The Distributional Impact of the Sharing Economy on the Housing Market

Revise and Resubmit, American Economic Review.

[show/hide abstract]

What is the impact of the sharing economy, pioneered by companies such as Airbnb, on the housing market? In this paper, I estimate the welfare and distributional impact of Airbnb on the residents of New York City. I develop a model of an integrated housing market, where a landlord can offer a housing unit for rent on either the traditional long-term rental market or the newly available short-term rental market. By estimating a structural model of residential choice and linking it to detailed Airbnb usage data, I estimate the effect of such reallocation on the equilibrium rent across different housing types and demographic groups. In addition, to evaluate the gains from direct home-sharing, I estimate a supply system featuring heterogeneous costs. Overall, renters in New York City suffer a loss of $178mm per annum, as the losses from the rent channel dominate the gains from the host channel. I find that the increased rent burden falls most heavily on high-income, educated, and white renters, because they prefer housing and location amenities most desirable to tourists. Moreover, there is a divergence between the median and the tail, where a few enterprising low-income households obtain substantial gains from home-sharing, especially during demand peaks. Thus, this paper delivers a more nuanced characterization of the winners and losers of the sharing economy, and provides a framework for understanding the consequences of regulating such technological innovations.

Algorithmic Pricing in Multifamily Rentals: Efficiency Gains or Price Coordination?

(with Gi Heung Kim), Revise and Resubmit, Journal of Political Economy.

[show/hide abstract]

This paper empirically evaluates the impact of algorithmic pricing on the U.S. multifamily rental market. We hand-collect data on management company adoption decisions of algorithmic pricing and combine it with a comprehensive database of building-level rents and occupancy from 2005 to 2019. We find strong evidence that algorithmic pricing helps building managers set prices that are more responsive to market conditions, with adopters lowering rents more rapidly than non-adopters during economic downturns. We also find that average rents are higher and average occupancies are lower in markets with greater algorithmic penetration during periods of economic recovery. Then, we estimate a structural model of housing demand to test for ''algorithmic coordination.'' Compared to a model of own profit maximization, our pair-wise tests favor a model of joint profit maximization among adopters of the same software. We estimate that the coordination channel results in an average markup increase of $25 per unit per month, impacting about 4.2 million units nationwide. Our findings have important implications for regulators and policymakers concerned about the potential risks and trade-offs of algorithmic pricing.

An Anti-IV Approach for Pricing Residential Amenities: Applications to Flood Risk

(with Alex Bell, Stephen Billings, and Shusheng Zhong), Working Paper, 2024.

[show/hide abstract]

Understanding how markets price housing amenities is essential to addressing residential socioeconomic disparities. Traditional methods have struggled with unobserved quality, often producing wrong-signed estimates. This paper introduces a novel "anti-instrument" approach to estimating amenity prices amid unobserved quality. Our identification strategy relies on an "anti-instrumental" variable that is relevant to the unobserved quality, but conditionally independent from the amenity of interest, which are properties opposite of typical instruments for amenities. We apply this method to estimate the implicit price of flood risk using detailed housing transaction records and house-level flood risk measures. Household income, used as an anti-instrument, successfully recovers the negative price of flood risk. We find that being located in a FEMA-designated floodplain reduces home prices by 2.8% on average. Drawing on more granular measures of flood risk, we find increasing price discounts for homes with higher flood probabilities, with the most risky locations suffering a discount of nearly 5%. These price estimates are crucial for evaluating returns on investment in local public goods and environmental policies.

Unobserved Heterogeneity, State Dependence, and Health Plan Choices

(with Ariel Pakes, Jack Porter, and Mark Shepard), Reject and Resubmit, American Economic Review.

[show/hide abstract]

In this paper, we propose a new approach for the structural estimation of switching costs in a panel data discrete choice model with individual choice-specific fixed effects, by leveraging the recent developments in moment inequality methods. We achieve this by comparing the choice probabilities of the same individual over two different points in time. With non-parametric i.i.d. distribution of the error terms, we develop a set estimator for the switching cost by constructing a set of improving choices over time. Next, when allowing for generic parametric assumptions on the error term, we develop conditional moment inequalities by making inference on the revealed preferences of a pair of choices. We show that in the CommCare Health Plan Choice context, they provide particularly informative upper bounds on the switching cost, rejecting a simplistic model when unobserved heterogeneity is omitted. In addition, if the errors are assumed to be logit, we make further improvements as the ratio of odds-ratio is independent of individual heterogeneity. Overall, these new sets of inequalities provide an intuitive and easy-to-implement test to assess the robustness of state-dependence models when concerns about unobserved heterogeneity loom significant.

The Value of Information: Why You Should Add the Second Order Conditions

[show/hide abstract]

When conducting estimation based on agent optimization, I show that one can improve the performance of the estimator when information such as the second-order condition is incorporated as moment inequality restrictions, especially when there are weak instruments. I run a simulation study to demonstrate the effectiveness of this approach in both continuous and discrete choice problems.

Diversity and Performance in Entrepreneurial Teams

(with Paul Gompers, Kanyuan Kevin Huang), NBER Working Paper, 2023.

[show/hide abstract]

We study the role of diversity on the performance of entrepreneurial teams by exploiting a unique experimental setting of over 3,000 MBA students who participated in a business course to build startups. First, we quantify the strong selection based upon shared attributes when students are allowed to choose teammates. Team formation based upon shared endowed demographic characteristics such as gender, race, and ethnicity is stronger than team formation based upon acquired characteristics such as education and industry background. Second, when team memberships are randomly assigned, greater racial/ethnic diversity leads to significantly worse performance. Interestingly, the negative performance effect of diversity is partially alleviated in cohorts where teams are formed voluntarily. Finally, we find that teams with more female members performed substantially better when their faculty section leader was female. These findings suggest that policy interventions targeting greater diversity should consider match-specific qualities in forming teams to prevent the potential negative impact of diversity. Our results on vertical diversity suggest that capital allocators could also play an important role in the mentoring and advising of minority entrepreneurs.

Venture Capital's Me Too Moment

(with Paul Gompers, Patrick Sweeney) NBER Working Paper, 2021.

[show/hide abstract]

In this paper, we document the historically low rate of hiring of women in the venture capital sector. We find that the high-profile Ellen Pao v. Kleiner Perkins gender discrimination trial had dramatic treatment effects. In difference-in-differences regressions, we find that the rate of hiring of female venture capitalists increased substantially after the trial and that the hiring was more pronounced in states that were more receptive to the exposure. We use the state-level mandated maternity benefits as an instrument for the receptivity to the treatment effects of the Pao Trial. We also show that the fraction of founders who are female increases after the Pao Trial, but that the increase is driven entirely by the hiring of female venture capitalists. There is no increase in the propensity of male venture capitalists to invest in female founders in the post-Pao Trial period.

Diversity in Innovation

(with Paul Gompers) NBER Working Paper, 2017.

[show/hide abstract]

In this paper we document the patterns of labor market participation by women and ethnic minorities in venture capital firms and as founders of venture capital-backed startups. We show that from 1990-2016 women have been less than 10% of the entrepreneurial and venture capital labor pool, Hispanics have been around 2%, and African Americans have been less than 1%. This is despite the fact that all three groups have much higher representation in education programs that lead to careers in these sectors as well as having higher representation in other highly-compensated professions. Asians, on the other hand, have much higher representation in the venture capital and entrepreneurial sector than their overall percentages in the labor force. We explore potential supply side explanations including both education attainment as well as relevant prior job experience. We also explore the correlation between diversity and state-level variations. Finally, we discuss how these patterns are consistent with homophily-based hiring and homophily-induced information flows about career choices. We end the paper by discussing areas for future research.

And the Children Shall Lead: Gender Diversity and Performance in Venture Capital

(with Paul Gompers) Journal of Financial Economics, 2021.

[show/hide abstract]

With an overall lack of gender and ethnic diversity in the innovation sector, we ask the next question: Does increased diversity lead to better firm performances? In this paper, we answer this question using a unique dataset of the gender of venture capital partners’ children. First, we find strong evidence that parenting more daughters leads to an increased propensity to hire female partners by venture capital firms. Second, using an instrumental variable set-up, we show that improved gender diversity, induced by parenting more daughters, improves deal and fund performances. These effects concentrate overwhelmingly on the daughters of senior partners than junior partners. Taken together, our findings have profound implications on how the capital markets could function better with improved diversity.

A Set-Valued Approach to Utility Maximization under Transaction Cost

(with Andreas H. Hamel), Decision in Economics and Finance 2017, (1-19).

[show/hide abstract]

A set optimization approach to multi-utility maximization is presented, and duality results are obtained for discrete market models with proportional transaction costs. The novel approach allows us to obtain results for non-complete preferences, where the derived formulae closely resemble but generalize the scalar case.

What Does Banning Short-Term Rentals Really Accomplish?

(with Chiara Farronato and Andrey Fradkin) Harvard Business Review, 2024.

[show/hide abstract]

Concerns that short-term rentals fueled by platforms like Airbnb have caused long-term rents to rise in major cities has caused some governments to place limits, including bans, on them. But research of New York City found that short-term rentals are not the biggest contributor to high rents, especially when it comes to the most vulnerable segments of a city’s residents. Given that short-term rentals have benefits, bans are a poor solution.

Diversity in Venture Capital

(with Paul Gompers, Kanyuan Huang, and William Levinson) The Palgrave Encyclopedia of Private Equity, 2023.

[show/hide abstract]

This review article first highlights some key statistics on diversity in venture capital. In particular, it establishes that the fraction of women in VC and entrepreneurship has remained quite low throughout the past three decades. Even as of 2019, fewer than 15% of new entrants into venture capital and entrepreneurship were women. In addition, it also documents that the racial/ethnic composition of this population is highly skewed toward Whites and Asian Americans, with only about 10% of them being under-represented minorities. This article also highlights that the performance implications of diversity in venture capital are likely context-dependent. Diversity improvements due to reduced structural frictions are likely performance-enhancing. In contrast, diversity achieved at the expense of sacrificing team-specific match quality degrades entrepreneurial performance. Such findings also suggest that policy interventions on the issue of diversity in venture capital, and in corporate settings at large, should be careful about how to design the methods through which diversity is achieved.