Canadian Wheat Board

Prairie strong, worldwide

Library

The CWB and barley marketing

Dr. Andrew Schmitz
Dr. Richard Gray
Dr. Troy Schmitz
Dr. Gary Storey
January 1997
Executive Summary
Issues and Objectives

The operation of the Canadian Wheat Board (CWB) as the single-desk seller of western Canadian feed and malting barley for export and domestic human consumption within Canada is at the center of ongoing debate and controversy in Western Canada. The key issues raised in the debate are as follows.
Several economic studies of the Canadian barley marketing system and several government processes directly addressed the national and international issues involved in the debate. These studies included a federal government Round Table process in 1992/93 that funded a study by Carter (1993) and led to a federal Ministerial decision by the Honourable Charles Mayer to create a Continental Barley Market (CBM) beginning August 1, 1993. This change in marketing structure was reversed by a federal court ruling on Sept. 10, 1993. Following this ruling, in 1994/95, the Canada-U.S. Joint Commission on Grains examined issues relating to the potential for harmonization of the Canadian and U.S. marketing systems. The results of the Commission were provided to a federally mandated Western Grain Marketing Panel (WGMP) in 1995/96 that examined all issues in the western Canadian grain marketing industry. The WGMP made several recommendations to the federal government that would (1) increase the operational flexibility of the CWB in procuring grain from producers, (2) provide payment alternatives to increase flexibility of cash flow, (3) change the governance structure of the CWB to allow for direct producer control of the organization through a board of directors with producer-elected representatives, (4) establish a full open market for feed barley, with participation by the CWB, and (5) continue the single-desk selling of malting barley by the CWB.

The current Minister of Agriculture, the Honourable Ralph Goodale, announced on October 7, 1996 that the Government of Canada would implement the majority of the operational and governance recommendations of the WGMP. The Panel's recommendation to create a full open market for feed barley sales while maintaining the single-desk status of the CWB in malting barley markets was not accepted. The Minister of Agriculture announced that aproducer vote should take place on this issue.

Recent public studies that have examined the economic issues surrounding barley marketing in Western Canada and North America have focused primarily on feed barley, with less emphasis on malting barley. The lack of focus on the interrelationship between these two different barley markets has limited the usefulness of earlier studies in determining the implications of various possible marketing arrangements for barley producers and for the livestock and malting industries. In addition, these studies are limited in scope because they had little or no access to actual CWB sale prices and contract terms. Finally, although problems of arbitrage within the western Canadian domestic feed barley market have been identified in some of these studies, very little has been done to formalize the concept of arbitrage in the context of CWB price pooling or to quantify the effects within a formal economic framework.

Using formalized and integrated economic analyses, the overall objective of this study is to evaluate the economic performance of the CWB with respect to the marketing of both feed and malting barley domestically and internationally. The specific objectives of this study are to:
World and Canadian Barley Markets

Canada is a major player in the world barley market. It is among the top three exporters of both feed and malting barley in the world. At times, total exports from the European Union, Canada and Australia have been in excess of 78% of total world barley exports. Canada and Australia together have had more than 50% market share of barley exports. For malting barley, Canada's export market share has been as high as 40%. From a market perspective, the Canadian domestic market for feed barley has been the largest component of total Canadian barley consumption.

Export Enhancement Program (EEP) subsidies and restitution payments have been central to the export of U.S. and E.U. barley, respectively. The European Union halted restitution payments in May 1995, but these were reintroduced in September 1996. For barley, these subsidies commonly exceeded US $60/mt. The United States has not subsidized barley exports since July 1995. However, the 1996 Farm Bill authorized EEP funding of US $350 million in fiscal 1996, $250 million in 1997, $500 million in 1998, $550 million in 1999, $579 million in 2000, and $478 million in 2001 and 2002.

Selected Previous Studies

Consensus cannot be reached as to the benefits and costs of the CWB as a single-desk seller. A number of studies have concluded that substantial benefits were associated with this system. Other studies that do not support these results have argued that, while price premiums may have existed, they were small relative to the added marketing and other costs associated with the CWB. The studies provide a basis for the modeling approach used in this study, however, a major limitation of almost all the studies is that they model the effects on the feed grain market separately from the malting barley market. Also, generally, no formal models have been developed for the malting barley market that parallel the analysis of the feed grain sector. Rather, assumptions have been made about the link between feed barley prices and malting barley prices. In other cases, malting barley price premiums have been calculated by comparing U.S. and Canadian prices. When modeling the behavior of the CWB, one cannot a priori assume that it acts in a perfectly competitive manner. The world barley market consists of relatively few sellers. Also, in view of the work by Haley et al. (1992), one must test whether the CWB has market power. Determining whether the CWB has market power is empirically difficult, unless actual contract pricing data are available. The issue of market power and the nature of competition is very important. As Johnson et al. (1994) pointed out, assuming competitive behavior (i.e. no market power) misses the major argument in the current debate over barley marketing.

Much of the confusion in the present barley marketing debate is based upon the lack of a clear distinction between additional revenues earned by a single-desk seller and the total efficiency or inefficiency of a single-desk seller versus multiple sellers. For example, it is theoretically possible for the CWB to earn price premiums and still have a situation in which producers could be worse off than they would be under multiple sellers. However, that situation could only occur if theCWB system resulted in higher costs. To highlight this point, in their study on wheat, Kraft et al. (1996) concluded that the CWB earned significant price premiums over multiple sellers. These price premiums were calculated (free on board) f.o.b. Vancouver and not atthe farm gate. If the marketing costs under the CWB were at least as low as under a multiple seller-situation, the farmer would do better under the CWB system. However, Carter and Loyns (1996) contended the single-desk system adds costs to those that would exist under a multiple-seller system, and these costs outweigh the premiums.

Theory of Single-Desk Selling

The CWB is a form of collective action by Canadian grain producers in an attempt to maximize returns by jointly providing marketing services and countervailing power against large multinational grain trading companies. The CWB's existence is a direct result of public policy as it requires federal legislation (i.e. the Canadian Wheat Board Act). The CWB is set up to operate as a producer marketing board and it has adopted the objective of maximizing returns from sales of wheat and barley. The CWB acts as the producer's agent through which all sales and payments are made. The theory of producer marketing boards has been discussed in several works including Bieri and Schmitz (1974), Just et al. (1979), McCalla and Josling (1981), Schmitz et al. (1981), and Just et al. (1982).

In theory, the CWB is a producer monopolist. It's not a middle man (i.e. where a firm attempts to exploit both producers and consumers) nor a monopsonist (i.e. where a firm exploits producers). In other words, the profits earned from sales by the CWB are returned directly to producers (i.e. producers are the "shareholders" of the CWB).

A major feature of the international barley market is that marketing boards, such as the CWB and the Australian barley boards, sell into a market in competition with multinational grain companies. Their behavior is influenced by state trading entities including the E.U. Cereals Management Committee and the U.S. Commodity Credit Corporation. State trading dominates the world barley market. Roughly one half of barley trade is dominated by single-desk sellers.

The marketing of grain in the United States is very different from the marketing of CWB grains. As pointed out by Hill (1992), large multinational trading companies dominate the export stage of the U.S. grain marketing system. The dominant multinational trading companies involved in the export of U.S. grain are Cargill (American-based), Continental (American-based but owned by a French family), Archer Daniels Midland (American-based) which has a joint export venture with Toepfer (German-based), Bunge (Argentinian-based), Louis Dreyfus (French-based), and several subsidiaries of large Japanese corporations whose headquarters are in the United States. All of these companies source grain from the United States and other origins. In essence, the U.S. multinational trading companies behave as middlemen with respect to the buying and selling of U.S. and other origin grains. They buy grain from optional origins and sell it to foreign buyers.

The ability of a single-desk seller to generate additional revenue through price discrimination is well founded in economic theory. There is general agreement that the CWB is able to price discriminate. However, there are other reasons why the CWB may be able to increase revenue above what would exist under multiple sellers. One reason, suggested by Carter (1992), is that the steady supply guaranteed by the CWB spreads the risk that grain companies face in dealing with the day-to-day transactions. If the CWB did not exist, higher variability in quantity, quality, and price might force these companies to manage risk through the futures exchange. These companies would incur additional costs in coordinating information and hiring experts in the futures market. Hence, the presence of the CWB may be a lower-cost solution to these companies than alternative risk spreading, such as the use of futures market options. The CWB may be able to extract premiums for many of these companies who are willing to pay for this lower risk. It may also be the case that the CWB can obtain premiums simply because the multinational trading firms may charge higher margins in a system where they did not have to deal with the CWB.

Test for Market Power

A key consideration in the debate over feed barley marketing in Canada is whether the CWB is able to price discriminate and, therefore, exert market power in world markets. To test for market power, actual CWB feed barley contract data by import market and sale date from 1980/81 through 1994/95 were examined. The data forsales made via Canada's ports on the West Coast during this period were aggregated on a f.o.b. vessel basis intothe following regions: 1) Japan; 2) the United States; and 3) the rest of the world (ROW). A mean difference test was then conducted to examine whether statistically significant differences existed among the prices in these markets.

As indicated in the results, statistically significant differences existed among the f.o.b. contract prices obtained by the CWB in these markets (Table 1). Thus, the CWB has been able to price discriminate. The CWB's ability to price discriminate has allowed it to capture a higher price than would otherwise exist if there were multiple sellers of western Canadian barley. Therefore, western Canadian feed barley producers have benefited from the CWB. The average difference between CWB contract prices for Japan and the United States, over the 1980/81 through 1994/95 period, was significant and averaged $25.29/mt (tonne). The difference between CWB contract prices for the U.S. and ROW markets was also significant, with an average price difference of $4.46/mt. The difference between CWB contract prices to Japan and the ROW markets was significant and averaged $20.73/mt.

Table 1: Mean Difference Test of CWB Prices Achieved for Feed Barley

Time Period Japan - U.S. U.S. - ROW(1)- Cdn $/mt - Japan - ROW
1980/81 - 1994/95 25.29* 4.46* 20.73*
1980/81 - 1985/86 1.46 4.32 13.99*
1985/86 - 1994/95 26.84* 4.47* 23.74*

* Statistically different from zero with a probability greater than 95%.
(1)ROW = rest-of-the-world.
Source: As calculated by authors.

As shown in the results, the introduction of the U.S. EEP and the resulting feed barley trade war between the United States and the European Union increased the degree to which the CWB price discriminated. The average difference between Japan and the United States rose from $1.46/mt in the early 1980s to $26.84/mt in the trade-war period. Similarly, the average difference between Japan and the ROW increased from $13.99/mt in the early 1980s to $23.74/mt.

Comparing the CWB against Multiple Sellers
In this study, data are used from every CWB sale of feed barley, 6-row malting barley and 2-row malting barley for the period 1985/86 through 1994/95. The data are compiled from CWB contract records. All prices are brought to a common basis point of either f.o.b. Vancouver or f.o.b. Thunder Bay. The sales data are aggregated into the following nine market segments: 1) Japanese feed market; 2) U.S. feed market; 3) all other offshore feed markets; 4) Canadian domestic 6-row malting market; 5) U.S. 6-row malting market; 6) offshore 6-row malting markets; 7) Canadian domestic 2-row malting market; 8) U.S. 2-row malting market; and 9) offshore 2-row malting markets.

The objective of CWB marketing is modeled as the allocation of the total quantity of barley that it received from producers in a given crop year across the above nine markets so as to maximize total sales revenue. In order to measure the impact that multiple sellers of Canadian feed and malting barley would have had on returns and trade flows, a comparison is made between the actual market structure (i.e., prices and quantities) observed under the CWB and the prices and quantities that would have existed if there were multiple sellers of Canadian feed and malting barley.

In this study, two economic models are developed to determine the extent of price discrimination by the CWB in world barley markets and the resulting benefits derived by western Canadian barley producers. The first model incorporates the market power of the CWB in world barley markets by assuming that the CWB allocates its sales in order to simultaneously maximize revenue across world feed barley markets, domestic and world 6-row malting barley markets and domestic and world 2-row malting barley markets. The equilibrium domestic feed barley price is assumed to be equal to the weighted average pooled price for CWB exports of all feed barley. Using actual CWB sales data for 1985/86 through 1994/95, the excess demand elasticity for each type of Canadian barley in each market is determined by the model, given the domestic demand elasticity for Canadian feed barley and the excess demand elasticity for Canadian feed barley in the non-Japanese offshore markets. The demand elasticities are used to generate demand curves for Canadian barley in each market. The second model replaces the CWB with multiple sellers of Canadian barley by assuming that multiple sellers would introduce perfect competition in feed markets and malting barley markets. Under this assumption, the law of one price would hold across all feed barley markets and would also hold across all malting barley markets. The first and second models are compared to determine the economic benefits or losses incurred under the CWB.

The key difference between the CWB system and a multiple-seller system is the ability to price discriminate. In the absence of any constraints on the quantity of feed barley, 6-row malting barley, and 2-row malting barley available for sale by Canadian producers, the law of one price would have to hold for all international and domestic barley sales in a multiple-seller environment. In the model, multiple sellers were assumed to be fully competitive, and this competition resulted in one market price for feed barley and one market price for malting barley at any point in time. This is a characteristic of all competitive markets.

Overall Impact
The impact of introducing multiple sellers on Canadian feed and malting barley prices and total Canadian producer revenue is shown in Table 2 for each year from 1985/86 through 1994/95. Overall, the returns from CWB single-desk selling are significantly higher than would be the case in a multiple-seller environment. During the time period, the CWB earned an additional average return of $72 million annually over the multiple-seller scenario.

Table 2: Impact of Introducing Multiple Sellers on Canadian Feed/Malting Barley Prices and on Total Canadian Producer Revenue

Crop Year Feed Barley Price 6-Row Malting Barley Price 2-Row Malting Barley Price Total Producer Revenue3
($/mt) ($/mt) ($/mt) ($ mln)
1985/86 (4.91) (95.70) (80.93) (104)
1986/871 (4.46) (63.16) (30.08) (96)
1987/88 (11.36) (84.08) (13.18) (156)
1988/89 1.10 (72.63) (59.20) (35)
1989/90 0.86 (37.18) (47.90) (19)
1990/91 (7.89) (28.28) (2.50) (102)
1991/92 (8.20) (9.48) (19.54) (99)
1992/93 (4.68) (12.50) (36.05) (66)
1993/94 (2.62) 1.23 (16.05) (48)
1994/95 6.62 (18.66) (35.51) 7
Average (3.52) (42.01) (34.06) (72)

Note: Brackets indicate a loss for multiple sellers.
(1) Includes the impact on the domestic feed barley market.
Base Case:
Assumes the elasticity of demand for Canadian feed barley in the rest of the world is -20
Assumes the Canadian domestic feed demand is -0.53
Assumes the malting barley price remains at a $15/mt premium to feed barley.
Source: As calculated by authors.

Impact on 6-Row Malting Barley
The introduction of multiple sellers in 1985/86 would have reduced the average price of Canadian 6-row malting barley by $95.70/mt. The annual average additional revenue or revenue "benefit" earned by the CWB for Canadian 6-row malting barley producers over the 10 year period 1985/86 through 1994/95 was $42.01/mt. The CWB earned a higher price for 6-row malting barley than multiple sellers would have earned in all years but 1993/94. In this year, a multiple-seller system would have earned a slightly higher price of $1.23/mt.

Impact on 2-Row Malting Barley
The introduction of multiple sellers would have reduced the annual average price for Canadian 2-row malting barley by $34.06/mt from 1985/86 through 1994/95. The CWB prices were higher in every year. The largest premium was $80.93/mt in 1985/86 and the lowest premium was in 1990/91 at $2.50/mt.

Impact on Feed Barley
From 1985/86 through 1994/95, the introduction of multiple sellers would have reduced the annual average price for feed barley in Canada by $3.52/mt (Table 2). The CWB returned the highest revenue benefits to Canadian producers relative to the multiple-seller scenario in 1987/88 ($11.36/mt). The CWB also obtained added revenue in 1985/86, 1986/87, and 1990/91 through 1993/94. However, the multiple-seller structure would have returned higher revenue relative to the CWB in feed markets in 1988/89, 1989/90, and 1994/95 ($1.10/mt, $0.86/mt and $6.62/mt, respectively).

As a caveat to the above, during the 1988/89 and 1989/90 crop years, Japan did not represent the highest value market for CWB sales of feed barley. Because of the global shortage of feed barley in those years, the CWB obtained a price that was significantly higher than the Japanese price (in excess of $15/mt) on approximately 500,000 mt of sales to offshore markets other than Japan. However, the results (Table 2) were based on a model with four regions of demand for feed barley (i.e. Japan, the U.S., the ROW, and the Canadian domestic market). By aggregating these higher-priced markets with the ROW, the difference between the Japanese price and the price in the ROW was reduced. This gave the appearance that the CWB was unable to price discriminate during the 1988/89 and 1989/90 crop years. To address this issue, the results for the 1988/89 and 1989/90 crop years were recalculated with the regions of feed barley demand being redefined. Specifically, Japan was aggregated with the ROW and replaced as a separate region by those markets in which the CWB obtained a substantial premium. With the introduction of multiple sellers, feed barley prices would have decreased by $1.70/mt in 1988/89 and increased by $0.07/mt in 1989/90. From 1985/86 through 1994/95, the introduction of multiple sellers would have resulted in an average annual decrease in Canadian feed barley prices of $3.88/mt.

Impact on Total Producer Revenue
If multiple sellers would have replaced the CWB, producers' revenues would have decreased by an average of $72 million per year for the period 1985/86 through 1994/95. Under a multiple-seller scenario, the change in revenue would have ranged from a loss to Canadian feed and malting barley producers of $156 million in 1987/88, to a gain of $7 million in 1994/95. The 1994/95 year would have been the only year to show a gain under the multiple-seller structure. Generally, the magnitude of the increases in total revenue attributed to the single-desk structure followed the same pattern as the per-unit EEP subsidies provided by the United States on feed barley sales to the ROW.

Costs of Single-Desk Selling
We identify the additional revenue from barley sales that the CWB derives from the marketplace though price discrimination. However, there still remains the issue of the costs associated with the CWB as a single-desk seller. For example, Carter (1993) and KenAgra (1996) identified price pooling and the lack of a clear price signal as costs to the western Canadian feedgrain industry. This study finds these costs to be overstated.

Pooling Price Arbitrage Losses
The CWB currently uses an annual pooled return to allocate sales revenue to producers. This mechanism does not provide a signal to producers that fully responds on a timely basis to changing market conditions within a given marketing year. If export market prices change substantially during a crop year, the prevailing pooled return will not reflect this change on a timely basis. This creates some economic losses because the export value of feed barley at a given point in time is not reflected in the CWB Pool Return Outlook (PRO), nor in the cash price in the domestic feed barley market in Western Canada.

Arbitrage losses resulting from the operation of the CWB's annual pool for feed barley are calculated by measuring the difference in the change in price of feed barley at two U.S. points (Great Falls, Montana and Devils Lake, North Dakota) relative to the price of feed barley at Lethbridge, Alberta from the beginning of the crop year for the period from 1988/89 through 1995/96. The efficiency loss from price pooling averaged $4.9 million per year for both the Great Falls, Montana and Devils Lake, North Dakota comparisons with mostof the losses concentrated in the 1995/96 crop year (Table3). The 1995/96 crop year was unique as during this year international prices rose dramatically after the beginning of the crop year due to a significant reduction in U.S. corn production at a time of low carryover stocksof feedgrains. This created a large price wedge between the Canadian domestic feed barley price (which reflected the PRO) and the Great Falls/Devils Lake feed barley prices.

Table 3: Canada-U.S. Barley Price Movements and Arbitrage Efficiency Losses

Arbitrage Efficiency loss*
- $ mln- 		Price movement difference**
- $/mt -
Great Falls Devils Lake Great Falls Devils Lake
1988/89 5.154 4.068 16.69 14.43
1989/90 2.404 5.334 11.60 17.46
1990/91 1.409 .813 (6.27) (3.82)
1991/92 .541 .624 (0.86) (2.71)
1992/93 .129 .286 (0.46) (1.45)
1993/94 2.355 2.088 (10.60) (9.60)
1994/95 6.006 6.875 17.30 19.08
1995/96 21.067 18.955 (33.05) (31.10)
88/89-
95/96 Ave.		 4.883 4.880 (0.71) 0.29

Source: Calculated from weekly average spot barley prices. Lethbridge barley price: high end of daily range data extracted from AGDATA database, Alberta Agriculture, Food And Rural Development; Devils Lake: local cash prices as reported in AGWEEK; Great Falls: feed barley prices (cash) USDA - Montana Grain Weekly Summary.
* domestic demand elasticity = -.53
**Price movements (Lethbridge less U.S. prices) for each week were calculated versus the average price in the first week of the September.

Price Variability in Feed Barley Markets
From a livestock producer's perspective, the issue of price variability is of major concern. Three measures of feed barley price variability are calculated in this study:

To compare barley price variability, Lethbridge off-Board feed barley prices were compared to the U.S. feed barley prices at Great Falls, Montana and Devils Lake, North Dakota. As shown in column 1 of Table 4, the average annual standard deviation in the Lethbridge cash price, from 1988/89 through 1995/96, was $7.88/mt. This indicates that the average September cash price in Lethbridge was on average $7.88/mt above or below the average price for the crop year. This compares to $7.88/mt and $7.23/mt measured at Great Falls and Devils Lake, respectively. Comparisons, among Canadian and U.S. feed barley prices for each month relative to the subsequent six months (column 2 of Table 4) from 1988/89 through 1995/96, show similar levels of variability. As well, substantial differences do not appear to exist in the variability of Canadian and U.S. feed barley prices relative to U.S. corn prices in the Pacific Northwest (PNW) (column 3 of Table 4). This analysis suggests that Canadian feed barley prices do not exhibit anymore variability than U.S. feed barley prices.

Table 4: Monthly Average Cash Price Variability 1988/89 - 1995/96

Crop Year Subsequent 6 Months(1)
- $/mt Standard Dev. - 		PNW Corn-Barley Basis
Lethbridge Barley 7.88 6.48 11.19
Great Falls Barley 7.88 6.04 10.72
Devils Lake Barley 7.23 5.57 11.62
PNW Corn 11.45 8.95 0

1The absolute average of the difference between the average feed barley price in each month relative to the average price in the subsequent 6 months.
Source: As calculated by authors.

Carter and Loyns
As shown in our empirical results and confirmed by the analysis in this report, it is clear that the CWB has been able to exercise market power to the benefit of western Canadian farmers. However, Carter and Loyns (1996) argued that there are extra costs due to the CWB's marketing of barley calculated at roughly $37/mt.

Our general conclusion is that while some of the costs addressed by Carter and Loyns are present in the Canadian system, they are not unique to CWB grain marketing and would be incurred by producers and government in the absence of the CWB as a single-desk seller. Part of the problem with the study by Carter and Loyns is that the methodology upon which their cost estimates were based was not spelled out. From a methodological standpoint, when the CWB is placed in the context of the entire Canadian grain regulatory framework, many of the costs that Carter and Loyns attribute to the CWB would disappear. It is possible that costs could be higher in the absence of the present regulatory framework and the CWB.

Conclusions
This study clearly establishes that the single-desk selling of barley creates more sales revenue for western Canadian farmers than would be created if there were multiple sellers due to the ability of the CWB to exercise market power on behalf of western Canadian farmers. The magnitude of the additional revenue created varies for different years depending upon a number of factors, including the occurrence and degree of export subsidization in feed and malting barley markets.

Given the dominance of the CWB as a marketer of malting barley in a relatively small world malting barley market, it is not surprising that the benefits of the CWB's single-desk status are largest for malting barley. In contrast to the situation with malting barley,the CWB is a somewhat smaller player in the world feed barley market and, as a result, its ability to exercise market power through price discrimination, while significant, has less of an overall impact on prices.

One of the common criticisms of the CWB marketing system is the lack of arbitrage between the CWB feed barley pool return, western Canadian feed barley prices and international feed barley prices. This study addresses this problem using formal economic analysis to estimate the associated losses. These losses are small relative to the price premiums earned by the CWB. Even so, they could have been significantly reduced by providing the CWB with added flexibility, including cash trading.

From a policy perspective, the results from this study provide additional information to policy makers and producers regarding difficult choices among alternative marketing structures for Canadian barley. Issues such as equity and producer risk management are not addressed. However, it is clear that the CWB has been able to earn additional revenue from the sale of western Canadian barley over what would have been achieved under multiple sellers.Many issues have been raised including those presented in a highly controversial report by Carter and Loyns. After an examination of the costs of single-desk selling identified by Carter and Loyns, it is clear that while some of the costs are present in the Canadian system, they are not unique to CWB grain marketing, and would be incurred by producers and government in the absence of the CWB as a single-desk seller. This does not mean these costs should be disregarded in policy analyses. Ways in which the Canadian grain marketing system can be made more efficient need to be constantly examined. Policies that would result in a reduction in these costs should be explored further.