The irrigation and drainage sector (I&D) is facing a variety of issues, the most controversial being poor water efficiency, big public contribution, lack of asset maintenance, and often socio-economic inequity. Numerous attempts have been made to try to find a way out of this vicious circle, and lately the idea of trying to involve the private sector through public-private partnership (PPP) has been growing, nurtured by encouraging ''pilot'' experiences.Recent decades have shown some concern and effort to improve the performance of the sector, with the focus switching from on-farm improvement, to farmer involvement through participatory irrigation management and irrigation management transfer, and to quality service-oriented organisations. PPP should really be viewed as a logical sequence to these successive efforts, with a differentiated treatment of the various functions of I&D projects: private participation hoped for in investment, the governance function strictly public, OMM functions (operation, maintenance and management) wide open to private sector participation, while the agricultural production function remains entirely private.These theoretical conclusions were tested with 21 examples of existing (15) or projected (6) PPPs in I&D, analysed and modelled in terms of demand, offer, content, contracts and risks, with the following main findings: (i) the PPP demand is mostly a government initiative wishing to reduce recurrent public subsidies; (ii) service providers are more reactive than proactive; (iii) while 2/3 of the case studies contracts concern private participation in one or more of the investment functions, 90% of the contracts concern private participation in one or more of the OMM functions; (iv) service (and management) contracts account for 13% of the total, whereas public service delegation (PSD) accounts for 81% of the contracts; and (v) while investment PPPs are understandably more sensitive to country risks (devaluation), PSD arrangements are just as understandably more averse to commercial risks (fee collection from farmers) than service or management contracts.Last but not least, the important point is not so much to find an ''absolutely private'' partner but rather a professional ''third party'' between farmers and government, whether it be public (e.g. a reformed and financially autonomous government agency) or private (e.g. a private service provider looking for business or a WUA turning into a private corporation Les dernières décennies ont montré la réalité de ces efforts pour améliorer les performances du secteur, l'accent étant d'abord mis sur les améliorations à la parcelle, puis à l'implication des irrigants à travers la Gestion Participative de l'Irrigation (Participatory Irrigation Management, PIM) et le Transfert de Gestion de l'Irrigation (Irrigation Management Transfert, IMT), puis enfin à la recherche de la qualité dans le service de l'irrigation. Le PPP doit être considéré comme une suite logique de ces efforts successifs, avec un traitement diversifié des différe...