Economics serving society

Collective production – do firms under-invest in quality and how does uncertainty on consumers’ evaluation of quality affect aggregate investment?

Fulvio Fontini, Katrin Millock and Michele Moretto

JPEG - 43.1 kb

Several products, in particular in the agri-food industry, share a common reputation. Prominent examples are French wines or Scottish whisky. Such common reputation can occur through the use of some input that is geographically delimited, such as terroir in wine-making, or specific climate conditions that favour a type of citrus fruit or yet still specific methods of production that have developed in an area over time. Examples are abundant in the agricultural sector: Florida grapefruits, Idaho potatoes, or Parmesan cheese. In these cases there is collective production and consumers cannot distinguish one particular individual producer from another. As a consequence, investment in quality becomes a public good – increases in quality benefit all firms through the collective investment in quality. The case of consumer demand being driven by an aggregate measure of quality can also occur when consumers care more about intrinsic characteristics of the production rather than the specific efforts of the firm per se. An example would be the case of consumer perceptions of the oil industry as ‘dirty’ in general, without distinguishing between individual firms’ investment in safety procedures of production. One frequent solution to the problem is to impose minimum standards of quality (1). Such a solution is still fraught with moral hazard problems and enforcement costs. In this paper, the authors model the collective game between firms using real option theory to analyse what incentives firms have to go beyond a minimum standard of quality.

In the model, quality is accumulated over time through a stochastic process. Consumer demand shifts according to average quality, but firms do not know consumers’ evaluation of quality, i.e., their willingness-to-pay for quality. The model combines these two sources of uncertainty to analyse the outcome of the dynamic game between firms in terms of level of investment in quality and the effects of information provision. The paper shows that in this case, the provision of information on consumers’ valuation of quality leads to non-trivial results.
The results show first that uncertainty on the accumulation of quality over time makes firms free ride and invest later on compared to the case when there is no uncertainty on quality. Incomplete information on behalf of firms on the willingness to pay for quality by consumers has a non-linear impact on firms’ investment in quality: the higher is consumers’ willingness-to-pay the more beneficial are the investments in quality to the firms, but the marginal effect is much smaller when consumers undervalue expected quality. It is as if firms are penalized more when consumers undervalue aggregate quality. There exists a minimum level of aggregate quality under which it is never optimal for firms to invest in quality, regardless of consumers’ valuation of quality. The results of the theoretical model thus support a result found in empirical studies of the wine industry with firms stuck in a “bad reputation – low quality” trap (2). When the aggregate quality is between this threshold and the standard trigger for investment, firms will never wish to pay for information on consumers’ actual willingness-to-pay for quality, even if its cost approaches zero. Furthermore, a third-party, such as the cooperatives that have been formed among producers in the agricultural sector – producer organisations - would never have incentives to release information on consumers’ willingness-to-pay for quality. This last non-trivial result stems from a « bad news » effect on firms’ investments in quality, where information would only lower the probability of firms investing in quality.
The authors do not allow for signalling by the firms but focus on the interactions between firms in the dynamic game using real option value theory. By doing so, they obtain analytical results extending several studies of collective reputation in homogenous goods all building on the seminal Tirole model of collective reputation (3). In this paper, consumer beliefs are not modelled explicitly, and hence not reputation as such, only collective production. The economic issue of free riding in the provision of a public good is the same though and the paper proposes a theoretical model that can be used for further analysis of the interaction between stochastic production and uncertainty on consumers’ valuation.

(1) Product state labels in the US, and the geographical identification and protected designation of origin labels in Europe
(2) Castriota and Delmastro (2015), in particular, find that firms can be stuck in a bad quality trap, under which only a positive random shock to aggregate quality (good weather conditions, for example) would induce more investment in quality.
(3) In particular, Winfree and McCluskey (2005) and McQuade, Salant and Winfree (2016).

Original title of the article : Collective Reputation with Stochastic Production and Unknown Willingness to Pay for Quality
Forthcoming: Environmental Economics and Policy Studies.
Available at:
© magicbeam -