(Cite as: 57 FR 22570, *22587)
Carbon Steel Wire Rod from Poland, 49 FR 19374 (1984), and its companion cases, and Georgetown, which affirmed Wire Rod, for
the proposition that the Department cannot determine that a countervailable subsidy exists where it has been established that no
market distortion exists. In Respondents' own words:
A government only confers a countervailable subsidy when it acts in such a way as to create a market distortion, that is, when its
action results in greater production by the recipient than would be the case in the absence of governmental action, or in a
lowering of prices by the recipient in competition with American producers.
Respondents' Joint Case Brief Concerning Alleged Stumpage Subsidies and Preferentiality, Vol. III-A, p. 111-24 (Apr. 21, 1992).
The first issue raised by Respondents' argument is whether the Department must conduct what has become known in this case as a
"market distortion test" in applying the countervailing duty law to imports from market economy countries; i.e., that a
countervailable subsidy cannon be found where there is a demonstration that the subsidy in question does not affect production
or price. Respondents rely heavily on statements made by the Department in Wire Rod, in which the Department determined that
the countervailing duty law did not apply to imports from nonmarket economy countries. In particular, Respondents cited
the following statement from Wire Rod:
We believe that a subsidy (or bounty or grant) is definitionally any action
(Cite as: 57 FR 22570, *22587)
that distorts or subverts the market process and results in misallocation of resources, encouraging inefficient production and
lessening world wealth.
49 FR at 19375. Respondents also rely heavily on the following statement made by the Department in the background section of
its Proposed Regulations:
Conceptually, the regulations are based upon the economic model articulated by the Department in its final determinations in
Carbon Steel Wire Rod from Czechoslovakia and Carbon Steel Wire Rod from Poland * * * and sustained by the court in
Georgetown Steel Corp. v. United States * * *. This model, which generally defines a subsidy as a distortion of the market process
for allocating an economy's resources, underlies the Department's entire CVD methodology.
54 FR 23366, 23367 (1989) (citations omitted). From these statements, Respondents conclude that a market distortion test is
required as a matter of law.
We do not agree with Respondents' contention that the countervailing duty law requires a market distortion test. First, by
relying on selected statements from Wire Rod, Respondents have misrepresented the nature of the issue involved in Wire Rod and
the thrust of the Department's statements therein. The issue in Wire Rod was whether Congress intended that the
countervailing duty law apply to imports from nonmarket economy countries. The Department, starting from the premise
that subsidies are a distortion of the market process,
(Cite as: 57 FR 22570, *22587)
reasoned that "(s)ubsidies have no meaning outside the context of a market economy", 49 FR at 19375, and that a market
benchmark is needed to identify a subsidy:
To identify subsidies in this pure market economy, we would look to the treatment a firm or sector would receive absent
government action. In the absence of the bounty or grant, the firm would experience market-determined costs for its inputs and
receive a market-determined price for its output. The subsidy received by the firm would be the difference between the special
treatment and the market treatment. Thus, the market provides the necessary reference point for identifying and calculating the
amount of the bounty or grant. (Emphasis added.)
Id. As this statement indicates, the Department never suggested in Wire Rod that a market distortion test, as proposed by
Respondents, would necessarily form part of the Department's analysis in a countervailing duty case involving imports
from a market economy country. To the contrary, to the extent that one can read anything into this statement, it is that in a
market economy case, the Department's analysis would be based on a comparison of a market-based benchmark and a
government-provided price and that the existence of what Respondents call a market distortion would normally be presumed.
Likewise, when the Department referred to Wire Rod and Georgetown in its Proposed Regulations, it meant only that its
countervailing duty methodology was based
(Cite as: 57 FR 22570, *22587)
on the use of market benchmarks to determine the existence and value of a subsidy. This reference certainly was not intended to
mean that it would be necessary to conduct the sort of market distortion test proposed by Respondents.
Amongst all of their quotes and citations, Respondents conveniently ignore what was the key statement in Wire Rod. After
describing the manner in which nonmarket economies operate, the Department stated:
In such a situation, we could not disaggregate government actions in such a way as to identify the exceptional action that is a
subsidy. Because the notion of a subsidy is, by definition, a market phenomenon, it does not apply in a nonmarket setting. To
impose that concept where it has no meaning would force us to identify every government action as a subsidy (or a tax). We are
not prepared to do this--we will not impose the *22588
(Cite as: 57 FR 22570, *22588)
market-based concept of a subsidy on a system where it has no meaning and cannot be identified or fairly quantified.
49 FR at 19376. In short, all the Department meant in Wire Rod was that it was meaningless to talk of subsidies in the context of
nonmarket economies. Thus, neither the Department nor the Court addressed the question of whether a market distortion test
would be required in a market economy countervailing duty investigation.
Thus, we do not agree with Respondents that Wire Rod and Georgetown establish
(Cite as: 57 FR 22570, *22588)
a precedent for the type of market distortion test they envision. Nevertheless, there remains the question as to whether the
Department is free to incorporate such a test into its countervailing duty methodology, or whether Congress precluded the
adoption of such a test. While we agree with Respondents that one of the reasons for having a countervailing duty law in the
first place is to combat the market distortions that subsidies may bring about, we do not believe that Congress intended that the
finding of a countervailable subsidy had to be based on the actual application of a "market distortion" analysis in individual cases.
The best evidence of this lies in the legislative history of the Trade Agreements Act of 1979 and the discussion therein of the
practice of the U.S. Department of the Treasury concerning regional subsidies and offsets. Prior to the enactment of the 1979 Act
and the transfer of responsibility for administration of the countervailing duty law to the Department, Treasury had a
practice of taking into account the effects of government subsidies on the competitive positions of firms receiving such subsidies.
For example, if a firm received a $100 million grant in order to build a factory in a disadvantaged region, but also incurred $50
million in additional costs as a result of locating in the disadvantaged region, Treasury reduced the subsidy by $50 million. In
theory, if the amount of the additional costs equalled or exceeded the amount of the subsidy, Treasury would find no subsidy at
all.
(Cite as: 57 FR 22570, *22588)
Although this practice did not amount to a determination of market distortion, it did reflect the view that it was appropriate in
certain circumstances to look behind the existence of a subsidy in an attempt to identify the net economic effect on the subsidy
recipient. Respondents have urged the Department to undertake just such a practice in relying on Nordhaus' arguments that there
is no net economic benefit or effect from stumpage programs. Congress was dissatisfied with Treasury's practice, and in enacting
section 771(6) of the Act, 19 U.S.C. section 1677(6), the "offset" provision, it clearly indicated that the administering authority
was not to engage in this type of analysis.
Commentators who have critiqued the Department's countervailing duty methodology, whether speaking of "market
distortion" or "entitlement", have often cited this example as a situation where the Department imposes countervailing duties
in the absence of any showing that a foreign producer's marginal cost has been affected. See, e.g., R. Diamond, Economic
Foundations of Countervailing Duty Law, 29 Va. J. Int'l L. 767, 788 (1989). Yet at the same time, these commentators
generally agree that the Department must find that a subsidy exists in this type of situation. See, e.g., id., note 59; and Cass, Trade
Subsidy Law: Can a Foolish Inconsistency Be Good Enough for Government Work? 21 L.& Pol'y Int'l Bus. 609, 641-42 (1990).
Respondents have essentially contended that the legislative history
(Cite as: 57 FR 22570, *22588)
does not preclude the Department from making its determination of the existence of a gross subsidy based on the same factors
which Congress indicated could not be taken into account for purposes of determining the existence of a net subsidy. Modifying
the example set forth above, if the additional costs incurred by the firm for locating in the disadvantaged region were $100
million, Respondents would presumably argue that the Department properly could determine that no gross subsidy exists.
However, this leads to an anomalous situation. If, as Respondents postulate, Congress intended that the amount of subsidy be
equated to the net economic effect on the subsidy recipient in a particular case, then the amount of subsidy existing in the first
example would be only $50 million. Yet the statute is clear, and Respondents do not contest, that under the law the Department
would have no alternative but to find a subsidy of $100 million.
Congress could not have intended such an anomalous result from the enactment of the offset provision. Given that this is the only
instance cited in the legislative history which discusses the type of analysis suggested by Respondents, we cannot accept the
notion that Congress would endorse or condone our engaging in any market distortion analysis in circumstances where the
subsidy is otherwise capable of being identified and measured following established statutory or regulatory principles.
Moreover, other than their references to Wire Rod and Georgetown and a few
(Cite as: 57 FR 22570, *22588)
other cases taken out of context, Respondents cite to no other cases supporting the proposition that the Department must apply a
market distortion test. They also do not cite to the countervailing duty laws of other countries or determinations
thereunder which would support the application of a market distortion test. There is nothing in the current GATT Subsidies Code
which mandates such a test. Finally, nothing in the draft subsidies agreement, MTN.TNC/W/FA, Part I, prepared as part of the
Uruguay Round of multilateral trade negotiations, supports a market distortion test. Indeed, that document describes a
countervailing duty methodology which is quite similar to existing Department practice.
In conclusion, therefore, we do not believe that the Department's precedents support the application of a market distortion test in
the circumstances before us here. Accordingly, we determine that the Department is precluded from measuring the benefit
conferred by stumpage programs on the basis of a market distortion analysis, such as the effect of stumpage prices on output.
However, we do consider it appropriate to comment on the information and argumentation placed on the record by Respondents
in connection with Dr. Nordhaus' study, if only because Respondents have used these analyses in an effort to show that it is
impossible for the Department to find that the provincial governments are subsidizing Canadian softwood lumber producers
through the provision of stumpage at administratively-set rates.
(Cite as: 57 FR 22570, *22588)
Comments on Dr. Nordhaus' Analysis
As an initial comment, it bears mentioning that although Respondents assert that Dr. Nordhaus' theoretical views concerning the
economics of stumpage markets are well accepted, they nonetheless fail to provide any independent support for this claim. See
Respondents' April 21, 1992 brief, Vol. III-B, Attachment III-2, p. 12. The Department, therefore, has no basis in the record to
accept the validity of Dr Nordhaus' analyses at face value, particularly insofar as other economists with at least as reputable a
name in the field of forestry economics appear to espouse somewhat conflicting points of view. See Coalition's April 21, 1992 brief,
Appendices 60 and 61.
As to the substance of Dr. Nordhaus' views, under his theory of economic rent, he contends that the harvest of *22589
(Cite as: 57 FR 22570, *22589)
stumpage from the provincial stumpage programs now in effect cannot exceed that of a competitive market. Furthermore, Dr.
Nordhaus contends that the price of stumpage under these provincial programs will be higher than what would prevail in a system
where stumpage prices were set competitively.
Dr. Nordhaus' analysis begins with stumpage harvest, where he notes that under provincial stumpage programs, timber harvest is
set on a sustained yield basis so that harvests can be maintained at that level into the future. At the same
(Cite as: 57 FR 22570, *22589)
time, he assumes that since a private market will seek to maximize value, the harvest under these provincial stumpage programs
must be less than under a competitive situation because the provincial stumpage programs are constrained to follow sustained
yield policies.
Dr. Nordhaus then goes on to describe stumpage prices and harvests under three scenarios: Net benefits, excessive stumpage, and
normal stumpage. "Net benefits" would exist when the stumpage charges are less than the commercial benefits that the
government is actually providing to the tenure holder. The effect, therefore, would be to confer a subsidy on the tenure holder
and to increase the harvest of timber. In his next scenario, "excessive stumpage," stumpage charges are so high that they raise the
cost of harvesting the timber above the market price for stumpage, thereby resulting in a decrease in stumpage harvested. Finally,
under "normal stumpage," stumpage charges are positive, but not so high as to turn profits negative within the normal range. Dr.
Nordhaus thus contends that normal stumpage has no effect on harvest.
First, we take issue with Dr. Nordhaus' contention that stumpage charges under the provincial stumpage programs will necessarily
be higher that those in a competitive market. Under provincial stumpage programs, each purchaser is given the right to harvest
timber upon payment of the administered stumpage charge. In a competitive market, although the seller of stumpage may set a
minimum price below which stumpage will not be sold, potential buyers will
(Cite as: 57 FR 22570, *22589)
still bid against each other for the right to harvest the stumpage. Any price determined in this fashion will almost always be higher
than an administered stumpage charge.
Second, we are unconvinced by Dr. Nordhaus' conclusion that the stumpage harvest under the provincial stumpage programs will
always be lower than the harvest under a competitive market. Depending on the objectives of either a provincial stumpage
program or a private owner, a forest can be managed to provide a high or low level of sustained harvest. For example, provincial
stumpage programs managed under a policy of sustained yield tend to use biological criteria for choosing rotation lengths, which
is the time period between the establishment and the harvesting of a timber stand. The use of such criteria tends to give a higher
level of timber harvest than the economic criteria which many private forest owners use to set rotation lengths. In addition to
rotation length, there are other factors which can affect the level of sustained harvest, such as the selection of tree species to
plant, the number of trees to plant per acre, and whether to clear-cut or use a single tree method of harvesting. Varying the level
of these and other factors will determine whether the level of sustained harvest is high or low.
Finally, with regard to the "residual value" or "economic rent" approach used by Dr. Nordhaus to arrive at his conclusions, we note
the following remark by G. Robinson Gregory, in his book "Resource Economics for Foresters," concerning
(Cite as: 57 FR 22570, *22589)
the application of the notion of economic rent to stumpage:
Neither rent theory nor the related appraisal procedure involve consideration of stumpage production costs or the possibility of a
reservation price on the part of forest owners. It is implicitly assumed that the supply of all productive factors other than timber
is perfectly elastic, that the supply of timber offered is perfectly inelastic with respect to price, and that the forest owner is in a
position to extract the rent. With these assumptions, the rent-based appraisal model of stumpage pricing provides little assistance
for analyzing the effect of changing timber supplies and/or production costs on either stumpage prices or on product prices.
See Coalition's April 21, 1992, brief, Attachment 61, p. 215.
The phrase "the supply of timber offered is perfectly inelastic with respect to price" means that the supply of timber does not vary
with price; the reference to "perfectly elastic" means that the price of a productive factor does not vary with the quantity
consumed. Finally, the "reservation price" is the price necessary to persuade an owner to sell stumpage.
The validity of the economic rent model proposed by Dr. Nordhaus depends on whether the supply of stumpage varies with price.
In this regard, Dr. Nordhaus has not taken account of the existence of intensive and extensive margins in timber harvesting. The
extensive margin means that, at any particular stumpage price, only certain categories of stands can be profitably harvested. As
the
(Cite as: 57 FR 22570, *22589)
price of stumpage drops, more and more stands become economically accessible, which allows the supply of stumpage to
increase. The intensive margin concept applies to trees within a stand that is currently economically accessible. It recognizes
that, within each stand, there are certain categories of trees that cannot be profitably harvested at a given stumpage price. If
stumpage prices are lowered, the intensive margin is expanded so that the formerly unutilizab1e trees within a particular stand
can be profitably harvested, thereby increasing the supply of timber. Consequently, Dr. Nordhaus' assumption that the supply of
timber does not vary with price appears to be at odds with the very way in which stumpage markets behave.
Dr. Nordhaus states that his theory of the economic rent of stumpage is a static analysis. That is, the analysis considers only one
period in time. (See Respondents' April 27, 1992 Brief, section 3, Attachment A at 9.) Indeed, Gregory agrees with Nordhaus when
he states that rent theory is applicable only in a stable, short-run analysis, which essentially implies a static, single period
analysis. (See Coalition's April 21, 1992 brief, appendix 61 at 215.) Nordhaus provides a two-period model to demonstrate that his
theory can be extended to include dynamic aspects (changes over time), but his example is really nothing more than a simple
extension of a theory which is essentially static in nature. The Department believes that a theory of stumpage supply which is
based on dynamic analysis, as opposed to a static analysis, and one in
(Cite as: 57 FR 22570, *22589)
which stumpage is not fixed in supply, provides a more realistic representation of stumpage harvest from provincial stumpage
programs. As such, the Department finds the economic rent theory of stumpage advanced by Dr. Nordhaus to be flawed in several
respects. By no means does it show that provincial stumpage programs are not countervailable subsidies.
In conjunction with their market distortion analysis, Respondents argue that, because Canadian forest products companies have
lower rates of return relative to other Canadian companies, provincial tenure holders are not receiving net benefits as defined by
Dr. Nordhaus under provincial stumpage programs. As such, according to Dr. Nordhaus' and Dr. Litan's analysis, if no net benefit
is being provided, no subsidy is being conferred. Further, Respondents argue that if Canadian firms were receiving preferential
treatment, presumably these firms would have supra-normal profits.
*22590
(Cite as: 57 FR 22570, *22590)
First, we do not consider a comparison of rates of return between industries to be relevant in determining whether a particular
industry or group of industries has received a countervailable benefit. Indeed, using such an analysis to prove whether a subsidy
exists turns our subsidy practice on its head. For example, many failing companies or industries with low or negative rates of
return have received massive amounts of government equity infusions, grants, and loans and still experienced negative rates of
return (see, e.g., Final Affirmative Countervailing Duty Determination: New Steel Rail, Except
(Cite as: 57 FR 22570, *22590)
Light Rail, From Canada, 54 FR 31991 (August 3, 1989). Following Respondents' reasoning, there would be no net benefits
because the company or industry had a lower rate of return than the rest of the industries in the country. Moreover, it is not
unusual for different industries to have different rates of return regardless of whether the government has intervened.
Second, we also note that in Dr. Litan's analysis comparing forest product companies' rate of return to other companies in
Canada, the 13 forest product companies listed include many vertically and horizontally integrated producers of both lumber
and pulp and paper; and that the remaining companies include financial service companies which are inappropriate to use as a
basis of comparison for manufacturing companies. Accordingly, we determine that comparing rates of return is not a valid
method for determining whether a subsidy is being conferred by stumpage programs.
In addition, Respondents argue that the results of their TSPIRS analysis shows that the government is providing no net benefits to
softwood lumber producers in accordance with Dr. Nordhaus' theory. First, as explained above, the Department does not measure
the net benefits provided to a subsidy recipient. Second, as explained below, the Department considers that Respondents have
failed to demonstrate that their modified TSPIRS analysis would accurately reflect the government's cost of providing the good in
this investigation.
(Cite as: 57 FR 22570, *22590)
Preferentiality Hierarchy
Respondents next contend that, to the extent that the Department decides to ignore the question of whether market distortion has
occurred in determining whether stumpage programs provide countervailable subsidies, it still cannot be blindly wedded to the
particular methodological formulation or sequence set forth in the 1986 Preferentiality Appendix. Respondents charge that, in its
Preliminary Determination, the Department engaged in a "rote application" of its preferentiality benchmarks without regard or
explanation as to whether the preferentiality benchmarks that it selected were appropriate to use in this case.
While Congress specified that the provision of goods or services at preferential rates is one example of a countervailable domestic
subsidy, it did not specify the manner in which preferentiality was to be determined. Although the Proposed Regulations are
intended to codify the methodology which the Department has used in particular cases to determine when goods or services have
been provided at preferential rates, Respondents stress that these remain only proposed rules and were never intended to
prescribe an "immutable formula" for all future cases. Citing IPSCO, Inc. v. United States, 687 F. Supp. 614 (Ct. Int'l Trade 1988),
Saudi Iron and Steel Co. v. United States, 686 F.
(Cite as: 57 FR 22570, *22590)
Supp. 914 (Ct. Int'l Trade 1988), dism'd on other grounds, 698 F. Supp. 912 (Ct. Int'l Trade 1988), and other countervailing
duty and administrative law cases, Respondents assert that the preferentiality hierarchy in the Proposed Regulations cannot
provide a stand-alone rationale for finding provincial stumpage fees to constitute the preferential provision of a good or service.
Rather, the Department must provide a reasoned explanation as to why any particular benchmark is "appropriate in light of the
facts of this case and the economic principles applicable to those facts" (emphasis in the original).
It is, of course, incontestable that the Department is obliged to make a reasoned decision reflecting the facts of the case and to
provide a complete explanation of the basis for reaching that decision. This principle was fully respected in the Department's
Preliminary Determination, just as it is now being respected in this final determination. However, to ensure that the facts of a case
are being considered in a manner which is fair and understandable to all of the parties concerned, our analysis cannot be
conducted in a methodological vacuum which ignores past practice or the agency's regulatory guidelines, be they "proposed" or
final. To do so would constitute little more than benchmark-shopping which, presumably, the Respondents would agree could
result in an arbitrary outcome.
As the Respondents correctly point out, the Department's methodological framework for valuing countervailable benefits
conferred through the provision
(Cite as: 57 FR 22570, *22590)
of goods or services is grounded in the statute: whether goods or services are provided or required to be provided by government
action to a specific enterprise or industry or group of enterprises or industries at "preferential rates." On this statutory foundation,
the Department built a hierarchical methodology for determining and measuring when goods or services are being provided at
preferential rates in the interest of maximizing administrative certainty and predictability in an area where (as Respondents again
correctly note) the statute did not provide considerable interpretative guidance. The fact that the Department indicated at the
time that it put forth both its Preferentiality Appendix and, later, its Proposed Regulations that these tests or rankings may not
always yield the most appropriate means of measurement in every case does not detract from their usefulness or applicability as
our primary methodological tool in such circumstances.
The ranking of the preferentiality hierarchy should not strike parties as unfamiliar or illogical as it reflects the fundamental
standard of measurement established by Congress in section 771(5)(A)(ii)(II)--i.e., preference. Thus, the most common test
which the Department has applied in determining preferentiality is whether the government (or government-directed supplier) is
providing a good or service at a price that is lower than the prices the government charges to the same or other users of that
product within the same political jurisdiction. Insofar as the exercising of price discrimination by
(Cite as: 57 FR 22570, *22590)
the same seller for the same product provides the clearest possible manifestation of whether preference exists, there is little need
to justify in each case why such a standard would be appropriate to determining whether goods or services are being provided at
preferential rates.
However, even in those cases where comparisons based on price discrimination within the jurisdiction cannot reliably be made,
the Department's sequential alternatives nonetheless flow naturally from the preferred test following certain fundamental
principles: (1) That preference is most commonly manifested through the behavior of the provider of the good or service; (2) that
domestic subsidies should be determined on the basis of comparisons within the same political jurisdiction; and (3) that prices
offer the most reasonable basis for making a *22591
(Cite as: 57 FR 22570, *22591)
comparison. While the first principle relates directly to the standard of measurement articulated in the statute, the other two
principles are identifiable philosophical threads which run through the entirety of our countervailing duty practice.
As a result, what the preferentiality hierarchy presents is not a "mandatory roadmap" but rather a series of conceptual guideposts
for evaluating the facts of a specific case in the context of Congressional intent and Departmental practice. The hierarchy of
benchmarks constitutes a preferred sequence insofar as, in most cases, it will faithfully reflect both the meaning of the term
"preferential" and the historical application of the countervailing duty law.
(Cite as: 57 FR 22570, *22591)
This does not mean that the ranking is "immutable"; it does mean that the Department will follow the ranking unless presented with
facts or arguments demonstrating that it is inappropriate, which was not the case here. We note also that Respondents,
themselves, expressed no objection to the ranking of the benchmarks as a general rule.
Having determined that a demonstration of "market distortion" is not a prerequisite for the identification of a countervailable
domestic subsidy, and having found that provincial stumpage programs are specific within the meaning of the statute, the
Department has applied its preferentiality hierarchy in the order described in the Proposed Regulations. In examining the
provincial stumpage programs under investigation, we have determined that the facts and information on the record permit a
finding of preferentiality which accommodates the law's and the Department's general predisposition towards a comparison of
actual prices within the relevant jurisdiction, without having to resort to other benchmarks which the Department has generally
recognized to be less consistent with the fundamental principles underlying the preferentiality standard.
This brings us to the next of Respondents' methodological arguments with respect to the identification and valuation of stumpage
subsidies--viz., that the cost benchmark is the only appropriate measure on the preferentiality hierarchy for determining whether
stumpage confers a subsidy.
(Cite as: 57 FR 22570, *22591)
Respondents, particularly BC, argue that the third alternative benchmark, the government's cost of providing the good or service,
is the preferred means for determining whether provincial stumpage programs confer a subsidy. They contend that a cost-based
comparison, when properly calculated, results in a finding of no preferentiality. As proof, they have submitted a modified TSPIRS
analysis (Timber Sales Program Information Reporting System--a three-part reporting system designed by the U.S. Department of
Agriculture Forest Service (Forest Service) to measure whether national forest timber is being sold at "below-cost" prices)
purportedly demonstrating that revenues exceed costs on a provincial basis in Alberta, British Columbia, Ontario and Quebec. BC,
in particular, argues that the facts in this case do not support a price-based comparison for that province, and that the
Department's current abandonment of the 1986 benchmark (i.e., cost) in favor of a comparison to competitive SBFEP prices is a
results-oriented contrivance designed to subvert the replacement costs instituted by BC under the MOU in order to obtain a
subsidy.
Respondents argue that a TSPIRS-type, cost-based analysis is the most appropriate methodology for determining whether
subsidies exist. This methodology, they contend, was designed by the Forest Service and the General Accounting Office to
determine whether "below-cost" sales of timber were occurring on national park sales. It has been officially adopted in the United
States as a measurement tool since 1989. To apply a different standard in this
(Cite as: 57 FR 22570, *22591)
case than that used by the Forest Service, Respondents claim, would be arbitrary and capricious. They further contend that if, as
was indicated in an April 20, 1992 memorandum from Marie Parker, Director, Office of Accounting, the Department had questions
regarding Canada's application of TSPIRS or concerns regarding certain data, these questions should have been raised in the
Department's questionnaire or at verification.
As explained above, the Department follows its hierarchy of benchmarks unless presented with facts or arguments demonstrating
that its application of this hierarchy is inappropriate. As the provincial preferentiality sections show, appropriate price-based
benchmarks exist in each of the provinces. Furthermore, the Department has long acknowledged that cost-based analyses pose
exceptional problems when they are attempted to be used to measure preferentiality in the provision of a natural resource. See
Preferentiality Appendix.
Even though endorsing the presentation of arguments regarding cost issues in the joint case brief, Alberta, Ontario and Quebec
have all raised concerns regarding the use of a cost-based benchmark in their respective provinces. At the hearing, Ontario stated
that, "* * * in the context of Ontario, a (cost) benchmark would not be an appropriate measure." See Hearing Transcript, Volume
II, p. 93. Ontario further stated that "* * * there's so much difficulty in Ontario on sorting out what kind of expenditures are
undertaken for what
(Cite as: 57 FR 22570, *22591)
purpose that it simply is not possible to come up with a result that isn't arbitrary." (Emphasis added.) See Hearing Transcript,
Volume II, p. 96. Alberta states that, "* * * it is clear that the Department need not, and under the Department's preferentiality
hierarchy should not, analyze the sale of timber harvesting rights in Alberta using any of its other less favored alternative
preferentiality benchmarks (i.e., private prices within Alberta, cost, and cross border)." (Emphasis added.) See Alberta Case Brief,
p. VI-31. Finally, Quebec, in its case brief on cost issues states,
Private stumpage prices in Quebec served as the benchmark in the Department's preliminary determination. Quebec endorses that
approach and has demonstrated that private market prices in Quebec are the appropriate benchmark to use to determine whether
public stumpage prices in Quebec are preferential, if a preferentiality analysis is employed. The Department's verification
confirmed this benchmark as the only one appropriate in the final determination, consistent with the Department's established
preferentiality hierarchy.
(Emphasis added.)
See Quebec Case Brief on Cost Issues, Tab 2, p. 2.
With regard to BC's argument that Lumber II and the MOU established cost as the appropriate benchmark for measuring stumpage
preferentiality, we note that the Department's use of a cost benchmark in the 1986 Preliminary Determination was predicated
upon the lack of appropriate price-based benchmarks and/or the
(Cite as: 57 FR 22570, *22591)
lack of information regarding the adjustments necessary to the price-based benchmarks. As outlined in each of the provincial
preferentiality sections, price-based nonpreferential benchmarks are now available in each of the provinces, as is the information
necessary for making all appropriate adjustments to the benchmarks. The MOU export tax rates and the provisions regarding
replacement measures were the result of negotiations. The fact that cost-based replacement measures for static negotiated export
rates were allowed under the MOU is immaterial to the Department's subsidy analysis in a countervailing duty
investigation.
Even assuming arguendo that cost were the appropriate measure of *22592
(Cite as: 57 FR 22570, *22592)
preferentiality in this case, the Department does not believe that the modified TSPIRS approach used by the Respondents is the
appropriate methodology for determining whether provincial stumpage programs are providing timber to the softwood lumber
industry at preferential prices.
First, and foremost, TSPIRS was not developed to determine whether a government was providing a subsidy to users of timber.
TSPIRS was developed in response to growing public and Congressional concern over the economics of timber sales and timber
management in the national forests, in particular, so- called "below-cost" sales of timber from national forest land. See Timber Sale
Program Information Reporting System: Final Report to Congress, U.S. Department of Agriculture, Forest Service, p. 5 (TSPIRS
Final Report). However, while the
(Cite as: 57 FR 22570, *22592)
term "below-cost sales" is found in both TSPIRS methodology and the unfair trade laws, its meaning in the two contexts is not
synonymous. In fact, as admitted in the Final Report to Congress regarding TSPIRS, "(t)here is no consensus on the definition of a
"below-cost" timber sale." See TSPIRS Final Report, p. 5. Furthermore, TSPIRS, as applied by the Forest Service, requires a
three-part analysis involving the consideration of a number of factors in order to determine whether the goals of the Forest
Service regarding national forest sales are being met. In fact, both the GAO and Congress have stated that an examination of any
one of the reports outside the context of the other two is inappropriate. See TSPIRS Final Report. The ad hoc TSPIRS analysis
conducted by the Respondents does not include all three parts.
Second, TSPIRS was designed to be applied strictly on a national forest basis. The Respondents' use of one section of this reporting
system on a province-wide basis represents a significant methodological modification of TSPIRS.
Third, several modifications and estimations made by Respondents when applying TSPIRS raise serious concern as to the results
of Respondents' TSPIRS analysis. Those of primary concern relate to in-kind services, so-called stewardship activities and backlog
silviculture.
All in-kind services are recognized as revenue in the current year even though the "benefit" provided by these services may occur
over an extended period of
(Cite as: 57 FR 22570, *22592)
time. The provinces excluded all stewardship activities from their cost pools. However, activities paid for or performed by tenure
holders that are considered by the provinces to be stewardship activities were recognized as revenues. Backlog silviculture
responsibilities were also not included in the provinces' TSPIRS cost pools. These backlog responsibilities arose out of the failure
of the provinces to perform certain silviculture activities necessary to maintaining the government's policy of sustained yield and
represent significant liabilities that would be included in a Forest Service TSPIRS analysis.
Finally, Respondents contention that the Department should have addressed its TSPIRS concerns in the questionnaires or at
verification misses the point. The Department did not request any TSPIRS information from Canada, BC or any other
province. This information was voluntarily submitted by Respondents. TSPIRS information was examined at verification at the
behest of Respondents, not at the Department's initiative. Respondents knew at the time of the Preliminary Determination that the
Department was focusing on price- based benchmarks to measure preferentiality. Irrespective of whether the Department
solicited additional information concerning the Respondents' proposed cost-based methodology the fact remains that, for the
reasons outlined in this discussion, they failed to demonstrate both that cost is the most appropriate benchmark for stumpage and
that their modified TSPIRS approach
(Cite as: 57 FR 22570, *22592)
would accurately reflect the government's cost of providing the good in this instance.
Lastly, in contrast to the methodological direction which Respondents advocate, the Coalition argues that a cross-border
comparison of adjacent U.S. and Canadian timber offers the preferred means for determining whether provincial stumpage
programs confer subsidies because a cross-border comparison is the only methodology which uses an undistorted benchmark
that most accurately reflects true commercial considerations. The Coalition contends that the other benchmarks provided for in
the preferentiality hierarchy are distorted in this case by reason of the log export restrictions and the large amount of timber
owned and administered by the provinces. In support of its position, the Coalition cites Leather from Argentina; Final Affirmative
Countervailing Duty Determination and Countervailing Duty Order. 55 FR 40212 (October 2, 1990) (Leather), as an
example of a case in which the Department chose to rely on an external benchmark to identify and measure a domestic subsidy.
The Coalition's reliance on Leather is misplaced. While the Department did resort to an external benchmark for measuring the
effect of the hide embargo in that case, it did not do so within the context of the Preferentiality Appendix since we did not consider
the embargo to be the government provision of a good or service. However, for the type of subsidy being conferred as a result of
the
(Cite as: 57 FR 22570, *22592)
hide embargo, we considered that use of an external benchmark was not only appropriate, it was the only method available to use
for valuing the subsidy benefit.
As stated earlier, it has been the Department's longstanding practice and preference to measure subsidies provided by a
government within the jurisdiction of that government. Thus, in the absence of clear and persuasive evidence that comparisons
made within the same jurisdiction would somehow yield skewed results, the Department will not stray from its methodological
preference. We do not find that the Coalition has presented such clear and persuasive evidence. Moreover, insofar as we have
sufficient and reliable nonpreferential price data in this case with which to compare stumpage prices within the relevant
provincial jurisdictions, we find that other factors which could adversely affect the comparability of adjacent U.S. and Canadian
timber (e.g., exchange rate fluctuations) merely underscore the appropriateness of remaining within the relevant jurisdictions.
Consequently, we have based our preferentiality determinations on actual price comparisons within each province, as explained
in greater detail below.
British Columbia
Pursuant to section 771(5)(A)(ii)(II) of the Act, the Department must examine
(Cite as: 57 FR 22570, *22592)
whether the BC's provision of stumpage to producers of certain softwood lumber products is at a preferential rate. As explained
above, the Department's preferred test for determining whether a good or service is provided at a preferential rate is to examine
whether the government provides the same good or service at a price that is lower than the price the government charges to the
same or other users within the same political jurisdiction, i.e., whether there is price discrimination by the government.
In our Preliminary Determination. we found that the Government of BC was providing stumpage at preferential rates. We based
this determination on our traditional measure of preference--price discrimination. A comparison of administratively-set
stumpage prices (primarily for major tenures) to competitively-bid stumpage prices in the *22593
(Cite as: 57 FR 22570, *22593)
section 16 Small Business Forest Enterprise Program (SBFEP) indicated that administratively-set prices, even after accounting for
differences in forest management and other obligations, are, on average, lower than the competitively-bid prices that the
Government of BC charges other timber harvesters. The Department considered that, since the SBFEP section 16 sales are based
on competitive market forces, these competitively-bid prices are nonpreferential, and as such, an appropriate benchmark.
In their April 21, 1992 briefs, Respondents argue that the issue in this case involves the provision of neither a good nor a service,
but rather a set of
(Cite as: 57 FR 22570, *22593)
rights and obligations:
Under the preferentiality benchmarks in the Proposed Regulations, any price comparison used by the Department to determine
preferentiality must involve either "identical" goods or goods that are sufficiently "similar or related" that adjustments for "cost
differences" will make comparisons meaningful. Application of the Department's price comparison benchmarks in this case,
however, is complicated by the fact that what is at issue here is not really a "good" or a "service," but rather a bundle of long-term
rights and obligations relating to access to timber. (See Respondents April 21, 1992 briefs, pp. 111- 44, 45.)
We find that we did in fact use our preferred benchmark appropriately. That is, just as in antidumping investigations, where the
Department may examine sales of the same goods but adjusts for differences in the terms and conditions of sale, in this instance
we examined the same good, i.e., softwood timber, and made adjustments which relate to the terms and conditions of the sale.
Respondents argue that the benchmark which the Department used in the Preliminary Determination (i.e., SBFEP section 16 sales)
is an inappropriate benchmark for this investigation. In their briefs they raise four main objections to the use of the SBFEP
benchmark.
Comparability of Rights and Obligations