Countercyclical Income Risk and Portfolio Choices: Evidence from Sweden
with Sylvain Catherine and Paolo Sodini
Journal of Finance, 2024
(Journal of Finance Dimensional Fund Advisors Prize Distinguished Paper awarded)

Using Swedish administrative panel data, we document that workers facing higher left-tail income risk when equity markets perform poorly are less likely to participate in the stock market and, conditional on participation, have lower equity shares. In line with theory, the relationship between cyclical skewness and stock holdings is proportional to the share of human capital in a worker’s total wealth and vanishes as workers get closer to retirement. Cyclical skewness also predicts portfolio differences within pairs of identical twins. Our findings show that households hedge against correlated tail risks, an important mechanism in asset pricing and portfolio choice models.

ETF Dividend Cycles Predict Money Market Fund Flows and Treasury Yield Changes
(previously circulated as “ETF Dividend Cycles”)
with Pekka Honkanen and Tong Zhou
Critical Finance Review, 2025

Exchange-traded funds (ETFs) collect approximately 7% of all U.S. corporate dividends, which they are required to redistribute to investors. How do the funds manage these dividend flows, and does such management have spillover effects on other financial markets? In this paper, we document a new stylized fact of the “ETF dividend cycle:” ETFs gradually invest in money market funds (MMFs) when they accumulate dividend receipts and periodically withdraw from MMFs when they distribute dividends. This cycle creates periodic liquidity shocks to MMFs and, consequently, to the Treasury markets as the affected MMFs liquidate some of their short-term Treasury holdings to satisfy ETFs’ dividend-driven withdrawals. As a result, ETF dividend cycles can explain flows to MMFs and fluctuations in Treasury yields.

Reluctant Entrepreneurs: Evidence from China’s SOE Reform
with Hanming Fang, Ming Li, and Zenan Wu
Management Science, Revise and Resubmit

We study the impact of state-owned enterprises (SOEs) on the quality of entrepreneurship in China. Using long series of firm registration and performance data, we document that the massive SOE downsizing in the late 1990s significantly improved the quality of entrepreneurship. Compared with entrepreneurs in other time periods, firms founded by the reluctant entrepreneurs induced by the SOE layoffs have better performances. To explain these results, we present a simple model of occupational choices where high-skilled individuals obtain a higher value than low-skilled individuals from the benefits offered by SOE jobs, leading them to select into the SOE sector in the pre SOE reform era. When the SOE sector was downsized, some high-skilled SOE employees were reluctantly unleashed into entrepreneurship. We also provide corroborating evidence for other implications of the model.

The Value of Payment Innovation to Households: Evidence from a FinTech Platform
with Junli Zhao

Many recent payment innovations feature embedding payment functionality into non-deposit assets. Leveraging a unique administrative dataset on payment-integrated money market fund products from Alipay, a digital payment provider in China, this paper examines the value of such innovation to households and the underlying mechanism. Using a structural demand model that accounts for both fund-level characteristics and investor-level heterogeneity, we estimate that households are, on average, willing to pay approximately 40 basis points annually to access integrated payment services. Valuation is higher among wealthy individuals and young males. We show that this heterogeneity is primarily driven by differences in consumption risk, suggesting that the primary source of value lies in the liquidity option value, rather than direct usage convenience. 

A Multifrequency Jump-Decay Approach to Volatility Forecasting
with Laurent Calvet and Jiawen Xu

Empirical evidence shows that volatility jumps upwards due to shocks to uncertainty and then gradually decreases as the uncertainty resolves, a pattern often called the shot-noise effect. In this paper, based on the MSM model in Calvet and Fisher (2004b), we develop a shot-noise volatility model (SN-MSM) that parsimoniously captures the jump-decay patterns and multifrequency properties of volatility. SN-MSM is obtained by introducing an asymmetric component into the transition matrix of a Markov-switching multifractal (MSM). SN-MSM generates extreme jumps and volatility decay patterns while preserving the fat tails and volatility clustering of MSM. We derive the closed-form likelihood function of our process and verify that maximum likelihood estimation produces accurate parameter estimates in Monte Carlo simulation. In an empirical application to five major currency series, the SN-MSM model exhibits superior performance than popular volatility models both in- and out-of-sample.

Seeking Skewness

Household Decisions under Pollution-Induced Health Risk
with Hanming Fang and Qian Li

Air pollution has a large detrimental effect on human health. Pollution-induced health risk incentivizes households to adjust their life-cycle consumption-saving decision. Using the Chinese Household Finance Survey data, we provide causal evidence that households facing high air pollution consume less but increase the proportion of safe asset. They also invest more in health and exhibit higher demand for insurance. We quantify the welfare cost of the air pollution-related health hazard using a stochastic dynamic overlapping generation model with endogenous health accumulation in a partial equilibrium framework. Our numerical results show that facing air pollution-induced multi-dimensional health hazard, consumption declines due to the crowding out effect of higher health investment and the fall in labor productivity. The overall welfare loss is around 20% in terms of consumption equivalence, over 85% of which is attributed to the more persistent health shock at polluted region.