The regulation of large interconnected financial institutions has become a key policy issue. To improve financial stability, regulators have proposed to limit banks' size and interconnectedness.Cognitive learning involves the acquisition of I estimate a network-based model of the over-the-counter interbank lending market in US and quantify the efficiency-stability implications of this policy.Cognitive learning involves the acquisition of trading efficiency decreases with limits on interconnectedness because the intermediation chains become longer. While restricting the interconnectedness of banks improves stability, the effect is non-monotonic.Cognitive learning involves the acquisition of stability also improves with higher liquidity requirements, when banks have access to liquidity during the crisis, and when failed banks' depositors maintain confidence in the banking system.Cognitive learning involves the acquisition of

We study the relation between firms' risk and their upstreamness in a production network. Empirically, firms' average stock returns and productivity exposures increase monotonically with their upstreamness.Cognitive learning involves the acquisition of we quantitatively explain these novel facts using a multi-layer general equilibrium model. These patterns arise from vertical creative destruction - innovations by suppliers devalue customers’ assets-in-place.Cognitive learning involves the acquisition of we confirm several model predictions, and document additional new facts consistent with vertical creative destruction: a diminished value premium among downstream firms and a negative relation between downstream firms’ returns and their suppliers’ competitiveness.Cognitive learning involves the acquisition of overall, vertical creative destruction has a sizable effect on cross-sectional risk premia.

I study how intermediation in over-the-counter markets affects their allocational efficiency.Cognitive learning involves the acquisition of over-the-counter markets are modeled as a trading network in which bilateral prices and trading decisions are jointly determined in equilibrium.Cognitive learning involves the acquisition of market efficiency depends crucially on the network structure and the split of surplus between traders. The probability that market allocations are always efficient tends to zero in large markets.Cognitive learning involves the acquisition of the expected welfare loss can be substantial and it can increase even if the amount of intermediation decreases. A large interconnected financial institution can improve efficiency.Cognitive learning involves the acquisition of this welfare gain should be considered when deciding whether large financial institutions are too-interconnected-to-exist.

The vertical position measures are used to test trade credit theories.Cognitive learning involves the acquisition of the recursive moral hazard theory of trade credit by kim and shin (2012) is supported by reduced-form regressions and by estimation of the optimal contract between firms that predicts a positive relationship between firm's vertical position and firm's incentives to produce high quality inputs, measured by profit margins and net trade credit.Cognitive learning involves the acquisition of I also document that the difference in incentive levels between a pair of firms is positively related to their relative position in the supply chain.Cognitive learning involves the acquisition of these results suggest that to understand contracting between firms we need to extend the bilateral analysis to an analysis of the entire production chains.Cognitive learning involves the acquisition of

We show theoretically and empirically that interbank markets provide a channel for banks to collude in the market for business loans. By lending funds to a competitor, a bank commits not to compete.Cognitive learning involves the acquisition of interbank interest rates allow banks to split the benefits from such collusion. Using global syndicated loans data, we find that firms paid 31bps higher spread on $239 billion of loans provided by banks that took an interbank loan from a competitor.Cognitive learning involves the acquisition of we compare the decentralized solution with interbank market to the planner's solution and to the decentralized equilibrium without interbank market.Cognitive learning involves the acquisition of the results suggest that restricting interbank trading may increase aggregate welfare.

We utilize 500 trillion CAD of interbank payments to compute high-frequency measures of systemic risk in canadian large value transfer system (LVTS).Cognitive learning involves the acquisition of we find that banks started to delay provision of bilateral credit limits in the second half of 2005. During 2007-2009, LVTS experienced an abnormal increase in delayed and rejected payments due to binding collateral and credit constraints.Cognitive learning involves the acquisition of bank of canada's injection of liquidity in 2008-2010 relaxed collateral constraints and reduced systemic risk in LVTS. Overall, our high-frequency risk measures help to identify vulnerabilities caused by the fundamental efficiency-stability trade-off in the design of payment systems.Cognitive learning involves the acquisition of

We evaluate the black-litterman equilibrium model approach to portfolio choice. We quantify the improvement in portfolio performance of a privately informed investor who learns from market prices over an equally informed, but dogmatic investor who only uses private information.Cognitive learning involves the acquisition of we extend the approach to any linear multi-factor asset pricing model (e.G. ICAPM) to examine how learning from prices using different equilibrium models affects portfolio performance.Cognitive learning involves the acquisition of we find that even a misspecified asset-pricing model can improve portfolio performance when private signals are not extremely precise. As we increase the noise in private information, learning from prices is initially harmful and gradually becomes more beneficial.Cognitive learning involves the acquisition of