Knowledge that Transforms

We study how capital account liberalization affects technological innovation. We provide robust evidence that industries more dependent on external finance have disproportionately higher innovation performance in economies with a more liberalized capital account. Among the components of capital account liberalization, although both equity market liberalization and outward FDI by domestic firms have sizable effects on innovation, they affect it differently. While equity market liberalization helps alleviate financial constraints by facilitating access to external finance, outward FDI by domestic firms promotes innovative activities by increasing internal finance from foreign operations. Further analysis indicates that the innovation-enhancing effects of capital account liberalization are limited mainly to countries with relatively well-developed financial systems and strong institutional quality, even in periods of financial crisis.

Using confidential loan officer survey data on lending standards and internal risk ratings on loans, we document an effect of large-scale asset purchase programs (LSAPs) on lending standards and risk-taking. We exploit cross-sectional variation in banks’ holdings of mortgage-backed securities to show that the first and third round of quantitative easing (QE1 and QE3) significantly lowered lending standards and increased loan risk characteristics. The magnitude of the effects is about the same in QE1 and QE3, and is comparable to the effect of a one percentage point decrease in the Fed funds target rate.

This paper establishes a sizeable negative effect of poor mental health on individuals’ net worth. In a representative panel of U.S. households, we find that a one standard deviation (or four unit) increase in Kessler’s K6 psychological distress level decreases net worth by 13.2 percent and increases by 5 percent the baseline risk of being in deficit net worth, where levels of debt outstrip the value of assets. Survival analyses further show that psychological distress accelerates the entry into and prolongs the stay in deficit net worth states, as well as increasing the probability of re-entry into deficit. Using a Blinder-Oaxaca decomposition, we find that differences in level of savings, medical debt and labor income predominantly explain the lower net worth and higher likelihood of deficit net worth of individuals with high psychological distress. Our findings highlight the significant longer-term implications of mental health on the net worth of individuals.

This paper examines the effects of bond liquidity on firms’ investments. We postulate that bond liquidity increases firms’ investment opportunities by reducing the cost of capital and improving access to financing. Using the variation in liquidity generated by several – both positive and negative – exogenous shocks, we find that firms respond to positive (negative) shocks by expanding (contracting) capital expenditures and acquisition activity. Further, by enhancing access to funding, bond liquidity facilitates acquisition financing and reduces the likelihood of investment delays. We also find a positive impact of bond liquidity on market valuations and profitability, suggesting that these investments are value-increasing.

We show that a bank's knowledge of an industry developed through its loan portfolio facilitates the bank's credit provision to other firms in that industry. This effect works beyond the bank's private information about the focal firm and is consistent with a cross information production where experience with other firms from a similar background reduces information asymmetry on the firm concerned. To tackle endogeneity, we develop an instrument for a bank's expertise in an industry based on historical, natural, and regulatory conditions. We provide further evidence using the 2007 housing market crash as a laboratory. We find that banks hit by the shock rebalance loan allocations to buffer borrowers in their expertise industries from a credit crunch. The effect of industry expertise is more pronounced for opaque firms and firms facing foreign competition pressure. Our findings suggest a spillover effect or economies of scale in banks’ information production. It helps explain the cost efficiency of financial intermediaries relative to direct lending and why, beyond relationship considerations, firms may prefer some banks over others.

This paper explores ways to improve the existing systemic risk measures by incorporating machine learning algorithms into the measurement. We aim to overcome the shortcomings of existing methods that rely on restricted modeling and are difficult to tap into various data resources. To this end, this paper unifies a dynamic quantification framework for systemic risk and links it to a two-step supervised learning problem, which allows for hierarchical structure of the systemic event and the return dependence. We leverage the generalization and predictive powers of machine learning to statistically model the tail events and the co-movements of the equity returns during the shocks to the macro-economy. Our results show that most machine learning algorithms enhance the systemic risk measure’s predictive power. Numerous comparative and sensitivity backtesting studies for United States and Hong Kong markets are conducted, from which we recommend the best machine learning algorithm for systemic risk measurement.

Credit ratings are fundamental in assessing the credit risk of a security or debtor. The failure of the Collateralized Debt Obligation (CDO) ratings during the financial crisis of 2007-2008 and the massive undervaluation of corporate risk leading up to the crisis resulted in a review of rating approaches. Yet the fundamental metric that guides the construction of credit ratings has not changed. We study the inadequacies of the old metric in simple models of investment and in structured finance portfolio optimization tasks, and we propose a new methodology based on a buffered probability of exceedance. The new approach offers a conservative risk assessment, with substantial conceptual and computational benefits. We illustrate the new approach using several examples and report the results of a structuring step-up CDO case study, with details available in an online Supplement.

We test for return co-movements among international commercial real estate markets. Our spatial econometric model estimates the market exposure to the performance of a reference portfolio. This benchmark portfolio contains all markets with a higher level of transparency, which reveals valuable information about the pricing mechanism. Empirical evidence suggests that these indirect effects transmit from more transparent to less transparent markets. We then study the predictive power of different familiarity-based channels to overcome entry barriers by predicting returns in less transparent property markets. The evaluation of the prediction performance indicates that observed price signals in highly transparent markets are attributed to less transparent markets, which we interpret as informational herding.

We examine whether media coverage affects credit rating change decisions by analyzing 732,426 newspaper items published by top U.S. media outlets on S&P 1500 firms. Our results show that negative media coverage has a strong association with credit rating change events, but positive media coverage does not. We find support for two channels that confirm this finding: the media's fundamental information content and the media's reputational pressure. Credit rating agencies appear to consider the fundamental information embedded in negative media coverage and recognize negative market sentiment when making rating change decisions.

According to the gradient allocation principle based on a positively homogeneous and subadditive risk measure, the capital allocated to a sub-portfolio is the Gâteaux derivative, assuming it exists, of the underlying risk measure at the overall portfolio in the direction of the sub-portfolio. We consider the capital allocation problem based on the higher moment risk measure, which, as a generalization of expected shortfall, involves a risk aversion parameter and a confidence level and is consistent with the stochastic dominance of corresponding orders. As the main contribution, we prove that the higher moment risk measure is Gâteaux differentiable and derive an explicit expression for the Gâteaux derivative, which is then interpreted as the capital allocated to a corresponding sub-portfolio. We further establish the almost sure convergence and a central limit theorem for the empirical estimate of the capital allocation, and address the robustness issue of this empirical estimate by computing the influence function of the capital allocation. We also explore the interplay of the risk aversion and the confidence level in the context of capital allocation. In addition, we conduct intensive numerical studies to examine the obtained results and apply this research to a hypothetical portfolio of four stocks based on real data.