Anthony J. Glass

Productivity Growth Decomposition Using a Spatial Autoregressive Frontier Model

A new spatial decomposition of total factor productivity growth into direct (own) and indirect (spillover) components is set out. We then apply the decomposition in the context of a spatial autoregressive production frontier analysis of 40 European countries over the period 1995–2008.

A Spatial Autoregressive Production Frontier Model for Panel Data: With an Application to European Countries

In this paper we merge techniques from the e¢ ciency literature with spatial econometric techniques. In particular, we combine calculation of e¢ ciency from the unit speci…c eects with the spatial autoregressive model to develop a spatial autoregressive frontier model for panel data. Features of the modeling include time-varying e¢ ciency and estimation of own and spillover returns to scale. The model is applied to aggregate production in European countries over the period 1995 2008. Among other things, we …nd that production in the sample average country is characterized by increasing returns to scale when we allow for returns to scale spillovers from other countries, and constant returns when these spillovers

The rise and fall of Railtrack PLC: an event study

Forming Railtrack was a key part of the privatization of British Rail (BR). Railtrack took over the ownership of BRs fixed infrastructure in April 1994 and its parent company, the Railtrack Group, was floated in May 1996 on the London Stock Exchange (LSE). Despite the group posting some excellent financial results in the early years, Railtracks record on infrastructure improvement and safety was frequently criticized. This apparent inconsistency between shareholder interests and public service obligations culminated in Railtrack being placed in administration in October 2001. In view of this apparent inconsistency and the High Court claim for additional compensation brought against the government by a group of 49 000 small investors, the reaction of the stock market to 19 key events is modelled. Among other things, we find when Railtrack announced after the fatal derailment of a high-speed train near the Hertfordshire town of Hatfield that there would be a 6-month programme of emergency track repairs, the Groups share price was marked down, but it did not plummet. Even though Railtrack was in panic mode, it appears that investors decided to hold onto their shares, believing the panic would have no long-term repercussions. This proved to be a huge error of judgement.

Bank performance and the financial crisis: evidence from Kazakhstan

During the first phase of the financial crisis in 2008/09, after Iceland and Belgium, Kazakhstan experienced the most significant bank failures as a share of bank system assets. Using rich monthly data for virtually the entire Kazakh banking industry for the period March 2007–December 2010, Stochastic Frontier Analysis (SFA) is used to fit several functions (cost, revenue, standard profit, alternative profit and input distance). Among other things, we estimate the effects of two measures of the quality and risk of the loan portfolio on the industry best practice frontiers and bank inefficiencies. We find that an increase in the volume of bad loans as a ratio of total lending has a desirable effect on the cost, input-distance and alternative profit frontiers, all of which is consistent with the ‘skimping’ hypothesis.

A spatial productivity index in the presence of efficiency spillovers: Evidence for U.S. banks, 1992–2015

We present the methodology for a new spatial decomposition of total factor productivity (TFP) growth. The relevant literature is underdeveloped as there is just one short study which proposes a partial spatial TFP growth decomposition. We develop this literature in four respects. The first two developments are methodological to go from a partial decomposition to a complete one. First, we augment the partial decomposition with a cost efficiency spillover growth component. Second, we introduce own and spillover allocative efficiency growth components. Third, we provide a more detailed coverage of the spatial decomposition of TFP growth. Fourth, in contrast to the traditional application to geographical areas (e.g., countries) in the relevant literature, we apply our decomposition using firm level data, which suggests that there can be an important role for spatial productivity analysis in OR. Our application is to large U.S. banks over the period 1992–2015. Among other things, we find for the average large U.S. bank that TFP growth since the financial crisis has become much more dependent on the bank itself and less so on spatial spillovers.

Relative Winners and Losers from Efficiency Spillovers in Africa: Implications for Regional Integration and Income Inequality

Studies that have analyzed the efficiency of developing countries have estimated non-spatial frontier models. We extend this approach by accounting for spatial dependence among African countries. In particular, we estimate a spatial Durbin stochastic production frontier model. We also make novel use of the efficiency scores from our spatial model to suggest a direction for regional integration policy for Africa that policy makers can consider. A previous suggestion to promote regional integration and income growth in Southern Africa has been to use South Africa as a regional integration hub and to encourage other countries in the region to improve economic links with the hub. We continue with this line of enquiry and although we conclude that there are currently no African countries that are ideal candidates to be a regional integration hub, we suggest three other countries that policy makers may consider using as hubs in the future. We therefore suggest that it would be prudent to consider implementing policies to expedite the readiness of these countries to act as integration hubs.

The Spatial Efficiency Multiplier and Random Effects in Spatial Stochastic Frontier Models

We extend the emerging literature on spatial frontier models in three respects. Firstly, we account for latent heterogeneity by developing a maximum likelihood random effects spatial autoregressive (SAR) stochastic frontier model. Secondly, to analyze the finite sample properties of a spatial stochastic frontier model we develop a Monte Carlo experimental methodology which we then apply. Thirdly, we introduce the concept of the spatial efficiency multiplier and show that the efficiency benchmark for a productive unit from the structural form of a spatial stochastic frontier model differs from the efficiency benchmark from the reduced form of the model.

Testing the Monotonicity and Curvature of a Translog Production Function with Spatial Autoregressive Dependence

In this paper we extend the tests of the monotonicity and concavity of a non-spatial translog production function to the case where there is spatial autoregressive dependence. The tests are then applied using data for 40 European countries over the period 1995-2008.

The Effects of Efficiency and TFP Growth on Nitrogen and Sulphur Emissions in Europe: A Multistage Spatial Analysis

It is common in firm level environmental efficiency studies for pollution to form part of the production technology. We omit nitrogen and sulphur emissions from the spatial analysis of production in European countries (1995 - 2008) because we find they are not significant inputs. Efficiency and TFP growth from the production analysis are then used in second stage spatial models of nitrogen and sulphur emissions in European countries. We find that to cut European sulphur emissions by a certain percentage requires a decrease in a composite measure of a country’s efficiency and TFP growth which is more than double the decrease needed to reduce European nitrogen emissions by the same percentage.