Published Papers

Using novel data from a crowdsourcing platform for ranking stocks, we investigate how investors form expectations about stock returns over the next week. We find that investors extrapolate from stocks' recent past returns, with more weight on more recent returns, especially when recent returns are negative, salient, or from a dispersed cross-section. Such extrapolative beliefs are stronger among nonprofessionals and large stocks. Moreover, consensus rankings negatively predict returns over the next week, more so among stocks with low institutional ownership and a high degree of extrapolation. A trading strategy that sorts stocks on investor beliefs generates an economically significant profit.

We show how data from online social networking services can help researchers better understand the effects of social interactions on economic decision making. We combine anonymized data from Facebook, the largest online social network, with housing transaction data and explore both the structure and the effects of social networks. Individuals whose geographically distant friends experienced larger recent house price increases are more likely to transition from renting to owning. They also buy larger houses and pay more for a given house. Survey data show that these relationships are driven by the effects of social interactions on individuals' housing market expectations.

We use survey data to measure households' propensity to default on mortgages even if they can afford to pay them (strategic default) when the value of the mortgage exceeds the value of the house. The willingness to default increases in both the absolute and the relative size of the home-equity shortfall. Our evidence suggests that this willingness is affected by both pecuniary and non-pecuniary factors, such as views about fairness and morality. We also find that exposure to other people who strategically defaulted increases the propensity to default strategically because it conveys information about the probability of being sued.

In this paper, I identify shocks to interest rates resulting from two administrative details in adjustable-rate mortgage contract terms: the choice of financial index and the choice of lookback period. I find that a 1 percentage point increase in interest rate at the time of adjustable-rate mortgage (ARM) reset results in a 2.5 percentage increase in the probability of foreclosure in the following year, and that each foreclosure filing leads to an additional 0.3 to 0.6 completed foreclosures within a 0.10-mile radius. In explaining this result, I emphasize price effects, bank-supply responses, and borrower responses arising from peer effects.

This research investigates whether a contextual factor-social density, defined as the number of people in a given area-influences consumers' propensity to share information. We propose that high- (vs. low-) density settings make consumers experience a loss of perceived control, which in turn makes them more likely to engage in word of mouth to restore it. Six studies, conducted online as well as in laboratory and naturalistic settings, provide support for this hypothesis. We demonstrate that social density increases the likelihood of sharing information with others and that a person's chronic need for control moderates this effect. Consistent with the proposed process, the effect of social density on information sharing is attenuated when participants have the opportunity to restore control before they engage in word of mouth. We also provide evidence that sharing information restores perceived control in high-density environments, and we disentangle the effect of social density from that of physical proximity.

Previous research on word of mouth (WOM) has presented inconsistent evidence on whether consumers are more inclined to share positive or negative information about products and services. Some findings suggest that consumers are more inclined to engage in positive WOM, whereas others suggest that consumers are more inclined to engage in negative WOM. The present research offers a theoretical perspective that provides a means to resolve these seemingly contradictory findings. Specifically, the authors compare the generation of WOM (i.e., consumers sharing information about their own experiences) with the transmission of WOM (i.e., consumers passing on information about experiences they heard occurred to others). They suggest that a basic human motive to self-enhance leads consumers to generate positive WOM (i.e., share information about their own positive consumption experiences) but transmit negative WOM (i.e., pass on information they heard about others' negative consumption experiences). The authors present evidence for self-enhancement motives playing out in opposite ways for WOM generation versus WOM transmission across four experiments.

We investigate the existence and importance of family welfare cultures, where the receipt of a welfare program by one generation causes increased participation in the next generation. Our context is Norway's disability insurance (DI) system. To overcome the challenge of correlated unobservables across generations, we take advantage of random assignment of judges to DI applicants whose cases are initially denied. Some appeal judges are systematically more lenient, which leads to random variation in the probability a parent will be allowed DI. Using this exogenous variation, we find strong evidence for a causal link across generations: when a parent is allowed DI at the appeal stage, their adult child's participation over the next five years increases by 6 percentage points. This effect grows over time, rising to 12 percentage points after 10 years. Although these findings are specific to our setting, they highlight that welfare reforms can have long-lasting effects on program participation, since any original effect on the current generation could be reinforced by changing the participation behavior of their children as well. The detailed nature of our data allows us to compare the intergenerational transmission with spillover effects in other networks and to explore mechanisms.

We estimate peer effects in paid paternity leave in Norway using a regression discontinuity design. Coworkers and brothers are 11 and 15 percentage points, respectively, more likely to take paternity leave if their peer was exogenously induced to take up leave. The most likely mechanism is information transmission, including increased knowledge of how an employer will react. The estimated peer effect snowballs over time, as the first peer interacts with a second peer, the second peer with a third, and so on. This leads to long-run participation rates which are substantially higher than would otherwise be expected.

This paper investigates whether peer effects play an important role in retirement savings decisions. We use individual data from employees of a large university to study whether individual decisions to enroll in a Tax Deferred Account plan sponsored by the university, and the choice of the mutual fund vendor for people who choose to enroll, are affected by the decisions of other employees in the same department. To overcome the identification problems, we divide the departments into sub-groups (along gender, status, age, and tenure lines) and we instrument the average participation of each peer group by the salary or tenure structure in this group. Our results suggest that peer effects may be an important determinant of savings decisions.

This paper analyzes a randomized experiment to shed light on the role of information and social interactions in employees' decisions to enroll in a Tax Deferred Account (TDA) retirement plan within a large university. The experiment encouraged a random sample of employees in a subset of departments to attend a benefits information fair organized by the university, by promising a monetary reward for attendance. The experiment multiplied by more than five the attendance rate of these treated individuals (relative to controls), and tripled that of untreated individuals within departments where some individuals were treated. TDA enrollment five and eleven months after the fair was significantly higher in departments where some individuals were treated than in departments where nobody was treated. However, the effect on TDA enrollment is almost as large for individuals in treated departments who did not receive the encouragement as for those who did. We provide three interpretations-differential treatment effects, social network effects, and motivational reward effects-to account for these results.

We hypothesize that there is a general bias, based on both innatepredispositions and experience, in animals and humans, to give greater weight to negative entities (e.g., events, objects, personal traits). This is manifested in 4 ways: (a) negative potency (negative entities are stronger than the equivalent positive entities), (b) steeper negative gradients (the negativity of negative events grows more rapidly with approach to them in space or time than does the positivity of positive events, (c) negativity dominance (combinations of negative and positive entities yield evaluations that are more negative than the algebraic sum of individual subjective valences would predict), and (d) negative differentiation (negative entities are more varied, yield more complex conceptual representations, and engage a wider response repertoire). We review evidence for this taxonomy, with emphasis on negativity dominance, including literary, historical, religious, and cultural sources, as well as the psychological literatures on learning, attention, impression formation, contagion, moral judgment, development, and memory. We then consider a variety of theoretical accounts for negativity bias. We suggest that 1 feature of negative events that make them dominant is that negative entities are more contagious than positive entities.

Using a field experiment in a 401(k) plan, we measure the effect of disseminating information about peer behavior on savings. Low-saving employees received simplified plan enrollment or contribution increase forms. A randomized subset of forms stated the fraction of age-matched coworkers participating in the plan or age-matched participants contributing at least 6% of pay to the plan. We document an oppositional reaction: the presence of peer information decreased the savings of nonparticipants who were ineligible for 401(k) automatic enrollment, and higher observed peer savings rates also decreased savings. Discouragement from upward social comparisons seems to drive this reaction.