Working Papers

These papers are working drafts of research which often appear in final form in academic journals. The published versions may differ from the working versions provided here.

SSRN Research Paper Series

The Social Science Research Network’s Research Paper Series includes working papers produced by Stanford GSB the Rock Center.

You may search for authors and topics and download copies of the work there.

Academic Area
Centers & Initiatives
Results for

Estimating Wage Disparities Using Foundation Models

Keyon Vafa, Susan Athey, David Blei
September2024

One thread of empirical work in social science focuses on decomposing group differences in outcomes into unexplained components and components explained by observable factors. In this paper, we study gender wage decompositions, which require…

Interest Rate Risk in Banking

Arvind Krishnamurthy, Peter M. DeMarzo, Stefan Nagel
September2024

We develop a theoretical and empirical framework to estimate bank franchise value. In contrast to regulatory guidance and some existing models, we show that sticky deposits combined with low deposit rate betas do not imply a negative duration for…

Price Experimentation and Interference

Wassim Dhaouadi, Ramesh Johari, Orrie B. Page, Gabriel Weintraub
September2024

In this paper, we examine biases arising in A/B tests where firms modify a continuous parameter, such as price, to estimate the global treatment effect of a given performance metric, such as profit. These biases emerge in canonical experimental…

Service Quality in the Gig Economy: Empirical Evidence about Driving Quality at Uber

Susan Athey, Juan Camilo Castillo, Bharat Chandar
September2024

The rise of marketplaces for goods and services has led to changes in the mechanisms used to ensure high quality. We analyze this phenomenon in the Uber market, where the system of pre-screening that prevailed in the taxi industry has been…

Going Beyond Black-Box Models by Leveraging Behavioral Insights: an Intent-Based Recommendation Framework

Yuyan Wang, Cheenar Banerjee, Samer Chucri, Minmin Chen
August192024

Modern recommender systems, which rely on large-scale machine learning (ML) models to predict the next item a consumer will engage with, often lack generalizability and understanding of consumer behaviors due to their black-box nature. To address…

Assessing the Costs of Industrial Decarbonization

Gunther Glenk, Rebecca Meier, Stefan J. Reichelstein
August2024

Companies in various industries are under growing pressure to assess the costs of decarbonizing their operations. This paper develops a generic abatement cost concept to identify the cost-efficient combination of technological and operational…

Can U.S. Treasury Markets Add and Subtract?

Roberto Gómez-Cram, Howard Kung, Hanno Lustig
August2024

The CBO cost releases of legislative proposals contain valuable news about surpluses priced in by Treasury investors. Using daily event windows, we find that cost releases with large negative cash flows lowered Treasury valuations by more than 20…

Fair Market Valuation of Electric Vehicle Batteries in Second Life Applications

Amadeus Bach, Simona Onori, Stefan J. Reichelstein, Jihan Zhuang
August2024

The rapidly growing number of lithium-ion battery packs deployed in electric vehicles (EVs) entails enormous economic potential for used EV batteries to be redeployed in a second life application, e.g., for behind-the-meter stationary energy…

Government Debt in Mature Economies. Safe or Risky?

Roberto Gómez-Cram, Howard Kung, Hanno Lustig
August2024

Governments and central banks can protect either taxpayers or bondholders from government spending shocks. When they choose to insulate taxpayers, government bond yields need to increase in response to unfunded fiscal expansions as the government…

Present Bias in Politics and Self-Committing Treaties

Bård Harstad, Anke Kessler
August2024

We study how international agreements can take advantage of domestic time-inconsistency problems in the context of environmental policies. For example, policymakers will prefer future policies to be sustainable, but find it tempting to raise…

The Incentives to (Not) Debate in Low-Information Races

Katherine Casey, Rachel Glennerster
August2024

Why are there few debates in low-information elections where they have the greatest potential to inform vote choices? Consistent with weak incentives to reveal qualifications or make policy commitments, we find only a quarter of Parliamentary…

Do Mergers and Acquisitions Improve Efficiency? Evidence from Power Plants

Mert Demirer, Ömer Karaduman
July2024

Using rich data on hourly physical productivity and thousands of ownership changes from U.S. power plants, we study the effects of acquisitions on efficiency and underlying mechanisms. We find a 2% average increase in efficiency for acquired…

The Consequences of Limiting the Tax Deductibility of R&D

Mary Cowx, Rebecca Lester, Michelle Nessa
July2024

We study the tax payment and innovation consequences of limiting the tax deductibility of research and development (“R&D”) expenditures. Beginning in 2022, U.S. companies are required to capitalize and amortize R&D rather than immediately…

The Gas Trap: Outcompeting Coal vs. Renewables

Bård Harstad, Katinka Holtsmark
July2024

We analyze a fundamental dilemma and time-inconsistency problem facing a climate coalition producing natural gas. In the short term, it is tempting to export more to outcompete coal. When this policy is anticipated, however, investments in…

LABOR-LLM: Language-Based Occupational Representations with Large Language Models

Tianyu Du, Ayush Kanodia, Herman Brunborg, Keyon Vafa, Susan Athey
June2024

Many empirical studies of labor market questions rely on estimating relatively simple predictive models using small, carefully constructed longitudinal survey datasets based on hand-engineered features. Large Language Models (LLMs), trained on…

Data-driven Error Estimation: Upper Bounding Multiple Errors with No Technical Debt

Sanath Kumar Krishnamurthy, Susan Athey, Emma Brunskill
May2024

We formulate the problem of constructing multiple simultaneously valid confidence intervals (CIs) as estimating a high probability upper bound on the maximum error for a class/set of estimate-estimand-error tuples, and refer to this as the error…

Democracy Corrupted: Apex Corruption and the Erosion of Democratic Values

Eduardo Rivera, Enrique Seira, Saumitra Jha
May2024

Democratic values are eroding just as citizens perceive increasing corruption, with numerous cases implicating the highest-level politicians. Could perceived increases in apex corruption be weakening democracy? We first present event study…

Book Value Risk Management of Banks: Limited Hedging, HTM Accounting, and Rising Interest Rates

Joao Granja, Erica Xuewei Jiang, Gregor Matvos, Tomasz Piskorski, Amit Seru
April2024

In the face of rising interest rates in 2022, banks mitigated interest rate exposure of the accounting value of their assets but left the vast majority of their long-duration assets exposed to interest rate risk. Data from call reports and SEC…

Contract Design in Influencer Marketing

Reto Hofstetter, Andreas Lanz, Navdeep S. Sahni
April2024

In influencer marketing, the differing incentives of advertisers and influencers necessitate a delicate balance between control and creativity. Investigating this trade-off, we explore the impact of contractual constraints on advertiser outcomes…

Market Access and Retail Investment Performance

Ed deHaan, Andrew Glover
April2024

We examine the effects of stock market access, and in particular trading hours, on retail investment performance. Using discontinuities around time zone borders, we find that plausibly exogenous decreases in waking trading hours are associated…