From April this year, people who retire in the UK with defined contribution pensions will have complete control over how they access their retirement savings. Taxes on withdrawals will be lower, with the first 25% of the money you take out being tax-free. As long as you are over 55, you will be able to withdraw your entire defined contribution funds and spend it or invest it as you like. Pensions minister Steve Webb stated “If people do buy a Lamborghini but know that they’ll end up just living on the state pension, that becomes their choice”. Mr Webb told the BBC: “We will guarantee people guidance, information, education but ultimately if they want to spend their money sooner rather than later, we are treating people as adults.”
How Do People Make Savings Decisions?
The Institute for Fiscal Studies (IFS) not only questioned the estimates of increased tax revenue that the new scheme would generate, but also raised concerns about increased freedom leading to more sub-optimal saving behaviors. The heart of the matter lies in understanding how people make savings decisions (and under which circumstances these decisions turn out to be optimal). Standard economic theories explicitly assume that accumulation and dissaving maximize a person’s lifetime utility function. They also make two important implicit assumptions: that households have the cognitive ability to solve the necessary optimization problems and that they have sufficient willpower to execute the optimal plan. However, there is extensive research on people using heuristics that can lead to systematic biases. Benartzi and Thaler (2007) studied the effect of shortcuts in decision making on retirement plans’ enrollment decisions, contribution rates and asset allocation.
Data from the UK Department of Work and Pensions tells us that even in plans where contributions are tax deductible, accumulations are tax deferred and employers fully match employees’ contributions, only 51% of eligible people sign up. There are some possible strategies to significantly improve enrollment: one of them is automatic enrollment, which involves changing the default option from “out” to “in” the retirement plan. The authors observed an “opt-in” enrollment rate of 20% after 3 months of employment and of 65% after 36 months. On the other hand, the “opt-out” approach effect was 90% immediately and 96% after 36 months. Not only participants join sooner in the automatic enrollment case, but more participants also join eventually. In terms of contribution rates, it is generally believed that workers without other retirement income sources are unlikely to achieve complete income replacement in retirement with typically low default saving rates. Also, when surveyed many employees believe they should be saving more: 68% of participants feel their saving rate is “too low”, 31% feel it’s “about right”, and only 1% think it’s “too high”. People extensively believe they should save more; still, enrolment and contribution rates are lower than optimal. In terms of asset allocation, people select suboptimal portfolios of investment using heuristics to avoid having to deal with complex differentiation strategies. When Nobel Laureate Markowitz, one of the founders of modern portfolio theory, was asked about how he allocated his retirement investments in TIAA-CREF account he said he should have computed the historic covariance of the asset classes and drawn an efficient frontier; instead, he confessed that he split his contribution 50-50 between bonds and equities. An anecdote from a chain of supermarkets operating in Texas illustrates how strong an impact peer effects can also have on investment decisions: participants’ behaviour in each supermarket was homogeneous, but heterogeneous across supermarkets. It turned out most of the employees believed the store butcher to be the investment expert, therefore turned to him for advice!
Institutional Solutions to Optimise Saving Behaviour
It is clear that we do not make optimal decisions (for instance, people aged in their 60s are generally believed to underestimate their life expectancy). Sometimes, we are even unaware of the extent to which our decisions are suboptimal (investors creating their own portfolio end up on average with a more risky and less diversified portfolio which underperforms professionally managed “lifestyle” portfolios). Two broad classes of interventions have been identified to nudge people into the right direction: education and plan design. Empirical evidence does not suggest that education is an adequate solutions. For instance, financial education programs offered free of charge had no impact on employees’ financial literacy. It is still possible to choose the features of the retirement savings plan to promote the desired objectives. Saving for retirement is a difficult problem; it is very important that individual’s freedom of choice to purchase a Lamborghini remains untouched. Nevertheless, investors are passive, make slow and infrequent changes, adopt naïve diversification strategies and in short need help. This help has to materialize in effective strategies to help people to make choices on their saving which are as close as optimal choices as possible.