Forecasting recovery rates on non-performing loans with machine learning

We compare the performance of a wide set of regression techniques and machine-learning algorithms for predicting recovery rates on non-performing loans, using a private database from a European debt collection agency. We find that rule-based algorithms such as Cubist, boosted trees, and random forests perform significantly better than other approaches. In addition to loan contract specificities, predictors that refer to the bank recovery process — prior to the portfolio’s sale to a debt collector — are also shown to enhance forecasting performance. These variables, derived from the time series of contacts to defaulted clients and client reimbursements to the bank, help all algorithms better identify debtors with different repayment ability and/or commitment, and in general those with different recovery potential.     

Institutions and economic growth in Portugal: a quantitative exploration

This paper presents a broad diagnostic of the level of institutional development in Portugal in the legal, corporate governance and financial systems. A comparative assessment suggests that Portuguese institutions are less developed than their European Union and East Asian counterparts, more developed than Greek institutions and on a level similar to that of Spanish institutions. We use data for a wide cross-section of countries since 1960 and correlate indicators of institutional development with the long-term average growth rate, identifying issues where reform is likely to significantly affect economic growth. We construct three new indices that measure the potential of institutional reform - the impact of reform on growth, the required reform effort and the efficiency of reform index - by taking into consideration the institutional distance between Portugal and the European Union. These indices measure, respectively, which reforms have the most payoff in terms of growth, which are less costly to undertake and which deliver the most growth per required effort. Our results strongly suggest that in a large number of issues, institutional reform may translate into substantially higher rates of economic growth. Of the ten most promising reforms, six are in the legal area, irrespective of which of the indices is considered. Whereas legal reform is promising at the aggregate and the microeconomic levels, in the financial sector aggregate indicators offer the wider scope for productive reform, while in the corporate governance area it is indices at the micro level that hold the most promise. These results support the view that a comprehensive reform effort is likely to deliver higher rates of growth in Portugal, allowing faster real convergence with the rest of the European Union.Received: January 2003, Accepted: February 2004, JEL Classification: O0, O5, K00, K4, G2, G3This paper is based on Firms, Financial Markets and the Law: Institutions and Economic Growth in Portugal, prepared for the conference Desenvolvimento Económico Português no Espaço Europeu: Determinantes e Políticas, organized by the Banco de Portugal. Financial support by the Banco de Portugal, NOVA FORUM and FUNDAÇÃO Para a Ciência e tecnologia and Polti through feder are gratefully acknowledged. Comments from Marcelo Rebelo de Sousa, two anonymous referees and the editor are sincerely appreciated. The usual disclaimer applies.     

Economic mechanisms for environmental protection: Business versus market transaction costs

This study presents theoretical arguments and evidence that attempt to show the contribution of economic mechanisms typical of a business for resolving problems of environmental protection. Firstly, a brief synopsis is given of the classic solutions that environmental economics proposes for correcting environmental externalities. Next, the benefit of expanding these two mechanisms is discussed so that they include those intermediate mechanisms for which it is not possible to clearly establish if they are developed within the domain of the market or the business. The study's focus resides in the fact that the explicit expansion of the range of solutions proposed from a macroeconomic point of view establishes a connecting point between environmental economics and business economics, an area in which academic attention to environmental matters has been significantly less. The present work forms part of research project SEC2002-00835, financed by MCYT-FEDER.     

Multiple Species Boreal Forests – What Faustmann Missed

Recent research in natural sciences shows that the dynamics in boreal forests are much more complex than what many models traditionally used in forestry economics reflect. This essay analyzes some challenges of accounting for such complexity. It shows that the optimal harvesting strategy for forest owners is history dependent and for some states of the forest, more than one strategy may be optimal.This paper confirms earlier literature on shallow lakes and coral reefs and shows that this kind of phenomena seem much more common than previously thought. They are valid for a wide range of ecosystems that cover large surfaces and they do not depend on the choice of some specific function to model the non-linearity. There are also indications that theses results could be obtained even for resources with concave growth if at least one species with non-linear growth affects their dynamics.     

A simple random assignment problem with a unique solution

Summary. All agents have the same ordinal ranking over all objects, receiving no object (opting out) may be preferable to some objects, agents differ on which objects are worse than opting out, and the latter information is private. The Probabilistic Serial assignment, improves upon (in the Pareto sense) the Random Priority assignment, that randomly orders the agents and offers them successively the most valuable remaining object. We characterize Probabilistic Serial by efficiency in an ordinal sense, and envy-freeness. We characterize it also by ordinal efficiency, strategyproofness and equal treatment of equals. Received: October 5, 1999; revised version: December 20, 2000     

Endogenous structural change and crisis in a multiple time-scales growth model

The slow and endogenous twist of economic macro-structure makes up an important evolutionary feature of capitalist economies, and may be at the root of structural crisis. In this line, a Goodwinian growth model with increasing returns and profit-sharing that tries to picture a simple scenario of the seventies crisis is considered. It is shown that the exhaustion of the Kaldor-Verdoorn “productivity law” can entail, in a nonlinear framework, a “catastrophic” bifurcation from a “high” to a “low” growth path. Slow/fast dynamical systems then allow one to formalize a multiple time-scales dynamics where the growth path is shaped by the structural framework in which it takes place, but has also a long -un feedback. Structural change and crisis appear as long term and endogenous outcomes.     

The effects of supportive management and job quality on the turnover intentions and health of military personnel

To be competitive in today's increasingly complex and rapidly changing envi‐ronment, organizations must retain personnel and promote the well‐being of employees. We examine the relationship of both support provided to person‐nel and job quality with employee health and turnover intentions among a sample of 450 military personnel. Factors involving the supportive manage‐ment of personnel (i.e., supervisory support, organizational support, and work‐life balance) and factors pertaining to job quality (i.e., work stimulation and job clarity) were indirectly related to health and to turnover intentions through the mediating influence of job satisfaction. © 2007 Wiley Periodicals, Inc.     

A Dynamic Leontief Model with Non-renewable Resources

In this paper we study a generalization of the dynamic Leontief input–output model. We extend the standard dynamic Leontief model with the balance equation of non-renewable resources. Obviously, the non-renewable stocks will decrease, exploiting primary resources. In this study we examine the controllability of this extended model by taking the consumption as the control parameter. Assuming balanced growth for both consumption and production, we investigate how long these scarce resources will cover the input needs of production and how the lifetime of the system depends on the balanced growth rate and on the consumption. In doing so, we apply classic results from control theory and on eigenvalue problems in linear algebra.     

More pessimism than greediness: a characterization of monotone risk aversion in the rank-dependent expected utility model

Summary. This paper studies monotone risk aversion, the aversion to monotone, mean-preserving increase in risk (Quiggin [21]), in the Rank Dependent Expected Utility (RDEU) model. This model replaces expected utility by another functional, characterized by two functions, a utility function u in conjunction with a probability-perception function f. Monotone mean-preserving increases in risk are closely related to the notion of comparative dispersion introduced by Bickel and Lehmann [3,4] in Non-parametric Statistics. We present a characterization of the pairs (u,f) of monotone risk averse decision makers, based on an index of greediness G _u of the utility function u and an index of pessimism P _f of the probability perception function f: the decision maker is monotone risk averse if and only if . The index of greediness (non-concavity) of u is the supremum of taken over . The index of pessimism of f is the infimum of taken over 0 < v < 1. Thus, , with G _u = 1 iff u is concave. If then , i.e., f is majorized by the identity function. Since P _f = 1 for Expected Utility maximizers, forces u to be concave in this case; thus, the characterization of risk aversion as is a direct generalization from EU to RDEU. A novel element is that concavity of u is not necessary. In fact, u must be concave only if P _f = 1.Received: 10 April 2001, Revised: 18 November 2003, JEL Classification Numbers: D81. Correspondence to: Michéle CohenAlain Chateauneuf, Michéle Cohen, Isaac Meilijson: We are most grateful to Mark Machina, Peter Wakker and two anonymous referees for very helpful suggestions and comments.