I am an assistant professor of finance at Arizona State University within the W.P. Carey School of Business. I joined ASU after completing my PhD in Financial Economics at MIT Sloan. My research focuses on the interaction of financial markets, labor markets, and innovation. Outside of academia, I am an avid reader and writer of poetry and literary fiction.
Investment composition matters for asset pricing. I develop a production-based model where firms invest in both tangible and intangible capital. The model predicts that, conditional on intangible-adjusted book-to-market, expected returns increase in investment composition---the difference between intangible and tangible investment rates---through differential exposure to displacement risk. Empirically, portfolios sorted on investment composition generate significant alphas, with annual premia of 4--5\% unconditionally and 9--10\% when conditioned on valuation. I validate the mechanism and apply the framework to explain the decline of the value premium as a compositional shift in investment.
We examine the role of process innovation in shaping firm investment and compensation. Empirically, investment, executive pay, and firm valuation ratios all rise with process intensity --- the share of innovation that is process-focused. To account for these patterns, we develop a dynamic agency model in which process innovation enhances firm-specific capital efficiency but exacerbates the hold-up problem for managers and skilled labor. The model not only explains the observed links between process intensity, investment, and compensation, but also predicts a convex relationship between compensation and process intensity. These predictions are supported by the data.
A factor's risk premium can be point-identified even when the vector of risk premia is not. We derive the necessary and sufficient condition---the kernel-orthogonality (KO) condition---and show it is equivalent to the existence of a population mimicking portfolio. When KO fails, standard estimators converge to a random variable rather than a constant, and $t$-tests spuriously reject zero risk premia. We develop a test to determine \emph{which} individual factor risk premia are identified, not just whether the entire model is identified. Applying our methodology to well-known models, we find that certain factors (e.g., consumption growth, intermediary leverage) fail KO while others (e.g., the market) pass.
Standard estimators of factor risk premia answer two questions jointly: Is factor $k$ priced? And does factor $k$ help explain the cross-section given other factors? I develop an estimator that separates them. The stochastic discount factor is recovered from test asset returns by minimizing Kullback--Leibler divergence subject to no-arbitrage, and each factor's premium is computed through $\lambda_k = -R_f \mathbb{E}[m f_k]$. The estimate for any factor depends only on that factor and the recovered SDF—not on which other factors the researcher specifies. Simulations calibrated to Fama--French factors show 70--90\% smaller errors than two-pass methods when priced factors are omitted; in a nonlinear Lucas economy, the estimator succeeds where Fama--MacBeth produces RMSEs 100x larger. Empirically, exponential tilting reduces holdout pricing errors by 44\% and produces stable estimates for macro factors where conventional methods yield contradictory results.