Abstract
The paper extends the use of Stochastic Frontier Analysis to assess the effectiveness analysis of public service producers to situations in which the resourcing level, and other relevant variables, such as the wages that are needed to attract labour of a given quality, are dependent upon the quality of output of the public service producer. It analyses the multiplier effects which the associated endogeneity of these resourcing variables produces and shows how its analysis can be used to extend the estimation of a conventional Production Possibility Frontier to a more general Achievement Possibility Frontier. The analysis makes use of multivariate skew normal distributions that generalise the assumption of Stochastic Frontier Analysis of a univariate half-normal distribution. The paper is a substantially generalised and revised version of a previous paper, and has been revsied and resubmitted to the Journal of Productivity Analysis at their request.
Original language | English |
---|---|
Pages (from-to) | 115-126 |
Number of pages | 12 |
Journal | Journal of Productivity Analysis |
Volume | 44 |
Issue number | 1 |
Early online date | 20 Jan 2015 |
DOIs | |
Publication status | Published - Aug 2015 |
Keywords
- Public services
- Effectiveness
- Endogeneity
- Stochastic frontier analysis
- Multivariate skew-normal