Working Papers

  1. Beyond Credit Scores (2026)
    with Deniz Aydın and Janis Skrastins.
    Draft coming soon
    Abstract

    Algorithmic lending has led to a rise in personalized pricing reflecting information beyond the credit score. We study the welfare effects of pricing on increasingly granular information. We use a randomized pricing experiment and causal machine learning to estimate demand and cost curves and embed them in competitive equilibrium. Finer pricing need not improve welfare: relative to uniform pricing, credit-score pricing reduces total surplus, as prices increase in adversely selected low-score pools, pushing out safest borrowers. Personalization improves on the score by separating observables predicting risk levels from those predicting selection, but uniform pricing dominates, as lending gains remain modest.

  2. Pre-Trends Ahead of Monetary Policy Surprises: Recovering the Response of Corporate Investment (2026)
    with Antoine Hubert de Fraisse, Peter Maxted and Kunal Sangani.
    Previously circulated as “Monetary Policy and Corporate Investment: Sufficient Statistics with Heterogeneous Firms.”
    PDF Slides
    Abstract

    How should high-frequency monetary policy surprises be used when the outcome of interest is only observed at a lower frequency? We first document a business-cycle endogeneity problem that arises when monetary policy surprises are aggregated to monthly or quarterly frequencies—surprise policy easings (resp. tightenings) tend to occur amid broader easing (resp. tightening) cycles. These low-frequency trends in pre-event Treasury yields are not reliably removed by standard controls without also weakening the relevance of the remaining policy variation. Motivated by this evidence, we next develop an alternative Q-theory-based approach for estimating the corporate investment response to monetary policy. Our approach combines high-frequency identification of firms’ valuation responses to policy surprises with low-frequency firm-level investment-Q sensitivities. Our baseline estimate is: a one-unit monetary tightening, normalized to a one percentage point increase in the one-year Treasury yield, reduces the annual tangible investment of non-financial, non-utility public firms by 0.38% of GDP, with 24% of this effect driven by a novel firm-heterogeneity channel.

  3. The Welfare Benefits of Pay-As-You-Go Financing (2025)
    with Paul Gertler, Brett Green and Renping Li.
    SSRN R&R · Review of Economic Studies
    Abstract

    The rapid expansion of digital financial products in low- and middle-income countries has increased access to credit but raises important questions about their welfare effects. Pay-as-you-go (PAYGo) financing is one such product, relying on lockout technology that allows lenders to remotely disable the collateral’s flow benefits when borrowers miss payments. This paper quantifies the welfare effects of PAYGo financing. We build a dynamic structural model of consumer behavior and estimate it using a large-scale, multi-arm pricing experiment conducted by a fintech lender that offers PAYGo financing for smartphones. We find that the welfare gains from access to PAYGo financing are equivalent to a 3.4% increase in income while remaining highly profitable for the lender. The welfare gains are larger for low-risk borrowers and those in the middle of the income distribution. Under plausible assumptions, PAYGo dominates traditional secured loans for all but the riskiest consumers. We explore contract design and show that variations of PAYGo contracts can deliver further welfare improvements.

  4. A Quantitative Analysis of Distortions in Managerial Forecasts (2024)
    with Yueran Ma, Tiziano Ropele and David Thesmar.
    SSRN R&R · Journal of Political Economy
    Abstract

    This paper quantifies the economic costs of distortions in managerial forecasts. We exploit a long panel of managerial forecast errors and investment decisions of a representative sample of Italian firms. Our analysis relies on three key features of the data: (1) on average, sales forecasts are unconditionally unbiased and the standard deviation of sales forecast errors is 17.0% (2) managerial forecasts are conditionally biased, i.e., errors are autocorrelated with an AR(1) coefficient of 0.331, suggesting underreaction (3) while investment responds significantly to sales-forecast, the elasticity of firms’ capital stock to sales forecast is significantly smaller than one, at 0.485. We quantitatively interpret these findings through the lens of standard Q-model with distorted expectations. Distortions in expectation reduce firm value by 1.5%, and lower aggregate productivity by 0.76%. These findings remain unchanged when we introduce additional frictions (e.g., fixed adjustment costs, financial constraints) and behavioral biases (e.g., optimism).

  5. Robustness Checks in Structural Analysis (2026)
    with Sylvain Catherine, Mehran Ebrahimian, Mohammad Fereydounian and David Thesmar.
    SSRN R&R · Journal of Finance
    Abstract

    This paper introduces a computationally efficient methodology for estimating variants of structural models. Our approach approximates the relationship between moments and parameters, offering a low-cost alternative to traditional estimation methods. We establish general convergence conditions, and show they are widely applicable, even for models with discontinuities. We also provide convergence rate bounds for Kernel and Neural Net approximations, with the latter demonstrating superior performance in higher dimensions. We apply this methodology to two standard structural models in financial economics: (1) dynamic corporate finance and (2) life-cycle portfolio choice. We demonstrate the reliability of our approach through simulations and then use it to explore identification, robustness to sample splits, moment selection, and model misspecification. These explorations are computationally infeasible with standard techniques, but become trivial with our methodology.

Retired Working Papers

  1. The Risk-Shifting Hypothesis: Evidence from Subprime Originations (2015)
    with Augustin Landier and David Thesmar.
    SSRN
    Abstract

    Using internal loan level data, we provide evidence consistent with risk-shifting in the lending behavior of a large subprime mortgage originator – New Century Financial Corporation – starting in 2004. This change follows the monetary policy tightening implemented by the Fed in the spring of 2004, which resulted in an adverse shock to the large portfolio of loans New Century was holding for investment. New Century reacted to this shock by massively resorting to deferred amortization loan contracts interest-only loans. We show that these loans were not only riskier, but also that their returns were by design more sensitive to real estate prices than standard contracts. New Century was thus financing projects with a high beta on its own survival, as predicted by a standard model of portfolio selection in financial distress. Our findings contribute to better characterizing the type of risk taken by financially distressed firms. They also shed new light on the relationship between monetary policy and risk taking by financial institutions.