A review of international experience

participation rate (%)
maximum employer match (% of pay)
regression line
SOURCE: Beshears and others 2010.
These results confirm the earlier conclusion: increasing the match rate on savings 
leads to small increases in savings plan participation. This conclusion holds for schemes 
with and without automatic enrollment.
Although automatic enrollment leads to unambiguous increases in savings plan par-
ticipation, its effects on savings plan contributions conditional on participation depend 
very much on the default contribution rate at which individuals are automatically enrolled. 
Just as the match threshold for savings plan contributions attracts the largest share of sav-
ings plan participants when there is a match, so too does the automatic enrollment default 
contribution rate when there is automatic enrollment. Contributions are higher with a 
higher default contribution rate under automatic enrollment than with a lower default 
contribution rate. The distribution of contribution rates for employees at a U.S. company 
that increased the default contribution rate in its savings plan from 3 percent of pay to 
6 percent of pay is shown in figure 15.7. With a default contribution rate of 3 percent, 
28 percent of plan participants contribute 3 percent of pay to the plan; another 24 per-
cent contribute 6 percent to the plan, the match threshold; and 41 percent contribute at 
a rate above 6 percent, primarily either 10 percent or 15 percent of pay (although these 

300 
MATCHING CONTRIBUTIONS FOR PENSIONS: A REVIEW OF INTERNATIONAL EXPERIENCE
two contribution rates are aggregated with other contribution rates in the figure). With 
a default contribution rate of 6 percent of pay, which coincides with the match thresh-
old, almost half of employees contribute 6 percent of pay to the plan, twice the fraction 
observed with a default contribution rate of 3 percent; the fraction of employees contrib-
uting 3 percent of pay to the plan is an almost negligible 4 percent.
FIGURE 15.7  Automatic enrollment for new hires and the distribution of savings plan contribution 
rates
% of employees
4 
1 
28 
3 
24 
18 
23 
9 
6 
4 
2 
49 
17 
13 
0 
10 
20 
30 
40 
50 
60 
0 1–2 3 4–5 6 7–10 
11–15 
contribution rate (%) 
hired under automatic enrollment (3% default)   
hired under automatic enrollment (6% default)
SOURCE: Beshears and others 2008.
A more extreme form of automatic enrollment is mandatory enrollment: individu-
als are automatically enrolled without the option of subsequently opting out. Most of 
the literature on defined contribution savings plans has focused on employer-sponsored 
401(k)–type plans in the United States, where voluntary participation is standard. In other 
contexts, participation in defined contribution savings schemes is mandatory. For exam-
ple, public sector entities in the United States that have a defined contribution scheme as 
their primary retirement savings plan (or one of their primary plans if participants have a 
choice of plans) tend to have mandatory enrollment with no option to opt out (Beshears 
and others 2011). Countries with defined contribution social security systems typically 
have automatic and mandatory participation, at least for workers in the formal sector. 
Whether to make participation voluntary or mandatory is an important policy question 
for defined contribution savings plans.
SIMPLIFICATION
One limitation of automatic and mandatory savings plan enrollment schemes is that these 
approaches work only in formal sector labor markets with developed financial institutions 
that can facilitate payroll deduction. In informal labor markets, these approaches are more 
difficult to implement. Lessons from the effect of automatic enrollment on increasing 

15.  MATCHING CONTRIBUTIONS AND SAVINGS OUTCOMES: A BEHAVIORAL ECONOMICS PERSPECTIVE 
301
participation rates in these contexts can inform the structuring of savings schemes in other 
contexts.
The success of automatic enrollment in employer-sponsored savings plans in the 
United States is predicated on two factors: (1) that most people recognize the need for 
retirement income above and beyond what they will get from social security and therefore 
want to save and (2) that automatic enrollment simplifies what individuals already want 
to do. Several pieces of evidence support the notion that people generally want to save. 
First, when asked, individuals typically state a desire to save.
2
Second, when asked to actively make a choice about whether and how much to save, 
most people choose to save. Carroll and others (2009) compare the savings outcomes in 
an employer-sponsored savings plan before and after employees were compelled to make 
an active choice about whether to participate in the savings plan. They find that when not 
required to make a choice, only 41 percent of newly hired employees enrolled in the sav-
ings plan. In contrast, when required to make an active choice about savings plan partici-
pation (which could include not participating), 69 percent enrolled. They conclude that 
most employees want to save but that an opt-in enrollment regime does not accurately 
reflect these preferences, because nonparticipation is consistent with both a preference not 
to save as well as with a preference to save accompanied by a delay in execution.
Third, very few people opt out of savings plan participation when they are automati-
cally enrolled. Choi and others (2002, 2006) show that savings plan participation is very 
persistent regardless of whether employees are automatically enrolled. In particular, only 
2–3 percent of automatically enrolled employees opt out of savings plan participation in a 
12-month period. That savings rates are high and persistent under automatic enrollment 
is further evidence that most people generally want to save.
An important caveat to these findings is that they yield evidence on saving prefer-
ences for a specific set of individuals in a very specific context: employees in U.S. firms 
with access to employer-sponsored savings plans. These findings say nothing about saving 
preferences outside the United States (although one would surmise that many individu-
als throughout the world also want to save; see for example, Soman and Cheema 2011) 
or about saving preferences in other types of savings vehicles. Most employer-sponsored 
savings plans in the United States offer an employer match, which may induce some oth-
erwise reluctant individuals to save. The evidence suggests that the effect of a match on 
savings plan participation is not large; nonetheless, a financial inducement is one way to 
shape saving preferences.
A potentially more important contextual factor is the level of trust individuals have 
that their savings will be secure. Guiso, Sapienza, and Zingales (2008) show that dif-
ferences in the level of trust across countries explain a sizable share of the cross-country 
variation in individual stock holding: in countries with higher levels of trust, citizens are 
more willing to invest in equities. Adopting a regulatory framework that increases trust in 
financial institutions and the financial system may be a prerequisite to successfully increas-
ing saving with any savings scheme.
3
The second factor accounting for the success of automatic enrollment is that it sim-
plifies the execution of what individuals already want to do—save. Indeed, automatic 
enrollment is an extreme form of simplification; individuals who want to save need 
not do anything. Psychologists have long recognized that choice complexity can affect 

302 
MATCHING CONTRIBUTIONS FOR PENSIONS: A REVIEW OF INTERNATIONAL EXPERIENCE
decision-making outcomes. One result is procrastination—individuals put off decision 
making as choices become more complicated (Dhar and Nowlis 1999; Iyengar and Lepper 
2000; Shafir, Simonson, and Tversky 1993; Tversky and Shafir 1992). 
Iyengar, Huberman, and Jiang (2004) show that in the United States, enrollment in 
employer-sponsored savings plan is negatively correlated with the number of investment 
options in the savings plans: having 10 additional options in the investment menu led to 
a 1.5–2.0 percentage point decline in participation.
4
 They hypothesize that having more 
investment options increases the complexity of choosing an asset allocation. Automatic 
enrollment decouples the choice about whether to save from the choice about how much 
to save or which asset allocation to select. The initial participation decision is simplified 
from one that involves evaluating myriad options to a simple comparison of two alterna-
tives: nonparticipation (consumption or saving outside of the savings plan) versus partici-
pating at a prespecified contribution rate with a prespecified asset allocation. Madrian and 
Shea (2001) and Choi and others (2004a) find that automatic enrollment has its largest 
impact on participation for workers who are least financially sophisticated—the young 
and people with lower levels of income. These are the individuals for whom the complex-
ity of the participation decision under an opt-in savings regime poses the greatest deter-
rent to participation (Beshears and others 2008).
If complexity is a deterrent to participation in a savings plan, then simplifying the 
task of savings plan enrollment, even if less extreme than automatic enrollment, should 
increase participation. Choi, Laibson, and Madrian (2009) and Beshears and others (2012) 
study the impact of a simplified enrollment process on outcomes in employer-sponsored 
savings plans. The intervention they evaluate, Quick Enrollment, gives employees a way 
to enroll in their employer-sponsored savings plan at a contribution rate and with an 
asset allocation preselected by their employer. Like automatic enrollment, this approach 
allows individuals to evaluate savings plan participation (at the preselected contribution 
rate and asset allocation) as a simple binary choice, without having to confront the multi-
dimensional challenge of choosing a contribution rate or an asset allocation. At the two 
firms studied, Quick Enrollment increased savings plan participation by 10–20 percent-
age points relative to a standard opt-in enrollment regime (figure 15.8). This finding 
suggests that complexity can be a significant deterrent to savings plan participation and 
that other measures to simplify the process of saving in this or other contexts could mate-
rially affect savings outcomes.
5
 Although the participation increases from this simplified 
approach to savings plan enrollment are not nearly as large as the estimated effects of auto-
matic enrollment, they are sizable and much larger than the estimated effects of matching 
contributions. Simplifying and streamlining the saving process can have a sizable impact 
on outcomes and may be a much more cost-effective approach to changing behavior than 
financial incentives.
Merely providing access to a simple and straightforward way to save may increase 
saving. Dupas and Robinson (2010) in rural Kenya and Aportela (1999) in rural Mexico 
find that increasing access to the formal saving sector leads to higher levels of saving. In 
the case of the field experiment evaluated in Dupas and Robinson (2010), the newly avail-
able savings account offered no interest and charged withdrawal fees, yet demand for the 
account was still high.

15.  MATCHING CONTRIBUTIONS AND SAVINGS OUTCOMES: A BEHAVIORAL ECONOMICS PERSPECTIVE 
303
EXECUTION AIDS
Even if individuals want to save, forgetfulness and procrastination may prevent execution 
of even the best-laid plans. Many strategies have been adopted to help individuals follow 
through on their savings goals. Research has identified a lack of planning as a primary rea-
son why individuals fail to achieve their goals (Gollwitzer 1999; Gollwitzer and Sheeran 
2006). 
Lusardi, Keller, and Keller (2009) study the impact of helping individuals form and 
implement a savings plan on savings outcomes. The intervention they study—a planning 
aid for savings plan enrollment at a U.S. employer—encourages individuals to set aside a 
specific time for enrolling in their savings plan, outlines the steps involved in enrolling in 
a savings plan (for example, choosing a contribution rate and an asset allocation), gives an 
approximation of the time each step will take, and provides tips on what to do if individu-
als get stuck. Provision of this planning aid increased enrollment in an employer-spon-
sored savings plan by 12–21 percentage points for newly hired employees (figure 15.9). 
This effect is two to three times the estimated impact of matching contributions on sav-
ings plan participation. Like simplifying the saving process, providing execution aids is 
extremely cost-effective.
In a series of field experiments conducted in cooperation with banks in Bolivia, 
Peru, and the Philippines, Karlan and others (2010) evaluate the impact of providing sav-
ings reminders (text messages or letters) on savings outcomes in bank savings accounts. 
They find that people who received reminders were 3 percent more likely to achieve a pre-
specified savings goal and saved 6 percent more in the bank sending the reminders than 
did people who did not receive reminders. They also find that reminders that highlighted 
individuals’ savings goals were twice as effective as generic reminders.
FIGURE 15.8  Quick Enrollment and savings plan participation: Firms C and D
% participating in savings plan
 
after Quick
Enrollment
34% 
0 
10 
20 
30 
40 
50 
Company C: 4 months after baseline 
Company D: 4 months after baseline 
after Quick
Enrollment
16% 
before Quick
Enrollment
9% 
before Quick
Enrollment
6% 
SOURCE: Beshears and others 2012.

304 
MATCHING CONTRIBUTIONS FOR PENSIONS: A REVIEW OF INTERNATIONAL EXPERIENCE
Kast, Meier, and Pomeranz (2012) evaluate the impact of providing text mes-
sage reminders on bank savings outcomes in Chile. They also find that individuals who 
received text message reminders saved substantially more than individuals who did not. 
For the populations in the developing countries targeted in the field experiments 
of these two studies, ongoing saving requires ongoing action—automatic enrollment and 
direct deposit are not relevant alternatives. These results suggest that limited attention can 
be an important impediment to saving in such contexts. Text messages are a cost-effective 
and scalable way to create attention shocks that motivate people to take action and follow 
through on prespecified savings goals.
The field experiments discussed in Karlan and others (2010) and Kast, Meier, and 
Pomeranz (2012) included treatment arms that offered individuals higher than market 
interest rates as an inducement to save. Neither study finds any statistically significant 
impact of a higher interest rate on savings outcomes. The higher interest rates were admit-
tedly much lower than the match rates that typically characterize matched savings schemes 
(in Kast, Meier, and Pomeranz 2012, for example, the high interest rate treatment group 
was offered an interest rate of 5 percent as compared to a then-prevailing interest rate of 
0.3 percent). Although these studies are not directly comparable to the studies discussed 
earlier on the impact of matching contributions on savings outcomes, the results support 
the general qualitative conclusion that financial incentives have at best modest effects on 
outcomes.
A growing body of literature examines a broad class of execution aids known as 
commitment savings products. In the most influential paper in this literature, Ashraf, 
Karlan, and Yin (2006) evaluate a field experiment in the Philippines that offered one 
such product to current or former clients of a local bank. In this field experiment, partici-
pating bank clients who opted for the commitment savings product voluntarily restricted 
the right to withdraw their savings until reaching either an individually chosen goal date 
FIGURE 15.9  Impact of planning aids on savings plan participation
% participating in savings plan
7% 
29% 
22% 
45% 
28% 
41% 
0
10
20
30
40
50
30 days after hire
60 days after hire
no help
8-step planning aid
7-step planning aid 
SOURCE: Lusardi, Keller, and Keller 2009.

15.  MATCHING CONTRIBUTIONS AND SAVINGS OUTCOMES: A BEHAVIORAL ECONOMICS PERSPECTIVE 
305
or an individually chosen goal amount. They show that there is a demand for commit-
ment: among people who were offered the option to open a commitment savings account, 
28 percent did so, even though it offered reduced flexibility and no higher interest than 
a standard bank account. Commitment products can have a sizable impact on savings. 
Relative to a control group not offered the commitment savings product, people offered 
a commitment account had bank balances that were 82 percent higher 12 months later. 
Corroborating work on commitment savings products in other countries includes Gugerty 
(2007), Ashraf and others (2011), Brune and others (2011), and Dupas and Robinson 
(2011). The reasons why commitment savings products are so effective at increasing sav-
ing are both internal (reducing the temptation to spend) and external (credibly telling 
others, primarily friends and family, that one’s savings are inaccessible).
Soman and Cheema (2011) evaluate one interesting variant of a commitment sav-
ings technology in a field experiment targeted at unbanked construction laborers in rural 
India who are paid cash wages. In this experiment, individuals earmarked a certain amount 
of their weekly wages as savings. A social worker visited participating households every pay 
day to set aside the earmarked savings amount into either one (nonpartitioned) or two 
(partitioned) sealed envelopes. The challenge in this field experiment was not to motivate 
individuals to set aside money for savings but to prevent them from raiding their savings. 
The authors show that partitioning earmarked savings into multiple “accounts” increased 
realized savings by 39–216 percent. They hypothesize that opening a savings envelope, or 
violating the partition, induces guilt. Having multiple accounts, or partitions, increases 
the psychological cost of spending money that has been set aside for a specific purpose. 
This simple, low-cost execution aid has obvious extensions to other contexts. For exam-
ple, having multiple retirement savings accounts may be more effective than relying on 
one type of savings account (for example, having both a retirement income account and 
a retirement health account may induce higher savings than a single generic retirement 
account).
Collectively, the research on execution aids suggests that many psychological imped-
iments stand in the way of carrying out even the best-laid plans to save. Financial incen-
tives do little in the face of such barriers. A more effective strategy is to directly address the 
barriers themselves.
Conclusions
A large body of literature has examined a wide variety of approaches to encouraging indi-
viduals to increase their savings. Traditional economic models point to financial incen-
tives, such as a matching contribution, as the logical mechanism for increasing savings 
plan participation. The research on matching contributions and savings plan participa-
tion is largely consistent with traditional economic models: a matching contribution does 
increase participation. But the quantitative impact of matching contributions on savings 
plan participation is small. The studies using the most credible empirical methods find 
strikingly similar results in a variety of different contexts using a variety of different data 
sources: a matching contribution of 25 percent increases savings plan participation by 
roughly 5 percentage points.

306 
MATCHING CONTRIBUTIONS FOR PENSIONS: A REVIEW OF INTERNATIONAL EXPERIENCE
The theoretical impact of matching contributions on the level of savings in tradi-
tional models depends on how much an individual would save in the absence of a match. 
The empirical results on this question finds results are inconsistent, although the most 
credible empirical work corroborates the predictions of traditional economic models.
Traditional economic models fail to characterize the most interesting features of the 
savings choices that individuals make. Savings rates cluster heavily around focal points, 
including the match threshold (as traditional economic theory would predict) and num-
bers that are multiples of five (something traditional economic theory would not predict). 
This finding suggests that the match threshold may be a much more important parameter 
in a matching scheme than the match rate.
Traditional economic models also fail to incorporate the many psychological fric-
tions that impede saving, including present bias, complexity, inattention, and temptation. 
In many cases, countering these frictions leads to increases in savings plan participation 
and asset accumulation that surpass the effects of a typical matching contribution, poten-
tially at a lower cost.
Notes
1.  Individuals 50 and older may also be allowed to make additional “catch-up” contributions of 
up to $5,000 a year.
2.  For example, Choi and others (2002 and 2006) report the results of a survey on retirement 
savings adequacy conducted by a large U.S. employer. Two-thirds of the responding employees 
stated a desire to save more than they were currently saving; one-third reported that they were 
saving about the right amount; and less than 1 percent responded that they were saving too 
much.
3.  There is no evidence on how financial incentives interact with the level of trust to affect sav-
ing. If financial incentives substitute for trust, the small impact of financial incentives on sav-
ing in the United States may reflect a high level of trust in the United States but might not rule 
out a larger effect of financial incentives in countries with lower levels of trust. Alternatively, 
trust may be a precondition for financial incentives to have any impact at all.
4.  This correlation is documented only among plans that do not have automatic enrollment.
5.  Research has documented sizable impacts of simplification in contexts other than saving, 
including school choice (Hastings and Weinstein 2008); health plan choice (Kling and others 
2008); mutual fund selection (Choi, Laibson, and Madrian 2010); and both college financial 
aid applications and college attendance (Bettinger and others 2009).
References
Andrews, Emily. 1992. “The Growth and Distribution of 401(k) Plans.” In Trends in Pensions 
1992, ed. John Turner and Daniel Beller, 149–76. Washington, DC: U.S. Government Print-
ing Office.
Aportela, Fernando. 1999. “Effects of Financial Access on Savings by Low-Income People.” Banco 
de México Working Paper.
Ashraf, Nava, Diego Aycinena, Claudia Martínez, and Dean Yang. 2011. “Remittances and the 
Problem of Control: A Field Experiment among Migrants from El Salvador.” Universidad de 
Chile Working Paper SDT 341.

15.  MATCHING CONTRIBUTIONS AND SAVINGS OUTCOMES: A BEHAVIORAL ECONOMICS PERSPECTIVE 
307
Ashraf, Nava, Dean Karlan, and Wesley Yin. 2006. “Tying Odysseus to the Mast: Evidence from 
a Commitment Savings Product in the Philippines.” Quarterly Journal of Economics 121 (2): 
635–72.
Bassett, William F., Michael J. Fleming, and Anthony P. Rodrigues. 1998. “How Workers Use 
401(k) Plans: The Participation, Contribution, and Withdrawal Decisions.” National Tax 
Journal 51 (2): 263–89.
Beshears, John, James J. Choi, David Laibson, and Brigitte C. Madrian. 2008. “The Importance 
of Default Options for Retirement Savings Outcomes: Evidence from the United States.” In 
Lessons from Pension Reform in the Americas, ed. Stephen J. Kay and Tapen Sinha, 59–87. New 
York: Oxford University Press.
—. 2010. “The Impact of Employer Matching on Savings Plan Participation under Auto-
matic Enrollment.” In Research Findings in the Economics of Aging, ed. David A. Wise, 311–27. 
Chicago: University of Chicago Press.
—. 2011. “Behavioral Economics Perspectives on Public Sector Pension Plans.” Journal of 
Pension Economics and Finance 10 (2): 315–36.
—. 2012. “Simplification and Saving.” Journal of Economic Behavior and Organizations.
Bettinger, Eric P., Bridget Terry Long, Philip Oreopolous, and Lisa Sanbonmatsu, 2009. “The 
Role of Simplification and Information in College Decisions: Results from the H&R Block 
FAFSA Experiment.” NBER Working Paper 15361, National Bureau of Economic Research, 
Cambridge, MA.
Brune, Lasse, Xavier Giné, Jessica Goldberg, and Dean Yang. 2011. “Commitments to Save: A 
Field Experiment in Rural Malawi.” Policy Research Working Paper 5748, World Bank, Wash-
ington, DC.
Carroll, Gabriel D., James J. Choi, David Laibson, Brigitte C. Madrian, and Andrew Metrick. 
2009. “Optimal Defaults and Active Decisions: Theory and Evidence from 401(k) Saving.” 
Quarterly Journal of Economics 124 (4): 1639–74.
Choi, James J., David Laibson, and Brigitte C. Madrian. 2009. “Reducing the Complexity Costs 
of 401(k) Participation through Quick Enrollment
TM
.” In Developments in the Economics of 
Aging, ed. David A. Wise, 57–82. Chicago: University of Chicago Press.
—. 2011. “$100 Bills on the Sidewalk: Violations of No-Arbitrage in 401(k) Accounts.” 
Review of Economics and Statistics 113 (3): 748–63.
Choi, James J., David Laibson, Brigitte C. Madrian, and Andrew Metrick. 2002. “Defined Contri-
bution Pensions: Plan Rules, Participant Decisions, and the Path of Least Resistance.” In Tax 
Policy and the Economy, vol. 16, ed. James M. Poterba, 67–113. Cambridge, MA: MIT Press.
—. 2004a. “For Better or for Worse: Default Effects and 401(k) Savings Behavior.” In Perspec-
tives on the Economics of Aging, ed. David A. Wise, 81–121. Chicago: University of Chicago 
Press.
—. 2004b. “Plan Design and 401(k) Savings Outcomes.” National Tax Journal 57 (2): 275–98.
—. 2006. “Saving for Retirement on the Path of Least Resistance.” In Behavioral Public 
Finance: Toward a New Agenda, ed. Edward J. McCaffrey and Joel Slemrod, 304–51. New 
York: Russell Sage Foundation.
Clark, Robert L., Gordon Goodfellow, Sylvester Schieber, and Drew Warwick. 2000. “Making 
the Most of 401(k) Plans: Who’s Choosing What and Why.” In Forecasting Retirement Needs 
and Retirement Wealth, ed. Olivia Mitchell, Brett Hammond, and Anna Rappaport, 95–138. 
Philadelphia: University of Pennsylvania Press.

308 
MATCHING CONTRIBUTIONS FOR PENSIONS: A REVIEW OF INTERNATIONAL EXPERIENCE
Clark, Robert L., and Sylvester Schieber. 1998. “Factors Affecting Participation Levels in 401(k) 
Plans.” In Living with Defined Contribution Plans: Remaking Responsibility for Retirement, ed. 
Olivia Mitchell and Sylvester J. Schieber, 69–97. Philadelphia: University of Pennsylvania 
Press. 
Dhar, Ravi, and Stephen M. Nowlis. 1999. “The Effect of Time Pressure on Consumer Choice 
Deferral.” Journal of Consumer Research 25 (4): 369–84.
Duflo, Esther, William Gale, Jeffrey Liebman, Peter Orszag, and Emmanuel Saez. 2006. “Saving 
Incentives for Low- and Middle-Income Families: Evidence from a Field Experiment with 
H&R Block.” Quarterly Journal of Economics 121 (4): 1311–46.
Dupas, Pascaline, and Jonathan Robinson. 2010. “Savings Constraints and Microenterprise Devel-
opment: Evidence from a Field Experiment in Kenya.” International Policy Center Working 
Paper 111.
—. 2011. “Why Don’t the Poor Save More? Evidence from Health Savings Experiments.” 
NBER Working Paper 17255, National Bureau of Economic Research, Cambridge, MA.
Dworak-Fisher, Keenan. 2008. “Encouraging Participation in 401(k) Plans: Reconsidering the 
Employer Match.” U.S. Bureau of Labor Statistics Working Paper 420. 
Engelhardt, Gary V., and Anil Kumar. 2007. “Employer Matching and 401(k) Saving: Evidence 
from the Health and Retirement Study.” Journal of Public Economics 91(10): 1920–443.
Even, William E., and David A. Macpherson. 1997. “Factors Influencing Participation and Con-
tribution Levels in 401(k) Plans.” Florida State University Working Paper.
—. 2005. “The Effects of Employer Matching in 401(k) Plans.” Industrial Relations 44 (3): 
525–49.
GAO (General Accounting Office). 1997. “401(k) Pension Plans: Loan Provisions Enhance Par-
ticipation But May Affect Income Security for Some.” Report to the Chairman, Special Com-
mittee on Aging, and the Honorable Judd Gregg, U.S. Senate, GAO, Washington, DC.
Gollwitzer, Peter M. 1999. “Implementation Intentions: Strong Effects of Simple Plans.” American 
Psychologist 54 (7): 493–503.
Gollwitzer, Peter M., and Paschal Sheeran. 2006. “Implementation Intentions and Goal Achieve-
ment: A Meta-Analysis of Effects and Processes.” Advances in Experimental Social Psychology 
38: 69–119.
Gugerty, Mary Kay. 2007. “You Can’t Save Alone: Commitment in Rotating Savings and Credit 
Associations in Kenya.” Economic Development and Cultural Change 55 (2): 251–82.
Guiso, Luigi, Paola Sapienza, and Luigi Zingales. 2008. “Trusting the Stock Market.” Journal of 
Finance 63 (6): 2557–600.
Hastings, Justine S., and Jeffrey M. Weinstein. 2008. “Information, School Choice, and Academic 
Achievement: Evidence from Two Experiments.” Quarterly Journal of Economics 123 (4): 
1373–414.
Huberman, Gur, Sheena S. Iyengar, and Wei Jiang. 2007. “Defined Contribution Pension Plans: 
Determinants of Participation and Contribution Rates.” Journal of Financial Services Research 
31(1): 1–32.
Iyengar, Sheena S., Gur Huberman, and Wei Jiang. 2004. “How Much Choice Is Too Much? 
Contributions to 401(k) Retirement Plans.” In Pension Design and Structure: New Lessons from 
Behavioral Finance, ed. Olivia Mitchell and Stephen Utkus, 83–95. Oxford: Oxford Univer-
sity Press.

15.  MATCHING CONTRIBUTIONS AND SAVINGS OUTCOMES: A BEHAVIORAL ECONOMICS PERSPECTIVE 
309
Iyengar, Sheena S., and Mark R. Lepper. 2000. “When Choice Is Demotivating: Can One Desire 
Too Much of a Good Thing?” Journal of Personality and Social Psychology 79 (6): 995–1006.
Karlan, Dean, Margaret McConnell, Sendhil Mullainathan, and Jonathan Zinman. 2010. “Get-
ting to the Top of Mind: How Reminders Increase Saving.” NBER Working Paper 16205, 
National Bureau of Economic Research, Cambridge, MA.
Kast, Felipe, Stephan Meier, and Dina Pomeranz. 2012. “Under-Savers Anonymous: Evidence on 
Self-Help Groups and Peer Pressure as a Savings Commitment Device.” IZA Discussion Paper 
6311, Institute for the Study of Labor, Bonn.
Kling, Jeffrey R., Sendhil Mullainathan, Eldar Shafir, Lee Vermeulen, and Marian Wrobel, 2008. 
“Misperception in Choosing Medicare Drug Plans.” Harvard University working paper, Cam-
bridge, MA.
Kusko, Andrea, James Poterba, and David Wilcox. 1998. “Employee Decisions with Respect to 
401(k) Plans.” In Living with Defined Contribution Pensions: Remaking Responsibility for Retire-
ment, ed. Olivia Mitchell and Sylvester Schieber, 98–112. Philadelphia: University of Penn-
sylvania Press.
Lusardi, Annamaria, Punam Anand Keller, and Adam M. Keller. 2008. “New Ways to Make People 
Save: A Social Marketing Approach.” In Overcoming the Saving Slump: How to Increase the 
Effectiveness of Financial Education and Saving Programs, ed. Annamaria Lusardi, 209–36. Chi-
cago: University of Chicago Press.
Madrian, Brigitte C., and Dennis F. Shea. 2001. “The Power of Suggestion: Inertia in 401(k) Par-
ticipation and Savings Behavior.” Quarterly Journal of Economics 116 (4): 1149–87.
Mills, Gregory, William G. Gale, Rhiannon Patterson, Gary V. Engelhardt, Michael D. Eriksen, 
and Emil Apolstolov. 2008. “Effects of Individual Development Accounts on Asset Purchases 
and Saving Behavior: Evidence from a Controlled Experiment.” Journal of Public Economics 92 
(5–6): 1509–30.
Mitchell, Olivia S., Stephen P. Utkus, and Tongxuan Yang. 2007. “Turning Workers into Sav-
ers? Incentives, Liquidity, and Choice in 401(k) Plan Design.” National Tax Journal 60 (3): 
469–89.
Munnell, Alicia H., Annika Sundén, and Catherine Taylor. 2001. “What Determines 401(k) Par-
ticipation and Contributions?” Social Security Bulletin 64 (3): 64–75.
Nessmith, William E., Stephen P. Utkus, and Jean A. Young. 2007. “Measuring the Effectiveness of 
Automatic Enrollment.” Vanguard Center for Retirement Research, Malvern, PA. 
Papke, Leslie E. 1995. “Participation in and Contributions to 401(k) Pension Plans.” Journal of 
Human Resources 30(2): 311–25.
Papke, Leslie E., and James M. Poterba. 1995. “Survey Evidence on Employer Match Rates and 
Employee Saving Behavior in 401(k) Plans.” Economics Letters 49 (3): 313–17.
Shafir, Eldar, Itamar Simonson, and Amos Tversky. 1993. “Reason-Based Choice.” Cognition 
49(1–2): 11–36.
Soman, Dilip, and Amar Cheema. 2011. “Earmarking and Partitioning: Increasing Saving by Low-
Income Households.” Journal of Marketing Research 48 (S1): S14–S22.
Tversky, Amos, and Eldar Shafir. 1992. “Choice under Conflict: The Dynamics of Deferred Deci-
sion.” Psychological Science 3 (6): 358–61.
VanDerhei, Jack, and Sarah Holden. 2001. “Contribution Behavior of 401(k) Plan Participants.” 
Investment Company Institute Perspective 7 (4):1–19.

311
CHAPTER 16