Part I
The structure of vertical relationships
Chapter one
Consumersā behaviour and producerādistributor relationships in convenience goods markets
Luca Pellegrini
Introduction
Unlike the much oversimplified description of standard textbooks, the relationship between producers and consumers is not direct but filtered by retailers, and often by other distributive agents. The question therefore arises whether such filtering affects the producerā consumer relationship and, if so, in which way. There seems to be no doubt that retailers do affect the producer-consumer relationship as there is no guarantee that the objective function of producers, selling their goods, is always compatible with the objective function of retailers, selling their services jointly with an assortment of goods. The development of large retailers, and the consequent increasing concentration of retailing, has made producerādistributor (PāD) relationships more and more central to the understanding, not just of marketing practices, but also of conduct and performance of consumer-goods industries.
There is much literature on the many aspects of vertical relationships in marketing channels. This literature is also very heterogeneous, aiming at different problems from different points of view (Gattorna 1978). Broadly speaking, the most common approach in marketing literature seems now to be centred on agentsā behaviour within channels. Channel structure and performance are analyzed as the result of channel membersā behaviour, in its turn a function of the underlying distribution of power among agents. In the tradition of industrial economics, the attention, with respect to convenience goods, is mainly focused on the aggregate effects of buying power, and to its welfare implications, while otherwise the literature is centred on vertical integration, either outright or through contractual arrangements, and it is mainly relevant for non-convenience goods.
Besides these two main streams of literature, a number of contributions, which were originally centred on the role of advertising, have been developed from a starting point somewhere in between these two areas of research, drawing both from marketing and industrial economics. The main purpose of this paper is to review the literature originated from this middle ground, trying to improve its heuristic power through the unifying role assigned to consumersā buying strategies.
The literature I shall refer to can be principally associated to four names, that is, Robert Steiner, Michel Porter, Mark Albion and Giampiero Lugli.1 Steiner, in particular, began to work in this direction almost fifteen years ago, and his ideas have been surprisingly neglected until the recent work by Farris and Albion (1980) and Albion (1983). However, they are essential to the understanding of PāD relationships and of the overall structure of consumer goods marketing channels. He seems to have been the first (Steiner 1973) to point out the role of advertising in conjunction with horizontal competition in retailing, a starting point for a far richer set of propositions he himself developed in later works (Steiner 1978a, 1978b, 1981, 1984). Porter (1974, 1976) also stresses the importance of the role played by retailers and the need to consider them when analysing industries along the classical paradigm of structure, conduct, and performance. He also introduces the definition of two main types of goods, convenience and non-convenience, a distinction which largely arises from the different way they are treated at the retail level. Albion and Farris build on both Steinerās and Porterās works, developing a detailed model of the interaction of producers and retailers. Their starting point is the role of advertising and its āhidden effectā, that is, the negative correlation between advertising and retailersā margins. Lugli (1976, 1984) proceeds along similar lines, but his analysis is centred on retailers. He shows how their behaviour affects producers, and he particularly stresses the destabilizing effect of retailersā strategies on the structure of consumer goods industries.
As a whole, the contributions of the four authors quoted, and of others which shall be quoted later on, provide all the elements for the development of a coherent model of PāD relationships in convenience goods markets. However, each author tends to analyse the same problem from a different perspective. Steiner and Porter are possibly more prone to assume producersā points of view than the retailersā, while Albion and Lugli do the opposite. Hence, it is probably worthwhile to try and find a common ground from which to reconsider their findings and give them unity. In the writerās opinion this is provided by consumersā behaviour, which is most often loosely brought into the picture, while it is at the centre of both producers and retailers sales efforts: the former try to increase brand loyalty against the same effort by the latter to increase store loyalty.
The aim of this paper is therefore to reconsider the existing literature using as a starting point producersā and retailersā perceptions of consumersā behaviour, and showing how the complex structure of marketing channels in convenience goods markets arises from these different perceptions.
As already stated, the analysis is strictly limited to convenience goods. Their main characteristic, as far as PāD relationships are concerned, is that consumers want to concentrate their purchases as much as possible. Retail assortments are therefore very wide, spanning over a large number of product categories, which makes either outright or contractual integration by producers virtually impossible. This, in turn, is why channel relationships in convenience goods markets differ from those of non-convenience goods, and why they should be analysed separately. In the first case producersā marketing strategies have to be defined recognizing the existence of independent and powerful retailers, whereas in the second case vertical agreements can be enforced to subject retailersā objective functions to those of producers (franchising, exclusivity, royalties, resale price maintenance, and so on, and their possible combinations, see, for example, Blair and Kaserman 1983).
The paper will be organized as follows. In the next section the environment which we refer to in the analysis is briefly sketched. Then consumers, retailers and producers are considered, showing the interrelationships of their behaviour. Finally, in the last section some general remarks are made, indicating issues worth further research.
Setting and sitters
The environment in which our characters act must be made explicit. Here a number of problems arise as to the complexity of such an environment. First, PāD relationships evolved together with the development of large retail chains, and, more generally, parallel to the so-called retail revolution. Even if we could assume a single path of evolution of retailing, PāD relationships should be described as a process and not as a static set of propositions. Second, there does not exist one particular structure of consumer goods industries, as they vary within a wide range of possible characterizations. Since the scope of this paper is limited to the description of the main features of PāD relationships and to their rationale, some assumptions are needed to reduce the complexity of reality to a manageable framework.
The first simplification involves the omission of all references to wholesalers: to consider their role would add to the complexity of the analysis without adding to the core of relationships on which this paper will be focused.
Next, something must be said about the structure of consumer goods industries and of retailing. The characterization of retailing seems the easiest, because the industry has developed along similar patterns in most countries. As far as convenience goods are concerned, when the retail revolution is completed. most researchers would maybe agree to the following characterization of retailing:
- retailing is differentiated into a number of shop types offering a different mix of services which are substitutes for one another; chain stores dominate the market, while small independent stores are confined to niches justified by the convenience they offer to consumers;
- although the number of chains competing in each shop-type segment of the market is small, competition is generally fierce. This is because:
(a) the market grows very slowly and market shares can be increased only at the expense of other firms;
(b) different shop types offer services which are close substitutes;
(c) within the same shop type product differentiation does not go very far and price remains the main competitive weapon;
(d) firms within the industry are organized in very different forms (multiples, voluntary chains, co-operatives) which make collusion more difficult;
- retailing is an unstable industry as retail innovation spreads very fast: its organizational nature makes it difficult to protect the initial advantage, and the industry is therefore periodically brought into periods of turbulence when an innovative form of selling gains market shares at the expense of the pre-existing ones.
To characterize consumer goods industries is much more difficult. For what we are going to say, we need a description of groups of firms operating on narrowly-defined product categories, and at this level the typology of market structures is very wide. In most countries a common feature of the different segments of consumer goods industry is their oligopolistic structure, but this in itself does not say very much. To shift on someone elseās shoulders the responsibility of a generalization, I shall keep to the following quotation from Caves and Porter (1977: 251): āA typical pattern in consumer goods industries is the presence of a small group of producers of a full line of nationally branded goods and a larger group of producers of unadvertised goods, regionally branded goods, and producers for private labelsā. Thus there are basically two groups, a group of firms producing leading brands with substantial market shares, and a fringe of producers with a lower degree of consumer franchise. Since we shall refer to narrowly-defined markets, firms need not be present in just one of them. Large firms are present in many markets, but they need not be in the same group in all of them. They will generally operate in the leading group, but in some markets they may also be in the fringe, especially if entering in the fringe is the first stage of a strategy of entry into a new market (Caves and Porter 1977). Therefore, the fringe does not necessarily include only small firms.
For the moment I shall not add more to this, leaving until the last paragraph some additional considerations on the interaction between horizontal competition within and betwe...