Le risque dans le contrat administratif ou la nécessaire reconnaissance de la dimension économique du contrat

Risks allocation in Government contract performance is a key issue for both economic analysis and legal doctrine. However one factor which restrains the dialogue between law and economics (in France at least) is that, while economics develops normative and pure theories of optimal risk allocation patterns, legal scholars focus on pure doctrinal analysis of administrative courts rulings. The paper tries to tackle that issue through an empirical analysis of French administrative contract law. It starts with the idea that it would be an error to consider that legal rules incorporate an economic theory of risk allocation. However that does not prevent positive contract administrative law to frame a pattern of risk allocation in the course on contract performance. To discover how this framework operates, one must take into account the diversity of rules, legal provisions and case-law rulings that apply to such contracts. The article therefore focuses on the following issues : standard contractual conditions (« cahiers des clauses administratives »), administrative courts rulings in contract litigation (« imprévision » and « sujétions imprévues »), and opinions of the « commissions spécialisées des marchés ». We argue that, contrary to the economic analysis view, that no single optimal pattern of risk allocation can in practice be set up. What the public authorities and administrative courts have to manage is to frame a pattern of risk allocation the less imperfect as possible, owing to the necessity to design some kind of compromise between several constraints : the firms should not have incentives to practice exaggerate risk premiums in setting the price ; the accounting officer should not allow cost overruns to be systematically allowed; the government should not compromise the delivery of work by the contractor; the funding of contracts by the Sate should be certain, and so on. However, the highest administrative courts (Cours administratives d?appel and Conseil d?État) rulings do not provide clear and explicit pattern of risk allocation: the decisions are taken on a case by case basis. An empirical analysis shows the absence of correlation between the facts that provoke contract litigation and the rulings of the administrative courts. It also reveals that the standard of « bouleversement de l?économie du contrat» is applied to the litigious contracts without considering the characteristics of the firms (size, financial capacity, etc.). We argue that improvements in the way public authorities deal with the risks issue are possible. The government should use contracts with incentive provisions and results commitments by the contractor ; it also should reform the regulation of price adaptation to the economic conditions changes in using output indexes rather than input indexes.