We investigate conditions sufficient for identification of average treatment effects using instrumental variables. First we show that the existence of valid instruments is not sufficient to identify any meaningful average treatment effect. We then establish that the combination of an instrument and a condition on the relation between the instrument and the participation status is sufficient for identification of a local average treatment effect for those who can be induced to change their participation status by changing the value of the instrument. Finally we derive the probability limit of the standard IV estimator under these conditions. It is seen to be a weighted average of local average treatment effects.

Evolution and Rationality Some Recent Game-Theoretic Results. Identification and Estimation of Local Average Treatment Effects

http://people.stern.nyu.edu/wgreene/FrontierModeling/Reference-Papers/Greene-JE2005-Heterogeneity.pdf

Reconsidering heterogeneity in panel data estimators of the stochastic frontier model

This paper uses the concept of a metafrontier to compare the technical efficiencies of firms that may be classified into different groups. The paper presents the basic analytical framework necessary for the definition of a metafrontier, shows how a metafrontier can be estimated using non-parametric and parametric methods, and presents an empirical application using cross-country agricultural sector data. The paper also explores the issues of technological change, time-varying technical inefficiency, multiple outputs, different efficiency orientations, and firm heterogeneity.

/pdf/metafrontier-frameworks-for-the-study-of-firm-level-59ca3drmoy.pdf

Metafrontier frameworks for the study of firm-level efficiencies and technology ratios

The Econometrics of Panel DataSpringer Handbook of Science and Technology IndicatorsPanel Data EconometricsThe Econometrics of Panel DataA Practitioner's Guide to Stochastic Frontier Analysis Using StataBenchmarking for Performance EvaluationEssays on Microeconomics and Industrial OrganisationHealth System EfficiencyInternational Journal of Production EconomicsEconometric Analysis of Model Selection and Model TestingInternational Applications of Productivity and Efficiency AnalysisAdvanced Robust and Nonparametric Methods in Efficiency AnalysisEconometrics and the Philosophy of EconomicsThe Measurement of Productive EfficiencyMeasuring Efficiency in Health CareFinancial, Macro and Micro Econometrics Using REconometric Analysis of Cross Section and Panel DataApplied EconometricsProductivity and Efficiency AnalysisEconometric Model SelectionProductivity and Efficiency AnalysisStochastic Frontier AnalysisThe Oxford Handbook of Health EconomicsThe Measurement of Productive Efficiency and Productivity GrowthNew Directions in Productivity Measurement and Efficiency AnalysisA Primer on Efficiency Measurement for Utilities and Transport RegulatorsPanel Data EconometricsProduction and Efficiency Analysis with RApplications of Modern Production TheoryThe Measurement of Productive EfficiencyNonparametric Econometric Methods and ApplicationAn Introduction to Efficiency and Productivity AnalysisHealth, the Medical Profession, and RegulationThe Analysis of Household SurveysData Envelopment AnalysisProgramming Collective IntelligenceEfficiency AnalysisProductivity and Efficiency AnalysisMeasurement of Productivity and EfficiencyProduction Frontiers

The Econometric Approach to Efficiency Analysis

A DEA-based stochastic frontier estimation framework is presented to evaluate contextual variables affecting productivity that allows for both one-sided inefficiency deviations as well as two-sided random noise. Conditions are identified under which a two-stage procedure consisting of DEA followed by ordinary least squares (OLS) regression analysis yields consistent estimators of the impact of contextual variables. Conditions are also identified under which DEA in the first stage followed by maximum likelihood estimation (MLE) in the second stage yields consistent estimators of the impact of contextual variables. This requires the contextual variables to be independent of the input variables, but the contextual variables may be correlated with each other. Monte Carlo simulations are carried out to compare the performance of our two-stage approach with one-stage and two-stage parametric approaches. Simulation results indicate that DEA-based procedures with OLS, maximum likelihood, or even Tobit estimation in the second stage perform as well as the best of the parametric methods in the estimation of the impact of contextual variables on productivity. Simulation results also indicate that DEA-based procedures perform better than parametric methods in the estimation of individual decision-making unit (DMU) productivity. Overall, the results establish DEA as a nonparametric stochastic frontier estimation (SFE) methodology.

Evaluating Contextual Variables Affecting Productivity Using Data Envelopment Analysis

Introduction In Chapter 1, we introduced a series of questions that the tools discussed in this book are designed to help answer. In this chapter, we provide the reader with the necessary theoretical underpinnings in order to answer these questions and to understand the models that are developed in later chapters. This is important as it is necessary to understand which is the most appropriate tool to use in which circumstance and what the limitations are of the different approaches. In some of the following sections the text is fairly technical, but these sections are useful as a general reference for the practitioner when modeling specific issues. For example, Section 2.5 on the functional forms of the production function provides the required formulae for some of the key economic issues discussed in many chapters. The study of the production, cost, and profit functions has a long history and the practical applications of modeling these functions are extensive. In line with the questions introduced in Chapter 1, a summary of the major objectives for studying these functions may include the following: (i) If a firm were to expand its operations by increasing its inputs by 10 percent, how much would output increase by? How much lower would its unit costs be (and thus how much lower could it reduce its prices by or increase its margins)? (ii) If a firm were to invest in new IT equipment, how many fewer manual employees would be needed to produce the same level of output? How many more IT personnel would be needed? (iii) Can we consider the use of certain inputs independently of others? (iv) Compared to last year, how much more output can be produced for a given level of inputs? (v) Compared to industry best practice, for a given input level, how much more output can an organization produce compared to its current output level?

A Practitioner's Guide to Stochastic Frontier Analysis Using Stata: Production, Distance, Cost, and Profit Functions

Government policies are frequently used throughout the world to promote productivity. While some of the policies are designed to work through technology enhancement, others are meant to exert the influence through effciency improvement. It is therefore important to have a program evaluation method that can distinguish the channels of effects. We propose a treatment eect stochastic frontier model for this purpose. One of the important feature of the model is that the participation in the treatment is endogenous. We illustrate the empirical application of the model using the data of large dams in India to study the eects on the agricultural production.

Treatment Effect Stochastic Frontier Models with Endogenous Selection

Introduction Until now, we have assumed that firms only incur technical inefficiency in the production of output, and the allocation of inputs is either at the most efficient level or inputs are exogenously given. If inputs are exogenously given, one may not address allocative inefficiency simply because the input allocation problem is assumed away. By contrast, if inputs are endogenous, then allocation decisions using some economic behavior have to be made. Such decisions may or may not be optimal. If the allocation is nonoptimal, then the producer is allocatively inefficient. In either case, endogeneity of inputs has to be taken into account in the model. In a cost minimizing framework, input allocation is optimal if producers allocate inputs in such a way that input price ratios equal the ratio of their marginal products. In such a case, actual cost differs from the optimal cost (barring the noise term) due to technical inefficiency. On the other hand, if input allocation is nonoptimal, cost will be higher because of both technical and allocative inefficiency. In this chapter, we examine the impact of both technical and allocative inefficiency on costs, assuming that output and input prices are exogenous and the level of inputs are the choice variables (i.e., the goal is to use inputs to minimize cost for a given level of output and input prices). In this setting, we are able to extend the questions we first set out in Chapter 4. In addition to examining whether, and by how much, a firm can reduce its costs, we are also able to identify how much of this cost reduction can be achieved through improvements in technical efficiency and how much can be achieved through an optimal mix of inputs. For instance, we can reconsider the health care questions first asked in Chapter 4, namely, how much lower could the cost of treatment be at certain hospitals while still treating the same number of patients for the same level of care and how can this be achieved?

Hung-Jen Wang

Papers

A Practitioner's Guide to Stochastic Frontier Analysis Using Stata: Production, Distance, Cost, and Profit Functions

Treatment Effect Stochastic Frontier Models with Endogenous Selection

Estimation of Technical and Allocative Efficiency in Profit Frontier Models Using System Models with Cross-Sectional Data: A Primal System Approach

Taiwan and the Changing Dynamics of Sino-US Relations

Deconstructing the Established Westphalian Architecture in Light of China’s Rise