The investor’s chief problem — and even his worst enemy — is likely to be himself.
— Benjamin Graham, author of Security Analysis and The Intelligent Investor

Welcome to Lake Wobegon, where all the women are strong, all the men are good-looking, and all the children are above average.
— Garrison Keillor, author of Lake Wobegon Days

Do you consider yourself a better-than-average driver? If your answer is “yes”, you would be in good company. In a 1981 study of US and Swedish drivers, between 77% and 88% believed themselves to be safer than the average driver. Similar findings of illusory superiority, a cognitive bias affecting an individual’s over-estimation of their own abilities relative to others, have been found in studies of peoples’ self-assessments of their academic ability, job performance and popularity.

Behavioural finance theory and research has shown that this illusory superiority, as well as many other important cognitive biases, carries over into the investment domain as well. I include myself with many individual investors, as well as investment professionals such as money managers, stock-traders and financial planners, who suffer from these behavioural and psychological influences, which can lead us into taking on too much (or too little) investment risk, and which ultimately endangers our financial plans and future retirement adequacy.

In the on-going debate on whether the Central Provident Fund (CPF) system should permit more flexibility in investment options for members, such investor behavioural quirks will therefore have to be taken into account.

In a 2010 article in the United States Social Security Administration’s Social Security Bulletin, the author, Melissa Knoll, summarises the behavioural and psychological factors that affect individuals’ retirement savings behaviour into four categories: (1) heuristics or mental shortcuts, and biases; (2) inter-temporal choices; (3) informational issues and (4) decision context, or framing.

In this essay, I will highlight investment decision-making idiosyncrasies in the each of these categories to demonstrate the issues that do-it-yourself retirement planners face in the presence of uncertainty and volatility.

Cognitive biases

Several studies have shown the notion of illusory superiority mentioned earlier to lead overconfident investors to trade more than rational investors, and by doing so reduce their net returns. Overconfident investors overestimate their knowledge about the value of the securities they purchase and are more likely to believe their personal assessments of a security’s value to be more accurate than other investors. This belief leads to differences of opinion, thus increasing the propensity to trade. They are also more likely to hold riskier portfolios than rational investors with the same risk appetite.

Another well-studied cognitive bias with a significant bearing on investing is that of loss aversion (which is in turn related to the endowment effect), identified by behavioural economists. Loss aversion refers to the tendency of people to care more about losses than for gains of the same magnitude. For example, I will feel the loss of misplacing $50 more strongly than the delight of picking up a $50 note in the street. In an investment context, someone evaluating equal chance outcomes of similar gains or losses would prefer to avoid losses and take the lower risk option.

These cognitive biases (overconfidence and loss aversion) can manifest in the same group of investors at the same time, leading to widely differing choices and outcomes in the same market conditions. Overconfident investors would take on the most risky positions, trade frequently and see their returns eroded through transaction costs even if they may have achieved their risk-adjusted return objectives. Loss-averse investors would take on too little risk for their own good and fail to achieve their return objectives.

Inter-temporal choices

Emotions play an important role in savings and investment behaviour. I might be deterred from ordering an expensive, fancy meal as much by the immediate pain of paying for a meal I might not enjoy (i.e., anger) as from a calculation of the future consumption foregone due to the high cost of the meal (i.e., regret). The emotions experienced at the time of consuming (or deferring consumption, i.e., saving) play a significant role in understanding the inter-temporal trade-offs people have to make.

Research conducted by the University of Connecticut for Prudential Financial (of America) evaluated the link between emotions and investment decision-making of retiree investors and pre-retirement investors (five years after and before retirement). Fear and regret were found to be the dominant emotions influencing investment decisions, with those influenced by fear preferring certainty and less likely to take managed risks. Those who experienced the strongest feelings of regret were typically dissatisfied with past events and decisions, and thus were reluctant to make future decisions in order to avoid feeling regret in the future.

Both these emotions of fear and regret are reflected in the study’s findings about survey participants’ reactions and choices when confronted with negative returns in the years immediately preceding their planned retirement date. Survey participants were shown a hypothetical scenario of a pre-retirement investor who hopes to have accumulated $350,000 at retirement date (Figure 1). Unfortunately, investment losses of 15% and 19% occur in the second and third year before the planned retirement date.

Figure 1. Hypothetical example of negative returns in the two years preceding retirement date shown to survey participants

The research found that the potential for significant investment losses during this critical period before the planned retirement date generated emotional responses of fear and regret, with almost two-thirds of survey participants (66%) indicating they would take all or some money out of the market if confronted with losses (Figure 2). Only 16% of participants would choose a course of action (by increasing market exposure and investing more in a market that has already declined by 31%) that would increase risk but also potentially increase the chance of achieving the stated retirement goal.

Figure 2. Survey participants’ investment decisions in a scenario of negative returns immediately preceding target retirement

These results may resonate with that of Singaporean investor behaviour in the aftermath of the Asian Financial Crisis in 1998. Faced with significant investment losses from a 38% decline in the Singapore stock market in 1998, and the effective freezing of $4.3 billion of Malaysian shares (owned by 180,000 Singaporean shareholders) traded on the CLOB marketplace, retail investor activity in the Singapore Exchange (SGX) has since been relatively muted with an active retail investor participation rate of only 8% in 2012, as compared with Hong Kong’s 25%.

Informational issues

A significant body of research (conducted outside of Singapore) on decision-making under conditions of imperfect information illustrates how uncertainty and a lack of knowledge can paralyse decision-making. For example, economists Annamaria Lusardi and Olivia Mitchell found that individuals who lack confidence and knowledge in matters such as financial planning tended to avoid making decisions. In contrast, individuals who were more knowledgeable about financial information were also more likely to have engaged in financial planning and making choices. Other research shows that people prefer to make choices when the risks are known, over those where the risks are unknown or unspecified.

Although Singaporeans generally do quite well in surveys of financial literacy, there remains a gap between what they would do ideally and what they actually do. For example, the National Financial Literacy Survey 2005 found that whilst 54% of respondents felt they should develop a financial plan once they started working, only 32% actually did so. Two-thirds of the respondents in the same survey indicated that they had no investments outside of the CPF scheme, with 40% citing reasons relating to lack of knowledge, inclination and risk appetite for not making financial investments (the rest said they did not have money to invest). Only 42% of pre-retirement respondents indicated that they knew how much CPF balances they would have at age 55.

How then do we bring together all these insights from the field of behavioural finance and psychology together in assessing the question about whether or not the CPF system should permit members to have more flexibility in investment options?

In providing CPF members additional choice or flexibility, will we end up creating more complexity in a domain where outcomes remain uncertain, and subject our retirement financial adequacy to the behavioural biases and blind spots that are our enemy within?

Decision contexts (or frames)

The way in which choices are presented matters, and how issues are framed can have meaningful implications for decision-making. In my view, the CPF system is currently well configured to provide for the needs of the vast majority of its members through its default investment option[1], whilst providing sufficient flexibility for those members who may wish to opt out of the default. CPF members already do have some choice in making investments with their CPF funds, as the CPF scheme permits members under certain conditions to withdraw amounts from their Ordinary and Special Accounts (OA and SA respectively) to make approved investments under the CPF Investment Scheme (CPFIS). Members who opt to make withdrawals under the CPFIS are, however, subject to a range of investment risks that are spelled out in a paper I co-authored with Peter Ryan-Kane, Mark Whatley and Will Rainey, titled “The Investment Risks in Singapore’s Retirement Financing System”.

The default investment option provided by the CPF scheme is a government-guaranteed security paying rates of interest that are subject to a minimum rate of 2.5%, but may rise above this level dependent on market rates of interest on bank savings and deposit accounts, and government bond yields. A major conclusion from our working paper above is that the CPF default investment option provides a long-term return that matches that of a 60:40 (global equity to bond portfolio) with substantially lower risk. By combining CPF members’ account balances with the government’s unencumbered assets, the combined fund can be invested with an extremely long-term perspective to pool society’s risks in the most efficient manner.

The way in which the CPF investment function has been set up makes it, in my mind, the “ultimate life-cycle fund” [2]. Unlike the typical life-cycle fund which reduces investment risk as the investor nears retirement (thereby lowering returns commensurately), the CPF default option provides all its members of all age groups a standardised rate of return in excess of the risk exposure. For those members nearing or already in retirement, the default returns provided by the CPF would thus be higher than what they might ordinarily be able to achieve on their own, without taking on an inappropriate amount of risk. The CPF system also better accounts for inter-generational implications of sequence risk, which we described in an earlier article.

CPF members’ savings held within the default scheme are protected from most investment risks, including protection from the behavioural and psychological risks from the enemy within. In my mind, the investment aspect of the CPF system requires little by way of adjustment, being well structured and designed to support the retirement financing needs of its members.

The next essay in this series will consider some of those areas in Singapore’s retirement financing system that I feel require some policymaker attention.

Christopher Gee is a Research Fellow at IPS. This is the third in a series of essays examining risks and returns in Singapore’s retirement financing system. The first and second essays can be found here and here.

——————
[1] CPF members’ balances that are not withdrawn (for housing, investments via the CPFIS, for medical expenses, etc.) are aggregated and invested in Special Singapore Government Securities, government bonds that pay rates of interest that are pegged to CPF interest rates.

[2] Life-cycle funds are a class of age-appropriate, diversified mutual funds, where the asset allocation between cash, bonds, equities and alternative investments is adjusted over time to become more conservative as retirement nears and continues whilst the investor is in retirement.