Throughout history farmers have expressed concerns about ‘middlemen’ capturing ‘pipeline’ markets that lie between their farms and consumers. Traditionally they have focussed on transport and storage markets but, more recently, their concerns are being extended to markets for information and marketing technology. Early literature in this area describes monopoly storage arguments and the so-called ‘hoarding’ problem which involved capture of storage either through ownership of facilities or, indirectly, through control of credit. Australian farmers, like their overseas counterparts, have also expressed concerns about being exploited by powerful groups in the marketing chain, in particular wool growers their worries about concentration amongst wool buyers.
In 1962, the Philp Report claimed there was collusion in wool sales by a Japanese cartel which was held responsible for ‘pies’ in the auction market. ‘Pies’ are groups of buyers who act in concert at particular auction sales to hold down prices. This report was influential in the establishment of the Reserve Price Scheme (RPS) for wool in the late sixties. Although the official rationale for establishment of this scheme was that it would stabilise wool prices, many people believe that fear of price control by foreign buyers finally turned the tables in favour of establishment of the RPS in the protracted debate amongst wool growers that preceded the scheme. Decades later, when the microeconomic reform process reached the Australian wheat market, ‘market power’ or ‘middlemen’ arguments were again prominent in submissions from growers opposed to the Commission on reform of the Australian ‘single desk’ wheat selling system.
At what stage does market power influence our agricultural export markets? That is, at what point are there too few buyers to ensure farmers get fair prices? Unfortunately, there are no simple answers since the ability to exercise market power rests on a range of market characteristics. In principle, a market with only one buyer may result in competitive prices if it is contestable while a market with large numbers of buyers may be fragmented into many sub-markets with little substitution and thus be exploitable. Thus, in considering buyer power in the Australian wool market it is important to determine how the ‘pipeline’ operates towards the raw wool end of the market and then, based on this understanding, to examine buyer behaviour at the auction market to see whether wool growers are likely to get a fair deal.
Fears about the structure of the wool buying industry have probably stemmed from the erroneous belief that the industry can act as a single agent and obtain very large profits by acting simultaneously as a monopolist (sole seller) to wool processors and as a monopsonist (sole buyer) to wool growers. In these circumstances, the wool buying industry's 'best' strategy would be to restrict supply by holding grower prices down and then to use the resulting scarcity of wool to obtain high prices from wool users. However, there is published evidence that the structure of the buying industry does not resemble a monopoly (Hansen and Simmons, 1995). Rather, it resembles a type of price leadership, with a few dominant firms competing with a fringe of relatively small firms. Hansen and Simmons showed the buying sector of the Australian wool market had an inner core of about eight large firms purchasing about 50 per cent of offerings in competition with about 75 relatively small firms. However, they also showed that with over 100 000 lots of wool sold in Australian auctions each season, competition for particular lots could become 'thin'.
In a later study, Simmons and Hansen (1997) examined the effect of buyer concentration on wool prices using two key assumptions. First, they assumed there were no institutional or legislative barriers to entry to the wool buying industry. Entry to and exit from the industry reflected legitimate business costs facing the individual firms involved. Their second assumption was that dominant positions of large firms resulted from cost advantages related to size.
The empirical results of the Simmons and Hansen study showed increased purchases by large buying firms relative to small buying firms are likely to lead to increased grower prices and that reduced purchases by large firms are likely to have the opposite effect. This was unexpected, since an increased market share by ‘big players’ in wool buying was a expected to result in lower prices for growers. However, it was discovered that growers benefited from the presence of the ‘big players’ in the market because increased buyer concentration in the face of fixed short term supply caused small buyers to reduce their purchases and hence their purchasing costs, at the margin, declined. Reduced marginal costs then translated into increased price limits by all firms at auction through competition. This is contrary to the conventional wisdom that increased wool buyer concentration works against grower interests.
When the Simmons and Hansen study was undertaken in 1997 the wool auction system had nine selling centres divided into three regions: Northern, Southern and Western. Auctions operated on Tuesday, Wednesday and Thursday for about 44 weeks of the year with sales with combing and carding wools auctioned simultaneously at each centre. The minimum auction volume at any one centre was approximately 20,000 bales and maximum around 60,000 bales.
Data were obtained from the Catalogue Information Database of the Australian Wool Corporation and their sample comprised 73 204 lots of Merino fleece wool of best, good and average styles sold at Adelaide, Brisbane, Fremantle, Geelong, Goulburn, Launceston, Melbourne Newcastle and Sydney. The empirical model included a variable to measure buyer concentration incorporated into a hedonic price equation describing the determinants of prices achieved at auction. The variable chosen was the Herfindahl Index (H-Index). The index is bounded between zero and one:
where n is the total number of firms making purchases during the week, Xi is the number of bales bought by company i and X is the total number of bales bought. The index captures changes in buyer concentration between selling centres and between weeks. A significant positive coefficient on the index variable in the hedonic equation in which it was embedded would provide support for the theoretical result that increased buyer concentration increased grower prices.
The statistical results did indeed provide a positive coefficient on the Herfindahl index. This indicated that Simmons and Hansen’s theory was consistent with observed behaviour.
Simmons and Hansen concluded that the structure of the buying industry in the Australian wool market was a type of price leadership that entails a core of relatively large firms surrounded by a competitive fringe of smaller firms, and that such a structure should be attractive to wool growers by promoting efficiency while preventing big firms from monopolising the market. The relatively low costs associated with core firms were likely to keep inefficient firms out of the industry while competition from smaller companies was likely to prevent under-pricing by large firms for specific lots. However, Simmons and Hansen argue this result should be qualified. If the concentration of big buyers was to become high enough to threaten the competitive fringe of smaller companies, then exploitive pricing by core firms could occur. However, the structure of the buying industry, as measured by market shares of individual firms, has been remarkably stable over its recorded history and in 2003 there still appears to be little risk of reduced buyer numbers.
Hansen, P. and P.R. Simmons 1995, ‘Measures of buyer concentration in the Australian wool market’ Review of Marketing and Agricultural Economics 63(2), 304-310.
Simmons, P.R. and P. Hansen, 1997, 'The effect of buyer concentration on prices in the Australian wool market', Agribusiness, 13(4), 423-430.