Sep 21, 2018
Abstract: Firms hold money market instruments issued by banks to finance non-pledgeable intangible investments. This liquidity demand pushes down the interest rate in money markets, feeding banks cheap leverage, so banks bid up asset prices in booms as their balance sheets expand. Higher asset prices induce firms to invest more in intangibles and hold more liquidity, which in turn leads to an even lower interest rate, faster growing banks, and higher asset prices. This paper models corporate savings gluts that arise endogenously from the interaction between firms and banks in asset and money markets. It offers a coherent account of several seemingly disconnected phenomena in the run-up to the Great Recession, and reveals how endogenous risk accumulates in asset prices during booms and materializes into stagnant crises. Government debt as an alternative source of liquidity stimulates intangible investments and economic growth, but it cannot reduce asset price volatility or the frequency of banking crises.
Presentations: WFA (Western Finance Association), European Winter Finance Summit (Best Paper Prize), EFA (European Finance Association),
CEPR ESSFM Gerzensee,
SED (Society for Economic Dynamics), Econometric Society (North America), 7th HKUST Macro Workshop, CMU/OSU/Pittsburgh/Penn State conference; earlier version at:
Columbia University, Federal Reserve Bank of New York, Finance Theory
Group (parallel session), LBS Trans-Atlantic Doctoral Conference
We provide a dynamic asset-pricing model of cryptocurrencies/tokens on (blockchain-based) platforms, and highlight their roles on endogenous user adoption. Tokens facilitate transactions on decentralized networks, and their trading creates an inter-temporal complementarity among users, generating a feedback loop between token valuation and platform adoption. Consequently, tokens capitalize future growth of promising platforms, accelerate adoption, reduce user-base volatility, and can improve welfare. Equilibrium token price increases non-linearly in platform productivity, user heterogeneity, and endogenous network size. The model also produces explosive growth of user base after an initial period of dormant adoption, accompanied by a run-up of token price volatility. We further discuss how our framework can be used to discuss cryptocurrency supply, platform competition, and pricing assets under network externality.
2018 AAM-CAMRI-CFA Institute Prize in Asset Management, CEPR ESSFM Gerzensee,
Finance Theory Group (FTG),
Fall Finance Conference at UT Dallas, Chicago Booth, CityU HK Finance Conf, 3rd Rome Junior Finance Conf, Emerging Trends in Entrepreneurial Finance (Best Paper Award),
2nd Private Markets Research Conference,
JOIM Conf on FinTech,
Finance UC International Conference, Norwegian School of Economics, LeBow/GIC/FRB Conf on Cryptocurrency, Shanghai Forum
We estimate the liquidity multiplier and individual banks’
contribution to systemic risk in an interbank network using a structural model. Banks borrow liquidity
from neighbors and update their valuation based on neighbors' actions. When
the former (latter) motive dominates, the equilibrium exhibits strategic substitution (complementarity) of liquidity holdings, and a reduced (increased) liquidity
multiplier dampening (amplifying) shocks. Empirically, we find substantial and
procyclical network-generated risks driven mostly by changes of equilibrium
type rather than network topology. We identify the banks that generate most systemic risk and solve the
planner's problem, providing guidance to macro-prudential policies.
NBER SI 2018, Macro Finance Society, Bank of England, Cass Business School, Duisenberg School of Finance, Koc University, London School of Economics, Stockholm School of Economics, OSU Fisher
WFA , 5th Joint Bank of Canada / Payments Canada Workshop on Payment Systems
Abstract: The ratio of long- to short-term dividend prices, "price ratio" (pr), predicts one-year stock market return with an out-of-sample R2 of 19%. It subsumes the predictive power of price-to-dividend ratio (pd). The residual from regressing pd on pr predicts one-year dividend with an out-of-sample R2 of 30%. Our results hold outside the U.S. In an exponential-affine model, we show the key to understand these findings is the (lack of) persistence of expected dividend growth. We also characterize the risk of time-varying expected return: (1) the expected return is countercyclical; (2) the response of expected return (rather than expected dividend growth) accounts for the impact of monetary policy on stock price; (3) shocks to pr are priced in the cross-section, which serves as an ICAPM test of pr as an adequate proxy for the expected return.
Presentations: 16th Paris December Finance Meeting, Econometric Society (North America), Fall Finance Conference at UT Dallas, Northern Finance Association (NFA), Örebro Workshop on Predicting Asset Returns
with Chen Wang, Sep 21, 2018
Abstract: Big data analysis requires expertise and creates a division of labor and knowledge. We delegate tasks to professionals who possess better information, but delegation carries an intrinsic form of uncertainty - its outcome depends on professionals' knowledge that is unknown to us. This paper studies delegation uncertainty in financial markets. The theory reconciles the growth of asset management industry and its lack of convincing performance. It also produces asset pricing implications supported by our empirical analysis: (1) CAPM alpha arises because investors partially delegate and they hedge against delegation uncertainty; (2) the cross-section dispersion of alpha increases in uncertainty; (3) managers bet on alpha, but alpha is immune to the rise of managers' arbitrage capital because delegation hedging is stronger when investors delegate more.
Presentations: CEPR ESSFM Gerzensee, Geneva Workshop on Financial Stability in a New Era, U Zurich
Abstract: This paper provides the first evidence on the distributional effect of Fintech credit. Using comprehensive data from Alibaba, one of the largest e-commerce platforms and fintech lenders, we estimate the causal impact of platform credit on the size distribution of small businesses. We find that platform credit promotes firm selection: credit leads to stronger growth of market share for online merchants that are already larger and better rated by customers. Credit increases a merchant’s market share especially when product demand is expanding, suggesting that the distributional effect is due to heterogeneous investment opportunities. Contrary to conventional wisdom, credit does not affect product pricing. Lastly, we analyze the information set of the platform as a lender, and document a dynamic effect: merchants’ credit score assigned by the platform is correlated with their market share and customer ratings. A feedback loop arises between market status and platform credit that accelerates firm selection in this fast growing entrepreneurial space.
Presentations: 2018 Finance, Organizations, and Markets (FOM) Conference, Bank of Canada, FDIC Annual Bank Research Conference, Fed Board, NTU, NUS