Measurement Issues, https://www.researchgate.net/publication/23749059.

[20] Sykes, W. et al. (2008), Understanding responses to pension forecasts: qualitative research.

[9] Wiener, J. and T. Doescher (2008), “A Framework for Promoting Retirement Savings”, The Journal of Consumer Affairs, Vol. 42/2.

[10] Yoon, H. and C. La Ferle (2018), “Saving Behavior Messaging: Gain/Loss Framing, Self/Family Orientations, and Individual Differences in Collectivism”, Journal of Advertising, Vol. 47/2, pp. 146-160, http://dx.doi.org/10.1080/00913367.2017.1408507.

Notes

← 1. Present bias (also called hyperbolic discounting) refers to the fact that people respond to urges for immediate gratification resulting in overvaluing the present over the future. As such, choices may be regretted in the future. Present bias can lead to self-control problems such as procrastination. Overconfidence occurs when individuals perceive the likelihood of good events occurring or their own ability and success at different tasks too highly, including the accuracy of their judgements. Rules of thumb or heuristics are adopted by consumers to simplify complex decision problems. When choosing from a wide range of options, people may choose the most familiar, avoid the most ambiguous or uncertain, choose what draws attention most (e.g. the first option on a list), or avoid choice, including sticking to the status quo. When estimating unknown quantities, people may anchor estimates to some relevant or irrelevant figure and adjust from there (OECD, 2018[4]).

← 2. For instance, procrastination (i.e. failure to follow through on good intentions), projection bias (e.g. underestimating how much one will need in the future), simple heuristics (e.g. approximating the appropriate contribution level by the maximum allowed by law or the minimum to get full matching contributions) and loss aversion (i.e. weighing losses more heavily than gains) affect how much people contribute to their DC pension arrangements. Setting default contribution rates at high levels, automatically increasing contribution rates over time, providing matching contributions, simplifying the contribution process and providing information about expected benefits are ways to overcome these issues. Similarly, choice and information overload, time-inconsistent preferences, heuristic decision-making (e.g. people allocating 1/n of their portfolio to each option proposed), framing effects (e.g. people focusing on the asset allocation rather than the risk and return profile of the different investment options offered if the asset allocation is shown), overconfidence, over-extrapolation (e.g. people making projections from just a limited number of observations), and loss aversion affect the way people invest their retirement savings and their ability to select the appropriate investment strategy. Simplifying choice by reducing the number of available investment options, establishing appropriate default investment strategies, and providing financial advice and financial education are ways to facilitate the choice of an investment strategy.

← 3. Denmark and Switzerland use the 1-year volatility; Hong Kong (China) and Sweden use the 3-year volatility; the European Union and European Economic Area, the Netherlands, Sweden, Switzerland and Turkey use the 5- year volatility; and Canada, Sweden and Switzerland use the 10-year volatility.

https://www.osc.gov.on.ca/en/SecuritiesLaw_ni_20161208_81-101-81-102_csa-mutual-fund-risk.htm; https://www.forsikringogpension.dk/media/4515/ipd-good-pensions-with-controlled-risk.pdf; https://www.forsikringogpension.dk/media/4423/henstilling-om-risikomaerkning-af-markedsrenteprodukter-opdateret-udgave.pdf; https://www.esma.europa.eu/sites/default/files/library/2015/11/09_1026_final_kid_srri_methodology_for_publication.pdf; http://www.mpfa.org.hk/eng/mpf_education/regulations/mpf_disclosure/utilizing_disclosure_tools/files/Fund_Risk_Indicator_E.pdf; https://www.afm.nl/~/profmedia/files/wet-regelgeving/beleidsuitingen/leidraden/risicoprofielen.pdf?la=nl-nl page 4 ; https://www.pensionsmyndigheten.se/forsta-din-pension/valj-och-byt-fonder/vad-ar-risk, https://www.oak-bv.admin.ch/inhalte/Regulierung/Weisungen/fr/Weisungen_05_2013_chiffres-cle_determinants_et_autres_renseignements_francais.pdf; https://www.spk.gov.tr/Sayfa/Dosya/1205 Section 1.1, pages 7 and 8.

← 4. The modified duration is defined as the average cash-weighted term to maturity of a fixed income security. It is a function of the maturity and coupons of a security, as well as of interest rates.

← 5. Sustainability risk is represented by a figure between 0 and 100, which indicates how well the fund's holdings may handle the financial risks linked to ESG factors. The holding analysis is performed by Sustainalytics based on stock market data and questionnaires and compiled by Morningstar at the fund level. More recent holdings carry a heavier weight than older ones and the same rating scale is used for all types of companies and industries. A lower value of the indicator can be interpreted as a lower risk that the fund may suffer unexpected losses linked to ESG issues. This risk indicator is updated monthly and is based on an analysis of each fund's holdings over the past 12 months.

← 6. The low-carbon risk indicator is shown for funds which are considered to have a low risk in the transition to an economy with low carbon dioxide emissions and limited exposure to fossil fuel. This indicator is updated quarterly by Morningstar and mostly applies to equity funds.

← 7. https://www.spk.gov.tr/Sayfa/Dosya/1205 Section 1.1, pages 7 and 8.

← 8. https://doctrine.amf-france.org/technique/multimedia?docId=workspace://SpacesStore/7f36f442-dd16-4779-b2a9-647b57c9009a_fr_1.0_rendition page 3 «Aucune couleur n’est utilisée pour distinguer entre eux les éléments placés sur l’échelle».

← 9. https://www.nvb.nl/themas/lenen-sparen-beleggen/risicometer-beleggen/; http://risicoprofieltoets.nl/; https://www.afm.nl/nl-nl/professionals/onderwerpen/downloadbestanden-informatieverstrekking/risicowijzer-beleggen.

← 10. Each provider willing to use the risk meter should therefore compute the volatility of its investment portfolio using the weight of each asset class in the portfolio, and the prescribed standard deviation per asset class and correlations between asset classes. https://www.afm.nl/~/profmedia/files/wet-regelgeving/beleidsuitingen/leidraden/risicoprofielen.pdf?la=nl-nl.

← 11. https://www.rbcgam.com/en/ca/article/staying-the-course-part-2-conservative-portfolios/detail/; https://www.rbcgam.com/en/ca/article/staying-the-course-during-periods-of-volatility/detail.

← 12. https://www.aegon.nl/file/82770/download?token=8ZCzMqcF; https://www.islenskilif.is/avoxtunarleidir/;

https://www.landsbankinn.is/Uploads/documents/Einstaklingsthjonusta/lifeyrissjodir/einblodungur_islif.pdf; https://www.samsunglife.com/individual/products/pension/PDP-PRREA030110M?active=1.

← 13. This measure was elaborated by the Australian Prudential Risk Authority (APRA), the Financial Services Council (FSC) and the Association of Superannuation Funds of Australia (ASFA) and supported by the Australian Securities and Investments Commission (ASIC).

← 14. https://www.superannuation.asn.au/ArticleDocuments/359/FSC-ASFA_StandardRiskMeasures_July2011.pdf.aspx?Embed=Y page 3.

← 15. https://www.forsikringogpension.dk/media/4515/ipd-good-pensions-with-controlled-risk.pdf page 9.

← 16. Assuming no change of fund and no further contribution for the entire period.

← 17. https://www.knf.gov.pl/dla_konsumenta/Informacja_dotyczaca_otwartych_funduszy_emerytalnych; https://www.pensionsmyndigheten.se/mina-tjanster/fondtorg/sok “Avgift” (i.e. charge) tab; https://www.dol.gov/sites/dolgov/files/EBSA/about-ebsa/our-activities/resource-center/publications/what-you-should-know-about-your-retirement-plan.pdf page 12.

← 18. For the purpose of this exercise, all standard costs such as guarantee costs, management and administrative fees, entry, exit and transfer fees are included in the calculation. The complete methodology can be found at http://www.covip.it/wp-content/uploads/INDICATORE-DEI-COSTI.pdf; http://www.mpfa.org.hk/eng/information_centre/publications/booklets_publications/mpf_investment/files/FFS_Leaflet_Eng.pdf, page 2; https://www.gob.mx/consar/prensa/nueve-indicadores-comparativos-para-elegir-afore?idiom=es.

← 19. The minimum return for each risk class is published monthly by the SFC and based on weighted historical performance data of the four AFPs and of the reference portfolio over a calculation period of 36 months, 48 months and 60 months respectively for the conservative, moderate and higher risk funds. Pension companies must cover any shortfall in the returns below the minimum return guarantee from their capital. Annual performance information is published monthly in nominal and in real terms by the SFC. Decree 2949 of 2010, and https://www.superfinanciera.gov.co/jsp/38581.

← 20. https://www.apra.gov.au/sites/default/files/draft-reporting-standard-srs-700.0-product-dashboard-december-2015_0.pdf; https://www.legislation.gov.au/Details/F2015L01008.

← 21. https://www.spk.gov.tr/Sayfa/Dosya/1205.

← 22. Comparison groups are determined by the Fund Performance Assessment Committee (FPAC) comprising one member of the PMC, three members of the Insurance, Reinsurance and Pensions Companies of Turkey (IAT), and three members of the Turkish Capital Markets Association (TCMA). For 2020, there are 22 comparison groups, for a total of 350 funds grouped, and 54 funds which were not assigned a comparison group. https://www.egm.org.tr/funds/fund-performance-assessment-system/comparison-groups/; https://www.egm.org.tr/funds/fund-performance-assessment-system/fund-performance-assessment-method/; https://www.egm.org.tr/funds/fund-performance-assessment-system/funds-that-are-not-included-in-the-comparison-groups/.

← 23. There is evidence that this type of approach can lead to herding behaviour from retirement income providers, where different providers adopt similar strategies in order to avoid being singled out as an underperformer. This ultimately can lead to providers converging towards sub-optimal investment strategies, and to a lack of competition for individuals, if all providers offer similar products. See Acharya et al. (2015[27]) and Stańko (2003[28]) for further reference.

← 24. March 2019 FCA Handbook on Collective Investment Schemes, Appendix 8 https://www.handbook.fca.org.uk/handbook/COLL.pdf.

← 25. This is of an even greater importance if the benchmark is potentially to be used as the reference to compute performance fees, as asset managers’ remuneration is then directly linked to the performance of the benchmark.

← 26. Ratings based on the weighted average of risk-adjusted performance figures for three, five and ten years; using proprietary metrics. https://www.gob.mx/consar/prensa/presentacion-del-informe-de-clasificacion-analista-morningstar-de-las-siefores?idiom=es; https://www.blackrock.com/us/individual/literature/fact-sheet/lijkx-lifepath-index-2035-fund-factsheet-us0669237648-us-en-individual.pdf.

← 27. https://www.almenni.is/avoxtun/avoxtunarleidir/.

← 28. Based on the monthly returns and standard deviations of each fund since October 2002, when the different categories of funds were launched, and excluding the 5% most extreme results (2.5% lowest and 2.5% highest returns).

← 29. https://www.spensiones.cl/portal/institucional/594/articles-13899_recurso_1.pdf.

← 30. Several European countries have adopted the format of the Key Information Document (KID) for investment options, which applies to any type of fund, both for pension and traditional saving. The KID was set out in the UCITS IV Directive to set a standard and streamline the information received by customers ahead of any investment in a collective vehicle. The KID has since been further expanded with the Packaged Retail Investment and Insurance-based Products (PRIIPS) regulation, aimed at protecting non-professional customers, as defined in the MiFID 2 Directive, when investing in packaged investments and insurance-based products. Pension products are excluded from this regulation, given their specificities and different savings horizon compared to most retail investment vehicles. However, many European pension providers offer fund solutions which can be used through a pension plan as well as other non-retirement specific savings vehicles. Therefore several providers actually do not differentiate their communication document, in particular their KID, and include the requirements from the PRIIPS regulation in pension investment strategy documentation. This is the case for instance of Swedbank Estonia’s key information document, see https://www.swedbank.ee/static/life-insurance/KID/20190701/KID_EST_PM_ENG_AGR_R.pdf.

← 31. http://www.mpfa.org.hk/eng/information_centre/publications/booklets_publications/mpf_investment/files/Fund_Booklet_Eng.pdf pages 34 and 35; https://minisite.mpfa.org.hk/MPFIE/en/#3.

← 32. https://pensyanet.cma.gov.il/Parameters/Index; https://www.pensionsmyndigheten.se/service/fondtorg/fond/734491.

← 33. https://www.colpensiones.gov.co/pensiones/Publicaciones/afiliados_colpensiones/doble_asesoria_entre_regimenes.

← 34. https://www.gov.uk/government/publications/pension-benefits-with-a-guarantee-and-the-advice-requirement/pension-benefits-with-a-guarantee-and-the-advice-requirement.

← 35. https://www.covip.it/wp-content/uploads/Deliberazione201703221.pdf.

← 36. https://fundfinder.sorted.org.nz/find-the-right-type-of-fund-for-you/.

← 37. The Pension and Disability Insurance Act (ZPIZ-2) specifies that supplementary pension providers must offer their participants a maximum of three investment strategies with different risk profiles based on their allocation to high risk assets, including a mandatory guaranteed fund. https://www.gov.si/teme/prostovoljno-dodatno-pokojninsko-zavarovanje/.

← 38. https://tatrysympatia.nn.sk/fondy/.

← 39. Using a sample of 6 954 individuals from the Czech Republic, France, Germany, Italy, the Netherlands, the Slovak Republic, Slovenia, Spain, Sweden and the United Kingdom.

← 40. Taking into account differences in age, gender, education, and financial literacy levels. The study also shows that information on liquidity and credit risks was not necessarily understood better by those shown the multi-dimensional risk graphic compared to those shown simple horizontal or vertical scales. Approximately 5% more respondents shown the multi-dimensional risk graphic declared having information on market risk than those shown the simple overall scales.

← 41. 43.3% of individuals shown the energy efficiency scale of A to E found that it was very easy or easy to understand, versus 36.2% of those shown the horizontal scale of 1 to 7.

← 42. Taking into account differences in age, gender, education, and financial literacy levels. Overall 70% of respondents shown the simple horizontal graphic correctly answered question linking risk to return, versus 60.7% of those shown the energy efficiency scale and 59.6% of those shown a multi-dimensional graphic of risk broken down by risk types.

← 43. Using a sample of 254 Italian investors. Unbundled risk measures provide three measures of risk for each investment option: 1) market risk via value-at-risk and volatility; 2) liquidity risk through a turnover ratio; and 3) credit risk via a credit rating agency rating and associated expected default probability.

← 44. The MySuper product dashboard was introduced in December 2013 for all superannuation default investment strategies. Consumer testing was performed online in 2013 on 54 Australian permanent residents or citizens. The product dashboard is intended to provide members with key information about the default option offered to them, and details the return target, the returns for previous financial years, a comparison of historical returns and the return target, the level of investment risk and a statement of fees and other costs. APRA’s Reporting Standard SRS 700.0 details its content and presentation requirements. The Choice product dashboard is planned to be required from superannuation providers for all investment options starting from 1 July 2023. Consumer testing was carried out online in 2014 among 120 Australian permanent residents or citizens.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.