Consumer credit-risk models via machine-learning algorithms

We apply machine-learning techniques to construct nonlinear nonparametric forecasting models of consumer credit risk. By combining customer transactions and credit bureau data from January 2005 to April 2009 for a sample of a major commercial bank’s customers, we are able to construct out-of-sample forecasts that significantly improve the classification rates of credit-card-holder delinquencies and defaults, with linear regression R²’s of forecasted/realized delinquencies of 85%. Using conservative assumptions for the costs and benefits of cutting credit lines based on machine-learning forecasts, we estimate the cost savings to range from 6% to 25% of total losses. Moreover, the time-series patterns of estimated delinquency rates from this model over the course of the recent financial crisis suggest that aggregated consumer credit-risk analytics may have important applications in forecasting systemic risk.     

Network effects, market structure and industry performance

This paper analyzes oligopolistic markets with network externalities. Exploiting a minimal complementarity structure on the model primitives that allows for pure network goods, we prove existence of non-trivial fulfilled-expectations equilibrium. We formalize the concept of industry viability, investigate its determinants, and show that it improves with more firms in the market and/or by technological progress. These results enlighten some well-known conclusions from case studies in the management strategy literature. We also characterize the effects of market structure on industry performance, which depart substantially from ordinary markets. The approach relies on lattice-theoretic methods, supplemented with basic insights from nonsmooth analysis.     

National culture and national savings: is there a link?

We study the effect of national culture on economic decisions, focusing on GLOBE cultural dimensions of uncertainty avoidance and future orientation. Specifically, we study the effect of divergence between cultural values and practices (societal aspirations), on the aggregate savings decision. Using the life-cycle model of savings as our basic model, we find that societal aspirations are important in explaining national savings behavior. In particular, we show that societal aspirations relating to future orientation and uncertainty avoidance have a positive effect on the rate of savings. We interpret our findings to indicate that such societal aspirations lead to mistrust in the societal arrangements and institutions, and induce savings as a means of securing the future and reducing uncertainty. To substantiate this interpretation, we utilize the microfinance industry; showing that high societal aspirations are associated with preference for savings through member-owned microfinance institutions (MFIs) over savings through non-member-owned MFIs.     

Supermodularity and Complementarity in Economics: An Elementary Survey

The literature on supermodular optimization and games is surveyed from the perspective of potential users in economics. This methodology provides a new approach for comparative statics based only on critical assumptions, and allows a general analysis of games with strategic complementarities. The results are presented in a simplified yet rigourous manner, without reference to lattice theory, for the special case of one-dimensional parameter and actions sets, with the emphasis being on wide accessibility. Detailed applications are presented for well-known models of consumer behavior, monopoly pass-through, Bertrand and Cournot competition, strategic R&D, search, and matching. Wherever appropriate, useful tricks for applications and comparative comments are inserted.     

Effects of One-Way Spillovers on Market Shares, Industry Price, Welfare, and R & D Cooperation

With one-way spillovers, the standard symmetric two-period R&D model leads to an asymmetric equilibrium only, with endogeneous innovator and imitator roles. We show how R&D decisions and measures of firm heterogeneity—market shares, R&D shares, and profits—depend on spillovers and on R&D costs. While a joint lab always improves on consumer welfare, it yields higher profits, cost reductions, and social welfare only under extra assumptions, beyond those required with multidirectional spillovers. Finally, the novel issue of optimal R&D cartels is addressed. We show an optimal R&D cartel may seek to minimize R&D spillovers between its members.     

Incorporating location,routing and inventory decisions in supply chain network design

This paper for the first time presents a novel model to simultaneously optimize location, allocation, capacity, inventory, and routing decisions in a stochastic supply chain system. Each customer’s demand is uncertain and follows a normal distribution, and each distribution center maintains a certain amount of safety stock. To solve the model, first we present an exact solution method by casting the problem as a mixed integer convex program, and then we establish a heuristic method based on a hybridization of Tabu Search and Simulated Annealing. The results show that the proposed heuristic is considerably efficient and effective for a broad range of problem sizes.     

Information Structure and the Tragedy of the Commons in Resource Extraction

Synopsis This paper considers the well-known Levhari-Mirman discrete-time model of resource extraction, and investigates the effects of the information structure of the dynamic game – open-loop, Markovian or history-dependent – on the equilibrium consumption path and the overall utility of the agents. Due to the special structure of the model, the open-loop regime yields a Pareto-optimal outcome. The Markovian regime leads to the most pronounced version of the tragedy of the commons. History-dependent behavior yields an outcome set that is intermediate between the other two cases, and that may include the Pareto-optimal outcome in some cases. The level of efficiency of equilibrium behaviour is thus U-shaped as a function of the level of information the agents’ extraction strategies are based on. The analysis suggests that in environments characterized by a dynamic (and no market) externality, forcing agents to commit to open-loop behavior would constitute welfare-improving regulation.     

Dividend Changes and Future Profitability

We investigate the relation between dividend changes and future profitability, measured in terms of either future earnings or future abnormal earnings. Supporting "the information content of dividends hypothesis," we find that dividend changes provide information about the level of profitability in subsequent years, incremental to market and accounting data. We also document that dividend changes are positively related to earnings changes in each of the two years after the dividend change.     

The association between individual audit partners’ risk preferences and the composition of their client portfolios

We explore whether audit partners’ attitude towards risk, as measured by their personal criminal convictions, are reflected in the composition of their client portfolios. Analyzing a unique dataset of Swedish audit partners’ criminal convictions, we find that the clients of audit partners with criminal convictions are characterized by greater financial, governance, and reporting risk than those of audit partners without criminal convictions. Also, clients of audit partners with criminal convictions pay larger audit fees, on average, than those of auditors without criminal convictions.