Catherine Cazals

Nonparametric Frontier Estimation: A Robust Approach

Most nonparametric methods for estimating production frontiers (data envelopment analysis and free disposal hall (FDH)) are based on envelopment techniques. Statistical inference based on these estimators is available but, by construction, they are very sensitive to extreme values or outliers. We propose a nonparametric estimator, which is more robust to these extreme values. It is based on a concept of expected minimum input function (or expected maximal output function). We show how this function is related to the efficient frontier itself. The resulting estimator is related to the FDH estimator but it will not envelop all the data. The asymptotic theory is provided. Our approach includes the multiple input and multiple output cases

Scale Economies and Natural Monopoly in the Postal Delivery: Comparison Between Parametric and Non Parametric Specifications

This paper synthesize different econometric models of the postal delivery, estimated using a french data set. These models explain the demand of labor (representing the essential part of the cost) by a vector of outputs (different types of delivered objects) and by geographical characteristics of the office. The data are constructed by a cros-section sample in the set of offices in France and there is no observable variation in the sample of the price of the labor.

An Econometric Study of Cost Elasticity in the Activities of Post Office Counters

Economists generally agree that scope and scale economies in the postal sector are concentrated mainly in distribution and counter activities. Numerous empirical studies highlight returns to scale in distribution (see, for example, Rogerson and Takis, 1993; Bradley and Colvin, 1993; Cazals, DeRycke, Florens and Rouzaud 1997; Roy 1999; Cazals, Florens and Roy 2001) but few studies are interested in counter activities.

Econometrics of Mail Demand

The investigation of mail demand reveals that it is a complex phenomenon. This is for a number of reasons. First and foremost, this complexity arises because the concept of “mail” encompasses an extremely wide range of products (first class letters, second class letters, parcels, newspapers, etc.) distributed to many different consumers (households, firms, and postal administrations, such as public corporations or government departments). Second, even if postal services in most countries are a legal monopoly for mail processing, some form of competition in terms of substitute products like the telephone, e-mail and fax must be taken into account.

Delivery Costs for Postal Services in the UK

This paper stresses the importance of the treatment of heterogeneity between delivery offices when estimating economies of scale in outdoor mail delivery activity. We show that models estimated with cross-section data can give biased results and that panel data models encompass cross- section models. In practice this means that empirical estimates of economies of scale in delivery will be underestimated when using cross-section data and researchers should, where possible, adopt panel estimation techniques. In line with these findings we have estimated cost functions for UK outdoor mail activities using panel data that take into account observed heterogeneity by using variables describing the environmental characteristics of delivery offices. Among these variables, the proportion of traffic delivered by non rural routes is introduced in a non linear way (with a stratification of the sample) into the cost models and we find strong evidence of UK outdoor mail delivery activities exhibiting returns to scale with respect to volumes per delivery point that lie in the range of less than 2 to over 4 according to the value of this proportion. These results are consistent with other studies on outdoor delivery costs in France and the USA and add to the growing international literature that substantial fixed costs are incurred in outdoor mail delivery activities.

A Panel Data Analysis of Inefficiency and Heterogeneity in the Postal Sector

This compilation of original papers selected from the 19th Conference on Postal and Delivery Economics and authored by an international cast of economists, lawyers, regulators and industry practitioners addresses perhaps the most significant problem that has ever faced the postal sector – electronic competition from information and communication technologies. This has increased significantly over the last few years with a consequent serious drop in mail volume.

The Effect of Unobserved Heterogeneity in Stochastic Frontier Estimation: Comparison of Cross Section and Panel with Simulated Data for the Postal Sector

In this paper, we examine the application of SFA method with time-invariant inefficiency and assess its estimation of inefficiency when applied to cross section and panel data. By using simulation methods, we look at the effect of unobserved heterogeneity on the estimates of inefficiency in both cross section and panel. In the presence of unobserved heterogeneity and significant variance in the inefficiency term, stochastic frontier estimation of inefficiency can be significantly different in panel and in cross section. This finding accords with analysis of actual data from the postal sector. We then suggest an estimation method for cost frontier when inefficiency is time-invariant and with unobserved heterogeneity.

Delivery Costs II

This paper presents several parametric log linear models for the outdoor delivery cost functions based on a survey of all the French delivery offices in 2001. The main empirical results derived from this set of estimations are the following:

Analysis of the Effect of Management Quality Variables in Delivery Cost Functions

Letter volumes in countries with advanced broadband networks have been in decline since the early to mid-2000s while, more recently, parcel volumes have started to grow quite rapidly. The main drivers of these trends, namely the substitution of physical letters with electronic modes of communication and increasing levels of on-line shopping, are expected to continue for some time. This raises two important challenges for postal universal service providers (USPs). The first is to manage operational changes to meet the evolving needs of consumers, such as changes in the quantity, shape, size and weight of mail sent and received. Second, USPs need to reduce costs and increase efficiency as quickly as is practically possible in order to help slow the decline in letters and to compete more effectively with other parcel providers.