Large‐Scale Risks and Technological Change: What About Limited Liability?

• Author: JULIEN JACOB,SANDRINE SPAETER
• Date: 01 February 2016
• DOI: http://doi.org/10.1111/jpet.12123
• Published date: 01 February 2016

LARGE-SCALE RISKS AND TECHNOLOGICAL CHANGE:

WHAT ABOUT LIMITED LIABILITY?

JULIEN JACOB

University of Lorraine

SANDRINE SPAETER

University of Strasbourg

Abstract

We consider a firm under strict liability that must choose between two

risky technologies, one being safer but costlier than the other one.

The total potential level of damage increases with the level of activity.

We show that, under limited liability, technological change is welfare-

improving and leads to full risk internalization when the firms are suffi-

ciently capitalized. Nevertheless, the percentage of firms adopting the

safer technology and full risk internalization is higher under unlimited

liability than under limited liability. We show how an adequate tax pol-

icy increases this percentage. We also determine the characteristics of

a second-best tax policy.

1. Introduction

Limiting the firm’s liabilities to its net value helps it to innovate and to levy funds for

research and development (R&D) by fostering a reassuring legal and economic environ-

ment. Nevertheless, it is common knowledge, since Shavell (1986), that a limited liability

regime may also induce either partial risk internalization by firms or over-investment in

risky activities. Indeed, and contrary to an unlimited liability regime, limited liability

creates an ex post payment ceiling for the firm: it will have to pay for the damage only up

to its net present value, and the firm’s shareholders do not have to compensate the vic-

tims through their own funds in the case of a huge accident. Thus some firms and their

shareholders may carry out projects driving excessive risk, some part being transferred

to society.

In order to countervail these negative effects, some works have focused on a regime

that extends the financial liability to the operators of a liable firm. An operator (a bank,

Julien Jacob, BETA, CNRS, and University of Lorraine (julien.jacob@univ-lorraine.fr). Sandrine

Spaeter, BETA,CNRS, and University of Strasbourg (spaeter@unistra.fr).

We thank Claude Fluet, Bruno Deffains, and Andr´

e Schmitt for comments on earlier versions. We

are also grateful to two anonymous referees for their comments and useful suggestions. Participants in

the Strasbourg-Nancy BETA-Seminar,in the Paris 1 Seminar on Environmental Economics and Natural

Resources, and in the Maastricht UNU-MERIT joint Seminar are also acknowledged.

Received January 26, 2014; Accepted February 27, 2014.

C2014 Wiley Periodicals, Inc.

Journal of Public Economic Theory, 18 (1), 2016, pp. 125–142.

125

126 Journal of Public Economic Theory

for instance) will have to pay for extra damage if the firm goes bankrupt.1Such a regime

should increase the available funds for compensation. But it may also lessen the incen-

tives of firms to invest in prevention, for their financial liability is transferred to the

other operators (Beard 1990; Pitchford 1995; Boyd and Ingberman 1997; Boyer and

Laffont 1997; Dionne and Spaeter 2003; Martimort and Hiriart 2006).

Some other economists and some jurists advocate for unlimited liability from a nor-

mative point of view (Halpern, Trebilcock, and Turnbull 1980; Hansmann and Kraak-

man 1991; Faure 1995): the potential injurer must always pay for all the damage caused

by its activity, so that risk is fully internalized. However, it can be complex to implement

it in practice. Indeed Alexander (1992, p. 389) writes: “To the extent that the proposal

could constitutionally be implemented, enforcement would raise substantial procedu-

ral obstacles.” For some practical reasons, the law could even be unenforceable (think

about out-of-state shareholders who must be identified) and implementation costs could

be higher than the net benefit for society.2In the particular case of large-scale (or catas-

trophic) risks, the efficiency of unlimited liability in terms of risk internalization can

still be put in doubt: the shareholders’ assets are not always sufficient to compensate the

whole damage. Besides, it can take a very long time before a legal court obtains all the

information and takes a decision about who should pay and how much. Hence limited

liability may not be welfare-inferior if all costs are taken into account.

The questions of technological change incentives can be split into two issues: the

incentives to design a new technology through the process of R&D, and the incentives to

adopt a new technology once developed (process of diffusion). A large number of papers

focus on the R&D process (and firms’ imitation abilities; see Parry 1995 and Fischer,

Parry, and Pizer 2003), or on pollutant technologies (Magat 1978, 1979). Diffusion is

also widely investigated. Downing and White (1986) and Milliman and Prince (1989)

are often referred as being the first papers having formally analyzed the issue of the

diffusion process of a green technology in the framework of pollution control.3All of

these analyses take the liability context as given, although the liability rule at stake does

play a significant role in the decision process as showed earlier by Shavell (1980, 1986).4

To our knowledge, the first studies that analyze the incentives induced by some lia-

bility rules to adopt safer technologies are provided by Endres and colleagues (Endres

and Bertram 2006; Endres, Bertram, and Rundshagen 2007, 2008; Endres and Friehe

2011b). Nevertheless, these authors consider that the firm is always able to fully compen-

sate the damage: the issue of partial risk internalization due to “judgment-proofness” is

put aside.

In this paper, we compare two liability regimes in terms of incentives to adopt asafer

(already designed) technology. We show that strict but limited liability does not always

lead to less risk internalization or poorer incentives for technological change compared

to strict and unlimited liability. We consider that the risk driven by the activity of the

firm is simultaneously affected by its technological choice and its level of production.

1See Klimek (1990) for some U.S. practical cases.

2See also Easterbrook and Fischel (1985) and Grundfest (1992).

3See Requate (1998), Requate and Unold (2003), Coria (2009), and Sanin and Zanaj (2011) for an

adoption process analysis when firms interact.

4Actually, liability and incentives to design/adopt a new technology were first studied in the context

of product liability (Viscusi and Moore 1993; Daughety and Reinganum 1995; Baumann, Friehe, and

Grechenig 2011). Endres and Friehe (2011a) consider non-point source pollution and analyze the

impact of liability rules on technological change. Product liability and non-point source pollution go

beyond the scope of this paper.