Overview

A. Background

B. Issues Addressed

C. Executive Summary

D. Organization of Report

E. A Note Concerning the Formats of Some Exhibits

Section I: Methodological Approach

Section II: Data Obtained

A. Walla Walla Grain Growers/Port of Walla Walla (Sheffler)

B. Louis Dreyfus/Windust

C. Pomeroy Grain Growers/Central Ferry

D. Central Ferry Terminal Association/Port of Whitman

E. Columbia Grain Growers/Lyons Ferry

F. Almota Elevator Company/Port of Almota

G. Port of Lewiston, Clarkston, and Wilma

1. Continental/Lewiston

2. Lewis and Clark Terminal Association/Ports of Lewiston and Clarkston

a. LCTA/Lewiston

b. LCTA/Clarkston

3. Port of Wilma

Section III: Study Findings

A. The TRAC Data

B. Locational Patterns of the County Elevators

1	River Elevation Interviews
2	Comparison of TRAC and Updated Off-River Origins, Port of Walla Walla/Sheffler
3	Summary of Methodology for Updating Windust TRAC Data
4	Comparison of TRAC and Updated Off-River Origins, Windust
5	Summary of Update Methodology, Pomeroy Grain Growers/Central Ferry
6	Comparison of TRAC and Updated Snapshots, Pomeroy Grain Growers/Central Ferry
7	Comparison of TRAC and Updated Off-River Origins, Central Ferry Terminal Association/Port of Whitman
8	Comparison of TRAC and Updated Off-River Origins, Columbia Grain Growers/Lyons Ferry
9	Comparison of Updated and TRAC Off-River Origins, Port Almota
10	Summary of TRAC Update Methodology, Lower Granite Pool River Elevations
11	Information Provided by Continental/Lewiston
12	Updated Off-River Origins, Continental/Port of Lewiston
13	Information Provided by LCTA
14	Comparison of Port of Lewiston Reported and Updated Totals
15	Implied Off-River Origins, LCTA/Port of Lewiston
16	Comparison of TRAC and Updated Off-River Origins, Port of Lewiston
17	Updated Snapshot of Off-River Origins, Port of Clarkston
18	Comparison of TRAC and Updated Off-River Origins, Port of Wilma
19	Changes in Off-River Origins, By State
20	Comparison of Updated and TRAC Origination Data, By State
21	Comparison of Updated and TRAC Origination Data for Idaho State Counties and Regions
22	Comparison of Updated and Historical Total Bushels, Port of Lewiston
23	Off-River Origins of Snake River Barged Grains, By Pool
24	Comparison of Updated and TRAC Data for Washington State
25	Country Elevators Under New Ownership

OVERVIEW

This report describes research undertaken and findings developed by Jack Faucett Associates (JFA) of Bethesda, Maryland, in accordance with work order number six of its support contract (number DACW72-95-D-0004) with the Institute for Water Resources of the U.S. Army Corps of Engineers. The study involved investigation of the impacts on grain transportation of a proposed breaching of four navigation and hydropower dams on the Lower Snake River in Washington State.

A. BACKGROUND

The Walla Walla District of the United States Army Corps of Engineers (USACE) (hereafter, "the Walla Walla District") is studying a proposal to breach four navigation and hydro-power dams on the Snake River between Kennewick, Washington and Lewiston, Idaho. The study project is entitled The Lower Snake River Juvenile Salmon Migration Feasibility Study. It will include an Economic Analysis Report (EA), along with other technical analyses.

The four dams involved are the Ice Harbor Dam, the Lower Monumental Dam, the Little Goose Dam, and the Lower Granite Dam. The purpose of the proposed breachings is to restore juvenile salmon migration rates. There is some evidence (hotly disputed among the various interested parties and fisheries experts) that breaching will restore the juvenile migration rates by restoring fuller water flows and rendering the juvenile salmon less susceptible to predators.

The Walla Walla District has tasked Planning Branch of the USACE/Portland District (hereafter, "the Portland District") to conduct the navigation analysis portion of the EA. The objective of the navigation analysis is to study the effects of the proposed breachings on the cost of transporting commodities that are currently shipped on the Lower Snake River. To accomplish this objective, the Portland District will evaluate two scenarios: (1) a no-project condition in which the dams are not breached and commercial navigation continues all the way to the vicinity of Lewiston, Idaho, and (2) a with-project condition, in which the four dams are breached, and the head of commercial navigation is lowered to the vicinity of Richland, Pasco, and Kennewick, Washington, near the confluence of the Snake and Columbia Rivers.

The Portland District has tasked the Corps’ Institute for Water Resources (IWR) at Fort Belvoir, Virginia to assist it with the navigation analysis. IWR will analyze current and projected navigation patterns on the Columbia-Snake River System (CSRS) and develop modal cost comparisons.

JFA is an Economic and Public Policy consulting firm, located in Bethesda, Maryland. JFA has a support contract with IWR, who has tasked JFA to assist it with its work on this project. Specifically, IWR has tasked JFA to analyze existing data concerning the inland origination points of grains that are trucked to Snake River elevators, then barged on the Snake and Columbia, primarily to export houses in the vicinity of Portland and Vancouver, Washington. This report memorializes the resulting research performed by JFA, and documents its findings.

JFA’s work revolved around information already compiled by the Portland District and a contractor, the Transportation Research and Analysis Corporation (TRAC). In 1993, TRAC developed a quantitative profile of barge commodity movements on CSRS. The TRAC data was provided to the study team by the Portland District.

B. ISSUES ADDRESSED

IWR tasked JFA to focus on a narrow issue: whether data developed in 1993 by the Portland District and a contractor still reflect the movement patterns of grains from off-river origins to Snake River grain elevators. This purpose featured two components: (1) whether data concerning off-river-origin-to-river-elevator movements still fairly represents the operations of the river elevators, and (2) whether the locational patterns of the country elevators identified as origination points by the TRAC data still fairly represents the current patterns. Where patterns had changed or other discrepancies appeared, JFA was tasked to explain those discrepancies and resolve them where possible.

C. EXECUTIVE SUMMARY

Through the research conducted in accordance with the SOW for this project, the study team arrived at conclusions concerning each of the issues identified in the previous subsection. The study team’s findings on each issue were based primarily on the centerpiece of its research methodology; conducting interviews with the river elevator operators.

The study team interviewed and obtained information from ten of the twelve major river elevator facilities on the Snake River (excluding facilities below the Ice Harbor dam.) The facilities from whom the team did not obtain information were both Cargill and Cograin facilities, one at the Port of Walla Walla/Burbank, the other in the Lower Monumental Pool, near the Windust facility. The facility and corporate management of the Burbank elevator refused to furnish any information to the study team. The elevator in the Lower Monumental Pool was initially was unknown by the study team. It was built after the TRAC data was developed and now handles grains originating from the Odessa Union Warehouse cooperative, formerly being shipped through the Almota elevator. The resources did not exist within this project to explore the operations newly discovered of this facility anew (without any TRAC data), and, in any case, the fact that the facility is owned by Cargill and Cograin suggested that the managers of it would not cooperate with the study team.

The study team estimates that its analysis of these ten of twelve total facilities, it accounted for approximately 91 percent of total grains moving through river elevator facilities on the snake. This estimate is based on the original TRAC data and represents the total volume of grains less the Burbank total and less the non-Burbank Odessa Union Warehouse total, divided by the total volume.

With regard to how well the TRAC did continues to represent the operations of the river elevator facilities, the study team believes that the it is no longer accurate. For one thing, the total volume of grains has increased substantially (22.3 percent.) Moreover, the volumes in the predominant states have changed in different directions. Volumes out of Washington have increased by 42.1 percent, while volumes out of Idaho have shrunk by 25.5 percent. These broad indications of changed movement patterns indicate substantial changes to county-level origination patterns and pool-level river elevator destination patterns.

As will be discussed in the next section, the study team’s methodology for evaluating the continued accuracy of the TRAC data was based on, essentially, updating it. Consequently, in arriving at the conclusion that IWR ought not utilize the TRAC data, the study team developed an alternative to it. That updated data is provided with this report, along with specific recommendations concerning how IWR ought to employ it, including several caveats.

As to the second issue, the study team found that the locational patterns of the country elevators have not changed significantly since the TRAC data was developed. A few consolidations and mergers have occurred, but the locational patterns and operations of the facilities have remained the same. The locational patterns of the country elevators does not seem to be a factor influencing the accuracy of the TRAC data vis-à-vis current movement patterns.

D. ORGANIZATION OF REPORT

This report is organized into three sections. Section I describes the methodological approach employed by the study team to address the issues raised in the SOW. Section II describes the information gathered during the study team’s work. Section III concludes the report by articulating the findings of the research.

E. A NOTE CONCERNING THE FORMAT OF SOME EXHIBITS

The study team used a technique known as "upsumming" to organize the data with which it was working. Many of the exhibits in this report will portray data in an "upsummed" scheme. In case the reader is not familiar with this method for organizing data, it is briefly explained here.

Upsumming is a technique by which data is presented in levels. Each level sums to the one above it. The levels are represented by indentations of one column within a spreadsheet.

A simple example will demonstrate the concept. Over the last year, the four river elevator facilities in the Lower Granite Pool have moved approximately 48,909,713 bushels of wheat and barley. Of these, approximately 26,909,713 were moved through the two facilities at the Port of Lewiston, 2,500,000 through the Port of Clarkston, and 20,000,000 through the Port of Wilma. In an upsumming scheme this total and its three constituent totals would be represented as follows.

The critical feature of an upsumming scheme is that the everything indented by one tab space sums to the total above it, indented one fewer tab space.

This indentation scheme can contain several or more layers. For instance, suppose we were to introduce an additional level of constituent information, such as the fact that two river elevators operate at the Port of Lewiston, in addition to brokered grains that are purchased on behalf of Continental, but which move through the LCTA/Clarkston facility. That additional level of information would be reflected in an upsumming scheme as follows.

Additional layers can be added as necessary. Many of the exhibits in this report will feature as many as eight levels of detail. Also, this type of structure will underlie references in this report to "lower levels" or "higher levels" of information.

SECTION I: METHODOLOGICAL APPROACH

This section will describe the methodological approach taken by the study team to evaluate the questions presented by the SOW. The study team relied on interviews with the river elevator operators identified by the client as points of contact to evaluate these issues. The study team also utilized the TRAC data, and other sources of information as supplements, where necessary.

Early on, the study team determined that its best approach to the identifying discrepancies between the TRAC data and the current movement patterns was simply to update the TRAC data. This conclusion followed from the realization that more than a few minor discrepancies existed and the operators themselves were not -- for the most part -- eager to invest a lot of time in discussing them. The study team’s conclusions concerning the "continued accuracy" of the TRAC data rest on a macroscopic comparison of it to the updated data. These comparisons were conducted at the total, state, and county levels for originations, and in terms of destinations for both pools and elevators.

It was clear also to the study team that the most fecund source of information concerning movements off-river-to-river grain movements is the river elevator operators themselves. The majority of JFA’s resources for this project, therefore, were dedicated to conducting interviews with the river elevator operators. Eleven river elevators on the Snake were contacted. They are listed in Exhibit 1. JFA interviewed both corporate and facility managers for each facility. A list of contacts is provided as Appendix C to this report.

The decision to rely primarily on the operators was based on two factors. First, and most definitively, no alternative could be identified. The TRAC data was highly detailed, identifying each individual country elevator. Short of reproducing TRAC’s methodology -- which was not possible within the budget and scope of this job -- no source of such detailed information was identified. The study team concluded, therefore, that interviews with the facility operators themselves and their parent companies was the only reasonable means of investigating the accuracy of the detailed TRAC data.

The study team correctly anticipated that at least some of the river elevator facility and corporate managers would be hesitant to cooperate. As it turned out, this was true to some extent of all the operators, even those that eventually provided updated totals for all of its off-river origination points. At the other extreme, one operator refused to cooperate in any way. This was the Port of Walla Walla/Burbank. No information concerning its operations are presented in this report.

The reticence of the facility and corporate managers of the elevators is not surprising. In some cases, it stemmed from antagonism toward the purposes of the study which JFA’s work is supporting. More often, it resulted from competitive concerns. Particularly among the facilities with larger operations, such concerns were uniformly expressed to the study team, and in some cases dispositive in an operator’s decision to yield no information or only limited information.

In anticipation of the interviewees’ predictable tendencies toward reticence, the study team carefully strategized for the interviews. The study team attempted to be sensitive to the types and amounts of information that the operators would feel comfortable providing to the team, and also toward the amount of time and effort required for them to provide us with certain types of data. Particularly with regard to the first point, the study team found that this sensitivity tended to foster a rapport, and in the end, maximize the cooperation of the operators. Also, the study team was careful to prioritize the information it attempted to obtain. In some cases, operators were willing to cooperate, but not to devote very much time to the effort.

The issues inquired into during these interviews centered on the TRAC data. Prior to the interviews, the study team analyzed the TRAC data and converted it into slightly different form, reflecting each interviewee’s operations. The result was a "snapshot" of the off-river origins of grains moving through each of the elevators. These snapshots provided the total bushels of wheat and barley received at the top level, totals by state of origination at the second level, county totals for each state at the third level (except for originations in Montana and North Dakota, which TRAC subdivided by regions), and finally, at the fourth level, a breakdown by individual off-river origins.

By providing the operators with the TRAC snapshot of their operations, the study team made their provision of additional information a simpler task for the interviewees. It also allowed them to conceptualize more easily of the type of data we hoped to obtain from them. Moreover, with regard to their hesitance to provide additional data, the snapshots presented a fait accompli to the operators. That is, it demonstrated the point that decision-makers were going to make a decision based, in part, on numbers already developed. By cooperating with the study, the operators had an opportunity to ensure that those numbers represented their interests as accurately as possible.

The study team initially telephoned each of the interviewees and explained the purpose of the call, the purpose of its study, and how it fit in to the Walla Walla District’s study. The study team stressed that the Corps had tasked us to analyze the data to ensure that it adequately reflected the full breadth of their operations so that accurate costs could be associated with the proposed project. The TRAC snapshot was then faxed to the facility manager, and he or she was asked to review it to see how well it reflected their current operations.

This initial contact was followed up several days later by another telephone call. At this point, the study team initially inquired into the operator’s thoughts about the data with an open ended question. This was followed up with more detailed questions. These questions were not pursuant to a formal survey, but rather designed to prompt discussion of each elevator’s current operations, vis-à-vis the TRAC data. Some of the questions asked included the following.

• Are there any off-river origins listed with whom you no longer conduct business?

• Are there any off-river origins from whom you receive grains who are not listed on this list?

• Are there any off-river origins listed whom you know to have ceased operations?

• Are there any country elevators listed whose ownership has changed, or which have been consolidated with other operations?

• Do the total bushel amounts of wheat and barley reflect your current operations?

• Do the origination percentages, by state, for the grains reflect your current operations? By county? By some of the larger originators?

• Is the ratio of barley to wheat bushels still reflective of your current operations? For the total received? For each state?

• Is the ratio of farm-to-river-direct shipments to shipments originating from a country elevator approximately the same? For the total bushels received? For each state? For each county?

• Have the locational patterns of the country elevators shipping grains to your elevator changed? Do you see any elevators that are missing from the data? Do you see any that have been sold to new ownership? Are there any with whom you no longer do business? Are there any that are out of business?

In some cases, quite lengthy discussions ensued from these questions. In others, the operators provided precise, current data, with little prompting necessary. In a few cases, the operators provided only general information, or refused to cooperate. Some means existed to compensate for uncooperative operators, but to some extent, the reliability of this report’s conclusions is limited by the failure of some river elevator operators to provide meaningful information.

In several cases, the study team supplemented the initial telephone/fax interviews with site visits. The budget supported only a brief visit to the northwest. The study team therefore exercised judiciousness about whom it chose to interview in person. It was decided that the Lower Granite Pool was the most critical area, about which the least information had been initially discovered. Due to scheduling conflicts, the study team was unable to meet with the manager of the LCTA facilities, but did spend four hours with four managers at the Continental Grain facility. This interview, and subsequent faxes greatly increased the study team’s knowledge of the Continental operations, the accuracy of the TRAC data, and importantly, the volumes of grains moving through both the Port of Lewiston and the Lower Granite Pool. The study team also met with the corporate manager of the Continental facility, whose office is in Portland. The study team also visited the elevator facility at Walla Walla/Burbank, based on their initial reluctance to provide information. The study team was unable to convince the facility managers to even provide a cursory review of the TRAC snapshot. In fact, they were rather upset by the fact we possessed the data, referring to it as "proprietary," "competition sensitive," and -- perhaps most directly to their point -- "none of [our] business." The facility manager referred the study team to a representative of the Corporate owner, Cargill. Several conversations with him were equally ineffective, and, in fact, he would not agree to meet with the study team. He implied that the data had been acquired by the Corps without the consent of the Cargill, stating that facility manager may have given it to the Port, and that the Port inappropriately furnished it to the Corps.

Based on these initial interviews and site visits, the study team developed initial updates of the TRAC "snapshot" data for each river elevator. These updates were then faxed again to the facility managers, who were asked to review them again to ensure that it reflected their current operations. In some cases, these second wave of interviews yielded additional information. This was probably the result of an increased rapport between the study team and the interviewees.

The updated snapshots accurately reflect the current operations of the elevators to varying degrees and to varying levels of detail, depending upon the amount of information the operator was willing to give the study team. In one case, no information was provided. In some cases, the operator fully updated all the information contained in the snapshot. In most cases, some amount of extrapolation or implication was involved in updating the TRAC data. Section II will describe the data collected and the updated snapshots in depth.

The updated TRAC snapshots served two purposes toward the ultimate objects of this report. First, it gave the study team a basis for evaluating the TRAC data as a source of data concerning current grain movements throughout the region. Second, it gave the study team a way of providing an alternative to the TRAC data as another basis for IWR’s analyses.

SECTION II: DATA OBTAINED

This section will describe the data obtained from the river elevator operators during the interviews conducted by the study team as described in Section I. For the most part, the section is organized by river elevator. One exception is the subsection addressing the entire Lower Granite Pool. The operations within it were addressed as a unit, because the study team possessed information that tied their operations together, and in some cases used information or data concerning one facility to assist it in updating the TRAC data for another facility in the Pool.

It is important to note that the updates to the TRAC snapshots provided in this section -- and upon which the study team’s conclusions are built -- are estimates. The updates are based in part on actual data, in part on estimates provided by the facility managers, in part of extrapolations, and in part on relationships found in the original TRAC data. The study team is confident of the accuracy of these estimates, particularly at the state and county levels, but nonetheless, the updates reflect estimates, not an accounting of actual totals.

Related to this caveat is the fact that the updates do not reflect activity within a given year. Rather, the updated snapshots portray a "representative year" of operations for the facility. In some cases the facility managers provided totals for May of 1997 through April of 1998, stating that the last year of operations was a representative year. In other cases, the facility manager recommended adjustments to the TRAC data where it no longer reflected his or her operations, but did not supply any actual data. The final result, aggregating all the updated snapshots, is a new data set that fairly represents the current movement patterns of grains to the river elevators, but which cannot be associated with a particular year.

A. WALLA WALLA GRAIN GROWERS/PORT OF WALLA WALLA (SHEFFLER)

The study team interviewed two representatives of Walla Walla Grain Growers (WWGG), one at a regional headquarters located at a country elevator in Walla Walla, and another the Sheffler river elevator facility. Exhibit 2 compares the original TRAC data with the updated version of it developed by the study team.

The study team is quite confident in the accuracy of the updated data. This confidence stems from the high level of cooperation offered by both of the representatives interviewed. The manager of the river elevator itself provided the study team with actual figures concerning the current bushels received from the off-river origins.

Like many of the elevators that are operated by cooperatives, the off-river origins of grains received at Sheffler have not changed much since the TRAC data was developed.

As Exhibit 2 demonstrates, all of the original TRAC origins were WWGG country elevators. They all still truck grains to Sheffler. In terms of current off-river origins, the sole difference between the original and updated TRAC data is the existence of an additional off-river origin for Touchet Valley Grain Growers (TVGG). This addition is the result of the merger between WWGG and TVGG.

The current quantity of grains moving to the Sheffler elevator is 89.5 percent greater than in the original TRAC data. This increase stems partially from increased shipments from WWGG country elevators, and partially from the addition of shipments from TVGG country elevators. Leaving aside the addition of TVGG grains, the total shipments of WWGG country elevators has increased 43.8%. The remainder of the increase is the product of the merger.

B. LOUIS DREYFUS/WINDUST

The study team was able to acquire fairly extensive information from the facility manager at Windust. This information was obtained in the course of four telephone interviews, and several exchanges of faxes. Specifically, the information provided by the Windust facility manager included the following.

• Total bushels of wheat and barley received in 1997,

• Percent breakdowns of originations by state,

• Information that Odessa Union Warehouse Cooperative no longer ships grains to the facility, and

• Revised and new totals for farm-to-river direct shipments of wheat originating in Adams, Franklin, Lincoln, and Grant counties.

Using this information, the study team updated the TRAC data. The totals and state percentages provided by the facility manager were used as controls in restructuring the lower level data. Grains no longer moving to the facility from Odessa Union facilities were distributed to other elevator origins within the same county according to their TRAC share of that county’s elevator origins. The farm-to-river information was input directly. Where the facility manager provided percentages only for total bushels, wheat to barley ratios were taken from the TRAC data. Finally, all the remaining origination totals were scaled to match the county level data provided by the operator. A summary of the update methodology is provided in Exhibit 3.

The overall volume of grains trucked to Windust has changed little, increasing by only 2.2 percent. Noticeable changes, however, were apparent in the pattern of county origins. This seems to have been driven primarily by the loss of Odessa Union grains and Windust’s increased business with other off-river origins, particularly in Spokane and Whitman counties. Exhibit 4 provides a comparison of the original and updated TRAC snapshots for the Windust facility.

C. POMEROY GRAIN GROWERS/CENTRAL FERRY

The facility manager at Pomeroy/Central Ferry indicated that overall, the TRAC snapshot still fairly represented the current operations of his facility. The off-river origins themselves had not changed significantly, with the exception that the Inland Empire Pea Growers (IEPG) had merged with Harvest States. This merger had not affected the amounts of grains moving from the former IEPG facilities to the river elevator.

In terms of the volume of grains moving to the elevator, the facility manager spotted several inconsistencies. The most critical difference was that a significant share of grains moving farm-to-river direct from Garfield county, Washington were missing from the TRAC snapshot. Initially the facility manager estimated that these grains consisted of 2.1 million bushels of wheat and 900 thousand bushels of barley. Later, upon learning that with that adjustment the total grains originating in Garfield county constituted 27.19 percent of grains being shipped to the elevator, the facility manager revised his initial estimate. He felt that all of the off-river origins in Garfield (including the previously omitted farm-to-river shipments) should be increased proportionally until the total originating in Garfield county represented 50 percent of the total originating in Washington State.

In terms of the percentages of grains originating in other Washington counties, the facility manager estimated that less than one percent of those originating in Washington came from Adams county. Based on this information, the study team continued to use the percentage from the original TRAC snapshot, 0.68 percent. The facility manager also indicated that 35 percent originated in Whitman county, and the remainder (14.32 percent) originated from Spokane county. In terms of the split between grains originating in Washington verses Idaho, the facility manager estimated a ratio of nine-to-one. This was similar to the original TRAC snapshot.

The study team used the county level information provided by the facility manager as controls in developing updates of individual off-river origin volumes. Percentage shares implicit in the TRAC data were used to divide county totals among them. A summary of the update methodology is provided in Exhibit 5.

Overall, the update shows 15.3 percent more grains moving through the facility. Increases are apparent in each county except for Whitman county, Washington, which declined by 40.7 percent. The addition of farm-to-river grains originating in Garfield county, Washington is the most significant difference between the TRAC and updated data. If left aside, the total volume of grains moving through the facility would have declined by 21.9 percent. Exhibit 6 provides a comparison of the original and updated TRAC snapshots for the Central Ferry facility.

D. CENTRAL FERRY TERMINAL ASSOCIATION/PORT OF WHITMAN

The manager of the Central Ferry Terminal Association (CFTA) river elevator at Port of Whitman provided significant information concerning its current operations, vis-à-vis the TRAC data. Specifically, he indicated that:

• Approximately eight million bushels of wheat and barley have been were shipped to the facility between June of 1997 and April of 1998.

• Another two-hundred thousand to four-hundred thousand bushels were expected to be shipped to the facility in April and May of 1998.

• The facility receives between eight and ten million bushels per year.

• About two-thirds of these grains are wheats, the other third is barley.

• No grains originating in Kootenai county, Idaho, or Garfield and Spokane Counties, Washington are still shipped to the facility.

• The only grains shipped to the facility originate in either Whitman or Columbia Counties, Washington.

• About 8 percent of those grains originate in Whitman County, Washington.

Given the small number of off-river origins featured in the TRAC snapshot, this information was sufficient to develop a highly accurate updated representation of the facility’s current operations.

The most noticeable change is the loss of all off-river origins in Kootenai county, Idaho, as well as Columbia and Garfield counties, Washington. All of the facilities off-river origins are currently in Spokane and Whitman counties, Washington. Exhibit 7 provides a comparison of the original and updated TRAC snapshots of the CFTA operations.

E. COLUMBIA GRAIN GROWERS/LYONS FERRY

The facility manager at the Lyons Ferry terminal was quite forthcoming in providing information. He indicated that the TRAC data omitted transfers of grain from Columbia County Grain Growers (CCGG) country elevators to the facility. The total omitted was almost 2.9 million bushels. He also provided new bushel counts for four off-river origin and indicated that farm-to-river volumes out of Walla Walla County had not changed significantly. Farm-to-river movements out of Columbia County, on the other hand, increased by 151%. All but 42,000 bushels originating at TVGG disappeared. Presumably, this is related to the merger between TVGG and WWGG.

Exhibit 8 provides a comparison of the original TRAC data and the updated version of it.

F. ALMOTA ELEVATOR COMPANY/PORT OF ALMOTA

The facility manager at the Port of Almota indicated that the off-river origins and bushels of grains moving through the facility vary significantly from year to year. Nevertheless, he thought that the TRAC snapshot still reflected a representative year of operations. The sole difference he could identify was that the elevator no longer receives grains from Inland Empire Pea Growers Association. He believed however, that an increase in farm origins from Spokane county compensated for the this loss of business.

The study team was careful to substantiate the facility manager’s general assertions with specific questions. Some of these questions included inquiries about percentages of originations in different counties, the ratios of wheat to barley, and the percentages of grains received from various country elevators. Throughout, he was unable to identify any other differences between the facility’s current operations and the TRAC snapshot.

The facility manager did provide some information concerning several mergers and changes of ownership of country elevators. Inland Pea Growers, for example, was acquired by Cenex Marketing and Supply, which is in the process of merging with Harvest States. Harvest States owns the Lyons Ferry river elevator, where the grains TRAC identified as moving from Inland Empire country elevators to Almota are now being shipped. The facility manager also indicated that the Garfield Union Warehouse Co. had been purchased by the corporate owner of the Port of Almota facility, Almota Elevator Company. This acquisition had not affected the amounts of grains moving from the Garfield facility to the Almota river elevator.

A comparison of the TRAC and updated snapshots of the Port of Almota’s elevator operations is provided in Exhibit 9. It shows that the TRAC and updated snapshots are almost identical. The sole difference lies in the loss of shipment from IEPG in Spokane county. According to the facility manager, these losses were compensated for by increased farm-to-river shipments out of Spokane county. The overall total volume of grain, therefore, has not changed, reflecting the facility manager’s opinion that the TRAC data still fairly represents his operations.

G. PORTS OF LEWISTON, CLARKSTON, AND WILMA

Grains moving on the Lower Granite pool are loaded onto barges from one of four river elevators. They are the Continental Grain facility at the Port of Lewiston (hereafter, "Continental/Lewiston"); two Lewis & Clark Terminal Association facilities, one at the Port of Lewiston (hereafter, "LCTA/Lewiston"), one at the Port of Clarkston (hereafter, "LCTA/Clarkston"); and the Columbia County Grain Growers facility at the Port of Wilma (hereafter, "Port of Wilma"). With varying success, the study team was able to acquire data from these four operations. The Continental/Lewiston and Port of Wilma facility managers furnished the most information, while the management of the two LCTA facilities were more reticent.

The study team updated the TRAC snapshots of the four river elevator facilities in the Lower Granite Pool as a unit. This approach was taken for three reasons. First, little more than general information was obtained about two of the elevators in the pool (the two LCTA facilities). Second, the original TRAC data seemed to portray the operations of the Port of Lewiston as a whole, not just for the Continental elevator. This complicated the study team’s work because the other facility at the Port of Lewiston was one of the LCTA facilities about which little information was obtained. Third, the study team was able to compensate for the information it was unable to collect from LCTA through the use of other information it obtained from Continental/Lewiston. This other information pertained to historical total wheat shipments out of the Port of Lewiston and through the Lower Granite Pool. This additional information allowed the study team to accurately imply fairly detailed, current snapshots of the LCTA operations.

A line-by-line summary of the methodology for the Lower Granite Pool river elevator operations is provided in Exhibit 10. A more detailed discussion of the derivation of the updates is provided for each elevator, in the following four subsections.

1. Continental/Lewiston

The Continental/Lewiston facility managers provided robust information to the study team. Among the facilities managers interviewed, they put -- by far -- the most effort into scrutinizing the TRAC snapshot. They concluded that the data undoubtedly did not represent their operations alone, but the operations of the entire Port of Lewiston, including the LCTA/Lewiston facility. Moreover, some volumes out of southern Idaho were missing from the TRAC snapshot. The facility managers were certain that these volumes had moved to their facility for at least a decade.

Continental/Lewiston provided the study team with detailed, current information concerning the off-river origins of grains shipped to their facility. This included breakdowns by region and farm-to-river versus country elevator origins within each region. It did not include breakdowns of wheat versus barley, but the facility managers confirmed that the TRAC snapshot fairly represented that ratio. The data they provided also did not yield details at the county level or concerning specific country elevators, as that information was considered too sensitive for disclosure. A summary of the information provided to the study team by Continental/Lewiston is provided in Exhibit 11.

The facility managers at Continental/Lewiston also provided two other important pieces of data to the study team. The first was total of barge loaded wheat at the Port of Lewiston over the last twenty years. The second concerned the amounts of wheat moving through the Lower Granite Pool annually between 1994 and 1997. This information is compiled by the Walla Walla District. Both these sets of data proved helpful to the study team’s efforts to analyze the other facility at the Port of Lewiston (LCTA/Lewiston), because the facility manager there was not as cooperative as the Continental/Lewiston managers.

Based on the information collected, he study team was able to develop a comprehensive updated snapshot of the Continental/Lewiston operation. The updated regional totals were used as controls and were broken down among counties using ratios derived from the TRAC data. The same approach was used in deriving totals for particular off-river origins and wheat to barley breakdowns for each. The breakdowns between farm-to-river and elevator originations were particularly useful in this effort.

Exhibit 12 provides the updated snapshot of the Continental/Lewiston facility’s operations. No comparison is made to the original TRAC data, because it reflected operations at both elevators at the Port of Lewiston. In the next subsection, comparison of the TRAC and updated snapshots for the entire Port of Lewiston (both the Continental and LCTA facilities) is provided in the next subsection, in Exhibit 16.

The facility managers at Continental/Lewiston also provided extensive information in terms of the locational patterns of the country elevators with whom they transact business. Like the other facility managers, they were unable to identify instances where new elevators had been built of existing elevators had ceased to function. They did, however, point out several elevators that were under new ownership. None of the consolidations or mergers they identified substantially impacted the volume of grains moving to the Continental/Lewiston facility from the country elevator in question.

2. Lewis and Clark Terminal Association/Ports of Lewiston and Clarkston

Lewis and Clark operates to facilities in the Lower Granite Pool; one at the Port of Lewiston and one at the Port of Clarkston. Unfortunately, the manager of both of these facilities did not cooperate extensively with the study team’s efforts to update the TRAC data. He did, however, furnish some precise, if general, information. It is provided in Exhibit 13. The following two subsections detail the derivation of the updated snapshot for each LCTA river elevator.

a. LCTA/Lewiston

The totals for the LCTA/Lewiston operations were developed based on four sources of information. They were the following:

• The bushels totals provided by the facility manager,

• The detailed information provided to the study team concerning the Continental/Lewiston operations,

• The total for grains shipped through the Port of Lewiston in 1994, 1995, 1996 and 1997, and

• The TRAC data.

The first step involved checking the accuracy of the total provided by the LCTA facility manager against the Continental/Lewiston and Port of Lewiston totals. Lewiston/Continental provided data indicating that 9,092,600 bushels were shipped though its facility in 1997. The LCTA/Lewiston facility indicated that in a representative year, about 16,000,000 bushels move through its facility. Assuming that 1997 was a representative year for Continental/Lewiston, 25,092,600 bushels move through the Port in a representative year, according to the facilities managers interviewed by the study team. This compares favorably with the total maintained by the Port itself. Exhibit 14 compares the total grains moving through the Port for the years 1990 through 1997, with this total.

The study team concluded that the Port of Lewiston totals were consistent with the data collected by the study team, particularly because the last two years of data are within 10 percent of the estimated total. This cross-check provided an important level of confidence, requisite to the next step of the methodology, in which the LCTA/Lewiston breakdowns were implied using the Continental/Lewiston totals.

The most critical factor in developing the LCTA/Lewiston snapshot was the knowledge obtained during interviews with Continental/Lewiston -- that the TRAC snapshot represented the operations of the entire Port, not just the Continental facility. Conceptually then, the study team recognized that by backing the Continental/Lewiston figures out of the Port figures, it could imply the more detailed LCTA/Lewiston figures.

This implication was accomplished by the following steps. First, the study team scaled the original TRAC data for the Port to reflect more recent total shipments to the facility. This was done with reference to the changes in the known totals of volumes moving through the Port, which was provided to the study team by Continental/Lewiston. This data was then reorganized to match the form of the data provided by Continental/Lewiston, which featured total grains broken down by farm-to-river and elevator origins, then further broken down by regional origin. The regions were Southern Idaho, Montana, Washington, Oregon, Utah, North Dakota, and Northern Idaho. Then the more detailed data concerning shipments to the Port of Lewiston were subtracted from the re-scaled TRAC data at the regional level provided by Continental. The balance was assumed to represent the operations of the LCTA/Lewiston facility. The regional totals were distributed to specific counties using the original TRAC data. The ratios of farm-to-river-direct to country elevator originations were held constant at the regional level. The study team limited this implication process to ascertaining county level data. Exhibit 15 provides the implied snapshot of the LCTA/Lewiston elevator’s operations.

This process resulted in a reasonably accurate, implied snapshot of the LCTA/Lewiston operations. Like all of the updated data developed by the study team, the higher the level at which it is analyzed, the more accurate it is. The overall bushels totals are probably very accurate, as they are based on information obtained from the facility manager. The state totals also are very accurate, because the facility manager provided some information at this level, and because the state totals for the Continental facility are accurate.

In order to understand how the operations at the LCTA/Lewiston and Continental/Lewiston have changed since the original TRAC snapshot, it is necessary to sum the updated snapshots for each facility. This is because the original TRAC snapshot reflects operations of the entire Port. A comparison of this snapshot to the summed data for both Port of Lewiston facilities is provided in Exhibit 16.

b. LCTA/Clarkston

The LCTA/Clarkston snapshot is also based on the precise but general information provided by the facility manager. He indicated that the only grains moving through the facility were either barley from Asotin County, Washington, or Wheats originating in Montana and North Dakota. Using these totals as controls, the study team distributed the wheat among the regions of Montana and North Dakota indicated in the TRAC data by reference to regional ratios for each state implicit in the TRAC data. Exhibit 17 provides the updated snapshot of the LCTA/Clarkston facility’s operations. The original TRAC data is provided because it differs so much from the original TRAC snapshot that no meaningful comparison can be made.

The Clarkston updated snapshot differs significantly from the TRAC snapshot. The study team attributed this to the fact that the two LCTA facilities in the Lower Granite Pool are just across the Snake River from one another and probably operate in tandem. The facility manager generally agreed with the validity of this assumption, although -- as already discussed -- he did not invest much time in looking over the original TRAC data, opting instead to provide the study team with precise but general information. Whether or not the assumed explanation is valid does not affect the study team’s confidence in the totals presented in the updated snapshot. They were derived directly from the facility manager, and both the pool totals of the updates of the four facilities in the Lower Granite Pool, as well as the Port of Lewiston totals for the two facilities there, are consistent with the Pool totals and Port totals that the study team obtained from Continental/Lewiston.

3. Port of Wilma

The data collected from the Port of Wilma was extensive, perhaps the most extensive of all the facilities. This allowed the study team to develop a highly accurate updated snapshot of the Port of Wilma’s operations. The facility manager provided the study team with the following pieces of information.

• The off-river origins and amounts of grains moving through the facility have not changed significantly over the last ten years,

• Currently, around 20 million bushels of grains move through the facility,

• Of these, 16 million are wheats, and 4 million are barley,

• The percentages of grains originating in the five states have not changed significantly,

• Total grains originating in Nezperce county, Idaho as a percent of total grains originating in Idaho, is now 22.55% instead of 27.55%.

• Total grains originating in Lewis county, Idaho as a percent of total grains originating in Idaho, is now 16.27% instead of 13.77%.

• Total grains originating in Idaho county, Idaho as a percent of total grains originating in Idaho, is now 16.27% instead of 13.77%.

• The totals of grains originating in all other Idaho counties, as a percent of the Idaho total, is still accurate,

• Seventy-five percent of the grains originating in Washington state and moving through the facility originate in Whitman county,

• Twenty-five percent of the grains originating in Washington state and moving through the facility originate in Asotin county,

• Approximately five percent of the grains originating in Whitman and Asotin counties, Washington are barley,

• The regional percentages of grain originations for North Dakota and Montana are still accurate,

• The ratio of farm-to-river direct shipments and country elevator shipments has changed from 3-to-1 to 3-to-2.

The overall volume of grains moving through the Wilma elevator is 53.3 percent greater in the update than in the original TRAC data. This increase is fairly uniformly spread among the various counties and off-river origins in the original TRAC data. Exhibit 18 provides a comparison of the original TRAC data to the updated, current snapshot.