When more choice is too much
The transition from defined benefit (DB) plans to defined contribution (DC) schemes in Europe has brought with it a wave of new responsibilities and investor choices for plan participants. While this new level of choice is seen by many as a major positive, European plan sponsors need to manage the transition carefully, so as not to scare off participants. The US experience offers European sponsors some vital lessons in managing DC transition, and striking the right balance between choice and guidance to help participants achieve their long-term performance goals.
At first glance, the US market should provide great encouragement to European DC plan sponsors. There has been tremendous growth in the number of DC plans, as well as the number of participants covered by these plans (see chart). The average number of investment options has also risen sharply over the past 10 years. Most DC plans started with only two investment options: usually, a balanced fund and a fixed income fund. The average DC plan in the US now offers 13 investment options. Investor choice is very much a reality.
So, what’s the problem? Simply put, investors make mistakes when they invest. The analytical capabilities and the emotional capacities of investors are particularly challenged during periods of bubbles and busts in the investment markets. In the US, based on our firm’s research findings, a significant number of plan sponsors added technology-oriented equity fund offerings to their DC plan fund line-ups during the late 1990s bubble. The result was that investor decisions were influenced.
This behaviour is not surprising. Psychologists have documented a social bias for attention; that is, people tend to pay attention to what others are paying attention to. Plan participants may simply buy the newest ‘hot fund’, ‘hot investment story’, or best recent investment performance which is gaining all the attention, rather than focus on long-term asset allocation.
Interpreting how our instincts impact our ability to make rational decisions falls into the remit of behavioural finance - a school of thought that disagrees with the conventional assumption by economists that investors make rational decisions. Books such as ‘Irrational exuberance’ by Robert Shiller and ‘Beyond greed and fear’ by Hersh Shefrin have emphasised the psychological and emotional factors involved in investing. The important premise of behavioural finance is that humans make mistakes, but they do so in a consistent manner. This not only presents fund managers with opportunities to exploit the valuation anomalies that are created, but also for DC plan designers to ensure that participants do not get seduced by fads and veer away from the rational path to long-term performance goals.
Despite the fact that the financial industry spends billions of dollars on investment education, procrastination and inertia still seems to govern investment decisions. A study by Americas and Zeds (‘How do household portfolio shares vary with age?’; December 2001) using the TIAACREF Plan, one of the longest running DC plans in the US, showed that 73% of plan participants made no changes to their plan balance over a 10 year period and that an additional 14% made only one change in 10 years. Although taking a long-term approach to one’s investments and being patient is highly commendable, retirement plan participants may not be rebalancing their portfolios to take into account market value changes that impact their original desired asset allocation nor do they rebalance to a more conservative allocation as they approach retirement age.
The reality is that people have human reactions to markets and often have little time to dedicate to managing and monitoring their investments. Instead, they use rules of thumb (heuristics) to deal with all the information they receive. This point was illustrated in an interesting study by Benartzi and Thaler (‘Naïve diversification strategies in defined contribution savings plans’; January 1999) that showed how a DC plan’s overall asset allocation is influenced by the number and types of funds offered in the line-up. Interestingly, a naive diversification heuristic existed when participants had only two funds to choose from (50:50 split), but as the number of options offered was significantly greater, decisions by participants were ultimately influenced by the number of funds offered in each asset class.
DC plan sponsors have been aware of this behaviour and recognised that most participants need some guidance in preparing for retirement. So-called default choices are common in many retirement plan designs in the US. However, Benartzi and Thaler (‘How much is investor autonomy worth?’; March 2001) demonstrated that most plan participants would select the average allocation as preferable to their own asset allocation when presented with that information. This shows that most plan participants had an aversion to being at the extreme allocations. Once again, it is a case of the herd instinct or social bias.
Remember the age-old adage: “Don’t put all your eggs in one basket!” However, the irony of offering DC plan participants numerous choices is that it may lead to confusion and irrational decisions. A recent study by Sheena Iyengar et al. (‘How much choice is too much?: Determinants of individual contribution in 401(k) Retirement Plans’; March 2003) addressed the issue of choice in 401(k) retirement plans. The study noted that, for every 10 investment options added to a plan, the participation rate drops by 1.5-2.0% in the plan. In other words, giving participants too much choice may negatively impact participation.
Investors should be encouraged to recognise the central lessons of behavioural finance; namely, we are all victims of over-confidence, cognitive dissonance, conservatism bias, anchoring tendencies, ambiguity aversion, and other heuristics. But how is the average plan participant protected from making these behavioural mistakes in their retirement plans? DC design should take into account how people behave. For example, to protect the overconfident participant, DC plans should consider limiting the number of fund choices. For the typical participant, avoid providing options in the plan that offer too much risk. To cover the loss aversion tendencies, they could allow participants to sign up for automatic increases from their pay to the plan.
Despite participants knowing that they are not saving enough, the average participant never gets around to increasing their contribution percentage. The importance of understanding the individual’s utility values was demonstrated by another study from Thaler and Benartzi (‘Save more tomorrow: using behavioural economics to increase employee savings’; August 2001). These two leading professors in the field of behavioural finance developed a prescriptive savings plan called Save More Tomorrow (SMT), in which employees commit in advance to allocate a portion of their future salary increases toward retirement savings. The results of the SMT savings plan were that the vast majority of employees stayed within the plan after three pay rises and the average savings rate increased from 3.5 to 11.6% over the course of 28 months. This success clearly illustrates that the SMT works because it uses the human behaviour of procrastination and inertia to positive effect.
Ultimately, plan sponsors influence participant behaviour through their plan design, fund line-up and how the information is communicated. We mentioned earlier that there is mounting evidence that the majority of plan participants opt for the default choice and match threshold for the plan. In terms of matching employee contributions, DC designs may want to consider that if their plan matches 3% dollar for dollar, a change to a matching of the first 6% at 50 cents on the dollars might have a better outcome as it could increase the contribution percentages to the plan.
Other studies showed that participants are likely to go with what is familiar to them when making their investment election in the plan. So if they do not know much about the markets or investing and company stock is offered then they will select their company stock because they feel that they are familiar with the company. Additional studies show that, when participants are electing their total asset allocation, they do not treat their employer stock like the equity/growth allocation it is and these participants end up with approximately a 15% higher overall allocation to equities (Benartzi and Thaler: ‘Naïve diversification strategies in defined contribution savings plans’; January 1999).
So, what are we seeing as best practices and trends in fund line up construction for defined contribution plans that can help to address some of these behavioural errors?
Many plan sponsors are adopting a tiered approach. The first tier is comprised of life style funds to make investing simple for those participants who don’t have the time, interest or skills to be able to make these decisions. These funds are meant for the delegators in the plan and help to protect plan participants from many of the behavioural errors that have been discussed. They are well diversified funds which rebalance over time toward a target date, usually a participant’s projected retirement date.
The second tier is the core funds for the plan and helps meet the needs of the self-directed investors in the plan. This portion of the fund line-up allows those self-directed participants, who are comfortable making their own asset allocation decisions, the opportunity to diversify across asset classes and styles. This tier of the fund line-up usually contains funds in the following categories:
o Capital Preservation Funds: GICs or money market type funds
o Income Funds: Bond funds
o Growth Funds: US and non-US Equity funds
o Company stock
The third tier, such as a mutual fund window or brokerage accounts, is used by those plan sponsors with a very diverse or sophisticated group of participants. This is for those participants with plenty of time and interest to study up on investments. It also quiets that vocal minority that is always asking for the next ‘hot fund’. This tier provides those self-directed investors access to asset classes or funds which might not be appropriate for the broader workforce.
What should European plan sponsors consider based on the US’s experience with DC plans?
o Limit the number of options. Don’t overwhelm less sophisticated participants as this creates a psychological barrier to their participation in the plan.
o Include life style funds for the delegator, thereby making it easier for them to participate. By providing well-diversified funds which are automatically rebalanced for them helps to take advantage of their natural inertia behaviour.
o Consider the demographics of the workforce as well as the company culture that exists. Remember the plan is for all employees, not just executives or for those who would like to be able to have every investment option.
o Include a self-directed account if there is a very diverse workforce. This will provide the more sophisticated investor with the options he/she wants, without making the plan confusing to the average participant.
But the US has not yet found all the answers. Plan sponsors worldwide still need to come up with models that better meet the needs of future pensioners and enable individual scheme members to better diversify, manage, and finance their individual pension risks. This is not only vital for this generation, but for those generations still to come.
Christian Elsmark is a director and head of investment services (EMEA) of T Rowe Price Global Investment Services of London and Pegi Almond is a relationship manager (Retirement Planning Services) at T Rowe Price in the US