Demand risk transfer and government's cost efficiency: Focusing on Korean waste treatment PPP cases.

• The fiscal efficiency of waste treatment PPP was studied. • The impact of demand risk taking on the fiscal expenditure was discussed. • VfM and sensitivity analysis was performed by using Korean cases. • Cost structure and demand risk should be considered for fiscal efficiency of PPP. This study aimed to examine the government's cost efficiency considering the high-risk/high-return mechanism of PPP. Faced with increasing demand but with limited budget, the Korean government has relied on the Public-Private Partnership (PPP) to provide waste treatment services for the last couple of decades to expand fiscal space. However, most of waste treatment facilities projects have been promoted using the BTO (Build-Transfer-Operate) method with high rate of return due to the demand risk that is transferred to the private. We performed a conversion analysis of a BTO to a BTL (Build-Transfer-Lease) method, in which demand risk is borne by the government, for 18 PPP waste treatment cases with actual operation records. The result of comparing the life-cycle costs of government in employing each project method shows that the BTL can provide 5.26% of Value for Money (VfM) compared to the BTO as the government takes the demand risk and lowers the required rate of return of the private. This implies that transferring the demand risk to the private sector may not always be the best option for the government. From the government's perspective, instead of transferring the demand risk to the private and providing a high rate of return, the government can retain the demand risk and reduce the rate of return, and it can be fiscally more advantageous considering the cost structure of each PPP method. Through Korean PPP waste treatment cases, this study suggests that policy makers who implement PPP should consider the government's strategic risk sharing by understanding the predictability of demand and the nature of cost structure of each PPP method. [ABSTRACT FROM AUTHOR]