Ex Post Evaluations in France and Norway, and Their Results

Ex Post Evaluations in France and Norway, and Their Results


David Meunier, Université Paris-Est, LVMT, UMR_T9403, Morten Welde, Norwegian University of Science and Technology


We present the French and Norwegian ex-post evaluation frameworks, outline questions that ex-post analysis should address, compare French and Norwegian results, provide indications on ex-post evaluation frameworks' tailoring and potential uses


Transport investments represent high budgets and a major share of public investment all over the world. Accordingly, before their implementation, projects are usually supported by many studies, including Cost Benefit Analysis (CBA) which aims to assess the projects’ value for money. Still, there is a general lack of organized feedback that would allow checking whether the virtual outcomes forecasted and announced in the ex-ante studies are actually observed or not when the project produces its effects.

The aim of this paper is to highlight how ex-post studies and an organized process for analysing them could be useful to project owners, project managers, CBA practioners and academics, besides the democratic accountability issues at stake behind this kind of ex-post exercise.

Very few countries have formal obligations requiring ex post evaluation. France is a country with a mandatory framework for evaluation of large transport infrastructure projects. During the last 3 decades, a high number of French road and rail projects have been subjected to ex post analysis. This has provided practitioners and decision makers with useful information regarding the performance of the projects and helped improve ex ante appraisal methodologies. Norway has performed similar analyses since some 10 years back. The French and Norwegian experiences combined makes for useful experience on how to design an ex post evaluation programme.

The paper presents the French and Norwegian evaluation frameworks; it outlines questions that ex-post analysis should address; and it compares the French and Norwegian results. The paper goes further than merely identifying discrepancies between ex ante estimation and ex post reality, it discusses also the impact of revealed information, it explores the formulated goals of the projects and whether the projects have been successful in achieving these. Goal-achievement may be difficult to assess using quantitative data alone.

One important dimension in project appraisal is risk and uncertainty: project outcomes are not mechanical and deterministic, thus we shall test some functional forms for the distribution of the main outcomes (traffic, costs), that could be useful for using ex-ante methods such as Monte Carlo simulations.
The paper will analyse project performance in two different European countries and provide useful lessons on the tailoring of ex post evaluation frameworks and on their potential uses. The results should be of interest both to countries that are currently carrying out ex post evaluations and to those that are considering doing so.


Association for European Transport