Public–private partnerships in the US and Canada: “There are no free lunches”1

1. This article builds on five cases that are forthcoming in Boardman, Poschmann and Vining (2005). A sixth case has been added as well as the US prison study. The theoretical framework is largely new.

Abstract

Governments in many industrialized nations have made concerted efforts to reduce their immediate expenditures and to reduce the cost of major infrastructure projects. Public–private partnerships (P3s) are one emerging method that might do so. Despite the increased use of P3s, there is little independent research on the effectiveness of P3s as a public policy instrument. This article considers the major rationales for P3s, including cost savings and keeping project financing off government budgets. It then presents a transaction cost model that suggests that P3s can often be prone to conflict, high contracting costs, opportunism and failure. Evidence from six major infrastructure projects and a summary analysis of US prisons is then presented. These cases confirm that contracting costs have been high, as predicted by the model. Specifically, high contracting costs reflect the presence of complexity/uncertainty, asset specificity, the potential for ex post bilateral opportunism and a lack of contract management skills by governments. Given these circumstances, the private sector can behave opportunistically at the expense of the public sector as there has sometimes been a political imperative to prevent projects from terminating. Public partners have also behaved opportunistically after projects are in place. Unless public sector managers recognize that they must design contracts that both compensate private sector partners for risk and then ensure that they actually bear it, P3s have little chance of being efficient or effective service delivery mechanisms.

Key words:

• Public–private partnerships
• transaction costs
• US
• Canada
• opportunism
• asset specificity
• complexity
• uncertainty

Acknowledgement

We would like to thank Michael Volker for research assistance.

Notes

1. This article builds on five cases that are forthcoming in Boardman, Poschmann and Vining (2005). A sixth case has been added as well as the US prison study. The theoretical framework is largely new.

2. For example, the US General Accounting Office includes conventional contracting out of government services and even privatization – the complete withdrawal of government provision and financing – as P3s (USGAO 1999). Additionally, the USGAO has treated non-profit entities as being “private” sector entities in P3s (USGAO 2004).

3. Specifically, we do not include the following relationships as P3s: (1) service contracts or other forms of contracting out by the public sector; (2) privatization in the form of the sale of public assets; (3) regulation (including franchise contracting) by public sector entities of privately owned natural monopoly facilities; or (4) the construction of facilities by the private sector and the leasing or sale of those facilities to the public sector based upon fixed, certain terms (including lease/purchase or turnkey agreements).

4. Governments cannot borrow infinite amounts of capital without affecting their credit rating. Raising funds for a P3 project may raise the cost borrowing for subsequent projects. Such costs should be included in the “full” cost of the P3.

5. The transaction cost language is more appropriate than agency language because P3s have the character of a relationship between independent organizational entities. Agency, or principal–agent theory, language is appropriate for intraorganizational hierarchical contexts.

6. Vining and Weimer (2005) distinguish between ex ante transaction costs, which can be called governance costs, and ex post transaction costs, which can be called opportunism costs or holdup costs.

7. Hall (1998) quotes the Chief Financial Officer of the private firm that operated the road as saying: “We haven't made any debt payments in so long I've forgotten how much we owe now.”

8. Assembly Bill 680. This section primarily draws on USCBO (1997) and USGAO (2004).

9. Chem-Security said the reasons for this included generators' pursuit of lower-cost options for waste disposal (NRCB 1994: 6–8).

10. If Chem-Security and BOVAR could have earned profits higher than the guaranteed rate of return, they would have had an incentive to control costs. However, Mintz (1995: 33 and Appendix) shows that even with some positive probability of profit, they have an incentive to over-invest.

11. Mintz (1995) estimates a weighted return on equity of 15.9 per cent for the period 1989 to 1994, far above the risk-free return.

12. This section draws on Poschmann (2003).

13. Note that the logic is flawed. The province's taking on of the financing necessarily brought risks and costs not featured in the government's analysis (de Bettignies and Ross 2004).

14. Per http://www.407etr.com (accessed August 28, 2004).

15. This discussion follows Loxley (1999).

16. This was the Auditor General of Canada's conclusion, and the government did not ultimately succeed in keeping the financing off-book (Receiver General for Canada 1995).

17. Nonetheless, the USGAO (1996) concluded that the evidence on cost savings was “inconclusive.”

Additional information

Notes on contributors

Finn Poschmann

Aidan Vining is the CNABS Professor of Business and Government Relations in the Faculty of Business Administration, Simon Fraser University, Vancouver. His research interests focus on public policy analysis and business strategy. Anthony Boardman is the Van Dusen Professor of Business Administration in the Strategy and Business Economics Division at the University of British Columbia. His research interests include privatization, contracting out and cost-benefit analysis. Finn Poschmann is Associate Director of Research, C.D. Howe Institute. His research interests include finance and intergovernmental fiscal relations.