In 2000, the Swedish government launched a new defined contribution component of its social security system that was dubbed the Premium Pension Plan. Like any service, whether provided by the public or private sector, this new plan had to have what Thaler and Sunstein (2008) have called a choice architecture that organizes how participants interact (if at all) with the system. Will there be choices? If so, how many? Is there a default fund? If so, what should it be? These issues are just some of the design features that are part of the choice architecture. 

Research over the last two decades has conclusively shown that features of the choice architecture that would be considered irrelevant in traditional economic analyses can have strong effects on behavior, for better or for worse. For example, the use of automatic enrollment as the default in optional defined contribution (DC) pension plans (as opposed to an opt-in system where the default is not to join) has been found to dramatically improve participation rates. 

However, most plans adopt a savings rate that is too low, and some participants are inadvertently nudged to save less than they would have on their own (Madrian and Shea, 2001). To deal with this problem Thaler and Benartzi (2004) created the Save More Tomorrow plan in which people sign up to have their savings rates increased automatically over time. This has proved effective in boosting people's retirement savings rates (Benartzi and Thaler, 2013). 

One question that has not received enough attention is whether nudges are long-lasting (though see Allcott and Rogers, 2014). One possibility is that people initially exhibit default behavior for reasons such as status quo bias (Samuelson and Zeckhauser, 1988), laziness, procrastination, and so forth, but over time get around to solve their optimization problem, and make corresponding changes to their initial choices. In such a world, the specific design by the choice architect has only a transitory effect. But if the effects of nudges are persistent then the choice architecture design can be critical with effects lasting decades. 

To inform economists and policy makers about whether the effects of nudges are persistent in one specific context, we study the choice architecture of the Swedish Premium Pension Plan, which was launched in 2000. Evaluating detailed post-implementation data enables us to estimate the persistence of the effects of the nudges. The data we study were provided by the Swedish Pension Agency and consist of all initial choices and subsequent rebalancing activities by the entire population of 7,315,209 retirement savers in Sweden during the period 2000 to 2016. Based on our analysis of these data, we conclude that the effects of nudging in this case were surprisingly persistent and seem to last nearly two decades, if not forever. 

This post is an excerpt of the paper available here.

Henrik Cronqvist is Professor of Finance at the Department of Finance, University of Miami, Richard H. Thaler is Ralph and Dorothy Keller Distinguished Service Professor of Economics and Behavioral Science at Booth School of Business, University of Chicago and Frank Yu is Associate Professor of Finance at China Europe International Business School.