Knowledge that Transforms

We propose direct multiple time series models for predicting high dimensional vectors of observable realized global minimum variance portfolio (GMVP) weights computed based on high-frequency intraday returns. We apply Lasso regression techniques, develop a class of multiple AR(FI)MA models for realized GMVP weights, suggest suitable model restrictions, propose M-type estimators and derive the statistical properties of these estimators. In the empirical analysis for portfolios of 225 stocks from the S&P 500 we find that our direct models effectively minimize either statistical or economic forecasting losses both in- and out-of-sample as compared to relevant alternative approaches.

We study the joint effects of uncertainty and investment spikes on corporate leverage adjustments. In our baseline finding we show that, while uncertainty has a nontrivial adjustment speed effect for overlevered firms, underlevered firms are not greatly impacted. In our second baseline finding, we document asymmetric effects of investment spikes on the adjustment speed between overlevered and underlevered firms. Specifically, during the investment-spike period, underlevered firms adjust their leverage very promptly, whereas overlevered firms tend to adjust their leverage much more slowly. Bringing these two baseline analyses together in a combined analysis, conditional on the level of uncertainty, the investment regime has asymmetric effects between overlevered and underlevered firms. Specifically, when uncertainty is fixed, regardless of whether uncertainty is high or low, overlevered firms facing investment spikes tend to adjust their leverage slower than their non-spike counterparts, but underlevered firms facing investment spikes tend to adjust their leverage substantially faster than their non-spike counterparts. Further, we find that elevated uncertainty uniformly increases the speed of adjustment for overlevered firms without investment spikes, but the influence for underlevered firms without investment spikes is much less evident.

This paper examines the optimal incentive scheme in motivating people to innovate under ambiguity. When an innovation’s prospects are ambiguous, the use of extrinsic, high-powered incentives can lead the agent’s beliefs about the project’s outcome to deviate from that of the principal’s, which consequently deters innovation. The deterrent effect, however, is alleviated for firms in which agents have strong intrinsic incentives to adhere to firms’ goals and missions. In equilibrium, extrinsic and intrinsic incentives are complementary, and firms that face greater uncertainty invest more in fostering intrinsic incentives. Hence, firms that pursue more exploratory and radical innovation invest more in creating corporate identity and culture.

This paper examines the impact of product failures on the pricing of bank loans using hand-collected data on product recalls. We find that banks tend to charge higher loan prices for firms involved in product recalls. Uncertainty as to a recall's ultimate impact on a firm's credit risk conditions banks’ loan-pricing reaction, as reflected in a firm's default risk, information asymmetry and governance deficiency, and by the damage to its reputation, arising from the recall. Further analysis reveals that the impact of product recalls on the cost of debt is stronger in firms that rely more extensively on bank financing, firms with more severe recalls, and those adopting passive recall strategies. However, medical device firms, for which product recalls are often considered a normal part of doing business, do not experience a rise in their bank financing costs following a recall. Finally, we find that recall firms experience a deterioration in their financial performance and a rise in product lawsuits post recall. Overall, our findings shed new light on the economic consequences of product failures through the lens of creditors.

We document a positive effect of product market competition (PMC) on venture capital (VC) staging. Employing large tariff rate reductions as an exogenous shock to PMC, we find that large tariff reductions lead to a greater likelihood of staged financing and a larger number of financing rounds. Cross-sectional analyses reveal that the effect is stronger for entrepreneurial firms that are younger, operate in high-tech and manufacturing industries, or receive investments from less reputable and inexperienced VCs. Our findings are consistent with the notion that by mitigating business uncertainty and survival risk, VC staging acts as a complement to PMC for enhancing entrepreneurial firms’ success.

Using a SVAR model with sign and zero restrictions, we propose a novel scheme to identify expectation, credit supply and mortgage rate shocks with the aim of exploring their role in the 2000s housing boom-bust cycles. Overall, credit supply and mortgage rate shocks are two major drivers of housing fluctuations instead of expectation shock. However, the relative importance of the three shocks varies considerably over different episodes. Specifically, gradual rise in house prices during 1999-2002 is mainly due to appreciation expectations. Such conclusion is reversed over boom-bust cycles. Compared with less than 10% contribution of expectation shock, credit supply and mortgage rate shocks become the two most important drivers of housing boom, with 20% and 24.5% contribution, respectively. In the bust, 20.2% and 23.1% of decline in house prices are associated with credit supply and mortgage rate shocks, respectively, while only 7.4% can be attributed to expectation shock.

This paper investigates how balance sheet opacity affects banks’ risk-taking behavior. We measure bank balance sheet opacity according to two metrics: the ratio of available-for-sale (AFS) securities and the ratio of off-balance sheet items. We show that balance sheet opacity is positively correlated with realized bank risk. Specifically, banks with more AFS securities have lower realized risk, while banks with more off-balance sheet items have higher realized risk. The correlation between opacity and risk depends on both macroeconomic variables and bank characteristics. The positive relationship between bank opacity and bank risk is weaker for better capitalized banks and banks that are subject to more market discipline. The relationship is also weaker during periods of favorable market conditions. Motivated by this analysis, we then investigate how regulation affects bank opacity. We show that higher capital requirements reduce bank opacity and bank risk through a portfolio rebalancing channel.

This study first investigates why only some banks use the internal models (IMs) introduced by Basel II that lead to more risk-sensitive capital ratios than standardized approaches (SA). I predict that banks opt for an IM if it allows economizing on capital requirements, given their underlying risk. I find support for this hypothesis by analyzing Mexican banks’ adoption of an IM for deposit maturity, a key input to measure interest rate risk, between 2006 and 2016. Secondly, I examine whether banks increase the duration of assets after IM adoption—high-maturity assets can be offset with deposits, leading to a lower risk exposure and additional capital savings. For the average maturity of total assets, I find no support for this conjecture. However, when a flattening of the yield curve spurs firms’ demand for high-maturity debt, micro data reveal that banks using the IM increase the duration of commercial loans more than those using the SA. These findings have broad implications for the design of internal risk models and of capital regulation.

Applying the difference-in-differences method, we show that high-speed rail (HSR) construction improves investment efficiency in China. The effect is more pronounced for companies with low information transparency and low media coverage, suggesting that HSR can mitigate information asymmetry while reducing external regulatory costs. Our analysis also shows that the effect is more powerful for non-state-owned enterprises, high-growth firms, and those companies located in core cities. Overall, our findings suggest that HSR can improve investment efficiency by shortening the travel time between firms and their investors.

We provide market-based evidence that a capital loss that is realized in the beginning of the year is less valuable than a loss that is taken at the end of the year. A simple binomial tree model that captures the resolution of tax rate uncertainty closely mimics observed market prices. Tax rate uncertainty arises from not knowing until the end of the calendar year whether the investor will have realized sufficient capital gains to fully benefit from losses harvested early in the year. We conclude that tax rate uncertainty influences investor behavior.