Explaining Non-participation in an Employee Savings Plan: Evidence from Administrative Data

This is an extension of the project “How do households learn to use a new financial product”. A first working paper (joint with Henriette Prast - Tilburg University) studies how a labeled bonus payment induces employees to take up a new employee savings plan. This extension studies how employees respond to an unexpected abolishment of an employee savings program.

In the fall of 2011, the Dutch government suddenly announced to discontinue a tax-favored employee savings plan. Employees could save out of gross (before taxes) wages and withdraw after four years. Given that the marginal tax rate between 31% and 52%, the effective return on savings was quite generous. The savings plan is illiquid: savers cannot withdraw in the first four years, and also not borrow against. Employees are eligible to participate if their firm offers the plan, and most firms do.

When the government announced the discontinuation of the savings plan, there was a window of three months in which employees could open a savings account, withdraw the annual amount in January 2012, and enjoy the subsidy without incurring the consequences of the illiquid nature of the plan. The Tax Authority estimated that around 267,000 new savers started in the period October-December 2011 – this is 13% of the total number of savers in that year.

We study the question of initial non-participation. Many of these last-minute savers could have participated into this savings plan in years prior, and the question is why they did not. Among the classical reasons for non-participation are liquidity constraints and time preferences. Impatient households do save little, and liquidity constrained households dislike illiquid savings accounts. Given that we have information on other individual savings, we can test for both reasons. Liquidity constraints and impatience are less likely reasons if we observe employees holding substantial amounts of savings in other accounts. Another reason for initial non-participation is lack of information. We control for the fraction of savers in the firm where the employee works, as well as follow employees that switch from a low participation firm to a high participation one. Finally, we can see which of the last-minute savers withdraw savings right away in the year after. Taken together this will give us a unique picture in the characteristics of non-participation.