Academic Commons

Theses Doctoral

Financial Intermediation, Heterogeneous Investors, and Asset Pricing

Cho, Jaehyun

This dissertation consists of three essays in financial intermediation, heterogeneous agents, and asset pricing. In the first essay, I extract mutual fund flows that respond to the active change in equity share of mutual funds and show that they have significant predictability of market return. These “market timing-sensitive (MT-sensitive) flows” have predictability of the overall market over the next two to twelve months, without evidence of reversal. This predictability holds even when controlling for other macroeconomic variables and market sentiment index. I report that mutual fund managers who enjoy MT-sensitive inflows outperform the managers with MT-sensitive outflows over the next quarter. Also, I show that investors whose mutual fund investments mimic MT-sensitive flows have market timing ability, and outperform investors with mutual fund investments in the opposite direction to MT-sensitive flows.
In the second essay, I analyze mutual fund investors' responses to changes in funds' allocations to emerging markets. I show that such flows predict positive abnormal returns in emerging markets at quarterly and annual horizons. When there is one standard deviation shock to the EMT-sensitive flows, a six-month equal-weighted emerging market return is expected to be 3.58% in excess of risk-free rate in the US, and 1.69% in excess of US stock market excess return. This predictability holds even when controlling for other macroeconomic variables. The evidence suggests fund investors collectively possess valuable information about emerging markets.
The third essay proposes a general equilibrium model with bounded rationality that explains both endogenous learning and price. If agents are bounded rational, in that they do not have complete processing capacity as assumed in rational expectations models, there is a role for endogenous allocation of resources to learning about the economy. Investors trade off learning about different elements, such as terminal dividend (asset fundamental) vs. market structure (aggregate demand schedules). I found that investors prefer to learn what others do not learn, and this explains why there is specialization in the investment. Investors tend to be fundamentalists when market is uncertain, but learning also depends on capacity, ratio of sophisticated investors, risk aversion, etc. I analyze the trade-off between these information sources, and the implications for price efficiency, risk, and return, in a general equilibrium.



  • thumnail for Cho_columbia_0054D_13027.pdf Cho_columbia_0054D_13027.pdf binary/octet-stream 1.16 MB Download File

More About This Work

Academic Units
Thesis Advisors
Tetlock, Paul C.
Ph.D., Columbia University
Published Here
October 15, 2015