While having choices gives individuals a sense of control, more choice does not necessarily lead to better outcomes or happier individuals, and too much choice can lead to paralysis. Choices relating to benefits are particularly complex given that they have long-term implications and should be viewed holistically. As such, adding choice, even if accompanied by financial education, does not necessarily lead to better results for members.
For example, employees in the professional services sector, perhaps the most financially literate of all the sector populations, exercise choice most frequently but have some of the lowest contribution rates and replacement ratios of all the sectors covered in this study. Dynamic default solutions that can adjust to an individual’s changing demands as they move through their financial lives provide a powerful way to ‘nudge’ members towards the optimal answers.
When Sheena Iyengar of Columbia Business School wrote her seminal book on The Art of Choosing, she observed that “Whenever possible, people reach for choice. We want to believe that seeing our lives in these terms will make us better off.” Choice is perceived as a powerful tool for controlling your environment. Even in situations where there is no advantage to having more choice because it raises costs in time and effort, choice is still instinctively preferred1.
In the famous Whitehall II Studies on stress in the civil service2, an extensive project that spanned several decades of medical and psychological analysis, researchers concluded that the removal of choice can actually lead to depression, listlessness and withdrawal. Even the perception or illusion of choice in these experiments had positive effects.
So, given how we feel about choice and given that for benefits to engage or make us physically, mentally or financially healthier they should be more tailored to our needs, is it also a given that employees should be allowed to structure their total rewards systems?
Research is beginning to show us that people perhaps value having choices more than they value making choices. In The Paradox of Choice: Why Less is More, Barry Swartz concludes that choice eventually “no longer liberates, but debilitates”. Unnecessary choice destroys value, forcing members to evaluate complex choices that could have easily been made for them. That’s why choice should not be offered for its own sake.
If the relationship between choice and people’s sense of well-being or control is complex, how do you begin to understand which choices relating to the total rewards system destroy value, and which have the potential to enhance value to the employee and improve engagement?
Unnecessary choice destroys value, forcing members to evaluate complex choices that could have easily been made for them. That’s why choice should not be offered for its own sake.
Let’s start by first understanding the scope of the problem.
Historically, South African employees were remunerated on a basic salary plus benefits approach. Generally, the cost of benefits was not known to employees, who trusted their employers to look after them until retirement and beyond. As service periods were generally long and employers paternalistic, this belief was often correct.
However, changing demographic shifts and legislation made defined benefit (DB) retirement plans and promises of postretirement medical scheme subsidies increasingly onerous. At the same time, many large South African industries faced a heavy burden of benefits, given that they were based in locations where companies had to build towns to be able to attract skills. Providing housing, schools, hospitals and sports facilities just added to the cost of benefits debate.
With an increase in competition for skills, employers saw the need to explain to employees the total cost to the company (TCTC) of employing them. Soon the TCTC became the norm and employees went from focusing primarily on salary to focusing on salary plus a number of benefits of certain cost but indiscernible value.
With this transition, employees were expected to take more ownership of their employee benefits, instead of relying on the employer to look after them for life. To enable this, some employers introduced choice into the employee benefits system.
But how well equipped are employees to make good choices? Unfortunately, it seems that when it comes to money, savings and wealth creation, individuals appear to be ‘wired’ to make bad choices for themselves! Some choices confronting employees are set out in the graphic below.
Let’s focus on retirement choices and the retirement investment choice in particular.
Recent surveys show that between 48% and 76% of funds offer individual member investment choice3. The reasons that trustee boards offer member choice across a spectrum of retirement fund-related decisions vary, and include allowing members to try to better their investment outcomes and even simply not knowing what their members need4. It is the latter rationale that suggests we should be concerned about the rise of interest in offering more choice. But can members themselves do a better job of driving their investment decisions and choices than their boards of trustees?
Olivia Mitchell and Stephen Utkus summed it up best when they concluded: “Being good at retirement savings requires accurate estimates of uncertain future processes including lifetime earnings, asset returns, tax rates, family and health status and longevity. In order to solve this problem, the human brain as a calculating machine would need to have the capacity to solve many decades-long time value of money problems, with massive uncertainties [as] to… cash flows and their timing”. If we consider just the computing challenges, the problem at hand raises serious doubts as to whether any lay-person could develop meaningful insights on the best course of action.
This is hardly a problem endemic to South Africans alone. In fact, it appears to be fairly universal in any country where the option to save, or invest for retirement, or determine your benefit structure is still subject to some element of individual choice. The presumption has been that the best way to address the problem is to simply increase the level of financial literacy of members. What we hope to illustrate though, is that even greater financial literacy doesn’t properly address the problem.
The professional and business services sector is the group that has shown the greatest interest in providing the most expansive opportunities for choice. Indeed, when compared to other sectors, professional and business services reflect the highest number of fund members who make member investments choices, as seen in the table below.
Many employers in the sector believe that for benefit structures to be really attractive to their ‘financially literate’ employees, the range of options at their disposal must be highly diverse and flexible. The fascinating reality, though, is that in spite of their members’ higher financial literacy, having this additional dimension of choice has not translated into better member outcomes.
The great irony is that these sectors typically reflect some of the lowest levels of ‘financial wellness’ in their members. As the graph on the next page highlights, professional and business services, which include financial services, have the lowest contribution rates towards retirement savings. This results in very low projected retirement benefits relative to pre-retirement earnings.
It’s a phenomenon that warrants close scrutiny, because it helps to highlight the dangers of giving members the choices they want to make for themselves.
A member’s retirement savings decisions are rarely centred around the goal of securing an adequate inflation-proof income in retirement.
Now consider the range of dynamics at play that may influence how members make their decisions around these choices. Essentially we can break this down into three primary categories:
Let’s consider each in turn.
Attitudes towards retirement saving
Ultimately, the point of saving for retirement is to secure an adequate inflation-proof income in retirement and this is termed the member’s liability. Due to market fluctuations and longevity improvements, this liability is not fixed and can vary rapidly and significantly.
A member’s retirement savings decisions are rarely centred around this goal. Behavioural finance research by Daniel Kahneman5 suggests that much of members’ decisionmaking in relation to their investment choice is guided by anchoring and past performance. In other words, decisions are strongly influenced by starting values or what may be happening in the markets today. Members entering an investment plan in times of market turmoil
will typically be strongly risk-averse, as opposed to members entering at the end of a bull market who will typically choose a much higher allocation to equities6. Worse still, members may be influenced by how well a strategy performed over a previous time period.
Most problematic is the reality that high returns from the best performing asset manager may have no relevance to how members’ liabilities may be changing, or even indicative of that manager’s future performance. As such, a decision made to address short-term market conditions may be completely wrong to meet a long-term income replacement target.
If we consider the employees in the professional and businesses services sector again, we find exactly this behaviour. When they made switching decisions, their timing was generally a knee-jerk reaction to recent events that invariably took the wrong view of the markets in the flight to safety7. Members typically switched to less aggressive strategies at the bottom of the market cycle and to more aggressive strategies at the top of a market cycle.
In addition, self-control for saving is elusive because we human beings are ‘hyperbolic discounters’. That means we place a lower value on future benefits and a higher value on the present. As such, we over-consume today and under-save for tomorrow8. Translation: young or highly indebted employees will opt for a lower portion of pensionable salary, for the minimum contribution rates and for the least financially demanding benefit structures despite the long-term consequences.
Default solutions that automatically adjust a member’s benefit structure, contribution rate and investment solution over the course of their lives in the fund clearly can keep members on track to meet their income requirements in retirement. Getting an individual to buy into the default solution is probably easier than many trustee boards realise.
When presented with too much choice, including retirement fund investment options, decision-makers tend to freeze up and make no decision at all.
How framing the key questions can impact on decision-making
Possibly the most important dynamic that is within the control of trustees is how choices are framed to members. This refers to something as simple as how many decisions and options a member is confronted with.
In Sheena Iyenger’s classic jam experiment, she demonstrated that if she loaded a sales table with as many as 24 jars, consumers were far less likely to make a choice or buy than a table exhibiting only 6 choices. The results flew in the face of product providers, who believed that more options allowed people to better tailor their choices to their specific needs. What in fact tends to happen is that when presented with complexity, decision-makers tend to simply freeze up – become inert – and make no decision at all9. The great irony here of course is that many financial service providers are selected for the very reason that they can offer members a vast array of portfolio options. But how useful is that if members won’t even make a decision when they see so much complexity?
This was exactly the experience of funds offering professionals a great variety of options. On the one hand the options either elicited the wrong basis for decision-making or no decision at all. At best, only 27% of fund members actually took advantage of the costly flexibility feature in their funds10.
If we understand this as a potential problem, then two things become immediately apparent. We can improve outcomes to members simply by:
The answer is not to remove choice in its entirety. Instead, we need to recognise that choice for its own sake simply does not lead to optimal outcomes in something as complex as retirement fund decision-making.
Why two heads may not be better than one
Finally we tackle the issue of how human interactions can influence decision-making, both at the board of trustee level and the member level. As Yale psychologist Irving Janis points out, once we move into the realm of interpersonal dynamics, getting the approval of another individual may be a more powerful motivator than getting to the right answer11. This dynamic can act to diminish the value of getting advice if individuals and trustee boards are not mindful of their motivators.
Here, the best defence may well be to provide members with decision-making tools that remove this dynamic. These tools would need to show the long-term effects of members’ decisions. Importantly, the tools should allow the user privacy and the opportunity to ask as many seemingly ‘stupid’ questions about the process and the decisions as necessary, without the embarrassment of having to reveal the true extent of their grasp of the key issues to an adviser or peer. It’s not necessarily an option that will suit all members – in fact, the appeal and practicality may be limited to individuals who at least believe they have a modicum of financial literacy, but who may also be grappling with an incomplete understanding. But in terms of getting the engagement and buy-in from the one set of members best equipped to benefit from the ‘choice’ framework, the tools provide an essential leap forward.
We started this debate with the statement that ‘choice’ is vital for instilling a sense of control and involvement in most human endeavours. We’ve already argued that it’s a critical factor in securing employee engagement. But, on further analysis, it is apparent that choice in the context of the design of an optimal retirement saving and benefit plan may not lead to this desired outcome, even among more financially literate members.
The answer then is not to remove choice in its entirety. Instead, we need to recognise that choice for its own sake simply does not lead to optimal outcomes in something as complex as retirement fund decision-making. But inflexibility can also result in sub-optimal outcomes. What we do need to begin to appreciate is the value of an automated default solution that introduces flexibility into the process, minimises the intrusion of sub-optimal human behaviours and creates outcomes that are more closely aligned to member needs as they move through the course of their financial life.
With such a default option on the table, at worst members can opt out, although we know that most employees won’t. But at the very least, we have used our best actuarial minds to try and solve one of the more complex investment and benefit design problems that individuals face in their lifetimes.
Richard Thaler classifies this form of ‘nudging’ as libertarian paternalism12. While seemingly an oxymoron, libertarian paternalism captures the dilemma perfectly. People generally hate to be told what to do at this most personal of human decisions, but they desperately yearn for some guidance – preferably very subtle – about what would serve them best. Perhaps the dynamic is best summarised in the words of the 1970s rock group Devo:
“Freedom of choice is what you got but freedom from choice is what you want!13”
1 lyengar (2010)
2 Bell et al (2004)
3 PWC South Africa (2012), Sanlam Employee Benefits (2012) and Old Mutual Corporate (2012)
4 Mitchell & Utkus (2003)
5 Kahneman (2011)
6 Mitchell & Utkus (2003)
7 Member WatchTM 2012 data set
8 Laibson et al (1998)
8 Iyengar (2010)
10 Member WatchTM 2012 data set
11 Janis (1972)
12 Thaler & Sunstein (2009)
13 Devo (1980). Freedom of Choice.