Random Effects Estimators with many Instrumental Variables

In this paper we propose a new estimator for a model with one endogenous regressor and many instrumental variables. Our motivation comes from the recent literature on the poor properties of standard instrumental variables estimators when the instrumental variables are weakly correlated with the endogenous regressor. Our proposed estimator puts a random coefficients structure on the relation between the endogenous regressor and the instruments. The variance of the random coefficients is modelled as an unknown parameter. In addition to proposing a new estimator, our analysis yields new insights into the properties of the standard two‐stage least squares (TSLS) and limited‐information maximum likelihood (LIML) estimators in the case with many weak instruments. We show that in some interesting cases, TSLS and LIML can be approximated by maximizing the random effects likelihood subject to particular constraints. We show that statistics based on comparisons of the unconstrained estimates of these parameters to the implicit TSLS and LIML restrictions can be used to identify settings when standard large sample approximations to the distributions of TSLS and LIML are likely to perform poorly. We also show that with many weak instruments, LIML confidence intervals are likely to have under‐coverage, even though its finite sample distribution is approximately centered at the true value of the parameter. In an application with real data and simulations around this data set, the proposed estimator performs markedly better than TSLS and LIML, both in terms of coverage rate and in terms of risk.     

Book reviews

Abstract

The newer deal: Social work and religion in partnership Ram A. Cnaan with Robert J. Wineburg and Stephanie C. Boddie (1999) Columbia University Press, New York $US22.50 (paperback)

Parenting between cultures — The primary school years Catherine Blunt (2000), Marymead Child and Family Centre, Kingston ACT $36.00

The answer within: A family in therapy reexamined Moshe Lang and Peter McCallum (2000), ACER Press, Melbourne ISBN 0-86431-333-0

Adoption and loss: The hidden grief Evelyn Burns Robinson (2000), Clova Publications, Christies Beach, South Australia 202 pp. $19.95 (paperback)

Managing to survive: Managerial practice in not-for-profit organizations Alun C Jackson & Frances Donovan (1999), Allen & Unwin, St Leonards, NSW Australia, ISBN 1 86448 728 3, 247 pp. (paperback)     

“It’s a Dignity Thing”: Nursing Home Care Workers’ Use of Emotions1

This article examines how nursing home care workers use emotions to construct dignity at work. Previous scholarship has shown how the financial and organizational characteristics of nursing homes shape and constrain emotion work among staff. Using evidence gathered during 18 months of participant observation in two nursing homes and 65 interviews with staff, this article analyzes how, despite obstacles, nursing home care workers generated authentic emotional attachments to residents. Surprisingly, some staff members said they particularly appreciated working with residents difficult to control. They felt accomplished when such residents successfully transitioned from life at home to life in institutional care. Emotions created dignity for staff and induced compliance among residents. Emotions are not only generated by organizations and imposed on workers; staff themselves produced emotions—sometimes in ways consistent with organizational demands, and sometimes not—and they consistently found in their emotions a resource to manage the strains of their work lives.     

Racial Threat or Racial Contact? How Race Affected Third‐Party Presidential Voting in the Antebellum North*

Objective. In this study, the “racial threat” and “racial contact” hypotheses are evaluated in relation to voting for the Liberty and Free Soil Parties in the North during the 1840s. Methods. Regression models are used to predict the effect of county‐level black populations on Liberty and Free Soil vote percentages in relation to types of employment. Results. Racial threat occurred in high manufacturing counties, but racial contact/threat emerged in more agricultural counties. The effects vary by party and region of the North. Conclusion. The effects of racial context on political behavior during this era are mixed, much like modern political studies have uncovered.     

Enhanced services and stipends for foster parents: effects on retention rates and outcomes for children.

Current national trends show that although the number of available foster homes is shrinking, the number of children and adolescents being cared for in the family foster care system is growing. This study demonstrates the significant benefits to both foster parents and the children in their care of providing enhanced services and stipends to foster parents.     

Fusion ballots as a candidate-centered reform: Evidence from Oregon

This article argues that fusion balloting, or the ability of multiple parties to nominate a single candidate for office, can be viewed as a candidate-centered and incumbent-oriented reform. Using the recent passage of fusion legislation in Oregon in 2009 and its subsequent use in 2010 state legislative elections as a test case, I study the factors behind supporting fusion legislation in both chambers of the state legislature and how these same factors affected, or did not affect, the incumbents that received cross-nominations in the 2010 elections. Overall, I find that party did not play a direct role in determining which legislators supported fusion, though female legislators were less supportive. When it came to obtaining nominations, the main finding is that incumbent candidates who voted yes to fusion were generally more likely to receive nominations. Still, legislators from the group that most opposed fusion, Democratic women, attempted to win cross-nominations once the 2010 elections approached. These findings suggest that candidate factors, more than party dictates, affected support for fusion and the ability of incumbents to receive cross-endorsements. Thus, the willingness of major-party candidates to adopt fusion can be understood through theories of candidate-centered elections and political parties.     

Optimal Portfolio Selection Using the General Multi-Index Model: A Stable Paretian Framework*

The problem of selecting optimal portfolios is examined using the general multi-index model. This model is useful because it allows investors to diversify across different types of assets and thereby exploit or hedge against a wide variety of economic conditions. The analysis is carried out in a stable Paretian framework with and without short sales. As such, it not only encompasses the mean-variance results for a variety of index models as special cases, but also provides a broad framework for applying the arbitrage pricing theory to portfolio decision making.     

Satisfactions and stresses in the sorting of mail

Abstract

The effects of two types of letter-sorting technology on job stress and satisfaction were compared. Manual sorting, which uses the traditional method of assigning letters to classified pigeon-holes, was judged to provide elements of discretion, autonomy, and social interaction which are largely absent in coding-desk work, in which a keyboard is used to enter postcodes. Between and within-subject methodology was used in the analysis of data from those engaged in sorting (n = 160). In addition, these groups were compared with supervisors (n = 36) and engineers (n = 26) also working in the sorting area.

The key findings were: (a) higher levels of satisfaction were found amongst manual sorters and those spending 50–75% of their time in code-sorting; (b) stress is closely related to satisfaction in the code-sort group but not in the manual group; (c) in manual groups and in supervisors stress was manifested in fatigue, but in the code-sort group stress took the form of anxiety. The results are discussed in terms of the impact of the type of work on fatigue, as well as on job stress and satisfaction.