The impact of local government fiscal gaps on public-private partnerships: government demand and private sector risk aversion

Abstract

While government fiscal gap is traditionally considered a demand factor for the use of public-private partnerships (PPPs) to deliver public services, a high level of fiscal gap may signal elevated financial risks to private partners and deter them from entering into PPP agreements. A causal mediation analytic framework is used to delineate the two distinct causal pathways. We develop a conceptual model and test derived hypotheses with data of Chinese prefecture-level cities during 2015–2017. The findings suggest that government fiscal gap has a positive impact on PPP adoption, through the mediating role of the debt position. The fiscal gap, as a risk factor, is negatively associated with PPP participation. Risk aversion of the private sector manifests more conspicuously as smaller PPP investment amounts than as a lower likelihood of PPP participation. The adverse effects of the fiscal gap associated with financial risks may entirely offset any positive impact.

Notes

Declaration of interest statement

No potential conflict of interest was reported by the authors.

Notes

1 Researchers use the term of societal capital organizations (shehui ziben) in China’s PPP context because, in addition to partnerships with private corporations and foreign businesses, governments also partner with state-owned enterprises that have various extents of public ownership. For a thorough review of engagement and influences of different types of societal capital organizations in China’s PPP projects, please refer to Xiong et al. (2021).

2 There are four centrally administered municipalities: Beijing, Tianjin, Shanghai, and Chongqing.

3 For a theoretical review of the Stata mediation package, please refer to Imai, Keele, and Tingley (2010); for a technical review, Hicks and Tingley (2011).

4 Values of simulated direct and indirect effects differ slightly from those derived directly from estimated coefficients. For example, in model 3, the simulated direct and indirect effect are respectively −2.371 and 0.861. But based on coefficient interpretation of the traditional linear mediation model, the direct effect is −2.362 (i.e., coefficient of the fiscal gap in the outcome equation) and the indirect effect is 0.859 [i.e., the product of coefficient of the fiscal gap in the mediation model (0.616) and coefficient of debt in the outcome model (1.394)]. The traditional coefficient multiplication method does not apply to non-linear models, such as model 1, so simulated effects are used and interpreted exclusively in this article.