Research Papers

Research Topics:

Subjective Beliefs and Asset Prices

Figure from the paper: CAPM-alphas of anomalies before (red)/after controlling for the CDR factor.

Asset Prices When Investors Ignore Discount Rate Dynamics

(Job Market Paper)

Why do cross-sectional asset pricing anomalies arise? This paper proposes and tests a unified hypothesis: some investors falsely ignore the dynamics of discount rates. In theory, these investors develop a positive bias in their expectation of future returns because they fail to understand a dynamic discount rate will offset part of the impact cash flow news has on stock prices. The biases differ across stocks, because investors incur these biases each period, and stocks differ in the timing (duration) and uncertainty (convexity) of their future cash flows, which are related to stocks' expected cash flow growth and volatility. Since these fundamental characteristics are dominant drivers of cross-sectional variation in asset payoffs, arbitrageurs are exposed to factor risk when taking the other side of these trades, and therefore mispricing persists. To test this hypothesis, I measure mispricing at the firm level and find that overvalued stocks exhibit large and long-lasting negative abnormal returns, even among stocks in the S&P 500. A tradable mispricing factor explains the CAPM alphas of 12 leading anomalies (9 out of the 11 in Stambaugh and Yuan (2017)) including investment, profitability, beta, idiosyncratic volatility and cash flow duration. These characteristics predict firms' future cash flow growth and/or volatility as well as errors of analysts' return forecasts, consistent with the hypothesis.

Figure from the paper: Aggregate return expectations of sell-side analysts (Blue), CFOs (black) and households (red).

Subjective Return Expectations

(2nd Job Market Paper)

Facing uncertainty and imperfect predictors of future returns, subjective return expectations could appear extrapolative or contrarian to past returns, depending on whether one believes a). the market is driven by future fundamentals or discount rate; b). good fundamental news leads to good future returns. In reality, Wall Street (sell-side analysts, buy-side analysts and professional forecasters) hold contrarian return beliefs while Main Street (CFOs, retail investors and households) extrapolate. Estimating the expectation formation model, I find that while Wall Street believe positive cash flow news would lead to lower future returns, Main Street believes the opposite. However, all of them believe fundamentals drive asset prices.

Link to paper

Figure from the paper: Cumulative CAPM-alpha after a shock to the slope of yield curve after 10 years, high-vol (black) vs. low-vol stocks

Macro Drivers of Low-Volatility Stock Returns

with Milan Vidojevic

Using a novel methodology, we provide evidence that low-volatility stocks persistently underperform following periods of increasing bond yields, inflation, and aggregate market variance. Long-term investors who prefer to hedge against these risks hold rational beliefs to hedge against them and therefore require a higher premium to hold low-volatility stocks.

Link to paper

Figure from the paper: entrepreneurs' own revenue growth projections, broken down by companies' age.

Subjective Growth Expectations of Entrepreneurs

with Adrien Alvero (work-in-progress)

Using a comprehensive and proprietary data set of 80,000 entrepreneurs' own revenue growth forecasts, we show that entrepreneurs have overly optimistic projections for future revenues, and consistently revise these projections downwards (constant disappointment). We relate these findings to entrepreneurs’ characteristics and find that entrepreneurs with high previous salary and more time invested in the start-up are less optimistic about future growth of the company. Finally, we use this dataset to confirm the previous findings that team strength is one of the best predictors of external funding.

Insurance Demand and Bond Markets

Figure from the paper: insurers' portfolio weights for corporate bonds (left) and treasury (right) for high and low (green) duration gap group.

The Constraints on Portfolio Rebalancing: When

Unconventional Monetary Policy Meets Insurance Capital


with Adrien Alvero

We examine the role of life insurers during episodes of Quantitative Easing (QE). To that end, we develop a new method to back out the duration gaps of life insurance companies based on their holdings and publicly available balance sheets information. We show that static capital regulation in the insurance sector actually could render the QE less effective: Those who face higher duration gaps did not rebalance more towards corporate credits, contrary to what the portfolio rebalancing channel predicts

Link to paper

Figure from the paper: Avg. Option Adj.Spread Baclays Long−term Corproate Bond Indices (AAA, AA, A) (right, red) vs. Avg. Previous Quarter Share of Bond Held by Life Insurers (par value held/ amount outstanding), R2 = 29.6%

Preferred Habitat, Aggregate Insurance Demand and Credit Spreads


I document that life insurers' aggregate demand for investment grade corporate bonds exhibits pro-cyclical pattern and strongly predicts future aggregate credit spreads. I develop a preferred-habitat model for corporate bond market featuring liability-driven insurers who face interest rate risk and capital regulation. Data are consistent with the predictions of the model: a). low duration bonds have higher Sharpe ratios than high duration bonds; b). bonds preferred by insurers exhibits lower average returns even after controlling for other characteristics; c). changes in life insurers' aggregate demand drives large fraction of time-series variations of firm-level credit spreads previously unexplained by structural credit risk models.