How default retirement funds enact long-lasting financial changes

Intervention Financial Products

Abstract

While nudges have proven to be effective, are they long-lasting? To investigate this, researchers studied the 2000 launch of a new component of Sweden’s social security system, the Premium Pension Plan. Building upon their understanding of choice architecture, the Swedish government included two distinct nudges in their design. The first was a default; individuals made no choice about which portfolio they wanted. The second, a ‘do-it-yourself nudge,’ encouraged investors to determine their own portfolios. This option was the focus of a well-funded national advertising campaign. Initially, two thirds of savers chose their own portfolios, however as ad funding declined after the first year, fewer people chose the ‘DIY’ option. In recent years, less than 1% of new savers created their own portfolios. Most people in the second condition stayed in this condition permanently. In summary, the effects of the nudging were steadfast and seemed to last years.

Rating: 5/5 (Large sample size, direct positive impact on participants)

The long-term effects of default versus DIY pension portfolios in Sweden
Condition Results
The Delegators: the default fund 
  • 33% enrolled in 2000
  • 99% of new savers enrolled in 2016
  • 27.4% switched to DIY
DIYers: investors who determined their own portfolios 

(nationally advertised through a campaign launch)

  • 66% enrolled in 2000
  • 1% of new savers enrolled in 2016
  • 2.9% switched to Delegators

Key Concepts

Choice architecture: The arrangement of how individuals interact with a system and make subsequent decisions.

Nudges: A subtle manipulation that changes our behavior in predictable and systematic ways. Nudges capitalize on our irrational behaviors and need to be easily avoidable without cost or fear of consequences.

Defaults: The option you’re automatically provided with when you make no choice. 

Delegator: The term used by the authors to describe people who were put into the default option.

DIYer: The term used by the authors to describe people who created their own investment portfolios.

The Problem

Can pension nudges have lasting effects?

Previous research has shown that nudges have strong influences over behavior. As an example, automatically enrolling participants in a savings plan, creating an “opt-out,”  compared to an “opt-in” situation) shows impressive results in increasing enrollment. While nudges have been implemented in pension plans, the question of whether the effects of nudges last remains. It’s possible people initially display default behavior for reasons other than the nudge, perhaps because they procrastinate or are lazy, but eventually get around to fixing their behaviors. In these situations, the choice architecture would only have a temporary effect.

Design

The Swedish Pension Plan: 2000-2016

To explore the above hypotheses, Cronqvist et al. conducted a review by looking at the Swedish Premium Pension Plan, chosen in part because of its detailed post-implementation data. These data, provided by the Swedish Pension Agency, let the researchers measure the continual effects of the nudges, and included all initial choices and rebalancing activities between 2000 to 2016. The sample size was all retirement savers in Sweden during this time, totaling 7,315,209 individuals. 

Delegators and DIYs 

The Swedish government incorporated two nudges into the system. The first, called the default nudge, is the fund that people were put into when they made no choice. Authors called these people “Delegators,” since they “delegated the management of their portfolios to the default fund.”1

The second nudge, called the “do-it-yourself nudge,” occurs when investors declined the default fund and chose to create their own portfolios. People in this group are referred to as “Do-It Yourselfers,” or DIYers. Notably, there was a high-budget advertising campaign that ran during the first year of the Pension Plan that encouraged Swedes to choose the DIYer option. The Swedish government wanted to make sure that everyone was enrolled in a pension plan and make this enrollment as easy as possible. The program launched in 2000. 

The EAST framework

These nudges align closely with the EAST Framework, which was developed by the Behavioural Insights Team in 2012. This framework recommends that interventions be Easy, Attractive, Social, and Timely. The pension plan hits each of these marks: it’s Easy to enroll, the advertising campaign was Attractive, the campaign and nudges took advantage of Social norms, and the campaign was Timely, as people are automatically enrolled in the midst of a life change (i.e. starting a career). 

Results and Application

The drop in DIYers

Advertising spending for the second nudge considerably dropped after the plan’s first year. This explains why, in 2003 (two years after the launch), only 9.4% of new savers became DIYers. This number has since dropped to below 1% in 2016 (the last year for which the researchers had data). Mathematical calculations show that “the odds of becoming a DIYer are about six times higher for people in the 2000 cohort compared to those in the 2001-2002 cohorts,” thus, the ads most affected people who were in a mindset to make a decision when the ads ran.1

Set it and forget it

Savers exhibit a “set it and forget it” mentality: most people who were initially DIYers stayed with that choice. Only about 3% of initial DIYers switched to becoming Delegators between 2001 and 2016. Conversely, 27.4% of the initial Delegators decided to change into becoming DIYers. Most of these changes happened during the first decade after their initial choices. While this shows that the nudge has had enduring effects, the low number of DIYers who switch into Delegators show that the advertising was even more persistent given DIYers mostly stayed with their decision. 

Long lasting nudges

These results support the claim that nudges’ effects are perpetual, however the authors note with the huge variety of default rules and nudges, results should not be projected onto all scenarios. On the other hand, we’ve seen how both groups “set it and forget it,” so it seems a fair assumption that nudges will last the longest in situations where people are on autopilot.

Industry Application
Investing At organizations where employees have the option to join an Employee Stock Purchase Plan (ESPP), a nudge could be that they’re automatically enrolled and can choose to increase or decrease their investment percentage (but not to opt-out).
Education During kids’ snack time at school, their default option could be to get a piece of fruit, though they could switch to another healthy snack if desired.
Insurance When buying a new car, the default option could be automatic enrollment in basic car insurance. Drivers could switch into whichever insurance policy they want, but would have to be enrolled in some type of policy.

Ethics

  • Given that this paper was a review of Sweden’s pension plan rather than the implementation of an intervention, different ethical expectations should be set.
  • Both conditions encouraged saving, which benefits all those involved.
  • Swedes’ opinions and perspectives are extrapolated from the data, but surveys or interviews would be a better way to grasp this.

Yes

Room for improvement

Insufficient information/Not applicable

Welfare
Does the intervention demonstrably improve the lives of those affected by it? Both conditions ensured that Swedes would be saving for retirement.
Does the intervention respect the privacy (including the privacy of identity) of those it affects? The privacy of savers involved in the pension plan has been maintained and respected.
Does the intervention have a plan to monitor the safety, effectiveness, and validity of the intervention? Data has been kept and maintained since the start of the pension plan in 2000. Since this paper is a review of the Swedish government’s intervention, it doesn’t have a plan to monitor the safety, effectiveness, or validity of the intervention.
Autonomy
Does the intervention abide by a reasonable degree of consent? Individuals have the choice to be in either condition, but they don’t have the choice to opt out of both.
Does the intervention respect the ability of those it affects to make their own decisions? People are free to choose either condition, and within condition #2, to choose their portfolio at will, but do not have the option to opt out of both.
Does the intervention increase the number of choices available to those it affects? Choices didn’t increase, but it can be argued that making informed choices became easier due to the campaign..
Equity
Does the intervention acknowledge the perspectives, interests, and preferences of everyone it affects, including traditionally marginalized groups? The review doesn’t mention the perspectives or interests of savers, however it deduces them from the data. It doesn’t mention anything about marginalized groups.
Are the participants diverse? Given that the sample size is over 7 million, it’s as diverse as it can be in a Nordic country.
Does the intervention help ensure a just, equitable distribution of welfare? It ensures that people are able to save for retirement, something that will benefit everyone.

Related TDL Articles

Why Don’t We Save Enough for Retirement? Interested in learning more about why it’s so hard to save for retirement? This article will answer your questions and provide links for further reading.

What Can George Costanza Teach Us About Making Better Investment Choices? Pride and regret have more of a role in investment decisions than you’d think, even more than profit and loss. This article also provides solid life advice from none other than George Costanza!

Nudges Work (and practitioners know exactly how well)

Governments and organizations are creating nudge units with increasing frequency. Two academics, Drs. Linos and DellaVigna, sat down with The Decision Lab to discuss the power of nudges and consider how they can be used in applied settings.

Sources

Cronqvist, H., Thaler, R. H., & Yu, F. (2018). When Nudges Are Forever: Inertia in the Swedish Premium Pension Plan. AEA Papers and Proceedings, 108, 153–158. https://doi.org/10.1257/pandp.20181096