HOW TO DESIGN INFRASTRUCTURE CONTRACTS IN A WARMING WORLD: A CRITICAL APPRAISAL OF PUBLIC–PRIVATE PARTNERSHIPS*

• Author: David Martimort,Stéphane Straub
• Published date: 01 February 2016
• Date: 01 February 2016
• DOI: http://doi.org/10.1111/iere.12148

INTERNATIONAL ECONOMIC REVIEW

Vol. 57, No. 1, February 2016

HOW TO DESIGN INFRASTRUCTURE CONTRACTS IN A WARMING WORLD:

A CRITICAL APPRAISAL OF PUBLIC–PRIVATE PARTNERSHIPS∗

BYDAVID MARTIMORT AND ST´

EPHANE STRAUB 1

Paris School of Economics, EHESS, France; Toulouse School of Economics, ARQADE, IDEI,

and IAST, France

We analyze how long-term uncertainty, for example, regarding future climate conditions, affects the design of

concession contracts and organizational forms in a principal–agent context, with dynamic moral hazard, limited liability,

and irreversibility constraints. The prospect of future, uncertain productivity shocks on the returns on the firm’s effort

creates an option value of delaying efforts, a course that exacerbates agency costs. Contracts and organizational forms

are drafted to control this cost of delegated flexibility. The possibility for the agent to delay investment in response to

uncertainty and irreversibility also elicits preference for unbundling different stages of the project through short-term

contracts. Our analysis is relevant to infrastructure sectors that are sensitive to changing weather conditions and sheds

a pessimistic light on the relevance of public–private partnerships in this context.

1. INTRODUCTION

Climate change raises specific and unprecedented challenges surrounding large investments in

infrastructure. The transport, water, and energy sectors are typically characterized by significant

sunk investments and technological choices that are locked in over several decades. Over such

periods, changing weather conditions are expected, albeit unpredictable. Certainly, the current

scientific uncertainty regarding future climate conditions, potentially coupled with the increasing

likelihood of very large catastrophic events, makes long-term decisions in such sectors quite

a challenge for practitioners.2Climate change might have significant implications for future

investments and maintenance of existing assets in key sectors, which in turn might affect the

nature of upfront investments.

Since they involve long-term contracts that often span twenty or thirty years, the so-called

public–private partnerships (PPPs) are particularly sensitive to climate change hazards. Over

recent years, a number of countries, including the United States and several European and

emerging countries, have increasingly relied on PPPs to respond to their investment needs in

sectors involving long-lived capital stock (LLKS).3Because it entails delegation to the pri-

∗Manuscript received September 2013; revised October 2014.

1We are extremely grateful to Marianne Fay and Daniel Benitez for initiating this research and providing much

detailed comments on a previous version and to PPIAF for providing financial support. An earlier version of this article

was circulated under the title “How to Design Public–Private Partnerships in a Warming World? (When Infrastructure

Becomes a Really Hot Topic).” We thank Perrin Lefebvre for outstanding research assistance and Antonio Estache,

Philippe Gagnepain, Paul Grout, Elisabetta Iossa, Margaret Leighton, St´

ephane Saussier, and Carine Staropoli for very

detailed discussions and challenging views. Two referees and the editor offered important comments that helped us

to improve exposition. Finally, discussions with seminar and conference participants at Tilburg University, IAE Paris,

Maastricht, and the conferences on “Contracts, Procurement and Public-Private Agreements” in Paris (May 2011), The

Economics of Public–Private Partnerships in IESE Barcelona (April 2012), and the APET Workshop in Brisbane (June

2012) were highly appreciated. All errors remain ours. Please address correspondence to: St´

ephane Straub, Toulouse

School of Economics, Universit´

e Toulouse 1 Capitole, 21 all´

ee de Brienne, Bat. F, 31000 Toulouse, France. E-mail:

stephane.straub@tse-eu.fr.

2Weitzman (2009) and Hallegate (2009).

3See, for instance, the exhaustive evidence and discussion in Engel et al. (2008) and Estache and Iimi (2011) and the

statistics reported by Engel et al. (2013) on the growing importance of PPPs in Europe and in the United States, with a

fivefold increase between 1998–2007 and 2008–2010.

61

C

(2016) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

62 MARTIMORT AND STRAUB

vate sector of key decisions over both the structure of initial investments and the subsequent

management of assets over a long period, this contracting mode has also been viewed as an at-

tractive response to important shortages in public funds. Beyond this public finance motivation,

the efficiency gains of PPPs have also been repeatedly emphasized in contexts in which rela-

tionships between public bodies and the private sector are plagued with agency costs, contract

incompleteness, and transaction costs, as pointed out by a burgeoning body of literature.4

Equipped with a model of long-term contracting tailored to the specificities that climate

change brings to the agency relationship between public bodies and firms, this article aims to

analyze the suitability of the standard PPP model in coping efficiently with climate change-

related uncertainty. In a nutshell, we argue that long-term contracting is plagued with new

agency costs of delegated flexibility that might be better controlled when parties wait until

uncertainty regarding climate conditions becomes verifiable to draft new arrangements. The

possibility for the agent to delay investment in response to uncertainty and irreversibility also

calls for unbundling different stages of the project (say building and maintenance, for instance).

This point thus leads to a rather pessimistic view of the benefit of PPPs and long-term contracting

in that context.

Although we chose to cast our contribution in the context of PPPs and climate change-related

uncertainty, as there is an important theoretical literature using similar modeling tools, and this

is also an area of considerable policy relevance, our results speak more generally to the issue

of long-term contracting relationships in contexts of large future uncertainty. We point in the

literature review below and in the final discussion section to some other areas where our results

may be relevant.

Climate change and infrastructure investment. The current process of anthropogenic climate

change, characterized by large and growing uncertainty of future values of environmental

parameters, will dramatically affect the environment in which long-term economic decisions

are made. Indeed, global scenarios related to climate change make it difficult to pinpoint future

outcomes more precisely than broad probability distributions and to rule out disastrous collapses

(IPCC, 2007).

Growing evidence points to a link between anthropogenic greenhouse gases concentration

and local extreme events, such as heat waves, floods, and precipitation.5However, uncovering

the exact channels and providing precise future projections appear to be beyond current sci-

entific possibilities.6As a result, climate risk assessments for specific businesses such as utility

providers are severely limited by the coarse spatial resolution of climate models and the ensuing

lack of clear understanding of how global climate change translates at the local level (IFC, 2010).

Climate change-related hazards are especially relevant to infrastructure on several fronts.

First, the accelerating rate of climate change implies that long-lived investments will have

to cope with a broader range of climatic conditions during their lifetime. Power plants are

typically built to last for at least 30 to 40 years and energy distribution networks and water

and transportation infrastructures for 30 to 100 years.7Such assets are thus likely to experience

large variations in average temperature conditions, precipitation, etc., over their life cycles.

This issue is particularly acute for developing countries, where there are both important needs

for infrastructure investment and wide expectation that the impact of climate change will be

stronger (World Development Report, 2010).

Second, the very nature of infrastructure investments implies crucial sensitivity to climate

hazards. Water collection and distribution networks as well as hydroelectric power plants8are

4Benz et al. (2001), Bennet and Iossa (2006), Hart (2003), Iossa and Martimort (2015), Martimort and Pouyet (2008),

Engel et al. (2010), among many others.

5See Stott et al. (2004), Kunreuther and Michel-Kerjan (2009), Pall et al. (2011), and Min et al. (2011).

6Stone and Allen (2005), Piao et al. (2010).

7Shalizi and Lecocq (2009), Hallegate (2009).

8See the case of the Brazilian Belo Monte dam, world’s third largest projected dam. http://www.globalpost.com/

dispatch/brazil/110203/drought-belo-monte-dam, last consulted February 23, 2011.