An Examination of the Market Structure of the U.S. Produce Industry

Recent literature, largely from the U.S. Department of Agriculture, Economic Research Service, indicates that substantial changes have occurred in the produce industry in recent years. With the rise of retail mass merchandisers and increased concentration in the retail food industry, the procurement power of these large firms reportedly has also increased. With direct buying and contracting, market intermediaries such as brokers and wholesalers allegedly are being bypassed. As a result, these market intermediaries ostensibly are also consolidating becoming fewer and larger with increased emphasis on servicing the food service industry. However, the findings of this study indicate that there is no convincing evidence that the market structure of the U.S. produce industry has markedly changed since the early 1980s. While supermarket concentration has increased noticeably, the same cannot be said for produce market intermediaries such as brokers and wholesalers. 1 An Examination of the Market Structure of the U.S. Produce Industry Marketing channels for fresh fruits and vegetables in the United States are distinctly different from those for other agricultural commodities. The primary marketing channel is composed of three major echelons: shipping-point markets, wholesale terminal markets, and retail markets. Shipping-point markets are located where the fruits and vegetables are grown. Firms in these markets include grower-packers or packer-shippers, brokers, wholesalers, and integrated wholesaler-retailers. Produce moves from the shipping points to export markets, wholesale terminal markets, and integrated wholesale-retail destinations. At terminal markets, wholesalers take possession of the fruits and vegetables, commonly repackage them under their own labels, and distribute to retail food stores and the food service sector which includes entities such as restaurants, hotels, and institutions. Examples of institutions are schools, hospitals, prisons, and the military. Merchant wholesalers and brokers/commission agents are the major players in terminal markets. The retail market includes food stores, restaurants, and direct fruit and vegetable markets (Calvin et al. 2001; Kaufman et al. 2000). In recent years, the produce marketing channel has purportedly become increasingly vertically integrated. Vertical integration refers to control over upstream suppliers or downstream procurers. This control can occur through ownership or contracts. In the case of fruit and vegetable markets, the industry is becoming more backward or upstream coordinated. Large supermarket chains in many cases have bypassed brokers and wholesalers, dealing directly with grower-packers often under contract (Calvin et al. 2001; Handy et al. 2000; Kaufman et al. 2000; Patterson and Richards 2003; Sexton, Zhang, and Chalfant 2003). 2 The market relationships among produce shippers, wholesalers, and retailers apparently have undergone substantial changes in recent years. Such changes first came to light in 1999 when the Federal Trade Commission and the U.S. Senate held hearings to discuss the effects of recent retail changes and trade practices on different industries, one of which was the produce industry. The hearings reflected all sides as shippers complained that wholesalers and retailers had too much market power while retailers and wholesalers indicated that new practices, such as fees and services, are just the cost of doing business in the current market environment (Calvin et al. 2001). Calvin et al. (2001) reported on the evolution of market channels from growers to retail buyers. The growth of mass merchandisers was linked to the decline in sales of produce to conventional retailers. Glaser, Thompson, and Handy (2001) studied the effects of retail consolidation along with changing customer preferences on shipper-wholesaler relations. Further, there has been some empirical work on market channel performance. Richards and Patterson (2003) studied the produce buying power exercised by larger retailers. They found that retailers are able to wield both buying and selling market power to some degree. Sexton, Zhang, and Chalfant (2003) studied the extent to which retailers are able to exercise oligopsony power in the procurement of fresh produce. There has also been research conducted on the use of contracts to increase vertical integration. MacDonald and Korb (2006) looked at the different types of contracts used to organize the production and use of agricultural commodities. Marketing contracts are the major type of contract used in the fruit and vegetable industry. Over half of fruit production and almost one third of vegetable production are produced under this type of contract. The actual use of 3 contracts for produce did not increase between 1994 and 2003. Regarding financial indicators, research has been carried out on how financial ratios change and how these changes reflect profitability and leverage. Chen and Zhao (2007) showed why more profitable firms have lower leverage ratios (more solvent) and how these ratios are affected by the choice of using internal or external funds for financing. The study concluded that the negative relationship between profitability and leverage ratios is a result of larger firms using internal instead of external funds. With the rise of retail mass merchandisers and increased concentration in the retail food industry, the procurement power of these large firms reportedly has also increased. After 1996 many retailers merged, leading to an increased percentage of sales by the top eight retailers – from 1987 to 2002 the increase was from 27 to 45.6% (Calvin et al. 2001; U.S. Department of Commerce, Bureau of the Census of Retail Trade 2005). With direct buying and contracting, market intermediaries such as brokers and wholesalers reportedly are increasingly being bypassed. As a result, these market intermediaries ostensibly are also consolidating becoming fewer and larger with increased emphasis on servicing the food service industry. If produce market intermediaries are adversely impacted by increasing oligopsony power of retail supermarkets, the evidence should manifest in financial time-series data representative of the impacted firms. Thus, a test of the adverse economic impact of oligopsony on upstream firms in the produce industry is possible through an analysis of the financial and census data representative of the impacted firms over the relevant time period. The overall purpose of this study is to determine whether produce wholesalers and brokers have been adversely impacted by the increase in mass merchandisers and retailers in the food industry. 4 The next section is devoted to an explanation of the financial ratios used in the analysis. Data, econometric analysis, and results are presented by financial standing of the firm. Next, census data are used to ascertain changes in market structure by type of wholesale firm over time. Lastly, conclusions are provided. Financial Ratios Ratios are one of the most important methods for analyzing the financial standing of a company. They provide managers and lenders ways to see relationships among items on financial statements. Two major sources of financial ratios are the Dun and Bradstreet Credit Services Industry Norms and Key Business Ratios (D&B) and Robert Morris Associates Annual Statement Studies (RMA). Robert Morris Associates is an organization of bank loan officers who obtain sample financial statements from over 375 industry groups subdivided by SIC code and more recently, NAICS code. Key ratios for two categories, solvency and profitability, are used in the analysis. An exact match of ratios from the two major published sources – D&B and RMA – was not possible. The ratios employed were matched as closely as possible. The D&B solvency ratio used in the study is Total Liabilities to Net Worth. The RMA ratio is the same but called Debt to Net Worth. Two sets of profitability ratios are used in the study, return on assets (ROA) and return on net worth (RONW). The D&B measures are after taxes while the RMA measures are before taxes, table 1. Analysis of Financial Ratios by Level of Financial Standing In order to test the hypothesis that the profitability and solvency of produce 5 wholesale intermediaries (agents, brokers, and merchant wholesalers; SIC code 5148 and NAICS code 422480) have declined in the face of increased procurement power of mass merchandisers, time-series date for the selected financial ratios are used. Ratios were gathered for a 20-year span (1984-2003) which covers the period over which structural changes reportedly have occurred in the produce market channel. The D&B data are based on financial reports from within given industries and span the same period. The categories employed are based on financial standing: upper, median, and lower quartiles. The median is the midpoint of all companies in the sample. The upper and lower breakdowns represent the midpoint of the upper and lower halves. The upper quartile encompasses strong ratio values, while the lower quartile encompasses weaker ratio values (Robert Morris Associates 1984-2003, Dun and Bradstreet 1984-2003). Each ratio is tested to see if it weakens over the study period (1984-2003). The regressors include time (TIME) and percentage change in real GDP. )GDP is included to account for the possible influence of the business cycle. The base year for real GDP is 2000 (U.S. Department of Commerce, Bureau of Economic Analysis). Data description and simple statistics are provided in table 1. Ordinary Least Squares adjusting for first order autocorrelation was used for each ratio and each dataset separately, tables 2 and 3. The results of each test include parameter estimates, t values, R, and F values. No evidence of multicollinearity was found. Results The test results are mixed. Overall, there seems to be only modest support for the hypothesis that produce intermediaries in recent years have suffered financial stress because of increased procurement pressures from mass merchandisers. The analyses using the D&B data 6 are more supportive of the hypothesis than those using the RMA data, tables 2 and 3. The trend model with D&B data shows some support for the hypothesis, table 2. The test is designed to see if the ratios conform to a trend over the study period. Congruous with the hypothesis, the coefficient for the TIME variable is positive and significant at the 0.01 level for the Total Liabilities to Net Worth ratio across all quartiles, a sign of increasing weakness. The same pattern of increasing weakness is shown for ROA and RONW across all quartiles – negative TIME coefficients. However, the TIME coefficients for only the lower quartiles are significant at the 0.01 level for ROA and RONW. The only significant median quartile coefficient (ROA) is significant at the 0.05 level. The F-values are significant at the 0.01 level or better for all trend equations involving significant TIME coefficients, table 2. Using RMA data, the TIME coefficients for the Debt to Net Worth ratio across all quartiles are not significant though the signs are as expected, table 3. The TIME coefficient for the ROA median quartile has the correct sign, negative, and is significant at the 0.05 level. The TIME coefficient for the upper quartile RONW is significant at the 0.10 level but has the wrong sign. The F-values for the two trend equations with significant TIME coefficients are significant at the 0.01 and 0.10 levels, respectively. All in all, the results for the test using RMA data are contradictory with no clear indication regarding the hypothesis, table 3. The addition of )GDP to account for the business cycle seems to add little information. In the few equations where the )GDP coefficients are significant at least the signs are correct, tables 2 and 3. In summary, the results using the two datasets (D&B and RMA) show mixed results regarding the financial strength of produce market intermediaries over the study period. For 7 additional insight an analysis of the census data is next. Census Evidence Using the census data, it is possible to shed light on the hypothesized decline in the financial condition of produce market intermediaries by comparing the number of establishments based on level of sales over the study period. The U.S. Census of Wholesale Trade (Establishment and Firm Size) compiles economic data for different industries every five years. The years used in this analysis are 1982, 1987, 1992, 1997, and 2002. Industries are classified by SIC code (5148 represents fresh fruit and vegetable wholesale intermediaries) in 1982, 1987, and 1992. For 1997 and 2002, the data are classified by NAICS code (424480). Census categories of produce wholesalers include merchant wholesalers, who buy and take title to the produce they sell, and agents, brokers, electronic marketers, commission merchants who collect a commission or fee for arranging the sale of produce owned by others. It should be noted that data reporting by level of sales for brokerage establishments/firms were discontinued after the 1997 census. Thus, the 2002 census contains data by level of sales for wholesale establishments/firms only. The total number of wholesale and brokerage establishments and corresponding sales continue to be reported (U.S. Department of Commerce, Bureau of the Census, 1982 2002). According to the hypothesis of this study, there should be a pattern of fewer and perhaps larger wholesale firms over the study period. Increasing consolidation is supported by the works of Calvin et al. (2001), Cook (2001), Handy et al. (2000), Kaufman et al. (2000), Richards and Patterson (2003), and Sexton, Zhang, and Chalfant (2003). The number of wholesale and brokerage establishments with lower sales (those under $5 8 million) declined 17.3% from 1982 to 1997, table 4. However, there was a large increase (71.5%) in the number of establishments with sales of $5 million and over. Overall, the number of establishments increased 8.1% from 1982 to 1997 and 5.5% with the period extended to 2002. Apparently, the effects of inflation aside, smaller firms are growing and merging into larger