Evaluation of Financial Structure

National Opinion Research Center, Chicago IL

Investigators

Abstract

Prop ID: 0318340 P I: Townsend, Robert M. Organization: National Opinion Research Center Title: Evaluation of Financial Structure Institutional Affiliation: University of Chicago The purpose of the research is to develop theory and empirical methods that can be used with micro and macro data to evaluate the financial structure of a given economy. The principal application is Thailand, where the PI has been fielding a large survey of households, small businesses, village funds, headmen, and joint liability groups. The proposal is distinguished by its three-pronged strategy: (1) take widely-accepted but rarely-tested models to the data, (2) engage in non-parametric fact-finding missions, thus moving back and forth between existing theory and data, finding patterns and documenting anomalies, and (3) create new theory, e.g., dynamic mechanism design. One topic concerns the micro-underpinnings of growth with increasing (and eventually decreasing) inequality as observed in Thailand l976-l996. Utilized is a dynamic stochastic infinite horizon model of transitional growth with endogenous financial deepening. Theorems establish overall concavity via endogenous randomization and allow computation. The model will be calibrated, estimated, and tested. As a non-ergodic transitional economy, entire paths are influenced by idiosyncratic and aggregate shocks, so a test statistic using Monte Carlo methods will be of independent interest. The model would be further compared via growth and inequality decompositions and predicted end-of-period income distributions to a model of wealth-constrained occupation choice. Likewise these models assume full access to the financial sector carries with it full risk-sharing in consumption and immunity against income effects in investment. The panel data would be used to gauge the performance of commercial banks, village funds, a government development bank, and the informal sector against these standards during and after the crisis, and in distinct regions. Tests, anomalies, and apparent policy remedies need to identify obstacles to trade: limited commitment, moral hazard, adverse selection and other impediments. Models of wealth-constrained occupation choice, some standard with potential default (limited commitment), others with exogenously incomplete contracts, and others with moral hazard are to be compared. Maximum likelihood estimation and mechanism design formulations are combined. Likewise, four polar but well-known models of joint liability groups positing moral hazard, monitoring, adverse selection, or limited commitment would be taken to data on repayment rates. Dynamic mechanism design models with both unobserved investment (moral hazard) and unobserved returns (hidden information) will be analyzed. This environment gives rise to imperfect risk sharing and to history-dependent partial credit guarantee schemes with insurance reserves, to be compared to those usually recommended. BROADER IMPACT: PhD students from Chicago and other universities are well integrated into the research, first in the classroom, and then as valuable research assistants and collaborators. Undergraduates from Chicago and other universities also contribute as summer interns, part time assistants, or full time RA's before entering graduate work. Women and minorities in the US and aboard play a key part. A field research, infrastructure unit with anthropological, geographic and ecologic components would be continued in Bangkok with subunits in the provinces. Enumerators include women and minority language groups (Khmer, Lao), and include current and future Thai students, some of whom visit the US for training (and US students visit Thailand). In addition, the Thai SES survey and comprehensive Village Census data are part of the construction of a broader data base/research archive, unified under a common GIS, enhancing infrastructure for research and education and facilitating dissemination. Policy implications follow at each step of the research algorithm. Executive summaries of completed papers and seminars in Thailand, the IMF, and IADB influence the formulation of public policy. A smaller, but parallel effort continues in the US with the Federal Reserve Banks of Chicago and Minneapolis, studying households and small businesses in African-American, Mexican, and Mong neighborhoods