| | The views represented herein are those of the authors, Colin A. Carter and R. M. A. Loyns, and we are responsible for the full contents of the report. Alberta Agriculture, Food and Rural Development provided financial and information resources to complete this work.
This section examines the most important aspect of single-desk selling of prairie wheat and barley: the costs of single-desk buying and selling. According to information from the CWB itself, that is an easy task. The CWB (1995) has stated that "the cost to farmers to run the CWB is less than five cents per bushel"-or $1.80 per tonne. However, this figure represents only the operating costs of the Board (i.e, salaries, office costs, legal costs, travel, public relations, etc.) and thus grossly understates the true economic costs. For example, grain elevation and handling costs are higher than they should be on Board grains. Previous studies have identified several economic costs, but there does not appear to have been any significant attempt to add them up. We attempt to do so here. As in any performance evaluation using the generalized framework of cost/benefit analysis, some of the costs are very difficult to quantify even though they are easily identified. The $1.80 per tonne operational cost provided by the CWB is the easy one.
There is a wealth of evidence demonstrating the CWB single-desk comes with "excess baggage" in the form of market inefficiencies and regulatory costs. In 1970, both Anderson and Irvine documented the foregone opportunities and cross-subsidization associated with failure of the CWB to buy and sell grain by protein categories. In the 1980s the Canada Grains Council identified regulatory costs due to grain grading, cost pooling, price pooling, protein pooling, and varietal licensing. (Canada Grains Council, 1982, 1989) In 1984, Loyns and Carter suggested several areas where deregulation would contribute to increased competition, productivity gains, and higher farm returns. The 1993 CBM study by Carter discussed the high costs of the CWB single-desk in barley.
In this chapter we quantify the single-desk costs associated with the following factors:
- administration of CWB;
- protein over-delivery in world markets;
- delays in varietal development;
- grade over-delivery in world markets;
- farm production inefficiency due to false market signals;
- excess domestic handling charges;
- excess cleaning;
- demurrage/dispatch/extra freight;
- overage credits;
- delays in farm cash flow and on-farm storage costs;
- grain company and processor costs, and;
- taxpayer costs due to government underwriting and food aid sales.
These factors, summarized in Table 7.1 below, are directly or indirectly a result of the single desk regulatory environment in which farmers and grain companies are forced to operate. Before discussing each of these costs we first explain why the CWB is responsible for them.
Why the CWB is responsible for system inefficiencies and associated costs
The CWB has a pivotal role in the Canadian grain industry and it has virtual control of most important aspects of prairie grain production. The CWB is a marketing agency, a regulator and a defender of government marketing policy. Farm and grain industry decision making is subordinate to CWB decision making and often the industry is responding to non-market incentives. Challenges to the CWB revolve around these conditions and the Board’s lack of accountability.
The CWB has argued (Brooks, 1993) that it is not responsible for any domestic marketing inefficiencies. This is not very convincing. The CWB’s financial liabilities are fully guaranteed by the federal government; it does not have a bottom line and, at best, has mainly a political performance reference. Therefore, the Board does not pursue improvements in the system the same way that market forces would.
Improvements that have been made in the Canadian grain marketing system usually come from outside the CWB. In many respects, the CWB is a follower and not a leader. For years, farmers argued that delivery quotas were distortionary before they were finally removed. Protein segregation was introduced after much criticism from outsiders. As another example, in the early 1980s, Kraft and Carter started publicly forecasting CWB final payments. The CWB and Prairie Pools were forceful in criticizing them for this activity. Ten years later, the CWB started making similar public forecasts.
To summarize, virtually all aspects of the Canadian grain marketing environment are highly institutionalized and controlled by the CWB. This has slowed down productivity gains in grain handling, increased marketing costs, interfered with farm-level decision making, and has reduced farm returns. This is not to say there has been a conscious attempt by the CWB to retard the system, but there are few political or economic incentives for the CWB to reduce many of the costs associated with the status quo. We now turn to these costs.
Cost estimates of SDS
Identification of costs does not always mean they have been measured or that they could be measured with precision. Some costs such as protein overages, foregone farmer interest, and storage costs are relatively easy to measure. Others require more information and time consuming measurement techniques. Examples here would include regulatory impacts on the handling and processing sectors. The KPMG (1995) report on grain logistics focused on this issue and found the Canadian system is plagued with inefficiencies due to such factors as lack of economic incentives, lack of accountability, excess segregation, and excess cleaning. They estimated that up to 20 percent of operational costs could be saved annually through reduced regulation, the introduction of transparent incentives, and improved accountability.
Whether or not all the costs have been estimated is not the issue. What is important is that the industry agree which costs are significant, and begin to reduce them. Table 7.1 summarizes the single-desk factors we have identified and the major entries in this table are discussed below.
Table 7.1 Cost Factors Associated With the CWB |
Estimated Cost |  | Wheat ($/mt) | Barley($/mt) |
Direct & Indirect Producer Cost Item: | Producer Impact | | |
| 1. CWB Administration | direct cost | $1.80 | $1.75 |
| 2. Protein Segregation | | | |
| -delays in implementation | lost revenue | n.e. | n.a. |
| -protein giveaway | direct cost | $1.60 | n.a. |
| 3. Varietal Development | | | |
| -delays in implementation | lost revenue | $3-$5 | $3-$5 |
| 4. Grade Giveaway | lost revenue | $1.25 | n.a. |
| 5. Lack of Arbitrage | | | |
| -poor price signals | lost revenue | n.e. | n.e. |
| -reduced U.S. sales | lost revenue | n.e. | n.e. |
| 6. Marketing Inefficiency | | | |
| -overage credits | direct cost | $1 | $0.5 |
| -excessive handling charges | direct cost | $3-$5 | $3-$5 |
| -demurrage/dispatch/extra freight | direct cost | $1.05 | $1.60 |
| -maltster free storage | direct cost | n.a. | $3.50 |
| -excessive cleaning costs | direct cost | $2.35 | $2.80 |
| -port congestion | lost revenue | $1 | $1 |
| 7. Farm Management Costs | | | |
| -production inefficiency | lost revenue | $4 | $4 |
| -excess malting barley | lost revenue | n.a. | $1-$3 |
| -delivery patterns | cross subsidy | ($2.60)-$2.60 | ($1.80)-$1.80 |
| -opportunity interest cost | lost revenue | $0.60-$1.60 | $.10-$.60 |
| -opportunity storage cost | lost revenue | $2-$4 | $2-$4 |
| -destabilization | lost revenue | n.e. | n.e. |
| 8. Grain Company and Processor Costs | | | |
| -country assembly | lost revenue | n.e. | n.e. |
| -segregation | lost revenue | n.e. | n.e. |
| -terminal throughput | lost revenue | n.e. | n.e. |
| -foregone value added | lost revenue | n.e. | n.e. |
| Society’s Cost Item: | Impact | | |
| 9. Taxpayer Costs | | | |
| -food aid sales | budget cost | $0.50 | n.a. |
| -government underwriting | foregone cost | $2.50 | $2.50 |
| -pool deficit | budget cost | $2.50 | $6.50 |
Note: n.e. denotes not estimated and n.a. denotes not applicable. Numbers in parentheses are negative costs (i.e., benefits). Downstream impacts on the livestock industry have not been included, nor have historical programs like the two-price wheat program which shifted production to eastern Canada. Some costs are transfers among farmers and some costs generate benefits to others in Canada and overseas.
Wheat protein give-away
There are two levels of protein give-away: one at the farm level, and one at the CWB customer level. Developing the process of compensating farmers for protein has been very slow and is still evolving. There is likely some give-away of protein by farmers but since the CWB now sells by protein levels, most of the value of protein should be reflected in the pools. For farm level give-away then, this is primarily a cross-subsidization problem and we have not tried to estimate it.
Protein value fluctuates considerably with market conditions. It can be quite valuable in some years. We see from Figure 7.1 that the incremental value of 1% of protein in the U.S. market (taken as the difference between 14% and 13% DNS, as reported by the Minneapolis Grain Exchange) has ranged from a low of $0.23 ($Cdn) per tonne in 1989 to a high of $34 per tonne in 1993, averaging about $9 per tonne over the 1984 to 1994 time period.
Figure 7.1:
As discussed elsewhere in this report, the CWB has failed to adequately segregate protein and pay farmers accordingly (see section III). In turn, the CWB often over-delivers protein content on export shipments. Unlike U.S. export shipments, the protein content of Canadian exports is normally well above the minimum delivery specifications. This practice has been verified by examining two separate data sources, from the Japanese Wheat Flour Institute and from the CGC. The Japanese Institute measures the protein content of all vessel unloads in Japan and the CGC does the same for all exports from Canada.
Actual Japanese deliveries against protein specifications are shown in Figures 7.2 for exports from Canada, the United States, and Australia. Like the CWB, the Australian Wheat Board over-delivers protein, but the U.S. system delivers what the customer pays for. CWB wheat shipments to Japan result in over-delivery of about 6/10ths of one percent of protein and, on average, this has cost prairie farmers approximately $5 ($Cdn) per tonne on Japanese sales (from 1984-94).
The over-shipments for No. 1 and 2 CWRS 13.5% to all destinations are shown in Figures 7.3-7.4. There are no minimum protein specifications on No.3 CWRS, but we see from Figure 7.5 that No. 3’s protein content is also quite high, on average.
Figure 7.3:
Figure 7.4:
Figure 7.5:
The cost of protein over-deliveries to all destinations was calculated using CGC quarterly bulletins on "Quality of Canadian Grain Exports", combined with protein premiums determined on the Minneapolis Grain Exchange. For 1984-1994, this amounts to an average of $1.25 per tonne on No. 1 CWRS and $2.53 per tonne on No. 2 CWRS, for a weighted average of $1.60.2 Additional details are provided in Table 7.2.
Table 7.2 Average Protein Give-Away on Total Exports: 1984-95 |
| #1 CWRS | #2 CWRS | Combined #1 and #2 |
| Average
$ Value
per 1% (%) | Protein
Over Spec
per 1% (%) | Average
$ Value
per 1% (%) | Protein
Over Spec
per 1% (%) | Average
$ Value
per 1% (%) | Protein
Over Spec
per 1% (%) |
| 84/85 | 1.58 | 0.18 | 1.56 | 0.23 | 1.58 | 0.18 |
| 85/86 | 3.10 | 0.21 | 4.90 | 0.38 | 3.64 | 0.26 |
| 86/87 | 1.34 | 0.17 | 1.55 | 0.29 | 1.41 | 0.22 |
| 87/88 | 2.14 | 0.30 | 2.68 | 0.41 | 2.41 | 0.35 |
| 88/89 | 1.32 | 0.52 | 1.01 | 0.51 | 1.22 | 0.52 |
| 89/90 | 0.10 | 0.38 | 0.14 | 0.67 | 0.12 | 0.48 |
| 90/91 | 0.32 | 0.15 | 0.63 | 0.31 | 0.36 | 0.17 |
| 91/92 | 0.22 | 0.15 | 1.15 | 0.20 | 0.28 | 0.16 |
| 92/93 | 1.33 | 0.15 | 0.88 | 0.13 | 1.24 | 0.14 |
| 93/94 | 6.31 | 0.18 | 7.89 | 0.23 | 7.02 | 0.20 |
| 94/95 | 2.54 | 0.17 | 3.96 | 0.24 | 3.43 | 0.21 |
| Average | $1.25 | 0.21% | $2.53 | 0.37% | $1.58 | 0.25% |
Source: Canadian Grain Commission "Quality of Canadiana Grain Exports"
Grade give-away
The CWB tends to overemphasize the importance of quality in the world wheat market, and this is part of the problem with the false signals the Board sends back to producers. As discussed elsewhere, only about 15 to 20 percent of CWB foreign customers are willing to pay for No. 1 CWRS quality, while 50% of Canadian wheat exports is graded No. 1. This means that what prairie farmers are growing does not closely match world demand. On average, about 65 percent of the western Canadian crop delivered to the Board as either No. 1 or No. 2, but some of this is sold out as No. 3. The overseas customer pays for No. 3 but receives what resembles No. 1 or No. 2. There is a famous anecdote in the trade about this, relating to a CWB sale to South Korea a few years ago. The Koreans purchased feed wheat but the Board couldn’t source the feed in the country, and instead forwarded No.3’s and No. 2’s and were paid for feed.
China is one of the largest CWB customers and, normally, it is only interested in buying the equivalent of No. 3 CWRS. However, the CWB often "over services" this market by forwarding No. 1 or No. 2, in place of No.3. This is perfectly acceptable to the Chinese because the contracts typically specify the minimum grade, or better. China normally buys No. 3 (or better), but only 1 out of 10 ships loaded (out of Vancouver) for China is predominantly No. 3. What this means is that high quality wheat is being sold for low quality prices.
What impact does this have on the producer and price pooling? First of all, there is an aspect of giving away quality to foreign buyers, not unlike the protein giveaways discussed elsewhere in this report. Secondly, the CWB method of price pooling likely results in the low quality wheat cross-subsidizing the high quality wheats. This follows because CWB selling No.1 for No. 3 lowers the average return to No. 1 but it does not affect the average return for No. 3. In order to bring relative prices back the true relationship, the average price of No. 3 must be lowered, so that the price of No. 1 is raised. The CWB has recently explained their approach to pooling as follows:
The actual pool return that is established for any particular class, grade, or protein level of wheat is determined by the price relationships that existed in world markets over the course of the crop year. The CWB keeps track of these price relationships to ensure that it will be able to reflect to farmers a proper return for each grade and class of wheat and barley. (Grain Matters, November-December, 1995).
Therefore, if the CWB preserves "price relationships that existed in world markets over the course of the crop year" they must be transferring revenue from low quality wheat farmers into the pockets of high quality wheat farmers, in order to keep relative price spreads at the "proper" level.
To illustrate this point, let's take a hypothetical example. Suppose the average world price of No. 1 is $200/mt, that of No. 2 is $190/mt, and No. 3 is $180/mt. Let the breakdown of CWB terminal receipts be 50% No. 1, 15% No. 2, and 35% No. 3 (ignoring feed). Further, assume that 15% of the No. 1 and 15% of the No. 2, received in the terminals, is forwarded overseas as No. 3. This means that actual foreign sales percentage breakdown is 42-13-45, but the farmers were paid for a 50-15-35 grade breakdown. If there was no cross-subsidization taking place, then the average prices would be $197, $188, and $180 for Nos. 1, 2 and 3 respectively. This would violate the rule the CWB uses to preserve the "proper" price spread between grades, as stated in the above quote. In this example, the spread has to return to that of $10 between grades. The only way the CWB can adjust these price relatives back to their "proper" relationship is if the price of No. 3 is lowered.
This hypothetical example may not be too far from reality, but the information required for comprehensive analysis of this issue is not readily available. This is yet another arbitrary practice of the CWB that has important cross subsidy implications. It requires further study but we have estimated 15-20 percent of the No. 1s and 2s, could be sold as No. 3s.
From 1984-85 to 1994-95, the average CWB price spread between No. 1 and 2 was $6/mt, and between No.2 and 3 $10/mt. Therefore, if 15 percent of No.1 and No.2 was sold as No.3, this means approximately 1.5 mmt of No.1 was downgraded to No.3 and 450,000 mt of No.2 was downgraded. The estimated giveaway on No.1 is therefore $24 million and $4.5 million on No.2. Averaged across all exports, the grade giveaway amounts to about $1.25/mt. This is quite apart from the cross-subsidization issue due to grade giveaway.
Excessive handling charges and overage credits
Once CWB grains are delivered to an elevator, a set of system charges are levied against that grain and deducted from the producer payment (which is based on net delivery valued at the initial price). The major deductions made at this level are country elevation and handling, dockage removal and transportation. Charges are later levied against the pool account for terminal handling, storage and interest charges, inspection, and CWB charges.
This very formal process has existed for years. These fees were filed by grain companies with the Canadian Grain Commission (CGC) until 1995 and these were the charges levied against deliveries. There was little variation across provinces or companies. The requirements for filing tariffs was changed by the CGC in 1995, but in effect the economics of this system has not changed significantly. The handling agreements between the CWB and grain companies provide for producer payments based on unloads at the elevator, but the handler is credited and paid for the grain unloaded at the terminal.
For decades, grain companies charged producers a "shrinkage" allowance for losses within the system despite the fact CWB usually pays "overages" out of the pools because the average quantity and quality of grain company terminal sales marginally exceeded their country purchases. These overage credits are one of the cost items identified in Table 7.1, listed under item 6 (Table 7.1). The calculations behind these overage credits are taken directly from the CWB Annual Reports and are summarized in Table 7.3. We emphasize these credits only reflect quantity overages and not quality overages. Quantity overages amount to $1 per tonne on wheat and $0.50 per tonne on barley.
This system is a product of the regulation which has existed over the years. Producers have no alternative on these charges except to avoid country elevation through use of producer cars. There is little competition among companies on CWB grains because of the way grain is called into the system, the way grain cars are allocated in the country and the way grain is allocated among terminals.
This issue is more or less summed up in the following quotation:
UGG was the first to act, raising its handling tariff to match Alberta Pools rates on January 17, Saskatchewan Pool followed suit on February 3, while Pioneer’s new rates are effective February 16....and,.....A terminal operator that charges less than its competitors is simply leaving money on the table. (The Western Producer, February 16, 1995).
Country elevators today under take a small amount of nonprice competition on their buying margins on CWB grains. They will negotiate on trucking or delivery allowances, and negotiate at the margin on grade or protein level, to secure extra business. Malting barley acceptance has also been a vehicle for negotiating on other grain business. But the grain companies typically do not negotiate handling tariffs on CWB grains because there is no incentive for them to do so.
Not surprisingly, system charges on CWB grains are higher than they would be in a competitive system; they are higher than exist in the system that is used in nonboard grains and much higher than in the U.S. Carter reported findings of excess handling charges on barley in the CBM report (Carter, 1993). This phenomenon was verified through more detailed study by the 1995 KPMG report on grain logistics. Grain handlers have stated there was a drop in marketing costs on oats of about one-third following deregulation of that market in 1989. The above quote from the Western Producer, attributed to two grain industry representatives (one a producer cooperative, one a private company) depicts how the lack of competition is played out and how the farmer pays the price. That one article in the Western Producer is worth more in documenting the problem than several studies by economists who are judged by regulators to be biased and/or uninformed.
An interesting aspect of any comparison of system charges on CWB and nonboard grains is the cost of risk management associated with both marketing channels. Grain companies manage risk in many ways and get paid for risk management from at least two sources: handling margins, and the spread between buying and selling prices. For CWB grains, grain companies have almost no risk of ownership, no storage and carrying costs, and no risk of price change. The farmers bear all of the costs associated with these risks. This suggests that system charges on CWB grains should be smaller, not larger, than on nonboard grains, but they are not.
Placing a cost estimate on the excess system charges is difficult. We have collected a great deal of system cost and charges information from the U.S. They are lower in almost every category and location than that from Canada. We have attempted to compare for recent years CWB and nonboard actual charges. And we have attempted to obtain data on oats and handling charges on barley sold to the U.S. during the CBM, in relation to charges under the CWB. Our investigation has produced nothing to reject the proposition that handling charges on CWB grains are considerably higher than they would be in a competitive system. Nor have we seen any evidence to suggest that the CWB does anything to have these charges reduced. They are cost-pass-through charges that appear to be taken for granted. Based on our analysis to date, we submit that producers pay between $3 and $5 per tonne more on handling CWB grains between the elevator and vessel than they would in a competitive system. This is a very conservative estimate. There is little in the present system to keep these costs down.
Some of our background data on comparative handling fees are summarized in Table 7.3, which reports comparable costs for movement of wheat from North Dakota to either the PNW or the Gulf, and from Saskatchewan to either the Pacific or St. Lawrence. These data indicate the Canadian handling fees on CWB grains are higher than in the U.S. by about $8 to $13 per tonne. In addition, they illustrate that handling fees on CWB grains are higher than on non-board grains. We also have information from the U.S. that indicates cleaning rates are 4 to 6% lower than five years ago, and that handling charges have fallen about 2 to 3%. The Canada Grains Council Statistical Handbook shows that cleaning charges on prairie grain have increased over 25% during this same period and terminal charges have increased between 8 to 10%. Primary elevator handling charges show a 2% decline.
Notes:
1. We used the Saskatchewan tariff (it would be $9.55 in Manitoba).
2. Carrying charges in country elevators.
3. Interest and bank charges, plus CWB administration.
4. Includes small port fees.
5. $Cdn/$U.S. exchange rate assumed to be 1.37.
6. 1995/96 Minot to Portland or Seattle.
7. Minot to Gulf.
8. n.a. signifies not applicable.
Sources: Canada Grains Council, Statistical Handbook, 1995.
Interviews with grain handlers.
Market development: varieties and lost opportunity
The Canada Grains Council (1982) was the first to point out that farmers lost money because the CWB insisted on selling only high quality wheat, even though world market opportunities were growing for medium quality wheats. Farmers lost out because the high quality wheats have relatively low yields and there are areas of the prairies that are more suited to higher yielding, different quality wheats.
Henning (1986) evaluated what would happen if there were a shift in Canadian export orientation from the traditional hard red spring varieties towards the medium quality varieties . The results indicated that:
...Canadian export earnings could be significantly increased by a greater emphasis on medium quality exports. When the effects over all classes are taken into account, by exporting more Class 2 wheat, export earnings could rise on the order of $100 to $250 million per annum (p.396.) Carter, Loyns and Ahmadi-Esfa-hani (1986) studied varietal licensing and Canadian wheat exports, and estimated the economic cost of this regulation to be 5-17 percent of annual net farm income, or $80-$272 million. Ulrich, Furtan, and Schmitz (1987) studied the same issue and came up with much higher costs. They found the adoption of higher yielding wheats was stopped even though these wheats had the potential to increase annual net farm income by 15-25 percent, or $200-$400 million which, they pointed out, was equivalent to Manitoba’s net farm income at the time. Like the other two studies, they argued that the CWB’s marketing emphasis on high quality wheats has been misplaced because of (developing country) growth markets for lower quality wheat. Ulrich, Furtan, and Schmitz argued the CWB could sell medium quality wheat into the same markets and at the same prices as the equivalent American or Australian wheats.
In 1987, Derek Byerlee also argued that the developing world would be the growth area for wheat. He was right. He emphasized the importance of the "tropical belt" of developing wheat importers and the overall importance of government policy in expanding wheat imports in these countries and...
urbanization will undoubtedly continue at a rapid pace; this will increase the consumption of wheat in these countries provided the money is available for purchase. Second, the demand will be for less than premium quality wheats.
Economic costs due to varietal regulation have also been an issue in barley. Gibney and Furtan (1983) and Ulrich, Furtan and Schmitz (1986) studied the regulation of barley varieties. Ulrich, Furtan and Schmitz found that Canadian society paid a high cost because the introduction of new barley varieties has been skewed away from high yielding feed varieties and towards malting varieties. This may partially explain why the Canadian malting industry vigorously supports the CWB.
The accuracy of these economic studies has been borne out with the passage of time. The growth markets have been in different quality wheats, and we have finally developed and licensed varieties that partially meet this demand. Similarly, prairie farmers are now producing malting barley varieties for export to the huge U.S. brewery market. This is a market that Canadians had written off because, as well placed officials in the CWB often commented (as recently as 1993), the U.S. would ultimately come around to using Canadian 2 row varieties.
The CWB will argue that it does not have any direct involvement in varietal development or regulation. It does, however, have several indirect effects. For example, the CWB was a vocal opponent of unlicensed U.S. varieties when they were in the system in the 1980s. The CWB openly discouraged the development and production of new varieties. In the absence of the CWB monopoly on exports, or in the presence of a dual market, competition among handlers and buyers would generate market development opportunities and new varieties. Varietal development and market development in the nonboard markets has been very rapid over the past decade.
What this means is that over the years the rate of development of new wheat varieties in Canada has been slow. In 1987, Leisle observed that "Only 34 new hard red spring wheat varieties have been approved over the past 63 years" (Conference on Millers, 1987). Moreover, the development of "other quality" wheats was resisted across the system. While there has been some improvement in this situation in the last few years, the reluctance to use U.S. varieties as recently as the Grandin fiasco in 1992-94, indicates that the CWB mindset is still resistant to change.
Grandin was legally brought into Canada in about 1992, grown for seed, and then refused license in 1994 even though there was an enormous amount of seed on farms. In 1995 it was given experimental status and even sold by protein segregation, but without contract registration. The CWB failed at getting it licensed.
This cost issue of varietal development and licensing had, and still has, its roots in two economic facts: farmers in some areas of the prairies have forgone economic benefits from not being able to grow marketable varieties; and the fastest growing and largest markets are outside conventional Canadian wheat varieties. Public statements by the regulators have held to the position that the high-quality-only policy was the correct one even though there is no evidence to support that view. Over time, it has slowly become recognized that this policy had costs.
The cost to the prairies of not producing these different wheats and not selling malting barley into the U.S. market has been enormous. This cost was avoidable. It occurred because the system was slow to change, unresponsive to market signals and top heavy with decision makers with no financial interest in how well supply is matched with demand. The result is explained by the structure, but the responsibility must rest with the decision makers that manage this industry on behalf of farmers and the "national interest". As the major marketing agency in this system, the CWB has to accept some of the responsibility for its lack of leadership and inflexibility when the evidence began to appear. The CWB has again, for the most part, followed rather than lead on quality change and it has, sometimes along with the grain companies and maltsters, helped to retard varietal change. This process looks much like the delays in adjusting to protein markets in the 1960s.
Given this evidence on costs due to lack of varietal development, it is possible to provide an estimate of the cost to prairie farmers. The annual costs for wheat have been estimated to range from about $100m to $400m. The period of time over which to assess the cost is, perhaps, more difficult. The need had been identified in the middle 1970s and some technology existed (e.g., HY 320). Not much happened in these markets until the late 1980s. Therefore, using the midpoint of the Ulrich, Furtan and Schmitz estimate of annual losses (about $300 million per year) on a conservative 12 year period (which coincides with the Henning estimate), it is easy to identify a nominal $3.6 billion dollar cost to prairie producers. Malting barley losses could easily be in the order of $20 to $50 million per year, depending on the time period. Of course, one should either discount or capitalize benefits and costs to one point in time. A conservative estimate of this cost, therefore, approaches $10 per tonne.
If we agree that this cost should not all be attributed to the CWB (they will surely argue than none of this cost is theirs), an estimate of $3 to $5 per tonne for wheat is credible without accounting for the time value of money. A similar number could be applied to barley for lost U.S. malting sales and for lost feed barley productivity if the Ulrich, Furtan and Schmitz argument of too much malting barley focus is believed.
Seaway cost pooling
Until August 1, 1995 pool accounts were operated as if the value of the grain in each pool was the same whether sold out of Pacific ports or out of Thunder Bay. Over a long historical period that assumption was reasonable. But during the late 1970s and early 1980s, two factors reduced the validity of this pricing method. First, the total costs of moving grain out of the Seaway increased faster than rail costs. This meant that shipments east netted less to the pools than shipments west under traditional shipping patterns. Second, Pacific markets became more attractive and shipments west increased. In about 1985 shipments west exceeded shipments east and this trend has continued. This development is illustrated in Figure 7.6 which shows the net receipts of grain at terminal elevators in Thunder Bay and the West Coast. The additional cost of shipments through the Seaway were shared by all pool participants until a policy change in 1995.
There is no doubt that this is another example of "cost pooling". It was discussed and analyzed extensively after being raised by two CWB economists in 1986 (Oleson and Brooks, 1986). This represents cross subsidization that occurred, and is still present although changed recently.
Removing Seaway costs from the pools reduced overall deductions by something in the order of $6 per tonne, based on 1993 figures. If we use this estimate of Seaway cost contribution to the pool, western prairie producers benefited by about $6/tonne from the 1995 change. A similar figure has been discussed in CWB news releases. On the other hand, eastern prairie producers have experienced an increased cost of $23.41, on average, or $17.50 on a net basis after subtracting the $6/tonne pool credit.
But the story is not finished. The Freight Adjustment Factors (FAFs) are in transition and there appears to be some pooling of Seaway charges under the new system. The FAF on durum and designated barley are very low, and are zero for some barley producers. This implies there is some utilization of these grains at or just east of Thunder Bay, which is not the case. Obviously some shipping costs are still pooled but it is too early to measure the magnitude.
Pooling of cross hauling costs
As shown in Figure 7.6, a shift in Canada’s export markets for grain from Eastern and Western European countries to the Asia-Pacific region has caused a shift in grain exports from Thunder Bay to the ports at Vancouver and Prince Rupert (West Coast). In the mid 1980s, receipts of grain at Thunder Bay and the West Coast were nearly equal. Prior to the 1986-87 crop year, receipts at Thunder Bay were well above those at the West Coast and subsequently the opposite case holds.
When producers deliver CWB grain, they are charged freight to the nearest port. If grain is shipped to a different, and higher cost location, the pool picks up the extra freight cost. These additional freight charges are recorded in the CWB Annual Reports. The difference between freight charged and freight paid is another example of cross subsidization within the pool. The importance of this aspect of cross subsidization appears to be related to increases in Pacific exports (see Figure 7.7).
The CWB annual reports lists additional freight to terminals and it is described as "...rail freight costs that in addition to Western Grain Transportation Act deductions, were required to cover the shift in movement of grain necessary to meet increased west coast exports" (CWB Annual Report, 1992-93). On average from 1986-87 to 1993-94, the charges per tonne were $1.02 for wheat, $0.20 for durum, and $1.24 for barley. There are no additional freight charges against malting barley. Instead of the CWB, the grain companies arrange malting shipments.
Pooling costs to livestock producers
The CWB pooling and assembly practices have two major impacts on domestic livestock producers: price signals are distorted, and procurement uncertainty is increased. As a result of pooling, feedgrain prices in the eastern prairies probably have been too high relative to feedgrain prices in the western prairies. Because, the CWB also distorts the absolute level of prairie feedgrain prices compared to world markets so this issue is very difficult to sort out. However, it seems obvious that "cost pooling" has important cross-subsidization impacts on livestock producers.
The long long weekend in August 1986
At the end of the 1985/86 crop year, the CWB left the wheat pool open and received substantial deliveries of lower grade wheats. It was alleged in a legal action by the Western Canadian Wheat Growers that $54.3 million of final payments for Nos. 1 and 2 CWRS and CWSWS were transferred away, to offset losses sustained by the Board on the lower quality wheat. An appeal court decision upheld the right of the CWB to determine the allocation of pool surpluses. This situation, like many others, indicates the power of the Board to redistribute revenue among farmers.
There are alternative ways that this case of cross subsidization could be quantified. Any method would likely show that No. 1 and 2 producers were penalized by several dollars per tonne. Allocating the reported loss against estimated volume suggests a figure of about $5/tonne.
According to protein values that year, it is possible that the cost to some high quality wheat producers was in the order of $20 per tonne.
We have not taken this particular cost factor any further because it is a historical cost. What is most relevant in this example is the way in which the cross subsidization arose and the legal judgement sustaining the decision. Obviously there are major differences between the way the CWB chooses to operate and what is required to achieve economic efficiency. The long weekend in August 1986 indicates that these differences can produce significant costs for some producers and, perhaps, significant benefits for others.
Estimates of farm management effects of pooling cash flow, farmer interest costs and effective prices
Regulated cash flow is one of the most important effects of the CWB pooling regime. Quotas determined when and how much a producer could deliver, whereas today, CWB contracts do the same thing. Delivery schedules determine cash flow on CWB grains. Delays in deliveries add to storage and interest costs. Flat prices over the crop year and late revenue represent a lower effective price, relative to early deliveries.
Delivery patterns that vary by crop or region mean different producers experience different impacts on their cash flow, costs and real prices. We have estimated some of these effects of pooling and the details are contained in appendix 7.A.
First, we attempted to estimate foregone interest associated with the almost eighteen month period over which CWB payments are made. To do so we compared payments as received over the ten year period 1984/85 to 1994/95 with receiving the same payment on January 1 of the particular crop year. The ten year period should average out seasonal and trend effects on prices. In non-board crops the receipt of payment in full by January 1 is plausible.
Our analysis provides a measure of the farmer interest opportunity costs (net of CWB interest paid), by pool, of the payment regime experienced over the ten years. Table 7.4 shows that the average cost to wheat producers was about $0.60/mt; amber durum about $1.60/mt, and $0.10 and $0.60 for barley and designated barley, respectively. Wheat has been as high as $1.89/mt and durum has been as high as $3.63/mt.
Table 7.4
Opportunity Costs of CWB Payment Schedules from 1984-1985 to 1993-94 |
Range (Crop Year) |
| Low | Year | High | Year |
| Wheat | $0.00 | (1985-1990) | $1.89 | (1984) |
| Durum | 0.00 | (1990) | 3.63 | (1987) |
| Barley | (0.46) | (1992) | 1.04 | (1989) |
| Malting Barley | 0.00 | (1986) | 1.50 | (1989) |
In addition to these interest costs, there are physical storage costs that would be avoided if delivery occurred earlier. Storage itself has some cost and there are physical and nuisance costs associated with moving grain on the CWB’s terms rather than the farmer’s own terms. Those who argue that farm storage is cheap and that the value of farm labour in the off season is low have probably never moved grain in the middle of winter on the prairies.
The storage rates paid by farmers out of pools provides one measure of the value of storage although they are probably too high. The proportion of volume stored and the length of time stored are also variables that affect any estimate of storage costs. We assume the January 1 date for early sale as in the interest cost analysis, which gives about 200 days of storage. If we bypass the volume question by using only storage rates available from tariff schedules posted with the CGC, it is easy to assess a commercial storage cost of at least $8.00/tonne for that period of time. If we assume that the commercial regulated rate is inflated, even though farmers pay these rates on storage costs charged against the pools, and discount the rates for both lack of competition and length of time, a conservative estimate of the physical costs of farm stored grain relative to early, unrestricted movement is $2 to $4/tonne depending on several factors. This estimate represents a measure of the cost of delivery restrictions faced by farmers who hold grain over into the second half of the crop year.
Cash flow is important as a farm management variable, especially if the farm has debt and if the cash flow is beyond the control of the farmer because it influences debt service capability. Delayed cash flow also has a real price impact on returns for any given (reported) CWB price. The level of the real price received differs from the reported CWB final returns (often incorrectly referred to as a price), depending on delivery patterns. We attempted to measure this impact as well. For wheat, over the crop year, the average difference between accelerated and delayed delivery patterns range between $1.80 for barley and $2.75 per mt for durum, in terms of the effective price received by growers. For wheat, the difference in effective price resulting from different delivery patterns was about $2.60/mt, these values are entered in Table 7.1 as pluses and minuses because they are cross-subsidies within the pool. Those farmers who deliver early enjoy a higher effective price. The "equal" prices paid by the CWB produce significantly different returns to producers and therefore the quality of the price signal varies across producers and crops.
The CWB and farm income stabilization
Price and income variability in agriculture have been targets of agricultural policy and programs for decades in Canada. Proponents of board or so-called "orderly marketing" in Canada argue one reason for maintaining the Board is for its stabilizing influence. The pricing regime does provide a form of stability to prices at the farm level. Fixed prices over parts, sometimes all, of the crop year certainly do "stabilize" prices.
Whether the regime associated with pooled prices stabilizes returns and contributes to improved decision making is open to question. Evidence produced by Watson and Parish in Australia showed that the wheat pool used in pricing under the AWB actually destabilized grower incomes (Watson and Parish, 1982). Also, Newberry and Stiglitz (1981) argue that "price stabilization can lead to unstable revenues, therefore reducing revenue variation is more relevant than reducing price variation".
Boyd and Love (1994) used the Newbery and Stiglitz argument to investigate the applicability of futures options combined with crop insurance as a stabilization tool on prairie grains. They used the experience with government programs as a basis for comparison. CWB payments schedules were one component of the programs considered. Their results are interesting: they found that government policy has been a poor stabilizer because some programs are volume related and some payments are not made for at least a year. Both factors can increase income when it is already increasing, and vice versa, thereby augmenting variation. Stabilization requires market and payment returns to move in opposite directions.
We analysed CWB payments alone in this context: are CWB payments stabilizing or destabilizing to farmer income? A simple test was used: compare movements in prairie farm cash receipts from grain with movements in CWB payments received. Data were examined on a calendar year basis over ten years. For Alberta and Manitoba the results are conclusive: 70 percent of the time changes in farm cash receipts and CWB payments are in the same direction (i.e., destabilizing) because of delayed payments. For Saskatchewan, one-half of the cases had cash receipts and payments moving the same direction (destabilizing) and half were in opposite directions (stabilizing).
This simple test for the income stabilization impact of CWB pricing indicates the pricing regime destabilizes grain producer income. The results are consistent with the Australian experience (1982) and with Boyd and Love’s findings (1994). We are not attributing a strong stabilization motivation to the existence of the CWB. The results imply some destabilization cost. With a tendency to destabilize cash flow and cash receipts, it would be difficult to argue other than an increased tax burden on farmers relative to the situation in which the farmer determines cash flow and receipt patterns. There are provisions for deferred payments on grain sales but those provisions are already reflected in the data through accounting for deferred receipts received and those issued.
Farm-level resource allocation: Technical and economic efficiency
Price signals are poor
The significance of the regime of initial, adjusted interim and final payments to farm management is the additional uncertainty imposed on a producer in having to make judgements (forecasts) on prices and relative prices when market information is, at best partial, or non-existent. In the period prior to introduction of the PRO’s (before mid 1993), announcement of the initial payments and delivery opportunities prior to a new crop year was an on-again and off-again process. During part of the 1970s initial payments and delivery estimates were announced prior to seeding. In the late 1970s that form of information was terminated. The dates at which initial payments have been announced have varied from March to late July. Deciding what to select as a price forecast for crops like nonboard barley, canola or flax is a difficult task. But explicit signals exist for these markets such as up-to-date cash and forward prices, futures prices, and basis contracts. These strategies are not available to producers of CWB grains.
Use of the initial price, even if it is available, is weak as a reasonable price signal. One measure of inadequacy of the initial price is provided in Table A4 which shows average payout over the 1984-85 to 1993-94 crop year. The data show the initial averaged as high as 87.4% of total payment on barley and as low as 78% on durum wheat. Within barley, the range was from 100% in the 1985/86 and 1986/87 crop years (no final payment), to 66% in 1991/92. Durum ranged from 100% in 1990/91 to 66% in 1991/92, to 48% in 1993/94. Wheat initial prices represented 100% of the final return in 1985/86 and 1990/91, and only 61% in 1993/94. Designated barley initials were 100% of final return in 1986/87 but only 61% in 1991/92. These levels, and their variability, indicate that the information value of initial prices is low.
There is a distinct pattern in these numbers. The 1985/86 and 1990/91 crop years were associated with changes in the U.S. Farm Bill, increased export subsidies by the U.S. and Europe, and the largest pool deficits ever experienced. The data indicate that a large pool deficit is followed by a very conservative initial price. In the case of designated barley in 1991/92, the initial price was lowered dramatically even though there had been a pool surplus on the 1990/91 crop. So the CWB price signals, such as they are, are further distorted by a poor CWB (government) forecast the year before, resulting pool deficits. Some of the market signals are actually perverse in that sometimes initials were lowered when markets were rising.
These data indicate farmers experience a major problem in arriving at a dependable estimate of the level of CWB prices to expect. Without a reasonable price forecast, the relative prices which are essential to economic decision making are not available. A senior CWB official has argued that this decision making process is made worse if producers try to assess CWB prices by comparison with U.S. spot prices (Brooks, 1994). Faith in how well the CWB may sell grain is one matter. Making production decisions in an environment in which part of the information required to determine price relatives is simply not available represents a fundamental economic problem. Resource allocation problems must necessarily follow. To illustrate the fact that grain price signals are poor in western Canada, we provide a set of four figures below (Figures 7.8-7.11). The first figure compares barley price movements between Lethbridge and Great Falls, while Figure 7.9 compares price movements in the Great Falls and Portland market. What these graphs show is that the Canadian-U.S. market is poorly arbitraged. The same message comes through in Figures 7.10 and 7.11 for wheat. Figure 7.10 is an example of a well arbitraged market and Figure 7.11 is an example of a poorly arbitraged market.
The quantitative impact of poor price signals is difficult to assess because different farmers are likely to respond differently to the lack of information. The rational economic producer will surely discount CWB prices for uncertainty. That means these producers will underproduce CWB grains. CWB committed producers operating on faith may actually overproduce by placing premia on CWB grains.
Economic efficiency seems an elusive target that may be hit only by coincidence. We have attempted to measure this impact by an analysis of the efficiency of agricultural production in each prairie province. The results indicate that efficiency levels are below their theoretical potential and are variable from province to province. There undoubtedly is a link between these results and the organization of the grain market which, among other considerations, provides very poor market signals to farmers.
In order to measure some of the effects of poor price signals we have used an economic model applied in other regulated farm situations. Brada and King (1993) used this same tool to examine inefficiencies in Polish agriculture. We have estimated the relative technical and economic efficiency across the three Canadian prairie provinces. Aggregating efficiency by years provides information on the average efficiency for each province, indicating which province is most efficient. The technical details of our procedure, data used, estimation technique, and detailed results appear in appendix 7B.
Our results support the view that resources are not being used in the most efficient way, from an economic standpoint. The regulatory environment gives rise to this situation, including CWB pricing and quotas, which are a major source of distortion to producer decision making. With fewer distortions, there would be a reallocation of resources that would tend towards equalization of the value of marginal products of inputs across all three provinces. This would raise the total value of output and prairie agricultural income.
We found that the regional difference between an input’s marginal product is significant for labor, livestock, and miscellaneous inputs. The marginal product of each of these inputs is significantly higher in Saskatchewan, compared to either Alberta or Manitoba. This result is consistent with the belief that some inputs are underutilized in Saskatchewan, relative to Alberta and Manitoba. CWB influence is much greater in Saskatchewan so this is no great surprise. The higher marginal product of livestock in Saskatchewan suggests there are too few resources devoted to livestock in that province, relative to Alberta and Saskatchewan. This is partly due to the fact that feed grain prices are highly distorted and the system encourages farmers to sell to the CWB for export, rather than sell to the local livestock industry.
To put a dollar cost on this inefficiency, we took a very conservative approach and set the foregone profits at 1% of prairie farm cash receipts. This amounts to about $100 million per year, which is about $4 per tonne exported. This is an incomplete measure based on our judgement but it surely is an understatement of the cost of poor market information.
Demurrage costs, excess cleaning and port constraints
When vessels at a port are not loaded on time a charge is levied against the shipper-demurrage. If a vessel is loaded ahead of schedule the shipper receives a credit called dispatch. The Canadian system is plagued by a number of forces like weather, labour situations, and long distances that result in more demurrage than dispatch. Delays also occur when the shipper has no direct financial interest in the product. The CWB annual reports record the amount of demurrage paid out of pool accounts.
We have reviewed demurrage/dispatch data on PNW and Gulf ports in the U.S. and we discussed the problem with many trade participants. There is logistics analysis available from a recent industry consultants report (KPMG, 1995). We have calculated a twelve year average of demurrage on CWB pooled sales. For the crop years 1982 to 1993, the average demurrage in the wheat pool was $0.18/mt., on barley it was $0.38/mt., durum had a $0.04 credit and malting barley had no demurrage on average. KPMG report that in 1993 demurrage was almost half the level of excess freight costs.
These data do not look significant compared with some of the other costs tabulated. But they understate the full demurrage impact because there is spillover into other grains. Perhaps significantly, the performance in malting barley in both excess freight charges and demurrage results in no costs being charged against the malting barley pool. Certainly a direct conclusion on the efficacy of the CWB in transportation logistics should not be drawn from these facts but private companies do all the work on malting barley. CWB logistics and forwarding services are a complete duplication of those available in the private trade and this duplication is inefficient and increases costs.
Canada's reputation for consistent, high quality grain is based partly on the fact that all CWB sales are cleaned and brought to export standards. The tolerances on the top grades are very tight; so tight in fact that until recently cleaning only occurred at port because dust and contamination from normal handling to the vessel would require that inland cleaned grain be recleaned.
Cleaning at port has two effects. First, the dockage is transported to port which reduces the effective capacity of the transportation system. Little attention was paid to this factor when the government picked up about half the freight costs. Second, the cleaning-at-port is being done within a system characterized by high wage rates and limited capacity, particularly at the Pacific ports.
The economics of port versus inland cleaning is a reasonably complicated research problem which has to consider transportation costs, relative value of screenings, relative wage costs, and cleaning economics. There have been some analyses done on this problem and they indicate that inland cleaning should increase. They also indicate that Canadian grains are overcleaned. The KPMG study in 1995 indicated that the top grades are typically loaded out at less than allowable dockage; that is good for buyers but, like protein giveaway, it costs prairies farmers. The KPMG study also indicates that the cost of cleaning to export standards is $2/mt.
Mercier (1993) concluded that cleaning U.S. wheat down to levels comparable with Canadian wheat (0.1 to 0.4 percent) is not economically feasible because the market benefits from doing so do not offset the additional net cost of cleaning. In a related study, Wilson, Scherping, Cobia and Johnson (1993) found the additional costs of cleaning from 1 percent of 0.3 percent (approximately the difference between U.S. and Canada) is about 4 1/2 cents ($U.S.) per bushel. This translates to approximately $2.25/mt on wheat and $2.83/mt on barley ($Cdn.).
Excess cleaning, port constraints, and demurrage are all interrelated. If we clean beyond trade standards, we may be giving something away. The optimum cleaning level is likely a continuously changing variable, and it may be a sale-by-sale variable conditioned by whose objective function is being maximized. Excess cleaning at port may or may not be economically rational. Cleaning at port, and excess cleaning, almost certainly add costs to the system and reduce producer returns. We have not gone further into these costs; we expect that there is a net cost of several dollars per tonne to prairie producers. We have no problem in attributing most of this cost to SDS as it exists. In a competitive system, cost minimization would occur; little terminal cleaning is done in the United States. And, a SDS organization focused on farmer returns and maximizing farmer returns rather than trying to show how well international markets are manipulated would have the answers to these domestic questions.
Taxpayer costs: Foreign aid sales and government interest guarantees
The CWB has recently stated (Grain Matters, Dec. 1995) that one of the major benefits to prairie farmers of the present organization of the grain market is the value of government underwriting of CWB borrowing. The CWB has attributed an annual value of $60 million to this benefit. Their argument is that as a government backed marketing agency, the borrowing rate on their line of credit is substantially lower than rates charged commercial organizations. We have confirmed that this proposition is true but it is still unclear how the CWB arrived at a $60 million annual value.
Our best guess of the justification for this number is an interest rate benefit of just under one percentage point on the CWB’s total credit line, including long-term outstanding credit sales to countries such as Brazil, Iran, Iraq, the FSU, etc. The Auditor General has indicated some credit is extended up to forty years. The CWB’s outstanding long-term debt is currently around $7 billion. In addition to the long-term credit
guarantees, we are told that the CWB line of credit also covers normal operations (payments to producers exceeding sales receipts for part of the crop year), and cash advances. We received an estimate from two independent and knowledgeable sources that an operating credit line of about $1 billion would be reasonable for a grain operation the size of the CWB. Therefore, we believe the $7 billion long-term debt to developing countries is somehow factored into the $60 million "benefit". According to Dominion Bond Rating Service (June 27, 1995) the CWB:
borrows to finance inventory ($1.387 billion outstanding) and to finance $7.0 billion in long term receivables owed by customers (mostly third world countries).
This suggests that if the Canadian taxpayer had only given more credit to Iraq, Iran, the FSU, etc., the underwriting "benefit" would far exceed $60 million per year!
CWB long-term credit sales are government guaranteed against default. We treat the $60 million as a cost to taxpayers ($2.50 per tonne in Table 7.1). Assuming there is a limit to the amount of debt the government can guarantee, this is $7 billion in guarantees tied up by the CWB.
Thus it clearly has a cost. The CWB probably "earns" a positive interest rate spread because interest earned on outstanding credit notes exceeds the short term rates paid by the CWB for their credit line. That result would be expected during a prolonged period of prolonged interest rate declines, although it becomes risky if interest rates turn around as some very well publicized public failures have shown.
The CWB’s food aid sales are a relatively small item but worth mentioning because the CWB price paid by the Canadian taxpayer for this wheat is well above world price levels. We assumed only 500,000 tonnes of wheat are sold this way each year and that the typical overcharge is $20 per tonne, which amounts to about $0.50 per tonne when averaged across all CWB export sales.
Finally, the entry under "pool" deficit is an estimate of the average cost of CWB deficits to the taxpayer. The deficits have averaged about $2.50 per mt on wheat and $6.50 per mt on barley, over the 1980-94 time period. However, they are quite variable and have been as high as $30 per tonne in wheat (1990/91) and $35 per mt in barley (1985/86).
Conclusions
The major conclusions of this section are:
- There is a sub-optimal allocation of resource in prairie agriculture and CWB regulation is one of the contributors because it interferes with farm-level decision making and the grain handling system.
- The creation of more efficient markets would lead to improved incentives for profit maximization, efficiency gains, and higher incomes throughout the system.
- The average yearly cost of the single-desk in wheat is at least $20/mt to the farmer and an additional $5/mt to the taxpayer. In the case of barley, the figures are slightly higher.
- All of our cost estimates are considered to be very conservative in level and we have not been able to measure a signficant number of additional costs.
- In its present single-desk form, the CWB contributes significantly to inefficiencies and costs in the Canadian grain system.
|
|