Emergency Application — Westmoreland Mining Holdings LLC, et al., Applicants v. Environmental Protection Agency, et al.

Supreme Court briefAug 16, 2024

Ask Donna

What actually matters in this document.

Text

TABLE OF CONTENTS

APPENDIX A

Order Denying Stay, State of North Dakota v. EPA,

No. 24-1119 (D.C. Circuit, August 6, 2024) ………………………………… App.1

APPENDIX B

42 U.S.C. § 7412 ………………………………………………………………… App.3

APPENDIX C

National Emission Standards for Hazardous Air Pollutants:

Coal- and Oil-Fired Electric Utility Steam Generating Units

Review of the Residual Risk and Technology Review,

89 Fed. Reg. 38508 (May 7, 2024) ……………………………………….…..App.34

APPENDIX D

Comments of Westmoreland Mining Holdings LLC,

EPA-HQ-OAR-2018-0794-5935………………………………………..…… App.120

APPENDIX E

Comments of Northwestern Energy,

EPA-HQ-OAR-2018-0794-5980……………………………………………...App.226

APPENDIX F

Comments of National Mining Association,

EPA-HQ-OAR-2009-0234-20531…………………………………………….App.251

APPENDIX G

Comments of Talen Montana,

EPA-HQ-OAR-2018-0794-5987……………………………………………...App.443

APPENDIX H

Declarations:

Exhibit 1 – Declaration of Patrick Barkey ………………………. App.484

Exhibit 2 – Declaration of Jeremy Cottrell …………………...…. App.518

APPENDIX D

APPENDIX

eee ese

@ WESTMORELAND

ne Tree, CO 80124

303.922.6463

www.wesimoreland.com

July 22, 2023

Docket ID No. EPA-HQ-OAR-2018-0794

ATTN:

Administrator Michael Regan

C/O EPA Docket Center (EPA/DC)

Docket ID No. EPA-HQ-OAR-2018-0794

U.S. Environmental Protection Agency

Submitted via Federal eRulemaking Portal (Regulations.gov)

RE:

Comments of Westmoreland Mining Holdings LLC (“Westmoreland”) on

the “National Emission Standards for Hazardous Air Pollutants: Coal- and

Oil-Fired Electric Utility Steam Generating Units Review of the Residual

Risk and Technology Review” 88 Fed. Reg. 24854 (April 24, 2023)

(“Proposed Rule”).

Dear Administrator Regan:

Westmoreland Mining Holdings LLC (“Westmoreland”) presents the following comments

regarding EPA’s proposed rulemaking titled “National Emission Standards for Hazardous Air

Pollutants: Coal- and Oil-Fired Electric Utility Steam Generating Units Review of the Residual

Risk and Technology Review,” published in the Federal Register at 88 Fed. Reg. 24854, on April

24, 2023 (hereafter, the “Proposed Rule”). Westmoreland supplies all coal required by the mine-

mouth powerplant in Colstrip Montana.

This proposed rulemaking has significant implications for coal fired power plants in

general and disproportionately impacts one facility in particular, the Colstrip Power Plant in

Colstrip, Montana. Under EPA’s proposal, the rule at issue would impose tens of millions of

dollars on Colstrip per year and require temporary shutdown to design and retrofit additional

emission controls, despite the fact that EPA has now determined multiple times that the existing

standards provide an ample margin of safety to protect public health and that more stringent

standards are not necessary to prevent an adverse environmental effect.

For these reasons and those discussed below, the Proposed Rule does not demonstrate that

its proposed technology-review-based emission standards are in fact “necessary” in terms of cost,

feasibility, utility, cost-effectiveness, and adverse collateral environmental impacts. Accordingly,

Westmoreland offers the following comments herein regarding specific issues with the Proposed

Rule and its supporting analysis.

App.120

In the event that EPA instead finalizes different limits than those in the Proposed Rule,

EPA must at that time provide a rationale for such inclusion and provide opportunity for public

comment on any such newly proposed limits.

Any Limits Proposed by EPA Must Appropriately Account for Cost, Feasibility, Utility, Cost

Effectiveness, and Adverse Environmental Impacts Associated With Proposed Technology

Major source emission standards under section 112 of the CAA are initially be based on

“the maximum degree of reduction in emissions of the hazardous air pollutants . . . that the

Administrator, taking into consideration the cost of achieving such emission reduction, and any

non-air quality health and environmental impacts and energy requirements, determines is

achievable for new or existing sources.” See 42 U.S. Code § 7412(d)(2)(a). EPA then must conduct

a “technology review” of these established limits every 8 years, in which EPA “shall review, and

revise as necessary (taking into account developments in practices, processes, and control

technologies), emission standards promulgated under this section.” See 42 U.S. Code § 7412(d)(6).

Importantly, during a technology review, EPA is not required to recalculate the maximum

achievable control technology floor. Ass'n of Battery Recyclers, Inc. v. E.P.A., 716 F.3d 667, 673

(D.C. Cir. 2013). Nor does EPA claim to do so in the Proposed Rule. When determining whether

changes are “necessary” as part of a technology review EPA is statutorily required to account for

the cost of any proposed changes. Jd. And courts have upheld EPA’s past practice of further taking

into account “feasibility, utility, cost-effectiveness, and adverse collateral environmental impacts

associated with this technology” when assessing whether to require additional limits as part ofa

technology review. Jd. at 674. And because EPA has relied on such factors in the past, it would be

unlawfully arbitrary for EPA to fail to consider such factors in this or other technology reviews

without providing a rationale for the reversal in policy. FCC v. Fox Te elevision Stations, Inc., 556

U.S. 502 (2009) (“An agency may not . . . depart from a prior policy sub silentio”). The Proposed

Rule purports to be solely based on EPA’s technology review authority, and thus must account for

these factors including cost, feasibility, utility, cost-effectiveness, and adverse collateral

environmental impacts.!

The Proposed

Rule is Not Cost Effective, and EPA’s

Determination

to the Contrary

is

Arbitrary and Unreasonable

The proposed tule seeks to impose an eight-figure cost-per-ton of emission reduction. Not

only is there no reasonable basis to find cost effectiveness of this magnitude to be cost effective,

but it is also so far afield from prior EPA cost effectiveness determinations as to be arbitrary and.

is in direct conflict with prior EPA determinations that such costs are in fact not cost effective.

The Proposed Rule itself estimates that lowering the {PM standard to 0.010 Ib/MMBtu

million

would only reduce industrywide metal HAP by about 6 tons per year but would cost $77.3

on a

1 Although EPA also has authority to make changes to an existing emission standard under Section 112 based

safety to protect public

“residual risk” analysis of whether an emission standard provides an “ample margin of

determination that the

2020

EPA’s

reaffirms

health.” See 42 U.S. Code § 7412(f), the Proposed Rule expressly

and that more

current MATS requirements already provided an ample margin of safety to protect public health

not rely on that

does

thus

and

effect,

al

environment

adverse

an

stringent standards were not necessary to prevent

24865-66.

at

Rule

Proposed

See

limits.

emission

basis for any of the proposed additional

3

App.121

to $93.2 million per year. See Proposed Rule at 24870, Table 3. Thus, the total cost effectiveness

of the proposed reduction is 12.2 million to 14.7 million dollars per ton of non-hg metal HAP

controlled. EPA has never endorsed such extraordinary costs for such meager rewards and doing

so here would be arbitrary and capricious.

Indeed, even the low end of EPA’s estimate—$12.2 million per ton of emission reduction

— dwarfs other cost-per-ton calculations previously rejected by EPA as not cost effective. See, e.g.,

85 Fed. Reg. 42074, 42090 (July 13, 2020) (rejecting an “estimated cost effectiveness ... [of]

approximately $14,000/ton,” because it “is above the range that the EPA has typically considered

cost effective for volatile HAP”); see also 78 Fed. Reg. 10006, 10020-21 (Feb. 12, 2013) (rejecting

more stringent limit of 0.04 Ib/ton, versus the then-current limit of 0.07 Ib/ton, reasoning that

achieving that additional increment of particulate reduction would not be cost effective on a costper-ton basis because “$268,000

per ton of PM

removed

... is [a] significantly higher cost

effectiveness for PM than the EPA has accepted in other NESHAP standards”) (citing 76 Fed.

Reg. 15704 (March 21, 2011) (rejecting $48,501 per ton of PM as not cost effective for PM emitted

by solid waste energy recovery units)). Likewise, in a report titled The Benefits and Costs of the

Clean Air Act from 1990 to 2020 (released in April 2011), EPA noted in the specific context of

PM, that “controls more costly than $15,000 per ton may not be cost-effective,” and acknowledged

that “[c]ontinued effort is needed to ensure that air pollution policies are pursued in the most cost-

effective manner possible.”?? And just earlier this year, EPA rejected, on cost-effectiveness

grounds,

imposition of lead controls that would,

effectiveness of $4.7M/ton,”

which

EPA

for at least one facility, result in “a cost

acknowledged

“is considerably higher than cost

effectiveness values we have historically accepted for lead.” See, e.g., 88 Fed. Reg. 11556, 11565

(February 23, 2023). The proposed rule’s cost effectiveness of $12-14 million per ton of regulated

pollutant emissions is not aligned with EPA’s policy, precedent, or obligation to pursue air

pollution policies “in the most cost-effective manner possible,” and would arbitrarily conflict with

every single prior cost-effectiveness determination EPA has made in analogous contexts.

The few instances in which EPA has found imposition costs per ton in the same order of

magnitude as the Proposed Rule to be cost effective is exclusively in the context of highly toxic

chemicals e.g. hexavalent chromium, or ethylene oxide. And even then, EPA only proposed such

high cost-per-ton of emissions reduced to be cost effective when such reductions were necessary

to provide an ample margin of safety under residual risk reviews. Those examples thus could not

be used to justify the Proposed Rule (and notably, EPA properly does not do so). In fact, EPA

expressly determined in 2020 that the current NESHAP

limit for coal fired power plants was

sufficiently protective of human health and the environment, and that no further reductions were

required. The Proposed Rule expressly agrees with its 2020 finding and makes clear that EPA is

“not proposing any revisions to the 2020 determination” “that the current MATS requirements

2 Report is available at https://www.epa.gov/sites/default/files/20 1 5-07/documents/fullreport_rev_a.pdf

to a

3 EPA assumes total {PM reductions of 2,074 tons/year (Proposed Rule at 24870, Table 3), which would equate

cost

not

found

specifically

cost effectiveness of $37,270 to $44,937 per ton of {PM, well over the costs that EPA has

additional

whether

on

focused

is

112

effective for {PM as well. But regardless, because an RtR review under Section

control of regulated HAP is “necessary” the cost effectiveness analysis should focus on the regulated HAP rather

than justification in terms of non-HAP reductions.

4

App.122

provided an ample margin of safety to protect public health in accordance with CAA section 112

... [and] that more stringent standards were not necessary to prevent an adverse environmental

effect.” (Proposed Rule at 24865-66.) It is therefore arbitrary and capricious for EPA to impose

such unprecedented costs on facilities, particularly when the reduction will only result in a

marginal reduction of emissions that even EPA agrees are not posing unreasonable risk to the

environment or public health.

Furthermore, EPA’s rationale for imposing such exorbitant compliance costs is a non-

sequitur. EPA justifies the costs “[bJecause much of the fleet is already” in compliance with the

proposed limit, and therefore the costs “are modest in the context of total control costs and

emissions of the coal fleet.” But the control costs of the coal fleet at large is not the relevant metric

because, as EPA points out, much of the fleet will not incur such high substantial additional costs.

Instead, the tens of millions of dollars of annual compliance costs will fall disproportionately on a

few facilities.

For Colstrip in particular, EPA estimates an annualized cost of $38,051,172.4 EPA

does not adequately justify its proposal of forcing a few facilities to incur massive compliance

costs, only to incrementally reduce emissions that have already been reduced to a level which EPA

agrees poses no danger to the environment or public health. The fact that the cost will be

disproportionately allocated to a few facilities, and most facilities will not incur any additional

costs, serves only to highlight the arbitrary and capricious nature of the Proposed Rule. After all,

it is arbitrary to justify imposition of cost-ineffective proposals by averaging the cost across all

power plants not in fact impacted by the massive costs, just as it would be arbitrary to conclude

that imposing cost-ineffective controls was cost justified in comparison to profits across all other

sectors of the US economy not impacted by the Proposed Rule.

Not only is EPA’s rationale arbitrary on its face, but it is also arbitrary because it reverses,

without explanation, EPA’s

prior acknowledgements that cost-effectiveness should account for the

cost effectiveness of imposing controls at each affected facility, and not simply on an aggregate

nationwide basis. See, e.g., 88 Fed. Reg. 11556, 11565 (February 23, 2023) (rejecting on cost

effectiveness grounds a lead emission requirement because it would have imposed costs on a single

facility “a cost effectiveness of $4.7M/ton,” which EPA acknowledged “is considerably higher

than cost effectiveness values we have historically accepted for lead.”).

EPA’s Cost Analysis is Unreasonable and Arbitrary

The other elements of EPA’s cost analysis underlying the Proposed Rule are likewise

arbitrary or otherwise unsupported in several key respects:

First, the monetized benefits that EPA weighs against the above discussed costs are based

entirely on non-HAP (i.e., PM2.s, Ozone, and COz).° This is not reasonable or appropriate, because

the statutory requirement that revised technology standards under section 1 12(d)(6) be “necessary”

4 $18,992,866 for one unit, and $19,058,306 for the other. See 2023 Technology Review for the Coal- and Oil-Fired

EGU Source Category, at 80, Appendix D, Docket ID. EPA-HQ-OAR-2018-0794-5789 (“Technology Review

TSD”)

5 See Proposed Rule at 24891 (projecting monetized benefits of $3.3 billion in present value, or expected annual

value of $390 million, based solely on anticipated reductions of PM2s, Ozone, and CO, that are expected to occur as

an indirect effect of this Proposed Rule).

5

App.123

must be understood in the context of section 112, which applies solely to HAPs (and in fact bars

the regulation of criteria pollutants as HAPs), and thus a finding that a specific control on a specific

regulated HAP is “necessary” must be justified by the benefits derived from additional regulation

of that specific HAP, and not based on benefits that might be associated with non-HAP.

Furthermore, EPA’s reliance on non-HAP reduction benefits to justify the cost of proposed

controls is arbitrary because it departs from EPA’s prior related rulemakings without adequate

explanation. Specifically, when assessing the appropriateness of taking into account benefits of

non-HAP reductions in the context of the Clean Air Act’s HAP regulations of fossil fuel fired

power plants under section 112 of the Clean Air Act, EPA found in 2020 that “equal reliance on

the particulate matter (PM) air quality co-benefits projected to occur as a result of the reductions

in HAP was flawed as the focus of CAA section 112(n)(1)(A) is HAP emissions reductions.”®

More specifically, “it would be highly illogical for the Agency to make a determination that

regulation under CAA section 112, which is expressly designed to deal with HAP, is justified

principally on the basis of the criteria pollutant impacts of these regulations. That is, if the HAP

related benefits are not at least moderately commensurate with the cost of HAP controls, then no

amount of co-benefits can offset this imbalance for purposes of a determination that it is

appropriate to regulate under CAA section 112(n)(1)(A).” Although EPA later rescinded the

particular rulemaking containing this analysis, EPA did not abandon this specific determination

regarding the appropriateness of using non-HAP co-benefits to justify section 112 HAP

regulations, instead making a specific finding that “under its preferred approach, the EPA finds

regulating EGU HAP emissions is appropriate without consideration of non-HAP emissions

reductions”.’ Although CAA Sections 112(n)(1)(A) and 112(d)(6) are separate statutory

provisions, both turn on whether regulation is “necessary” in the specific context of section 112,

and thus the cost/benefit analysis must be treated consistently. Accordingly, it is unlawful and

arbitrarily inconsistent with prior EPA actions to justify the costs and burdens of the Proposed

Rule by resort to non-HAP co-reductions.

Second, the Proposed Rule states that its calculation of estimated benefits related to

decreased CO2 emissions are based on an interim Social Cost of Carbon estimate developed by an

Interagency Working Group in February 2021 ® Those estimates, relied on in the Proposed Rule,

are in turn “identical to those reported in the 2016 SC-GHG TSD (WG 2016a) adjusted for

inflation to 2019 doliars.”” It is not reasonable for EPA to rely on these social cost of carbon

6 See “Proposed Rule: National Emission Standards for Hazardous Air Pollutants: Coal- and Oil-Fired Electric

Utility Steam Generating Units—Reconsideration of Supplemental Finding and Residual Risk and Technology

Review,” 84 Fed. Reg. 2670, 2676; see also “Final Rule: National Emission Standards for Hazardous Air Pollutants:

Coal- and Oil-Fired Electric Utility Steam Generating Units—Reconsideration of Supplemental Finding and

Residual Risk and Technology Review,” 85 Fed. Reg. 31,286, 31,299 (May 22, 2020) (“finalizing the determination

outlined in the 2019 Proposal”).

7 See “National Emission Standards for Hazardous Air Pollutants: Coal- and Oil-Fired Electric Utility Steam

Generating Units—Revocation of the 2020 Reconsideration and Affirmation of the Appropriate and Necessary

Supplemental Finding,” 88 Fed. Reg. 13956, 13967 (March 6, 2023).

8 Proposed Rule at 24890 (citing “Technical Support Document: Social Cost of Carbon, Methane, and Nitrous Oxide

interim Estimates under Executive Order 13990,” (Feb. 2021), available at https://www.whitehouse.gov/wp-

content/uploads/202 1/02/TechnicalSupportDocument_SocialCostofCarbonMethaneNitrousOxide.pdf).

9 See “Final version of the RIA for the Proposed EGU MATS RTR,” At 4-49, Table 4-11, Docket ID. EPA-HQ-

OAR-2018-0794-5837,

6

App.124

estimates, given the many issues identified with the arbitrariness and unlawfulness of these very

same social cost of carbon estimates by stakeholders and states in other litigation to which EPA

has been a party, without addressing the flaws that have been identified to EPA in such lawsuits.!°

Specifically, it is arbitrary for EPA to fail to address relevant considerations, and these many

challenges to the validity and reliability of the 2021 social cost of carbon estimates are relevant

as

considerations which EPA must address before deciding to rely on those estimates.'' And

further detailed in the complaints and motions for preliminary injunction specifically referenced

in footnote 10 above, which are hereby incorporated by reference into these comments, EPA

reliance on the IWG2021 cost of carbon estimates is unlawful and arbitrary, because the IWG2021

expressly violates statutory prohibitions, is based on a fundamentally flawed methodology that

does not take into account statutory considerations, ignores decades of best regulatory practices,

and sub silentio departs from regulatory documents that remain in force. For example:

a. The underlying 2021 Interim Working Group assessment (IWG 2021) reflects a policy and

value judgment to consider the anticipated global effects of greenhouse gases, not just their

anticipated effects within the United States. [WG2021 at 14-16. But reliance on global

impacts directly conflict with the text and structure of the Clean Air Act, both by

considering factors not permitted by Congress (Le., global impacts) and neglecting to

consider relevant statutory factors and their unmistakably domestic focus. In adopting the

CAA, Congress made several specific findings, all of which are explicitly domestic. See

42 U.S.C. §7401(a)(1)-(4) (referring to “the Nation’s population,” the “responsibility of

States and local governments,” and “cooperative Federal, State, regional, and local

programs to prevent and control air pollution”). Congress also made clear that CAA

so

programs are designed to protect and enhance the quality of the Nation’s air resources

as to promote the public health and welfare and the productive capacity of its population.”

the

Id. §7401(b)(1) (emphasis added). To implement this design, the CAA authorizes

Administrator of EPA to set air pollution standards for new motor vehicles and stationary

sources, “which may reasonably be anticipated to endanger public health or welfare.” Id.

§§ 7411(b)(1)(A), 7521{a)(1). Nowhere is the Administrator authorized to consider global

effects in setting these vitally important standards. The structure of the CAA likewise

confirms that agencies may not consider global effects. When Congress addresses global

the

effects through the CAA, it does so through a specific process in Section 115. There

reason

Administrator is instructed to instate a formal State SIP plan process when he “has

or

cause

to believe that any air pollutant or pollutants emitted in the United States

contribute to air pollution which may reasonably be anticipated to endanger public health

or welfare in a foreign country.” 42 U.S.C. §7415(a). When EPA wishes to consider foreign

the other

effects of air pollution, it can only through this specific process—not through

10 See Missouri v. Biden, No. 4:21-cv-00287 (E.D. Mo., complaint filed, and motion for preliminary

injunction filed

April 22, 2021, and motion for

May 3, 2021); Louisiana v. Biden, No. 2:21-cv-01074 (W.D. La., complaint filed

preliminary injunction filed July 29, 2021).

11 See FCC vy, Prometheus Radio Project, 141 S. Ct.

1150, 1158 (2021) (When an agency acts, it must “reasonably

see also Michigan v. EPA,

consider[] the relevant issues and reasonably explain[]” its actions) (citations omitted);

of the relevant factors”

ion

considerat

a

on

rests

it

if

only

576 U.S. 743, 751-52 (2015) (“[Algency action is lawful

omitted).

citations

and

quotations

and “important aspects] of the problem.”) (internal

7

App.125

provisions of CAA, which are explicitly focused on domestic effects. This conclusion is

further buttressed by the fact that when Congress wishes for agencies to consider

international effects, it does so explicitly. For example, in the Energy Independence and

Security Act of 2007’s amendment to EPCA, Congress created an International Clean

Energy Foundation entirely separate from the CAFE and energy efficiency programs to

study international implications of greenhouse gases. See 42 U.S.C. §17352(a)(3). By

creating such a separate authority Congress demonstrates that it knows how to clearly

instruct agencies to consider international effects. Cf Morrison v. Nat’l Australia Bank

Lid.,

561

U.S.

247,

255

(2010)

statute

(“When a

gives

no

clear indication

of an

extraterritorial application, it has none. ”); ¢f Corrosion Proof Fittings v. E.P.A., 947 F.2d

1201, 1209 (5th Cir. 1991) (international considerations cannot be implied when statue

“provides a laundry list of factors to consider when promulgating a rule under section 6,

including ‘the effect [of the rule] on the national economy” and “[i|nternational concerns

are conspicuously absent from the statute”). Accordingly, reliance on the [WG2021

explicitly and unlawfully conflicts with the text and structure of the Clean Air Act. Agency

action is unlawful if it is “not in accordance with law” or “in excess of statutory jurisdiction,

authority, or limitations.” 5 U.S.C. §706(2)(A) & (C).

. EPA’s reliance on IWG2021 ignores specific factual findings and decisions made by the

previous EPA Administration and other federal agencies. In the first place, WG2021’s

reliance on global impacts, in addition to being unlawful, as described above, also

arbitrarily departs from decades of prior EPA practice and the directives of OMB Circular

A-4. See Arden Rowell, Foreign Impacts and Climate Change, 39 Harv. Envtl. L. Rev.

371, 389 (2015) (observing that “agencies have used [the domestic scope] in their cost-

benefit analyses for decades”). Indeed, IWG2021 concedes that “[b]enefit-cost analysis of

U.S. Federal regulations have traditionally focused on the benefits and costs that accrue to

individuals that reside within the country’s national boundaries.” [WG2021 at 14. Cf

California v. Bernhardt, 472 F. Supp.

3d 573, 600-01

(N.D.

Cal. 2020)

(“While the

Executive branch holds the power to issue executive orders, an agency cannot flip-flop

regulations on the whims of each new administration. The APA requires reasoning,

deliberation, and process. These requirements exist, in part, because markets and industries

rely on stable regulations.”). But [WG2021

does not actually acknowledge that it broke

with OMB Circular A-4’s clear and reasoned directive to “focus on benefits and costs that

accrue to citizens and residents of the United States.” Circular A-4, at 15. Instead,

IWG2021, like its predecessor, attempts to argue that its global focus is somehow

consistent with Circular A-4’s domestic effects framework. See Rowell, supra, at 397 0.138

(“The IWG’s choice of a standardized global SCC thus deviates from [Circular A-4’s]

guidance in two important ways: in focusing on foreign impacts, and in encouraging

agencies to value, calculate, and make decisions based upon a global SCC that does not

require any separate identification, much less consideration, of domestic impacts.”). By

misrepresenting Circular A-4’s clear guidance, the IWG2021 arbitrarily failed to “display

awareness that it is changing position.” See Fox Television Stations, 556 U.S. at 515 (“An

agency may not, for example, depart from a prior policy sub silentio or simply disregard

rules that are still on the books.”). Moreover, the IWG2021 completely ignores the

8

App.126

extensive discussion and findings regarding discount rates and uncertainty in Chapter 7 of

the EPA’s RIA supporting the repeal of the Clean Power Plan. See EPA, Regulatory Impact

Analysis for the Review of the Clean Power Plan, at 7-2 through 7-8 (June 2019), Docket

ID EPA-HQ-OAR-2017-0355-26743; EPA, Regulatory Impact Analysis for the Review of

the Clean Power Plan: Proposal, at 42-46 (Oct. 2017), Docket ID EPA-HQ-OAR-20170355-0110.

Moreover,

[WG2021

arbitrarily departed

from decades

of prior Executive

Branch cost/benefit practice regarding discount rates. As embodied in OMB Circular A-4,

Executive Branch agencies have long utilized standard discount rates of 3 percent and 7

percent. See Arden Rowell, Foreign Impacts and Climate Change, 39 Harv. Envtl. L. Rev.

371, 385 n.88 (2015) (“Agencies typically discount at 3% and 7%, the rates set forth in

OMB Circular A-4, which provides well-established executive guidance to agencies

performing cost-benefit analysis.”). The TWG fundamentally alters this longstanding

practice without a reasoned justification. Circular A-4 explains that the 7 percent discount

rate is essential to reasoned cost/benefit analysis because “it is a broad measure that reflects

the returns to real estate and small business capital as well as corporate capital” and

“approximates the opportunity cost of capital.” Circular A-4, at 33. Therefore, the 7 percent

discount rate “is the appropriate discount rate whenever the main effect of a regulation is

to displace or alter the use of capital in the private sector.” /d. Neither the Proposed Rule

nor IWG2021

assert that the fundamentals

of economics

have changed

in a manner

justifying abandoning the longstanding 7 percent discount rate. Instead, to reach a higher

SC-GHG valuation, the Proposed Rule simply ignores the effects that regulations have on

“the use of capital in the private sector” by instead stating that the use of 3% or lower

discount rate is appropriate due to “intergenerational impacts” (Proposed Rule at 24894).

But this capital measure, as long recognized by Circular A-4, and Executive Branch

agencies, is an indispensable variable when calculating regulatory costs. By directing

agencies to ignore it, the [WG2021 estimates direct agencies to systematically ignore an

important aspect of any regulatory cost/benefit analysis. EPA’s failure to address these

prior fact findings by itself renders EPA’s reliance on [WG 2021 Estimates arbitrary and

capricious. See F.C.C. v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009) (agency

required to “provide a more detailed justification than what would suffice for a new policy

created on a blank slate” when “its new policy rests upon factual findings that contradict

those which underlay its prior policy”); see also Dep ’t of Homeland Sec. y. Regents of the

Univ. of Cal., 140 S. Ct. 1891, 1912 (2020) (agency’s “prior judgment” or factual finding

cannot be ignored

alternative).

“without

any

consideration

whatsoever”

of the prior

finding

or

Similarly, EPA’s reliance on IWG2021 ignores the relevant fact that other agencies have

rejected [WG2021 and its approaches. For example, 1WG2021 does not even acknowledge

the existence

of the Council

on Environmental

Quality’s

final rule regarding

the

appropriate calculation of uncertainty under NEPA, which finalized revisions to its binding

NEPA regulations to “simplify the definition of effects by striking the specific references

to direct, indirect, and cumulative effects and providing clarity on the bounds of effects

App.127

consistent with the Supreme Court’s holding in Public Citizen.”'? As agencies have

observed, CEQ’s NEPA regulation “explains that agencies should not consider effects that

are ‘remote in time, geographically remote, or the result of a lengthy causal chain.’””?

Instead, “[u]nder this standard, the mere fact that an effect might not occur ‘but for’ the

project is not sufficient to trigger a NEPA analysis; rather, there must be a ‘reasonably

close causal relationship’ between the proposed action and the effect, ‘analogous to

proximate cause in tort law.’” Jd. This proximate cause standard is fundamentally at odds

with the expansive approach to causation embedded in the [WG2021, which are much

closer to the but-for standard that CEQ explicitly considered and rejected. The [WG does

not even acknowledge the existence of CEQ’s standard—or that it is compelling agencies

to use a fundamentally different causation standard when calculating environmental effects

for cost/benefit purposes than the standard they must use when calculating the same effects

under NEPA. Furthermore the Bureau of Land Management found that the IWG2016 on

which the IWG2021 is based “is not appropriate” because “‘it is intended to model effects

on the welfare of future generations on a global scale by additional carbon emissions

occurring in the present,” “[m]onetizing only certain effects on social welfare can lead to

an unbalanced assessment,” and “[r]eporting the SCC in isolation would be misleading.”

BLM, Record of Decision, WY-060-EA13-147 (Nov. 30, 2017) (regarding West Antelope

Il South Lease Modification WY W-177903; see also WildEarth Guardians vy. Bernhardt,

2020 WL 6799068, at *11 (D.N.M. Nov. 19, 2020) (affirming BLM’s refusal to use SCC).

d.

The IWG2021 fails to justify the use of a global rather than domestic scope in calculating

costs. Focusing on global effects is particularly unreasoned in the context of greenhouse

gas emissions. As one court has noted, “Greenhouse gases, once emitted from a specific

source, quickly mix and disperse in the global atmosphere and have a long atmospheric

lifetime. Current research on how greenhouse gases influence global climate change has

focused on the cumulative environmental effects from aggregate regional or global sources.

But there is limited scientific capability in assessing, detecting, or measuring the

relationship between a certain GHG emission source and localized climate impacts in a

given region.” Washington Env’t Council v. Bellon, 732. F.3d 1131, 1143-44 (9th Cir.

2013). Thus, calculating the global effect of any discrete regulated activity is inherently

arbitrary. Jd. (noting that “the effect of collective emissions from the Oil Refineries on

global climate change is ‘scientifically indiscernible”).

e.

IWG2021

fails to consider all important aspects of the problem by failing to consider the

positive externalities of energy production. [WG2021 arbitrarily focused only upon energy

production’s negative aspects without discounting costs based on the concomitant benefits

affordable energy has on economic development, social welfare, economic and energy

independence, and international peace. Instead, it continued to rely upon flawed integrated

assessment models (IAMs) that arbitrarily refuse to consider potential benefits from a

12 CEQ, “Update to the Regulations Implementing the Procedural Provisions of the National Environmental Policy

Act,” 85 Fed. Reg. 43304, 43375 (July 16, 2020).

'3 Qualifying Facility Rates & Requirements Implementation Issues Under the Pub. Util. Regul. Pol’y Act of 1978,

173 FERC § 61158 n.790 (2020) (quoting id.).

10

App.128

the input of [WG2021 (id.

warming climate. The DICE Model, which accounts for 1/3 of

g-related diseases but

at 10), focuses exclusively upon increased mortality from warmin

Dudley et al., The

ignores decreased mortality from wintertime mortality. See Susan E.

on the Benefits and

Office of Management and Budget’s Draft 2010 Report to Congress

Center, see also Arden

Costs of Federal Regulations, at 12, GW Regulatory Studies

L. Rev. 371, 383 (2015)

Rowell, Foreign Impacts and Climate Change, 39 Harv. Envtl.

models, the IWG

(“[L]acking any principled mechanism for distinguishing between the

each model as one third

decided to simply average them. This had the effect of weighting

refuse to consider

of the SCC.”). Both [1WG2021 and its underlying models systematically

science to a particular

economic and health benefits, indicating a determination to bend

policy outcome.

There is a “significant mismatch”

between the conclusions of the IWG2021

and the

Ct. 2551, 2575 (2019).

administrative record. See Dep’t of Commerce v. New York, 139 S.

ted models, needs

IWG2021 itself recognized that its SC-GHG Estimate is based on antiqua

ork. But instead of

updating to reflect changed circumstance, and based largely on guessw

out the SC-GHG

addressing these serious shortcomings, the Working Group rushed

able reason for this

Estimates in a hurried process with no outside input. The only discern

up with the most

rush was to comply with EO 13990’s demand that the IWG come

Texas v. United States,

important number in American regulatory policy in a month. Cf

that a similar time frame “did

2021 WL 723856, at *41 (S.D. Tex. Feb. 23, 2021) (noting

l reasons can be a

not leave much time for reflection and analysis”). Although politica

accountability by refusing to

legitimate basis for agency action, agencies cannot avoid

decisionmaking, Dep't of

frankly acknowledge the influence of such reasons on their

nect between the decision

Commerce, 139 S. Ct. at 2575 (“[W]e cannot ignore the discon

d to exhibit a naiveté from

made and the Case explanation given [and] are ‘not require

ther, the evidence’—that

which ordinary citizens are free.’”) (cleaned up). “Altoge

-- “tells a story that does not

1WG2016 was at a minimum, insufficient and out of date

decision.” /d. Because IWG2021

match the explanation the [Working Group] gave for [its]

ledging the political

dresses up a political decision in the cloak of science without acknow

drivers of the decision, it is arbitrary and capricious.

21 uses to produce social

_ Both the underlying models themselves and the process 1WG20

acknowledged that the task of

cost values from them are inherently unreliable. [WG2021

ting to predict global “changes

assigning “social costs” to greenhouse gases involves attemp

ty damage from increased

in net agricultural productivity, human health effects, proper

s, risk of conflict, environmental

flood risk natural disasters, disruption of energy system

at 2. This includes “spillover

migration, and the value of ecosystem services.” IWG2021

and global migration.” Jd. at 3. In

pathways such as economic and political destabilization

unknowable contingencies as the

other words, this task involves attempting to predict such

ational conflicts and human

likelihood, frequency, scope, and severity of future intern

Group also admitted that its

migrations for 300 years into the future. Id. The Working

opments in human technology and

calculations involve attempting to predict future devel

195

gies performed by the world’s

innovation for centuries to come, future mitigation strate

11

App.129

nations, and global atmospheric concentrations due to greenhouse gas emissions. /d. at 1516. The underlying models estimates of climate impacts out to the year 2300 represent “a

significantly longer time scale than is generally included in such analyses, which do not

typically extend further than fifty years.” Rowell, supra at 386. This time horizon is

inherently arbitrary and cannot serve as a reasoned basis for regulatory prediction given

the impossibility of forecasting that far into the future. For example, estimating costs

incurred almost 300 years in the future would require considering how well someone living

in 1720 could have projected the societal values and vulnerabilities of 2020, including

markets, levels of economic output, technologies, and lifestyle preferences. Market seers

of 1720 would have overemphasized the importance of protecting against a future threat to

the supply of horses, given that horses and other animals provided the sole source of

locomotion on land other than walking. As noted above, the models are also flawed because

they

systematically

undervalue

or entirely

ignore positive

externalities

from

energy

production. And the [WG added another layer of arbitrariness in its method of using these

arbitrary models to come to the SC-GHG Estimates. Each model produces different SCGHG values. Instead of accounting for the relative flaws and strengths of each model, the

IWG 2021 simply averages the SC-GHG values produced by these models to reach the

definitive SC-GHG Estimates that agencies must follow. See Rowell, supra at 383

(“[G]iven that the different IAMs predicted different monetized impacts, how should the

IWG calculate a single, standardized SCC? One option, of course, would have been to

create multiple SCCs representing the predicted outcomes of each model. ... Instead,

lacking any principled mechanism for distinguishing between the models, the [WG decided

to simply average them. This had the effect of weighting each model as one third of the

SCC.”). In other words, the SC-GHG Estimates are just a simple average of three flawed

models— nothing more. See id. at 383 n.75 (noting that “[tJhis approach has not been met

with universal approval” in the expert community); Cf Settling Devotional Claimants v.

Copyright Royalty Bd., 797 F.3d 1106, 1109 (D.C. Cir. 2015) (an “arbitrary splitting of the

baby” is not reasoned decision making).

. Unreliability is further introduced in the [WG2021 approach to calculating the present

dollar value for the “social cost” of a greenhouse gas (i.e., the “discount rate”). Jd. at 16—

22. IWG2021 calculated the social cost of each gas at four different values using three

different discount rates — 5%, 3%, and 2.5%, and a 95% probability distribution for the 3%

rate. Using these different discount rates, the “social cost” of carbon dioxide ranges from

$14 per metric ton to $152 per metric ton, depending on the discount rate selected. Jd. at 5.

The Working Group admits that “the range of discount rates reflects both uncertainty and,

at least in part, different policy or value judgments.” /d. at 27. According to the Working

Group, “the choice of a discount rate ... raises highly contested and exceedingly difficult

questions of science, economics, ethics, and law.” Jd. at 17 (emphasis added). But rather

than explain the proper discount rate to use in this action, EPA arbitrarily averages the

discount rates together without attempting to explain which is more appropriate or why an

average would appropriately address the “highly contested and exceedingly difficult

questions of science, economics, ethics, and law” identified by [WG2021.

12

App.130

Third, regardless of the validity of using the interim social cost of carbon estimates, it is

arbitrary for EPA to take credit for CO2 decreases associated with the Proposed Rule when EPA

is simultaneously rendering any such emission emissions irrelevant in its contemporanecously

proposed GHG NSPS rulemaking requiring carbon capture on these very same sources.!* EPA is

required to account for other rulemaking actions it has proposed when evaluating a given proposed

rulemaking.!° Accordingly, EPA cannot double count anticipated CO emission reductions from

this rulemaking when this rulemaking would not in fact cause such reductions, or at least not to

the same magnitude, when EPA’s other related rulemakings for this same source category are

accounted for.

Fourth, EPA’s estimated benefits related to decreased Ozone emissions is flawed because

it accounts for decreases in Ozone exposures generally, without accounting for the fact that EPA

has in another contemporaneous rulemaking determined that Montana emission sources do not

significantly contribute to any out of state Ozone concentrations above EPA’s national ambient air

quality standards for ozone, and thus any benefits related to reduced ozone related to Colstrip

emissions cannot be assumed to occur outside Montana.!®

Fifth, although EPA estimates that a vastly disproportionate majority of costs and emission

reductions (e.g., of PM) will be localized at Montana power plants,!” EPA does not appear to limit

its modeling of benefits from reduced PM emission exposures to populations actually within a

scope that could be affected by such plants, or otherwise account for the localized nature of benefits

from reduced emission exposures to populations near relevant facilities EPA anticipates PM

emission reductions to actually occur. And it would be arbitrary to account for health benefits

nationwide from reduced exposures without first demonstrating that such populations are actually

geographically capable of benefitting from any reduction in emissions from Colstrip and the other

facilities from which EPA anticipates the Proposed Rule to force PM reductions.

"4 See e.g., “New Source Performance Standards for Greenhouse Gas Emissions From New, Modified, and

Reconstructed Fossil Fuel-Fired Electric Generating Units; Emission Guidelines for Greenhouse Gas Emissions

From Existing Fossil Fuel-Fired Electric Generating Units; and Repeal of the Affordable Clean Energy Rule,” 88

Fed. Reg. 33240 (May 23, 2023).

5 See e.g., Portland Cement Association v. EPA, 665 F.3d 177 (D.C. Cir 2011) (holding in the context of EPA

publishing related but separate NSPS and NESHAP rulemakings that “Basing its decision on a premise the agency

itself has already planned to disrupt is arbitrary and capricious” and “Reasoned decision-making requires an agency

to examine the relevant data and articulate a satisfactory explanation for its action[s]. The impending [regulatory

change regarding an] undeniably related source category is clearly a relevant factor or an important aspect of the

problem that must be considered.” And “Since agencies have an obligation to deal with newly acquired evidence in

some reasonable fashion, or to reexamine their approaches if a significant factual predicate changes, an agency must

have a similar obligation to acknowledge and account for a changed regulatory posture the agency creates—

especially when the change impacts a contemporaneous and closely related rulemaking.” And “The refrain that EPA

must promulgate rules based on the information it currently possesses simply cannot excuse its reliance on that

information when its own process is about to render it irrelevant.”) (internal quotation marks and citations omitted)

16 See “Federal ‘*Good Neighbor Plan’ for the 2015 Ozone National Ambient Air Quality Standards,” 88 Fed. Reg.

36654 (June 5, 2023) (e.g., at 36710, estimating the highest contribution from all Montana combined to still be

almost an order of magnitude below EPA’s de-minimis screening threshold or 0.7 ppb).

17 Eg. 2023 Technology Review for the Coal- and Oil-Fired EGU Source Category, at 80, Appendix D, Docket ID.

EPA-HQ-OAR-2018-0794-5789 (“Technology Review TSD”).

13

App.131

Sixth, EPA’s

cost analysis does not appear to account for lost revenues during the downtime

required to engineer and retrofit additional control technologies to comply with the Proposed

Rule.

skews EPA’s

Such downtime would be significant, and ignoring this relevant factor arbitrarily

EPA’s cost

cost assumptions extremely low and as a result makes them significantly flawed.

y retrofit

analysis likewise does not appear to account for other indirect impacts of such temporar

associated

induced shutdowns, including unavailability of power during the retrofit, security risks

coal

with even temporarily reduced electric generation capacity, and increased costs to the utilities,

suppliers, and ratepayers purchasing needed makeup power on the open market.

ing the

Seventh, EPA fails to provide any consideration for the life of sources in determin

related

feasibility or cost effectiveness of the Proposed Rule. This conflicts with EPA policy in

ing whether a

rulemakings, where EPA has expressly accounted for the life of sources in determin

recently

particularly costly control technology can be cost justified. For example, in EPA’s

same coal

proposed greenhouse gas NSPS emission standards under section 111(d) for the very

life of

fired power plants subject to the Proposed Rule, EPA proposed to account for existing

compliance

sources by exempting or otherwise requiring alternative more cost effective

alternatives for sources that were scheduled to retire before January 1, 2040." It is arbitrary for

arbitrary

EPA to fail to account for the life of the plant in one rule and not the other. It is further

Rule

Proposed

to fail to account for the life of a facility where EPA’s own proposed rules (both the

costs of such

and EPA’s contemporaneous NSPS rules targeting coal fired power plants) impose

a magnitude as to affects the feasibility and cost justification of coal fired power plants

continuing

that

operations beyond 2040.'%° This is especially true here, where EPA’s cost analysis assumes

the equipment,

large capital expenditures (¢.g. a baghouse) can be annualized based on the life of

have a much

without even attempting to account for the fact that such large expenditures would

the fabric filter

higher annualized cost if the facility were forced to close sooner than the life of

GHG NSPS.

control system, due to costs imposed by EPA’s contemporaneously proposed

forcing the choice

Accordingly, EPA must at minimum provide compliance flexibilities to avoid

be able to recoup the

between unjustified cost and premature retirement for sources that would not

additional control device capital expenditure investments over a longer cost period.

eam

Eighth, EPA entirely fails to consider, quantitatively or qualitatively, the downstr

s, the Proposed

economic costs of the Proposed Rule. As discussed elsewhere in these comment

Rule threatens the viability of Colstrip, especially

when

considered in concert with EPA’s

on Colstrip, its

contemporaneously proposed GHG NSPS. In addition to the direct costs imposed

Colstrip would also impact

owner/operators, its workers, and the Colstrip community, impacts on

Westmoreland Mining Holdings, which operates the Rosebud mine that supplies

coal to Colstrip.

s From New, Modified, and

18 See e.g., “New Source Performance Standards for Greenhouse Gas Emission

es for Greenhouse Gas Emissions

Reconstructed Fossil Fuel-Fired Electric Generating Units; Emission Guidelin

the Affordable Clean Energy Rule,” 88

From Existing Fossil Fuel-Fired Electric Generating Units; and Repeal of

Fed, Reg, 33240 (May 23, 2023).

ed costs are likely to force a shift away

19 EPA has already recognized that its proposed NSPS standards and associat

-OAR-

Adequacy” at 5, table 1, Docket ID. EPA-HQ

from at least some coal power plants. See e.g. “TSD-Resource

shutdown of over 39 GW of coal fired power plant

the

to

GHG NSPS will lead

4023-0072-0034 (predicting that the

capacity by 2040).

177 (D.C. Cir 2011).

20 See fn.11 above, citing Portland Cement Association v. EPA, 665 F.3d

14

App.132

App.1382

As discussed in detail in the Barkey Report (attached hereto), a closure of the Rosebud mine would

be devastating for the State of Montana.”! Among other things, closure would lead to a Montana

economy that:

a.

Loses over 1,800 jobs, with an average earnings per year exceeding $62,000;

b.

c.

Experiences a shortfall of more than $40 million in state government revenue; and

Faces a loss of revenue to business in excess of $330 million per year.

Impacts Summary?

CATEGORY

UNITS

Total employment

Personal income

IMPACTS

BY YEAR

2020

2022

2022

2023

2024

2025

Jobs

-1,716

-1,852

-1,893

-1,872

-1,819

-1,754

Millions

-$115.4

-$95.6

-$101.4

-$109.2

-$113.0

-$114.8

Disposable personal income

“$81.1

-$86.1

-$92.9

-$96.4

-$98.0

-S98.6

State Tax and Nontax Revenue

-$39.0

-$410

-$42.7

-$43.8

-$44.4

-$44.7

Output

-$325.9

-$347.8

-$355.2

-$353.4

-$346.8

-$338.5

$64,921

-1,909

Earnings per Lost Job

Dollars

$61,998

$60,722

$61,828

$62,896

$63,940

Population

People

-568

-998

1,337

-1,592

1,777

These economic costs to the state of Montana and its citizens are critical considerations

that EPA failed to consider and that, once considered, would not allow EPA to reasonably

or rationally finalize the Proposed Rule as being “necessary” under section 112(d)(6),

particularly when much of the putative benefits are limited to Montana, where much of the

impact of the regulation will be felt.

Ninth, EPA’s focus on contextualizing costs associated with the Proposed Rule (including

retrofits and costs associated with temporary shutdowns, among others) in terms of the electric

utility industry as a whole arbitrarily ignores the disparate impacts of the rule’s costs, which will

not in fact be spread out industrywide. At Colstrip in particular, multiple of the utilities with an

ownership share in Colstrip are forced to exit before 2030 or in some cases 2025, due to their state

PUC mandates to shift away from coal. This places a disproportional funding obligation on the

remaining owners (post 2025). Likewise, the sole merchant utility owner is most impacted. And

cost impacts are particularly significant for merchant utility owners who are unable to pass through

costs to ratepayers.

The Proposed Rule Does Not Adequately Demonstrate the Achievability and Feasibility of

the Proposed Limits.

21 The Barkey Report was prepared in 2019 in the context of administrative litigation concerning a Montana Major

Facility Siting Act certificate revision that purported to permit Colstrip to use coal from non-Rosebud mines.

Accordingly, the impacts on the Colstrip community would, in fact, be far greater if Colstrip were to close because

the Barkey Report only accounts for the closure of the mining operations in support of Colstrip, whereas both the

powerplant and its dedicated Rosebud mine may not continue to operate if the Proposed Rule is finalized.

Moreover, although the terms of the report itself were focused on a specific time horizon, Westmoreland attests to

the accuracy and relevance of this recent econometric study under 2023 conditions; if anything, the impact would be

greater today on a monetized basis due to inflation.

15

App.133

Under 42 U.S.C. § 7412(d)(2), EPA’s promulgation of new emissions standards can only

be the “maximum degree of reduction in emissions” that EPA determines are “achievable” for new

or existing sources after balancing factors such as cost, non-air-quality health and environmental

concerns, and energy implications. EPA claims the Proposed Rule is “achievable” here for new

and existing sources under this standard but, in reality, EPA reached this (incorrect) determination

using an arbitrary and self-serving set of criteria and data. EPA also acknowledges that the

Proposed Rule will require modifications to the Colstrip facility but does nothing to analyze how

such modifications affect its “achievability” analysis.

First, EPA supports its achievability analysis based on a document titled “2023 Technology

Review for the Coal- and Oil-Fired EGU Source Category.”’? Therein, EPA states that when

assessing the emission rates actually achieved in practice by existing coal fired power plants, it

“removed units from [its] analysis that ... will shut down or no longer burn coal/oil by December

31, 2028.” But, at the same time, EPA included “EGUs [already] planning to convert to natural

gas by the proposed compliance date” that are not subject to the rule, and thus concluded that since

those EGUs “already meet a 6.0E-03 lb/MMBtu potential standard” they will have “no associated

upgrade costs or emission reductions assumed in this analysis.” As such, EPA has cherry picked

the available data to support its belief that lower emissions standards are “achievable.” EPA on the

one hand arbitrarily excludes from consideration facilities that likely do not meet the agency’s

preferred emissions rate based on something these facilities might do in the future, while on the

other hand including facilities that support EPA’s desired outcome despite acknowledging such

facilities will not even be subject to the Proposed Rule. Such analysis does not evidence objective

and reasoned agency decision making.

Second, and similarly, § 7412(d)(3)(A) requires EPA to exclude from its minimum

“achievability” analysis for existing sources those that “within 18 months before the emission

standard is proposed or within 30 months before such standard is promulgated, whichever is later,

first achieved a level of emission rate or emission reduction which complies, or would comply if

the source is not subject to such standard, with the lowest achievable emission rate [] applicable to

the source category and prevailing at the time[.]” But EPA has not excluded any sources from its

technical analysis of achievable rates on this basis. This again results in a skewed analysis towards

a lower “achievable” emissions standard using a method explicitly prohibited by the CAA. See 5

U.S.C. § 706(2) (reviewing court shall hold unlawful and set aside agency actions that are “not in

accordance with law” and “in excess of statutory [] limitations”); see also Bethesda Health, Inc. v.

Azar, 389 F. Supp. 3d 32, 41 (D.D.C. 2019) (setting aside as arbitrary and capricious agency action

that contradicts its own regulations). And although this requirement is only expressly listed with

respect to setting the initial MACT

floor, it would be unreasonable and arbitrary to interpret

requirements imposed under technology review not to be subject to the same requirement since

EPA could otherwise easily circumvent the statutory limit at will simply by adding such sources

back to its achievability analysis under each successive technology review period. In any case EPA

has not identified any rationale for exempting Section 112(d)(6) technology review determinations

from this same constraint. In accordance with the CAA, EPA must determine what the LAER rate

2 Docket ID. EPA-HQ-OAR-2018-0794-5789

16

App.134

is for this category, and then exclude from its achievability analysis any sources that already meet

the lower emissions limit contemplated in the Proposed Rule.

Third, in EPA’s “Technical Memo,” EPA acknowledges (at 9) that the Proposed Rule

would require installation of fabric filter controls (such as a baghouse) at Colstrip. But—contrary

to EPA’s obligations under the CAA—EPA fails to consider the economic or technical feasibility

of doing so. See § 7412(d)(2) (EPA must “tak[e] into consideration the cost of achieving such

emission reduction, and any non-air quality health and environmental

impacts and energy

requirements”). Installation of a baghouse at Colstrip’s two generating units would likely costs

tens of millions of dollars to design, install, and maintain—assuming, that is, retrofitting the units

with a baghouse is even technically possible in the first place given spatial constraints, changes in

fluegas rates, temperatures, constituents of the locally permitted coal, and compatibility with the

existing venturi-type wet-scrubbers. EPA considers none of these limitations here, despite being

obligated to do so by statute. What’s more, EPA has in the past accounted for feasibility of

standards imposed under a technology review, and thus cannot depart from that practice without

providing any rationale. See, e.g., Nat’l Ass'n of Battery Recyclers, Inc. v. E.P.A., 716 F.3d 667,

674 (D.C. Cir. 2013); FCC y. Fox Television Stations, Inc., 556 U.S. 502 (2009).

The Alternative Potential Standard is Not Justified.

In addition to all the reasons EPA has not adequately justified a 0.01 lb PM/MMBtu

standard, there are additional reasons EPA cannot and should not consider any stricter limit such

as the alternative 6.0E-03 Ib. PM/MMBtu potential standard EPA requests comment on.

First, EPA acknowledges that any limit below 0.01 lb PM/MMBtu would be more stringent

than the PM limit applicable to new sources. See Proposed Rule at 24857 & n.5. But EPA has

identified no rationale—and thus it would be arbitrary and capricious— to impose a stricter limit

on existing sources than on new sources, especially where it costs more, and requires greater

resources and downtime, to retrofit existing units so that it can achieve the same standards as a

new source. See Watson Lab'ys, Inc. v. Sebelius, 2012 WL 6968224, at *19 (D.D.C. Oct. 22, 2012)

(agency decision arbitrary and capricious where “it produces absurd results that are contrary to the

purpose of the [statute].”), vacated as moot, 2013 WL 11250319 (D.C. Cir. June 10, 2013).

Second, EPA’s

titled “Particulate Control

technical memorandum

Cost Development

Methodology” notes (at 2) that suppliers of control technology that will be needed to comply with

a stricter than 0.010 Ib/MMBtu standard will not provide any performance guarantees below a

0.010 Ib/MMBtu threshold. And even at the 0.010 lb/MMBtu level, suppliers only guarantee such

performance “depending on the application.” It would be arbitrary and contrary to the evidence

before the agency to finalize an industrywide requirement below 0.010 Ib/MMBtu level in the face

of such acknowledged unavailability of vendor guarantees.”’ It is critical to obtain vendor

guarantees from suppliers when constructing or retrofitting facilities and associated control

23 Indeed, the fact that even at the 0.010 lb/MMBtu level, suppliers only guarantee such performance “depending on

the application” likewise means that it would be arbitrary and contrary to the evidence before the agency to finalize

an industrywide requirement even at that level, because a level below what control technology vendors can

consistently guarantee on an industrywide basis cannot be said to be “necessary” from a strict technological

feasibility perspective, absent some other justification such as residual risk.

17

App.135

systems, to ensure that the purchased equipment can comply with applicable standards. It would

be arbitrary and capricious for EPA to ignore this relevant factor by mandating emissions standards

that hold source operators to standards that industry suppliers needed to achieve those standards

will not even guarantee. See FCC v. Prometheus Radio Project, 141 S. Ct. 1150, 1158 (2021)

(When an agency acts, it must “reasonably consider[] the relevant issues and reasonably explain[]”

its actions) (citations omitted); see also Michigan v. EPA, 576 U.S. 743, 751-52 (2015) ([A]gency

action is lawful only if it rests on a consideration of the relevant factors” and “important aspects]

of the problem.”) (internal quotations and citations omitted). Furthermore, EPA has previously

acknowledged the importance of being able to obtain vendor guarantees when setting the emission

standards, and thus it would be arbitrary for EPA

to ignore that consideration here without

providing adequate rationale for the reversal in course.”

Third, EPA’s own technical supporting documents concede that emission limits below

0.010 Ib/MMBtu are not likely attainable using an electrostatic precipitator (ESP) unless (1) a

separate fabric filter is installed or (2) the ESP is itself converted to a fabric filter.*” And yet these

same documents also note that such conversions are not universally possible because ESPs vary

in size. Id. EPA makes no effort to analyze the extent of this reality at facilities implicated by the

Proposed Rule. Nor does EPA consider the significant increases in energy disruption if a greater

percentage of facilities must be idled during necessary retrofits, if such retrofits are even possible.

There are a variety of costs involved in this disruption that are each meaningful, cumulatively very

substantial, and none considered by EPA’s existing analysis in the Proposed Rule. And EPA

conducts no feasibility analysis to conclude that a limit below 0.010 Ib/MMBtu is physically

attainable by any substantial number of sources that currently rely on an ESP. And EPA

acknowledges that the substantially higher compliance costs on the industry compared to the

proposed 0.010 Ib/MMBtu standard, but nowhere provides rationale for how such higher costs

could qualify as cost effective.”° It would be arbitrary and capricious to impose this alternative

lower standard without having accomplished any of this necessary prerequisite analysis.

Fourth, Given EPA’s acknowledgement that emission limits below 0.010 lb/MMBtu are

generally unattainable by a significant portion of existing facilities absent retrofit, and EPA’s

selection of only certain time periods of data to review, the proposed alternative standard appears

to be an attempt to set a standard that seeks to improve upon the best-performing period of

reporting from various plants. But that is not a lawful use of a technology-based review under

section

112(d)(6), which

does not provide

for EPA

to update

standards

in order to drive

technology, but instead only allows EPA to update standards as “necessary” to account for already

24 Fg. “Standards of Performance for New Stationary Sources: Electric Arc Furnaces and Argon-Oxygen

Decarburization Vessels in Steel Plants” 49 Fed. Reg. 43838, 43,840 (Oct. 31, 1984) (in the context of setting PM

not

limits for electric arc furnaces, concluding “However, the Agency has determined that the mass standard should

be lowered. This is because it was determined that, to guarantee fabric filter compliance with [a more stringent grain

loading standard], vendors might increase capital costs of fabric filters as much as 25 percent [docket references

omitted]. This increase in costs would result from the increased air-to-cloth ratio and other designed factors needed

to ensure continuous compliance with a more stringent emission limit.”),

25 @.g., “Particulate Control Cost Development Methodology,” at 2, Docket ID. EPA-HQ-OAR-2018-0794-5834

26 See 2023 Technology Review for the Coal- and Oil-Fired EGU Source Category, Appendix D, Docket ID. EPAHQ-OAR-2018-0794-5789 (estimating annual costs would increase by an order of magnitude)

18

App.136

existing historic “developments in practices, processes, and control technologies,” not to improve

on such developments.

EPA Should Not Eliminate the Alternative Option to Demonstrate Compliance Via Tests of

Individual or Total Regulated Metal HAPs Rather than fPM.

EPA proposes to eliminate the existing alternative option to test for the specific metal

HAPs subject to the NESHAP for this source category (i.e., 40 C.F.R. Part 63, Subpart VDUUUU),

and proposes to only rely on the surrogate of {PM going forward. Proposed Rule at 24886. But this

proposal is arbitrary because EPA’s own data demonstrates that ratios of metal HAP to {PM can

significantly vary2’, and thus that it is possible for a source to be in compliance with the limits on

individual metal HAP regulatable under Subpart UUUUU, even if emitting amounts of total {PM

above the proposed 0.010 !b/MMBtu standard.”* It is particularly arbitrary to remove individual

metal HAP measurements as a compliance option for Colstrip, given that EPA’s own data appears

to indicate that Colstrip’s coal sources and unique control system have a lower ratio of metal HAP

to PM than most other sources analyzed by EPA.”?

Finally, EPA’s stated rationale for eliminating individual and total metal HAP standards is

arbitrary. Specifically, EPA claim that companies currently demonstrate compliance almost solely

through fPM testing is irrelevant because EPA has not proposed to retain the standard which

companies currently demonstrate compliance with through {PM testing. It would be arbitrary to

eliminate compliance demonstration options on this basis without first demonstrating that

companies will also not use (or need to use) any option other than {PM to demonstrate compliance

with the Proposed Rule’s lower standards. EPA has wholly failed to attempt such rationale here,

ignoring a highly relevant factor in determining whether individual and total metal HAP standards

should continue to be included in the rulemaking.

The Proposed Rule Does Not Appropriately Account for Environmental Impacts Of the

Controls Required to Achieve the Proposed Emission Standards, Including Increased Air

Pollution and New Non-Air Waste Streams.

Section 112(d) of the CAA explicitly states that in making its achievability determinations,

EPA must “tak[e] into account the cost of achieving such emission reduction, and any non-air

quality health and environmental impacts and energy requirements.” (emphasis added). And

although this requirement is only expressly listed with respect to setting the MACT floor, it would

be unreasonable and arbitrary to interpret requirements imposed under technology review not to

be subject to the same requirement, and in any case EPA has not identified any rationale for

exempting RtR determinations from this same constraint. EPA has failed to reasonably account

for such environmental impacts from the Proposed Rules for multiple reasons.

27 See “Quarterly MATS Metals and PM2.5 data” Docket ID EPA-HQ-OAR-2018-0794-5789_attachment_1

(showing Metal to PM ratios ranging all the way from 0.000808448 to 0.025126035).

28 Motor Vehicle Mfrs. Assn. of United States, Inc. v. State Farm Mut. Automobile Ins. Co., 463 U.S. 29, 43 (1983)

State Farm, 463 U.S. at 43, 103 S.Ct. 2856 (finding agency action arbitrary and capricious if the agency explained

its decision in a way that “runs counter to the evidence before the agency”).

29 See “Quarterly MATS Metals and PM2.5 data” Docket ID EPA-HQ-OAR-2018-0794-5789_attachment_l

(showing Metal to PM ratio of ratio of 0.00225 for Colstrip, as compared to an industry average of 0.00413).

19

App.137

Specifically, EPA does not appropriately account for the potential to actually increase

some emissions of fossil fuels at the Colstrip facility based on the apparent need, under the

Proposed Rule, to install a baghouse in line with the wet scrubbers at Colstrip. Specifically, EPA’s

“Particulate Control Cost Development Methodology” documents notes that installation of a

baghouse downstream of a wet scrubber would require additional power generation due to

“increased fan power to account for the added fabric filter pressure drop and, as applicable, air

blowers

and transport

air drying

equipment

for the

SO3

mitigation

system.”

Despite

this

prediction, EPA currently accounts for the extra power needed only as an additional economic

cost. EPA fails to account for the environmental impact that additional non-PM emissions would

be associated with the need to supply the additional power for a baghouse. Nor would any generic

acknowledgement of indirectly increased non-PM emissions be adequate to demonstrate reasoned

decision making, because EPA cannot conclude that such increases in non-PM criteria pollutants

and HAPs are justified by PM decreases without substantive analysis. See Sierra Club v. Mainella,

459 F. Supp. 2d 76, 100 (D.D.C. 2006) (“Merely describing an impact and stating a conclusion of

non-impairment is insufficient[.]”

The Proposed Rule Fails to Account For Related Rulemakings.

When determining whether, and to what extent, changes to the existing MATS NESHAP

standards may be necessary, EPA fails to account for the effects of its concurrent proposal to

impose carbon capture and sequestration (CCS) on these very same existing coal fired power

plants? As the D.C. Circuit has held in the context of another occasion on which EPA

contemporaneously proposed changes to section 111 NSPS standards and section 112 NESHAP

standards for related source categories, “Reasoned decision-making requires an agency to examine

explanation

the relevant data and articulate a satisfactory

for its action[s].

The

impending

[regulatory change regarding an] undeniably related source category is clearly a relevant factor or

an important aspect of the problem that must be considered.”?! EPA’s

failure to consider the related

impacts of this contemporaneous proposed rule affects the reasonableness of many aspects of the

proposed rulemaking.

First, Source owners will need to determine how to make investments to comply with both

the GHG NSPS and a potential CCS mandate in an integrated manner, taking into account the

specific regulatory requirements that EPA imposes, market conditions, ete. But EPA’s seriatim

consideration of two tremendously impactful rulemakings that mutually ignore one another makes

doing so infeasible. As a matter of rationality and simple fairness, EPA should exercise its

30 See “New Source Performance Standards for Greenhouse Gas Emissions From New, Modified, and Reconstructed

Fossil Fuel-Fired Electric Generating Units; Emission Guidelines for Greenhouse Gas Emissions From Existing

33240

Fossil Fuel-Fired Electric Generating Units; and Repeal of the Affordable Clean Energy Rule,” 88 Fed. Reg.

(May 23, 2023) (GHG NSPS)

31 Portland Cement Association v. EPA, 665 F.3d 177 (D.C. Cir 2011). See also id. (“Basing its decision on a

premise the agency itself has already planned to disrupt is arbitrary and capricious” and “Since agencies have an

if a

obligation to deal with newly acquired evidence in some reasonable fashion, or to reexamine their approaches

a

for

account

and

acknowledge

to

obligation

similar

a

have

must

agency

an

significant factual predicate changes,

closely

and

ous

contemporane

a

impacts

change

the

when

ially

creates—espec

agency

the

posture

changed regulatory

related rulemaking.” And “The refrain that EPA must promulgate rules based on the information it currently

possesses simply cannot excuse its reliance on that information when its own process is about to render it

irrelevant.”) (internal quotation marks and citations omitted).

20

App.138

discretion to determine that any revisions to the MATS must be done in conjunction with the GHG

NSPS and with compliance deadlines that are coordinated with the GHG NSPS.

Second, EPA states in this concurrent proposed GHG NSPS rulemaking that CCS, and the

additional control technologies required for CCS to work, may both cause co-reductions in {PM

as well. Accordingly, it is arbitrary for EPA not to account for the impact of these planned [PM

reductions on whether any additional controls on {PM were “necessary” for purposes of section

112(d)(6) at all.

Third, EPA fails to account for the proposed GHG NSPS rule in terms of compliance

deadlines and availability for compliance extensions. Any sources that are required to install

additional control technology as a result of both rules (such as Colstrip, which may be required to

install both a fabric filter baghouse by the Proposed Rule, and CCS by the proposed GHG NSPS,

as well as a series of upstream controls required to make both work, assuming it is technically

possible to retrofit) will require additional time to engineer and design such systems to work in

concert than a source would need simply to engineer and install systems required only for a single

applicable rulemaking.

Fourth, it was arbitrary for EPA to deny the request for comment extension for the

Proposed Rule to account for the interplay and effect of these two related rulemakings on each

other. As raised in a prior extension request for this Proposed Rule, EPA published the proposed

GHG NSPS halfway through the comment period for this Proposed Rule. And this newly proposed

GHG NSPS fundamentally changed the technical and cost analysis with respect to the Proposed

Rule, because if both rulemakings are finalized, sources will have to assess how this different

control requirements required by each rulemaking interact with and affect each other rather than

simply analyzing the Proposed Rule requirements by themselves as was required during the first

portion of the public comment period for the Proposed Rule. This is particularly important for

Colstrip which may be required to install both a fabric filter system under the Proposed Rule and

CCS by the proposed GHG NSPS, as well as a series of upstream controls required to make both

work, assuming it is technically possible to retrofit, all of which require additional analysis to

determine feasibility and cost of designing each new system to work with the other newly required

control system. EPA’s unexplained denial arbitrarily prevented facilities, including Colstrip, from

fully analyzing the technical interplay of these rules during the public comment period. In any

case, EPA’s failure to itself assess the technical effect of the GHG NSPS on the feasibility and

cost of the Proposed Rule was itself arbitrary and inconsistent with reasoned decision-making.

EPA Cannot Publish Alternative Emission Limits Without First Issuing for Public Comment

Additional Proposed Standards and their Basis.

The rulemaking procedures at section 307(d) of the CAA specifically require that a

proposed rulemaking must “include a summary of —{A) the factual data on which the proposed

rule is based; (B) the methodology used in obtaining the data and in analyzing the data; and (C)

the major legal interpretations and policy considerations underlying the proposed rule” and “Ail

data, information, and documents referred to in this paragraph on which the proposed rule relies

21

App.139

shall be included in the docket on the date of publication of the proposed rule.”>? Furthermore, any

final “promulgated rule may not be based (in part or whole) on any information or data which has

not been placed in the docket as of the date of such promulgation.” Relatedly, EPA has “an initial

burden of promulgating

and explaining a non-arbitrary,

non-capricious rule” including

an

obligation to “explain how the standard proposed is achievable under the range of relevant

conditions which may affect the emissions to be regulated.” Accordingly, EPA cannot finalize

any emission standard other than those analyzed in the Proposed Rule absent a new proposed rule

providing the opportunity for public comment on the necessity, appropriateness, feasibility, and

cost effectiveness of any such newly proposed limits. To do otherwise would be unlawful and

arbitrary.

Sincerely,

Ga

ee

ae

Jeremy Cottrell

General Counsel & Secretary

Westmoreland Mining LLC

32 42 U.S.C. § 7607(d)G3)

33 42 U.S.C. § 7607(d)(6)(C)

of anew source performance

34 National Lime Ass'n y. E. P. A., 627 F.2d 416, 433 (D.C. Cir. 1980) (in the context

burden of promulgating and

initial

“an

that

holding

7607(d),

§

U.S.C.

42

to

subject

standard rulemaking procedure

by failing to explain how the

explaining a non-arbitrary, non-capricious rule rests with the Agency and we think that

emissions to be

standard proposed is achievable under the range of relevant conditions which may affect the

EPA, 755 F.3d 1010,

v.

Council

Def.

regulated, the Agency has not satisfied this initial purden.”); see also Nat. Res.

burden of

affirmative

its

of

part

as

1023 (D.C. Cir. 2014) (“EPA retains a duty to examine key assumptions

assumption

that

justify

must

EPA

therefore

and

promulgating and explaining a nonarbitrary, non-capricious rule

omitted)

ellipses

and

marks,

question

internal

(citation,

even if no one objects to it during the comment period.”)

22

App.140

Attachment: Barkey Report

23

App.141

The Economic Impacts of a Rosebud Mine Closure

By Patrick M. Barkey, Ph.D.

October 4, 2019

October 4, 2019

__________________

(date)

______________________________________

________________________________

Patrick

M. Barke

Barkey, Ph

Ph.D.

atri k M

D

1

App.142

Table of Contents

Executive Summary

Introduction and Overview

Rosebud County Profile

The Rosebud Mine Closure Scenario

Research Approach

The Economic Impacts of a Rosebud Mine Closure

Summary and Conclusions

References

Appendix 1: REMI Tables

Appendix 2: Barkey Vita

2

App.143

EXECUTIVE SUMMARY

Since the 1970s, the economy of Colstrip, Montana has been dominated by the operations of the

Rosebud coal mine and the mine-mouth coal-fired electric generating plant that have operated

continuously in this small community in eastern Montana since that time. Located on the northern

portion of the Powder River Basin just south of the generating station, the Westmoreland’s Rosebud

mine has delivered coal to the generating units via conveyor to fuel their production. The mine and the

generating units have thus been closely linked for decades, with production in each facility tied to the

other.

Recent events have called into question the future of that relationship. In a future scenario where the

Colstrip Steam Electric Station (SES) takes deliveries of its coal via rail or truck from a separate, possibly

non-Montana, source, the continued operation of the Rosebud mine would no longer be viable. In the

event that coal production at the facility were to end, the community, the region and the state as a

whole would lose an important driver of its economic vitality.

This study is intended to assess the economic implications of the closure of the Rosebud mine, in a

scenario where the Colstrip SES continues to operate with coal from another source. If the mine were

to close, its wages paid to workers, purchases from vendors, and tax payments to governments would

largely cease. The spending of those who receive those payments as income or revenues can be

expected to be significantly curtailed, with further knock-on effects on economic activity. As the process

continues, a new, lower level of economic activity emerges that shrinks the overall economic pie by

more than that of the mine itself.

We assess the ultimate impact of the Rosebud mine closure with the use of an economic model that

traces the interactions between the mine and the rest of the economy. Specifically developed for this

purpose and calibrated with Montana-specific data, the REMI model was leased from Regional Economic

Models, Inc. (www.remi.com) for use in this study. Since its development in the early 1980s, the REMI

model has been the subject of dozens of peer-reviewed economic articles and has been employed in

hundreds of applications.

A mine-closure scenario for the Montana economy was constructed for this study that consisted of

three broad components. The first was the removal of the mine itself from the economy. Using

information obtained from the mine owners on production, employment, payroll, taxes and vendor

spending (Westmoreland, 2019), the direct contribution of its operations was removed from the

economy. A second component was the remediation activities that would commence and continue

after mine closure.

The third component were the operating changes, if any, at the Colstrip SES that might be expected to

occur with coal deliveries by rail to that facility from a non-Montana source. While little information

was available to detail actual plans and potential investments, a simplified scenario can at least

represent the basics of how fuel source changes would affect economic flows.

3

App.144

SUMMARY OF FINDINGS

Our basic finding is that the shutdown of the Rosebud mine would be a significant economic event, both

for the Colstrip community and to the remainder of the state. The loss of 337 jobs at the facility 1,

producing 6.5 million tons of coal annually, would ultimately lead to a state economy that:

x

x

x

x

x

x

Suffers a loss in excess of 1,800 jobs, across a broad mix of industries, with average earnings per

job lost in excess of $62,000 per year;

Has annual income received by households fall more than $100 million short of the amount that

would have occurred had the mine remained in operation;

Experiences a shortfall of more than $40 million per year in state government revenue;

Sees local governments collect $8 million less per year in property tax revenue, most of which

occurs in Rosebud County;

Faces a loss of revenue to businesses in excess of $330 million per year compared to levels that

would be maintained under continued operation;

Has a population that is smaller by more than 1,900 people;

Impacts Summary 2

IMPACTS BY YEAR

CATEGORY

UNITS

2020

2021

2022

2023

2024

2025

Total employment

Jobs

-1,716

-1,852

-1,893

-1,872

-1,819

-1,754

Personal income

Disposable personal income

State Tax and Nontax Revenue

Output

Millions

-$95.6

-$81.1

-$39.0

-$325.9

-$101.4

-$86.1

-$41.0

-$347.8

-$109.2

-$92.9

-$42.7

-$355.2

-$113.0

-$96.4

-$43.8

-$353.4

-$114.8

-$98.0

-$44.4

-$346.8

-$115.4

-$98.6

-$44.7

-$338.5

Earnings per Lost Job

Population

Dollars

People

$61,998 $60,722 $61,828 $62,896 $63,940 $64,921

-568

-998

-1,337

-1,592

-1,777

-1,909

These findings take into account the positive impacts that come from the $165 million that is expected

to be spent in the decade following the closure on environmental remediation, as well as the modest

increase in rail and material handling jobs that could be expected to occur as the Colstrip SES moves to

use of non-Montana coal.

1

Since a permanent mine closure would occur after the announced date of closure for the smaller, older

generating units of the Colstrip SES, the loss of employment and production directly linked to mine closure is

slightly smaller than current levels, reflecting mine activity serving the remaining Colstrip SES units.

2

Impacts in the table correspond to mine closure in year 2020. Impacts extend beyond the years shown. The

nature and magnitude of the findings would not be materially affected by a different choice of start year for events

associated with mine closure.

4

App.145

1. INTRODUCTION AND OVERVIEW

This is an analysis of the economic implications of the closure of the Rosebud mine located just south of

the city of Colstrip in Rosebud County in eastern Montana. The mine has continuously served the coal

needs of the four units of the 2,100 megawatt Colstrip Steam Electric Station (SES) since the 1970s,

delivering coal to that facility via a 4.5 mile conveyor belt. The possibility that the SES would satisfy its

coal needs from a different source in the future is the motivation for conducting this study, since the

nature of coal markets and the limited transportation access of the mine to those markets would render

its continued operation impossible.

The author of this report is Patrick M. Barkey, Ph.D. He was retained by BakerHostetler to conduct this

analysis and to write this report. He is being compensated for this work. The amount of his

compensation is not dependent on the outcome of the project. He has not testified in deposition or as

an expert witness in any litigated matter in the past.

In addition to publicly available data (see References) the study has made use of operating and other

information (Westmoreland, 2019) provided by Westmoreland Mining, LLC, the owner and operator of

the Rosebud mine.

Ample evidence exists to support the conclusion that coal mining in Montana makes sizable

contributions to the state economy (Barkey and Polzin, 2010; Barkey (2012); Bureau of Business and

Economic Research (2015); Barkey (2018)). The high rates of employee compensation, the large

component of inputs sourced from within the state, and the higher than average tax contribution of the

coal mining industry all play a role in making coal production an important driver of activity in the

economy as a whole.

The prosperity and relative affluence of the small community of Colstrip is testimony to that fact.

Incorporated in 1998, the city of just over 9,000 people is home to one of the largest industrial facilities

in the state, and has an assessed property tax base as large as urban areas many times its size. It has

been largely immune to recessions, and has median household income almost 75 percent higher than

the state as a whole. Located in the eastern portion of the state where economic growth has lagged

that of the more populous urban areas to the west, these accomplishments stand out.

Yet the base on which that prosperity is built upon is narrow. Events which impact the coal-fired electric

generation industry have an outsized importance for Colstrip, certainly. But they also have implications

for the entire state, as this report demonstrates.

5

App.146

2. ROSEBUD COUNTY PROFILE

Rosebud County, the home of the

Westmoreland Rosebud mine and the

community of Colstrip, is the 25th

largest county by population, and the

fourth largest by area in Montana. It is

traversed by Interstate 94, which

passes through Forsyth, the county

seat, as the major transportation

artery between Billings and Bismarck,

North Dakota. A large portion of the

Northern Cheyenne Indian Reservation

lies within Rosebud County.

Like many other counties in eastern

Montana, agriculture plays an

important role in the economy. More uniquely, the southern and central portions of the county sit over

the Powder River Basin, a rich coal-producing seam that produces 43 percent of all U.S. coal (Energy

Information Administration, 2019). The two largest coal mines in Montana are either completely

contained within the County (Rosebud

Rosebud County Profile

mine) or straddle its border (Spring

Value

State

Creek mine). The county is also home

INDICATOR

Rank

to the 2,100 megawatt Costrip Steam

Total Population

9,063

25

Electric Station (SES) coal-fired power

generator, one of the largest industrial

Percent Change Since 2010

-1.8%

45

facilities in the entire state.

Median Age

36.5

49

Percent Aged 65 or Older

14.4%

27

As a whole, the county shares some of

Percent with Bachelors Degree or

the same characteristics of other less

Higher (Age 25+)

18.8%

50

populated counties in the eastern

Percent Without Health Insurance

portion of the state. It has seen slight

Coverage

13.8%

42

population decline since the Census of

2010. Its adult population is ranked low – 50th out of 56 counties – in the percent with at least a 4-year

college degree. The outmigration of younger people has skewed its population slightly older than the

state average, and much older than many of the more populous urban areas.

Rosebud County covers more than 5,000 square miles of territory, and the county-level data can mask

some of the unique characteristics of the community. It is useful to examine the city of Colstrip and the

remainder of the County separately. This can be accomplished with data from the American Community

Survey, produced by the U.S. Census Bureau.

Population by Age, Race and Educational Status

Examining the population by race and Hispanic origin, we can clearly see the prominence of the

American Indian population in the southeast portion of the state. While the city of Colstrip’s share of

American Indians is high in comparison to the state average, the balance of the County is almost 44

6

App.147

percent American Indian. Colstrip

also has a higher fraction of its

population who consider

themselves to be Hispanic or Latino,

which potentially straddles multiple

racial classifications.

The age profile of Colstrip reveals

some of the importance of the

Rosebud mine and the Colstrip SES

as employers in the community.

The city of Colstrip is dominated by

prime-aged workers, more than

either the rest of the County or the

entire state. The population cohort

aged 45-64 years makes up more

than 40 percent of the population,

almost double the share of the rest

of the County.

The city has a markedly lower

fraction of those 65 and older in its

population, and a higher fraction of

children. The latter could be

influenced by the Colstrip school

system which, thanks to the strong

tax base, has outstanding resources

that few other schools in the eastern

portion of the state can match.

7

App.148

There are also significant differences between

Colstrip and the balance of Rosebud County in

educational attainment. The city has an

extremely low fraction – just 1.2 percent -- of its

adult population that lacks a high school

diploma or GED. In contract, the rest of the

county has more than double this fraction than

the state as a whole. Differences in other levels

of educational attainment for the city and the

balance of the County are not as marked,

although the city of Colstrip does score slightly

higher in terms of participation in College.

8

App.149

Employment and Income

The importance of the strong mining

and industrial employers in both the

city and the County can be seen from

the occupation and industry

classifications of the working

population. Compared to the state

average, Rosebud County workers –

both in Colstrip and without – are

much more prominent in those

classifications. More than a third of

Colstrip workers are employed in

agriculture, mining or construction.

The state average is less than 20

percent.

The other side of the coin finds Colstrip

with a much smaller presence of

workers for professional business

services employers, which include

finance, real estate, administrative and

professional and technical services

companies. Colstrip also has a smaller

number of workers working for

government, perhaps due to the fact

that neither tribal government or the

county seat are located there.

In terms of occupations, Colstrip

workers are more prominent in

construction trades, production and

transportation activities, and much less

likely to be in service and sales

occupations.

9

App.150

Perhaps the starkest contrast between

Colstrip and the surrounding County is in

measures of income. Household income

includes income of all household members,

with households encompassing both

individual and family households. As can be

seen from the figure, Colstrip has a very high

fraction of its households with income

towards the upper end of the distribution. To

state it another way, just under 14 percent of

Colstrip households make less than $50,000

per year. By contrast, nearly half the

households in the state make less than that

amount.

The fact that the jobs at both the Colstrip SES

and the Rosebud mine pay salaries that

extended into six figures in many cases is

apparent from the fraction of Colstrip

households with income higher than $100,000

per year. The balance of the county is

dramatically different than the city, with

nearly a third of households having annual

income of $25,000 or less.

Summary measures of income underscore

these differences. Median household income

in Colstrip, at $88,452, is more than 75 percent

higher than the state average. Rosebud

County (balance of county data are not

available for these measures) is much closer to

the state average for both median household

income and median family income. The latter

excludes single person households. The

comparatively larger presence of children in

Colstrip brings the differences in per capita

income closer than the other two measures

between different geographies.

10

App.151

The differences in income and other

characteristics between Colstrip and the

balance of Rosebud County produce

significant differences in poverty

outcomes as well. As shown in the

figure, less than 5 percent of all families

in Colstrip were below the poverty line

in 2017, compared to nearly 15 percent

in the rest of the County. It is

particularly unfortunate that poverty

rates are so alarmingly high for two

categories most at risk – American

Indian families and female headed

families. For these groups, as much as a

third or more fall under the poverty

line, and the differences in these

fractions is not appreciably different for

different areas.

It is apparent from the Census data that the energy and natural resource jobs connected – literally, by

conveyor – to the electric generation activities in Colstrip have made an enormous impact on the

economic status of many who live there. There is another aspect to the presence of the Colstrip SES and

the Rosebud mine in the economy that is not captured by this data. That is the economic stability the

relatively steady operation of these facilities has produced.

11

App.152

Employment in Rosebud county has fluctuated in a very narrow range at roughly 4,000 since the early

1990s, as shown in the figure above. During this interval of time the national economy experienced 3

recessions, the last of which produced a pronounced decline in the state economy. During the early

1980s Rosebud County employment showed strong gains as the construction of the two newer, larger

generating units at the Colstrip SES was completed, again during a period of considerable economic

turbulence in the national economy.

The capital intensive nature of wealth generation in Colstrip, in a county with less than 10,000 people,

stands out in a state that is not home to many industrial facilities of the same size. County Gross

Domestic Product, or GDP, is a measure of economic output. In year 2015, the most recent year for

which data are available, Rosebud County GDP was almost $90,000 per capita, easily the highest of any

county in the state.

Conclusion

The presence of the Colstrip SES and the Rosebud mine within and immediately adjacent to the city of

Colstrip has fostered the growth and development of a community like few others in the state. As a

result of the high wages jobs, significant vendor spending and tax contributions of these facilities, the

Colstrip community:

x

x

x

x

has median household income of $88,452, more than 75 percent higher than the state;

has poverty rates of just 4.9 percent for families, much lower than either the state (9.1 percent)

or the remainder of Rosebud County (14.9 percent);

has just 1.2 percent of its adult population lacking a high school diploma or equivalent,

compared to 7.0 percent for the state and 16.2 percent for the balance of the County;

was not visibly impacted by any of the three most recent national economic recessions,

including the Great Recession of 2008-09;

12

App.153

Its economic success stands in even greater contrast to its eastern Montana peers, who have seen much

more evident economic booms and busts, and generally have incomes less that the state as a whole. In

this portion of the state that has enjoyed less prosperity than average, its achievements stand out as a

remarkable exception.

13

App.154

3. RESEARCH APPROACH

Regional economic impacts occur because of events or activities that create new expenditures within

that region. New spending – that is, spending which is over and above existing expenditures, and which

does not displace other spending elsewhere in the region – not only adds to economic activity in its own

right, but it also induces further spending as the recipients of wages, sales, and tax revenues spend a

portion of their income in the local economy. Changes in the path of investment, migration, prices, and

wages are also possible.

This study uses an economic model, calibrated to represent the interactions specific to the Montana

economy, to estimate the economic impacts resulting from events that comprise the Rosebud mine

closure scenario. Leased from Regional Economic Models, Inc., the REMI model is one of the best known

and most respected analytical tools in the policy analysis arena. It is constructed using national, state

and sub-state area data on income, employment, spending, prices, trade flows and population from

public data sources, including the Bureau of Economic Analysis, the Bureau of Labor Statistics, the

Census Bureau. A full description of the data sources and estimation methodology used to construct

and calibrate the model can be found at https://www.remi.com/wp-content/uploads/2019/08/DataSources-and-Estimation-Procedures.pdf .

The REMI model forecasts employment, income, expenditures, and populations for regions based on a

model containing over 100 stochastic and dynamic relationships, as well as a number of identities. A full

explanation of the design and operation of the model can be found in Treyz (1993).

The REMI Modeling Methodology

The REMI model is a state-of-the-art policy analysis model that brings together three powerful tools to

address a wide range of applied research questions: an input-output model that captures the

connectivity between industries in the economy, an econometric model that accounts for the

adjustment in markets to changes in economic opportunity and rates of return, and a demographic

model that can relate changes in migration to age-structure of the population and labor supply. The

model is based on extensive and detailed data on income, employment, prices, population, and

spending in the national, state and sub-state economies (REMI, 2019).

The basic philosophy of the REMI model is that regions of the country compete for investment, jobs and

people. When events – such as a new business opening or an existing business closing -- cause changes

in a regional economy that impact economic opportunity or rates of return, spending and population

flows occur between regions to bring the economy to a new resting point. The REMI model used in this

analysis pertained to the state of Montana as a whole, with the U.S. economy as a backdrop.

The model itself has been extensively documented and evaluated in a number of peer-reviewed

economic journals (see, e.g., Treyz, et. al., 1992; Crihfield, 1992; Rickman and Schwer, 1993). It has been

used in hundreds of published studies on topics including energy, tourism, taxes, transportation, and

economic development. An extensive bibliography of REMI studies can be found at the REMI web site,

www.remi.com .

14

App.155

The basic approach of using the REMI model to produce the results for this study is illustrated in Figure

1, below. The analysis started with a baseline projection for the Montana economy, using the status quo

assumption that the Rosebud mine continues to operate. Next, the analysis employed the REMI model

a second time, simulating an alternative scenario where the events described in the Rosebud mine

closure scenario occur and the mine’s contributions are removed from the economy.

Figure 1. Policy analysis using the REMI model

The difference between economic activity in baseline scenario and the alternative scenario is the

ultimate economic impact of the events precipitating closure of the Rosebud mine.

The REMI model utilizes historical data on production, prices, trade flows, migration, and technological

advances to calibrate the relationship between five basic blocks of the state economy, as shown in

Figure 2, below.

15

App.156

Figure

Schematic model

Figure 2

2 Schematic

model of

of REMI

REMI Linkages

Linkages

(1) Output and Demand

Sn ad Leet

Government Spending

Tas

Output

MeN

a

—_}

a

Consumption

P

{

T

Investment

Exports

Real Disposable Income

1

a

1

t

4

}

L

1

(S) Population and

(2) Labor and

Labor Supply

Capital Demand

t

(5) Market Shares

-

t

Migration

Population

Employment

4

|

|_|, Participation

(| abor nse 6

⊳

;

Optimal Capital

Labor

Domestic

International

Stock

Productivity ‘

Market Share,

, Market Share

Rate

}

|

⋅

|__|

(4) Compensation, Prices, and Costs

Employment

⊳

Opportunity

:

∁

⋅

≏∏⋮∎⇟−⇁

i acetate

©

Composite

Compensation

⋅

⊳−⊧↽∘⊔∐∘≖∘⋂∁∘≙↡⊰

Rate

:

|

4

t

Hausing Price

sng

∎

Gansumer Prices

∎∙

eS

Real

Compensation Rate

y

16

16

App.157

’ Gam

I

site Prices

paste

“

These blocks are:

(1) Output and Demand;

(2) Labor and Capital Demand;

(3) Population and Labor Supply;

(4) Compensation, Prices and Costs; and

(5) Market Shares.

The differences in production, labor demand, and intermediate demand associated with the closure of

the Rosebud mine perturbs this system, causing changes and adjustment that feedback on each other

and eventually come to a new equilibrium. This new equilibrium, or resting point for the economy, is the

alternative projection depicted in Figure 1.

The underlying philosophy of the REMI model is that regions throughout the country compete for

investment, jobs, and people. When events occur in one region that change economic opportunity,

demand, or productivity, they set off a chain reaction of events that cause resources to flow toward

better investment and production opportunities, followed over time by workers and households toward

better employment opportunities and higher wages.

At its heart of the REMI model is an 82-sector input/output matrix that models the technological

interdependence of production sectors of the economy, as well as extensive trade and capital flow data.

Together, these components enable the estimates of the shares of each sector’s demand that can be

met by local production. Simplified illustrations of the schematic model in Figure 2 are provided on the

following pages, in figures 3 through 7.

Figure 3 OUTPUT LINKAGES

17

App.158

Figure

AND CAPITAL

Figure 44 LABOR

LABOR AND

CAPITAL DEMAND

DEMAND LINKAGES

LINKAGES

--

===.

o>.a

-

Swag”

Seheeed

~

os

Optimal Capital

Empl

Stock

,

Labor Output

ploymen

Ratio

z∎

∎

∙

∎

∙

∎∙−∙−∙∙

∎∙∽∙∎∙∙

∎∙∙∎∎∙∙∎∙∙∎∎∎∙∎∎∎∙∙≺∶

Wage

-

Rate

Veena”

⊞

Precccnccnsccccnccss

∙

∎

∎

∙

∙∎

-

Figure

Figure 55 Demographic

Demographic Linkages

Linkages

ee

em,

u

a

‘

Steere

eee

eT

Pe

Participation

-*”

ba

Employment Opportunity

»,

Real Wage Rate

af

∙∙∙∙∙

∙∙−−−−−−∙∙−↽∙∙

18

18

App.159

eee eeeeee

Population

Figure 6 WAGES, PRICES AND PRODUCTION COSTS LINKAGES

Figure 7 MARKET SHARE LINKAGES

The use of the model involves (a) carefully constructing an input scenario that faithfully represents the

changes in the economy that are the subject of analysis, and (b) using the model to produce an alternate

projection of the economy that incorporates those changes. The model’s purpose is to trace and

quantify how events in specific sectors or industries ultimately move the overall economy to a new level

of activity.

Conclusion

The REMI model is a powerful, flexible, well-documented and well-respected tool for conducting policy

analysis in general, and the analysis of the Rosebud mine closure in particular. We now turn to the use

of the model for this study.

19

App.160

4. THE ROSEBUD MINE CLOSURE SCENARIO

In the event that the Colstrip SES switched sources for its coal away from the Rosebud mine, the mine

would lose its only significant customer. Because of this, it would lose its ability to operate in an

economic manner. This study constructs, analyzes, and describes a scenario that includes the closure

and reclamation of the Rosebud mine.

While any number of events can occur to affect the future, the objective of this study is to explore and

quantify the economic implications of a specific event, namely, the closure of the mine. The outcome of

this analysis is not a forecast – it is the difference between two different forecasts of the state economy:

a forecast that continues its operation, and a forecast that includes the closure of the Rosebud mine.

Because the closure of the two smaller, older generating units (Units 1 and 2) of the Colstrip SES is

scheduled to take place in the very near future, the baseline projection of the economy entails slightly

reduced levels of activity at the Rosebud mine than currently exist today. The changes in the economy

that would occur in the event that the mine ceased operation are measured against this lower baseline.

The first step in this analysis is to carefully construct the Rosebud mine closure scenario. This is meant

to describe how an economy with no Rosebud mine would differ from the baseline. In the language of

economic impact analysis, these are the direct impacts.

Additionally, there are other economic activities that occur that are not part of the production of coal

but are nonetheless linked to the cessation of activities at the mine. These include (a) the

environmental remediation activities that would commence after production ends, and (b) the new

transport and material handling activities that can be expected to occur at the Colstrip SES due to the

shift to rail delivery of its fuel source. These are often referred to as indirect impacts.

These changes to the economy are presented to an economic model (REMI) that translates them into

changes in the broader economy as spending, production, income and employment changes propagate.

These total impacts include the direct and indirect impacts, and also include the other changes in

economic activity that are induced by the mine closure.

In this analysis, no changes in the (post Unit 1 and 2 closure) operation of the Colstrip SES are assumed –

its production levels and thus electricity markets are assumed to be unaffected by the change in fuel

source. It is implicitly assumed that Montana coal is not used for the Colstrip SES in the Rosebud closure

scenario. In the event of a fuel switch at the Colstrip SES, the Rosebud mine is almost certain to close.

We can attach no such certainty to the ultimate source of whatever new fuel is acquired, but the

proximity and capacity of the Wyoming mines in the Powder River Basin make it quite plausible that

non-Montana coal would be used.

Direct Impacts: Rosebud Mine Operations

The Rosebud mine has produced output in the neighborhood of 8.5 million tons of low-sulphur,

subbituminous coal for use almost exclusively to the Colstrip SES since 2012 (see graph). The coal is

delivered by conveyor to the electric generating units 4.5 miles away. After the closure of Units 1 and 2

of the Colstrip SES, this output will fall to approximately 6.5 million annual tons.

20

App.161

Rosebud Mine Annual Production, Tons

16,000,000

14,000,000

12,000,000

10,000,000

8,000,000

6,000,000

4,000,000

2,000,000

0

In the scenario where the fuel for the Colstrip SES comes from another source and the Rosebud mine

closes, this production is lost to the economy.

The post Unit 1 and 2 closure operations of the

Rosebud Mine Operational Information

Rosebud mine will employ 337 Westmoreland

employees, in addition to a small number of non- CATEGORY

UNITS

2020

company contractors, on an annual basis. As can

be seen from the table, those jobs pay total

Headcount employment Jobs

337

compensation of approximately $40.6 million per Employee compensation $ Millions

$40.6

year, an average compensation level of

Mine production

Million tons

6.5

approximately $120,000 per employee.

Additionally, the company pays vendors and suppliers for everything from diesel fuel to scientific

services. Spending for inputs, some of which the company itself supplies, will be on the order of $31

million annually.

Operations of the mine also entail tax and royalty payments to all levels of government. Of special

importance are the taxes and royalties that are paid by all coal producers to state and local government.

Taxes Applied to Montana Coal Production

Tax Description

Coal Severance Tax

Gross Proceeds Tax

Resource Indemnity and Ground Water Assessment Tax

State Royalty Assessment

Federal Royalty Assessment

Federal Black Lung Tax

Federal Reclamation Tax

State Funds

State & Federal Funds

Federal Funds

Recipient

Tax Rates

15% of Contract Sales Price

State Government

5% of Contract Sales Price

Local Government

.4% of Contract Sales Price

State Government

Statutorily Set by DNRC

State Government

State & Federal Government 12.5% of Contract Sales Price

$0.55 per ton

Federal Government

$0.28 per ton

Federal Government

21

App.162

These taxes comprise an important

component of the Rosebud mine’s

direct impact. Based on recent

history, we have used the estimates

for mine-related tax and royalty

payments shown in the table, adjusted

to incorporate the closure of Units 1

and 2 of the Colstrip SES.

Rosebud Mine Coal Tax Payments ($ Millions)

CATEGORY

REMITTED TO

AMOUNT

Coal Severance Tax

State

$18.2

Gross Proceeds Tax

Local

$6.0

Resource Indemnity

and Ground Water

State

$0.5

Closure of the mine ceases production Assessment Tax

State Royalty Assessment

State

$0.5

ends these payments. The

Federal Royalty Assessment

State and Federal

$12.0

government spending and services

supported by these payments is assumed to be reduced accordingly. The alternative, raising taxes on

other taxpayers, was considered a less reasonable scenario.

Rosebud Mine Closure: Indirect Impacts

The closure of the Rosebud mine would be closely associated with two other events with economic

implications. One is the environmental remediation activities to return the land to a natural state.

Planning and bonding such activity are pre-requisites for permitting.

Westmoreland currently has a $165 million bond set aside for remediation. We assume that the

remediation activity funded by this bond will take place over a 10-year period commencing immediately

after mine closure. Since the source of funds for this activity comes from the company owners, this

activity adds jobs, income, and other spending flows to the local economy.

22

App.163

BNSF Southeast Montana – Northeast Wyoming Network Map

A second category of indirect impacts has to do with investments and operations at the Colstrip SES to

accommodate the delivery of coal by rail instead of conveyor. No information was made available as to

the size and scope of the capital investments needed at the Colstrip SES site. We did, however, make a

projection of the new rail activity that could occur. While not definitive, it does serve to illuminate the

order of magnitude of impacts that are associated with rail transport.

We begin with the assumption that Colstrip SES coal is sourced from Campbell County, Wyoming, where

the largest coal mines in the world are located immediately adjacent to rail transport. The BNSF rail

network map suggests an approximate rail distance of just under 300 miles along these corridors,

travelling through Sheridan, Lockwood, Forsyth, and finally south to Colstrip. Approximately 200 miles

of that route lies within the state of Montana, and at a cost of $0.03 per ton-mile for rail transport, it

represents a demand for $58 million for rail transport services for the 6.5 million tons of annual coal

consumption for the Colstrip SES.

This demand for rail transport services does increase rail employment and investment in the state, and it

is incorporated into the analysis.

23

App.164

Conclusion

The Rosebud mine has served the coal needs of the adjacent Colstrip Steam Electric Station (SES)

generating facility for more than 40 years. If this relationship were severed and those needs were

satisfied elsewhere, the mine’s operation would cease. This section has described the changes in

economic activity that could reasonably be expected in a mine closure scenario, including the cessation

of mining operations, the environmental remediation of the mine site, and changes in coal delivery at

the Colstrip SES. All of the changes are measured against a baseline that incorporates the closure of the

two smaller, older units of the Colstrip SES.

As described previously, these changes comprise the Rosebud mine closure scenario that is used to

produce an alternative projection of the state economy that reflects those events. We now turn to an

examination of those outcomes.

24

App.165

5. THE ECONOMIC IMPACTS OF A ROSEBUD MINE CLOSURE

An assessment of the economic implications of the closure of Westmoreland’s Rosebud Mine adjacent

to Colstrip, Montana involves (a) assessing the economic changes directly and indirectly associated with

the event, and (b) using an economic model to track how those changes propagate across the broader

economy to estimate the ultimate impact of the closure.

The economic impacts reported here include all changes in the economy that can be expected to occur

in a Rosebud mine closure scenario described in the previous section. As will be seen, the impacts

extend beyond the industry and workers directly affected, and are several orders of magnitude larger

than the activity of the mine itself. The impacts induced in the economy beyond the mine and beyond

Colstrip occur because of the spending flows from coal production that are received as income and

revenue by individuals, businesses and governments.

The impacts summarized in this report are not a forecast. They are the difference between two

forecasts of the future for the state economy – a status quo forecast, and a forecast that incorporates

the changes that would occur in a mine closure scenario. Since many measurements of economic

activity are in dollar terms, all such measures have been corrected for inflation and can be understood

as measures of dollar in terms of their purchasing power in the year 2019.

The closure scenario assumes that the fuel switching at the Colstrip SES and the mine closure take place

in the year 2020. This is clearly not realistic – an actual sequence of events that would produce a mine

closure would doubtless take years to occur. But the specificity of the assumption is useful in

understanding the results. The findings would be fundamentally unchanged for a different choice of

specific timing.

Impact Summary

Our basic finding is that the closure of the Rosebud mine would be a significant economic event, both

for the community of Colstrip and the rest of the state. The loss of 337 mining jobs and the production

of 6.5 million tons of coal in Colstrip ultimately contribute to a total job loss statewide of more than

1,800 jobs, as shown in the Table. The magnitude of the induced impacts on the rest of the economy

occur in part because of the high-paying nature of the mining jobs, and the outsized impact that mining

activity has on state and local tax revenues.

Impacts Summary

IMPACTS BY YEAR

CATEGORY

UNITS

2020

2021

2022

2023

2024

2025

Total employment

Jobs

-1,716

-1,852

-1,893

-1,872

-1,819

-1,754

Personal income

Disposable personal income

State Tax and Nontax Revenue

Output

Millions

-$95.6

-$81.1

-$39.0

-$325.9

-$101.4

-$86.1

-$41.0

-$347.8

-$109.2

-$92.9

-$42.7

-$355.2

-$113.0

-$96.4

-$43.8

-$353.4

-$114.8

-$98.0

-$44.4

-$346.8

-$115.4

-$98.6

-$44.7

-$338.5

Earnings per Lost Job

Population

Dollars

People

$61,998 $60,722 $61,828 $62,896 $63,940 $64,921

-568

-998

-1,337

-1,592

-1,777

-1,909

25

App.166

Other measures of economic activity reflect on different aspects of the mine closure’s impact.

x

Personal income received by Montana households would be reduced by more than $110 million

per year, compared to levels that would occur if the mine remained in operation. In terms of

after-tax, or disposable income, this is a decline of almost $100 million in purchasing power of

Montana households.

x

The revenue received by state government would be more than $40 million less each year due

to the closure of the mine, reflecting revenue from both tax and non-tax sources. Revenue

changes occur directly from the payments of the mine in natural resource taxes, and indirectly

from closure-induced changes in income and population.

x

Output, defined as gross receipts of business and non-business organizations would be more

than $340 million lower due to mine closure across the state;

x

The state of Montana would have fewer people, amounting ultimately to more than 1,900, in

the wake of events that led to closure of the Rosebud mine as described in this report. Those

shortfalls widen with time as individuals and families move in response to changes in economic

opportunities.

Not all of the changes in the economy that would take place in the aftermath of a Rosebud mine closure

are coincident with the actual closure events. We report on changes for a six year period – projected

economic impacts beyond six years can be found in the Appendix.

The changes summarized in the table refer to the economy of the state as a whole. While it can be

expected that the largest portion of these impacts are experienced in Colstrip and the surrounding

region itself, there are also impacts felt in other parts of the state. These occur primarily due to the

cutbacks in state government that are caused by the revenue impacts of the closure.

These impacts reflect all of the direct and indirect impacts associated with mine closure described in the

previous section – the negative as well as the positive. They include, for example, the positive impacts

that occur due to mine remediation activities as well as the additional rail activity that would occur in a

mine closure scenario. The fact that the net impacts are negative is because of the dominant effect of

the mine closure itself.

Employment Impacts

The dynamic nature of the economic changes that can be expected to occur in a mine closure scenario

can be seen by examining impacts over a longer period of time, as we present for total employment

impacts shown in the Figure below. There are several factors at work in the behavior of total

employment in response to mine closure-related events.

26

App.167

Employment Impacts by Year

0

-200

-400

-600

-800

-1,000

-1,200

-1,400

-1,600

-1,800

-2,000

The first observation is simply to note that the job impacts which commence in year 2020 when closure

takes place change over time. While many factors are at work in contributing to this outcome, three

stand out as being particularly important.

x

Short-term construction overshoot. In the years immediately following the closure, the

economy will have excess capital – both in structures and equipment. Construction will be

initially depressed as that excess supply is gradually absorbed. Thus impacts build for an interval

beyond the year that closure takes place, recovering slightly thereafter.

x

Productivity impacts. Over medium and long intervals of time, the productivity changes that can

be expected to occur that reduce labor requirements in capital intensive industries such as

mining result in smaller employment impacts for any physical output loss. Thus employment

impacts moderate slightly over longer time horizons.

x

Migration and demographic impacts. The migration of individuals and families not only happens

more slowly that other changes, the demographic changes that migration induces show up in

more complex ways. This is the main reason why population impacts continue to exhibit

variation decades after the closure event.

Another important aspect of the economic implications of Rosebud mine closure events can be seen

from a more detailed examination of job impacts, as shown in the table below.

27

App.168

State Employment Impacts

NUMBER OF JOBS

INDUSTRY

2020

2021

2022

2023

2024

2025

State and local government

-540

-602

-633

-646

-650

-646

Mining

-384

-382

-369

-358

-347

-337

Construction

-256

-310

-318

-301

-271

-237

Retail trade

-151

-159

-165

-164

-161

-156

Accommodation and food services

-75

-83

-90

-95

-98

-99

Health care and social assistance

-68

-69

-70

-69

-67

-66

Other services, except public administration

-59

-57

-56

-55

-53

-50

Professional and technical services

-33

-37

-38

-38

-37

-36

Wholesale trade

-26

-26

-26

-24

-23

-22

Arts, entertainments and recreation

-20

-20

-20

-19

-18

-18

Transportation and warehousing

-14

-14

-13

-11

-9

-8

Administrative and waste services

-9

-11

-11

-11

-10

-9

Manufacturing

-11

-11

-11

-10

-10

-9

Other private

-70

-72

-73

-70

-66

-62

-1,716

-1,852

-1,893

-1,872

-1,819

-1,754

TOTAL

Employment Impacts by Industry, 2022

State and local government

-633

Mining

-369

Construction

-318

Retail trade

-165

Accommodation and food services

-90

Health care and social assistance

-70

Other services, except public administration

-56

Professional and technical services

-38

Wholesale trade

-26

Arts, entertainments and recreation

-20

Transportation and warehousing

-13

Administrative and waste services

-11

Manufacturing

-11

Other private

-73

28

App.169

Perhaps the most striking aspect of the industry profile of employment impacts is the fact that the

sector of the state economy that makes the largest contribution to the total is state and local

government, not mining. There are several reasons for this outcome.

The first is that mining is a capital-intensive industry, and the loss of output affects relatively fewer

workers as a result. By contrast, government is primarily services with a relatively larger labor

requirements for each dollar of output. A second reason for the large state and local government

employment impact has to do with the disproportionate tax burden borne by natural resource

industries in general, and coal mining in particular. As we detail below, the Rosebud mine directly pays

more than $25 million to state and local government in taxes and royalties.

A third reason for the impact on government has to do with changes in population that occur due to

mine closure-related events. This clearly affects tax revenues, but it also impacts important nontax

revenue sources such as transfers from the Federal government. Those transfer flows change because

their apportionment formulas are linked to population.

More generally, the employment impacts demonstrate how closure related events propagate through

the rest of the economy. Removing almost $100 million in spending power from the state economy

lowers demand and thus labor needs for a wide range of industries, including retail trade, health care

and construction.

Personal Income Impacts

While reductions in employment associated with mine closure events clearly produce a loss in wage

income for Montana households, the reduction in economic activity that occurs due to closure also

produces other outcomes affecting income. This can be seen for a more detailed look at personal

income impacts shown in the Table below.

The Table presents an income accounting that is in standard use by Federal statistics agencies, applied

to the changes in income that can be expected to occur in a mine closure scenario. As can be seen,

while the bulk of changes due to closure events do occur in earnings – more specifically, in wages and

salaries – property income is affected as well. The latter exceed $9 million per year in year 2025. These

reductions occur due to population declines, as well as the lower demand for housing and commercial

real estate that result from lower levels of economic activity.

The definition of personal income, as can be seen from the Table below, is net of pension and social

security contributions and thus is closer to income available for spending than raw compensation.

Disposable income, which nets out personal income taxes, is even closer to that concept.

29

App.170

Personal Income Impacts

INCOME PER YEAR ($ Millions)

CATEGORY

TOTAL EARNINGS

Total wages and salary disbursements

SUPPLEMENTS TO WAGES AND SALARIES

Employer contributions for employee pension and insurance

funds

Employer contributions for government social insurance

Proprietors’ income w/inventory valuation and capital

consumption adjustments

LESS:

Contributions for government social insurance

Employee and self-employed contributions for government

social insurance

Employer contributions for government social insurance

PLUS:

2020

2021

2022

2023

2024

2025

-106.4

-112.5

-117.0

-117.7

-116.3

-113.9

-77.6

-81.2

-84.2

-85.0

-84.5

-83.3

-18.9

-20.4

-21.7

-22.4

-22.7

-22.7

-12.5

-13.5

-14.3

-14.8

-15.0

-15.0

-6.4

-6.9

-7.4

-7.6

-7.7

-7.7

-10.0

-10.9

-11.1

-10.4

-9.2

-7.9

-13.9

-14.7

-15.4

-15.8

-15.8

-15.7

-7.5

-7.8

-8.1

-8.2

-8.1

-8.0

-7.7

-6.4

-6.9

-7.4

-7.6

-7.7

Adjustment for residence

0.4

1.2

1.2

1.0

0.8

0.5

Gross in

-1.5

-0.9

-1.0

-1.1

-1.3

-1.5

Gross out

-1.9

-2.1

-2.2

-2.2

-2.1

-2.1

EQUALS:

Net earnings

-92.1

-96.5

-100.4

-100.9

-99.7

-97.6

PLUS:

Property income

-3.1

-3.8

-5.6

-7.1

-8.3

-9.2

Dividends

-1.1

-1.3

-1.9

-2.4

-2.7

-2.9

Interest

-1.4

-1.7

-2.7

-3.5

-4.1

-4.7

Rent

-0.6

-0.7

-1.1

-1.3

-1.5

-1.6

Transfer payments

EQUALS: Personal Income

LESS:

Personal current taxes

DISPOSABLE PERSONAL INCOME

-0.5

-1.1

-3.1

-5.0

-6.8

-8.5

-95.6

-14.5

-81.1

-101.4

-15.3

-86.1

-109.2

-16.2

-92.9

-113.0

-16.6

-96.4

-114.8

-16.8

-98.0

-115.4

-16.7

-98.6

Output Impacts

Income to business and non-business organizations is significantly impacted by the events described in

the Rosebud mine closure scenario. We show how economic output, or gross receipts for goods and

services, if affected for major industry categories by Rosebud mine closure events. The definition of

output in this table differs slightly from the more generally used definition – in the REMI model, output

is equal to gross receipts with the exception of retail and wholesale trade industries, where markup is

used instead.

The figure below gives a slightly different portrayal of impacts by industry than the pattern of impacts

for employment. In terms of output, mining itself is the industry that experiences the largest decline.

The sizable impacts felt by other businesses, most notably construction, retail trade and health care,

also are of note. These impacts give more information on the kinds of business and non-business

activity that would ultimately feel the effect of a closure of the Rosebud mine.

30

App.171

State Output Impacts

OUTPUT PER YEAR ($ Millions)

INDUSTRY

Mining

2020

2021

2022

2023

2024

2025

-167.6

-170.0

-169.6

-169.2

-168.8

-168.5

State and local government

-53.7

-60.2

-63.9

-65.8

-66.6

-66.8

Construction

-43.2

-53.6

-56.1

-53.9

-49.3

-43.9

Retail trade

-11.6

-12.5

-13.3

-13.5

-13.5

-13.4

Health care and social assistance

-7.5

-7.8

-8.1

-8.2

-8.1

-8.1

Wholesale trade

-5.7

-6.0

-6.0

-5.9

-5.7

-5.5

Accommodation and food services

-4.5

-5.0

-5.5

-5.8

-5.9

-6.0

Other services, except public administration

-4.0

-3.9

-3.9

-3.9

-3.8

-3.6

Manufacturing

-3.9

-3.9

-3.8

-3.5

-3.2

-2.9

Professional and technical services

-3.4

-3.8

-4.0

-4.0

-4.0

-3.9

Transportation and warehousing

1.8

2.1

2.6

3.1

3.6

4.0

Arts, entertainments and recreation

-1.3

-1.3

-1.3

-1.3

-1.3

-1.2

Administrative and waste services

1.1

1.0

0.9

1.0

1.0

1.1

Other private

-22.3

-22.8

-23.3

-22.5

-21.2

-19.8

TOTAL

-325.9

-347.8

-355.2

-353.4

-346.8

-338.5

Output Impacts by Industry, $ Millions, 2022

Mining

-169.6

State and local government

-63.9

Construction

-56.1

Retail trade

-13.3

Health care and social assistance

-8.1

Wholesale trade

-6.0

Accommodation and food services

-5.5

Other services, except public administration

-3.9

Manufacturing

-3.8

Professional and technical services

-4.0

Transportation and warehousing

2.6

Arts, entertainments and recreation

-1.3

Administrative and waste services

0.9

Other private

-23.3

31

App.172

State and Local Revenue Impacts

The outsized role of coal mining and other natural resource industries in the state revenue base makes

events which affect production and sales of especial importance to Montana. A closure of the second

largest coal mine in the state can be expected to have a noticeable impact on revenues. Since Federal

tax revenues flow outside the state and have only a limited impact on Federal spending, we focus on the

impacts at the state and local level.

The revenue impacts that are estimated by the REMI model are classified by the same categories

defined by the Census of Governments conducted by the U.S. Census Bureau. As can be seen from the

tax and nontax revenue impacts of the events associated with the Rosebud mine closure shown in the

Table and Figure below, some of these categories directly correspond to specific taxes, such as the

individual income tax. The general sales tax category, of course, registers no impact because the state

of Montana has no general sales tax.

The second largest revenue impact to state government is not a tax source. It is the broad category of

intergovernmental revenue that is almost totally comprised of transfers from the Federal government.

This reflects everything from highway funding to health care programs, and is affected by changes in

income and population that go into the apportionment formulas. The largest impact is in the selected

sales tax category, where mining royalties (both from state land and the remittance of a portion of

royalties from federal land) and severance taxes are included.

Aside from a small levy to support the University system, there is no statewide property tax in Montana.

Those impacts are not included in the Table.

32

App.173

State Revenue Impacts

REVENUE PER YEAR ($ Millions)

CATEGORY

Intergovernmental Revenue

2020

2021

2022

2023

2024

2025

-7.9

-8.9

-9.8

-10.4

-10.8

-11.2

General Sales Tax

0.0

0.0

0.0

0.0

0.0

0.0

Selective Sales Tax

-20.6

-20.9

-21.1

-21.1

-21.2

-21.2

License Taxes

-0.5

-0.6

-0.6

-0.6

-0.6

-0.6

Individual Income Tax

-2.8

-3.0

-3.2

-3.2

-3.2

-3.2

Corporate Income Tax

-0.8

-0.8

-0.9

-0.9

-0.8

-0.8

Other Taxes

-1.3

-1.3

-1.4

-1.5

-1.5

-1.5

Current Charges

-1.4

-1.4

-1.6

-1.6

-1.6

-1.6

Miscellaneous General Revenue

-1.2

-1.3

-1.4

-1.4

-1.5

-1.5

Utility Revenue

-0.1

-0.1

-0.1

-0.2

-0.2

-0.2

Liquor Store Revenue

-0.2

-0.2

-0.2

-0.2

-0.2

-0.2

Insurance Trust Revenue

-2.2

-2.3

-2.5

-2.6

-2.6

-2.7

TOTAL

-39.0

-41.0

-42.7

-43.8

-44.4

-44.7

State Revenue Impacts, $ Millions, 2022

Intergovernmental Revenue

-11.2

General Sales Tax

Selective Sales Tax

0.0

-21.2

License Taxes

-0.6

Individual Income Tax

-3.2

Corporate Income Tax

-0.8

Other Taxes

-1.5

Current Charges

-1.6

Miscellaneous General Revenue

-1.5

Utility Revenue

-0.2

Liquor Store Revenue

-0.2

Insurance Trust Revenue

-2.7

Property tax impacts are almost totally felt by local governments, and can be expected to be significant.

Aside from the six largest cities, the city of Colstrip has the highest assessed value for property tax in the

state. Actual tax impacts depend on a number of factors that are difficult to project, including the

33

App.174

setting of mill rates of remaining property. But it is possible to describe the kinds of revenue declines

that could be expected to occur in the mine closure scenario addressed by this report.

The results reported here were produced by combining the impacts on residential and nonresidential

capital stock produced by REMI model with effective tax rates by property class and county across the

state. Because changes in capital occur slowly in response to economic events, the property tax impacts

display a lagged response to mine closure events.

$ Millions

Property Tax Impact by Region 2020-2040

0

Other

Regions

-1

-2

-3

-4

-5

-6

Eastern Region

-7

-8

2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040

The results are shown for five regions of the state in the Figure above. As can be seen, the property tax

impact for the eastern region of the state, which includes Rosebud County where the mine is located,

dominate the overall impacts. The impacts in the east exceed $7 million in lost annual revenue per year.

Population Impacts

The movement of people and families to places of greater economic opportunity is an important part of

this analysis. Those movements are coincident with, and often caused by, changes in business

investment towards and away from regions in response to profitability and return. It can be expected

that declines in opportunity in the small community of Colstrip would produce considerable migration

from the community. Even at the state level, population impacts would be noticeable.

34

App.175

Population Impacts by Age

POPULATION IMPACT

AGE COHORT

2020

2021

2022

2023

2024

2025

Ages 0-14

-140

-251

-343

-418

-477

-523

Ages 15-24

-126

-206

-254

-274

-276

-267

Ages 25-64

-302

-539

-734

-889

-1,007

-1,094

Ages 65+

0

-2

-6

-11

-18

-26

TOTAL

-568

-998

-1,337

-1,592

-1,777

-1,909

Population Impacts by Age, 2025

Ages 0-14

-523

Ages 15-24

Ages 25-64

-267

-1,094

-26

Ages 65+

The impacts on population caused by mine closure events are dominated by working age people and

their children, as can be seen from the figure above. This has implications for demand for local

government services, especially K-12 schools. Population impacts continue to evolve in subsequent

years.

35

App.176

Conclusion

The closure of the Rosebud mine would be a significant economic event. Even after taking into account

the new economic activity associated with remediation of the mine site, as well as additional rail jobs

created from rail delivery of coal to the Colstrip SES, the net result of a fuel switch would be:

x

x

x

x

x

x

a loss in excess of 1,800 jobs in the state economy, across a broad mix of industries;

a reduction in annual income received by households of more than $100 million per year,

compared to levels that would have occurred had the mine remained in operation;

a shortfall of more than $40 million per year in state government revenue;

a decline in local property tax revenue in excess of $7 million per year, most of which occurs in

Rosebud County;

a loss of revenue to businesses in excess of $300 million per year compared to levels that would

be maintained under continued operation;

a state population that is smaller by more than 1,900 people;

36

App.177

6. SUMMARY AND CONCLUSION

This study is intended to assess the economic implications of the closure of the Rosebud mine, in a

scenario where the Colstrip SES continues to operate with coal from another source. If the mine were

to close, its wages, vendor purchases, and tax payments to workers, businesses and governments would

largely cease. The spending of those who receive those payments as income or revenues can be

expected to be significantly curtailed, with further knock-on effects on economic activity. As the process

continues, an equilibrium level of activity emerges that shrinks the overall economic pie by more than

that of the mine itself.

We assess the ultimate impact of the Rosebud mine closure with the use of an economic model that

traces the interactions between the mine and the rest of the economy. Specifically developed for this

purpose and calibrated with Montana-specific data, the REMI model was leased from Regional Economic

Models, Inc. (www.remi.com) for use in this study. Since its development in the early 1980s, the REMI

model has been the subject of dozens of peer-reviewed economic articles and has been employed in

hundreds of applications.

A mine-closure scenario for the Montana economy was constructed for this study that consisted of

three broad components. The first was the removal of the mine itself from the economy. Using

information obtained from the mine owners on production, employment, payroll, taxes and vendor

spending, the direct contribution of its operations was removed from the economy. A second

component was the remediation activities that would commence and continue subsequent to mine

closure.

The third component were the operating changes, if any, at the Colstrip SES that might be expected to

occur with coal deliveries by rail to that facility from a non-Montana source. While little information

was available to detail actual plans and potential investments, a simplified scenario can at least

represent the basics of how fuel source changes would affect economic flows.

SUMMARY OF FINDINGS

Our basic finding is that the shutdown of the Rosebud mine would be a significant economic event, both

for the Colstrip community and to the remainder of the state. The loss of the facility’s 337 jobs,

currently mining 6.5 million tons of coal annually, would ultimately lead to a state economy that:

x

x

x

x

x

x

Suffers a loss in excess of almost 1,900 jobs, across a broad mix of industries;

Has annual income received by households fall more than $100 million short of the amount that

would have occurred had the mine remained in operation;

Experiences a shortfall of more than $40 million per year in state government revenue;

Sees local governments collect $7 million less per year in property tax revenue, most of which

occurs in Rosebud County;

Faces a loss of revenue to businesses in excess of $300 million per year compared to levels that

would be maintained under continued operation;

Has a population that is smaller by more than 1,900 people;

37

App.178

Impacts Summary

IMPACTS BY YEAR

CATEGORY

UNITS

2020

2021

2022

2023

2024

2025

Total employment

Jobs

-1,716

-1,852

-1,893

-1,872

-1,819

-1,754

Personal income

Disposable personal income

State Tax and Nontax Revenue

Output

Millions

-$95.6

-$81.1

-$39.0

-$325.9

-$101.4

-$86.1

-$41.0

-$347.8

-$109.2

-$92.9

-$42.7

-$355.2

-$113.0

-$96.4

-$43.8

-$353.4

-$114.8

-$98.0

-$44.4

-$346.8

-$115.4

-$98.6

-$44.7

-$338.5

Earnings per Lost Job

Population

Dollars

People

$61,998 $60,722 $61,828 $62,896 $63,940 $64,921

-568

-998

-1,337

-1,592

-1,777

-1,909

These findings take into account the positive impacts that come from the $165 million that is expected

to be spent in the decade following the closure on environmental remediation, as well as the modest

increase in rail and material handling jobs that could be expected to occur as the Colstrip SES moves to

use of non-Montana coal.

38

App.179

REFERENCES

Barkey, PM and PE Polzin (2010) The economic contribution of Colstrip Steam Electric Station units 1-4,

Missoula.

Barkey, PM (2012). The economic impact of increased production at the Spring Creek mine, Missoula:

Bureau of Business and Economic Research, University of Montana.

Bureau of Business and Economic Research, University of Montana (2015). The economic implications

of implementing the EPA Clean Power Plan in Montana: Final Report. Missoula: Bureau of Business and

Economic Research.

Barkey, PM (2018). The economic impact of the early retirement of Colstrip Units 3 and 4: Final Report.

Missoula, Bureau of Business and Economic Research, University of Montana.

Brucker SM, Hastings SE, Latham WR III (1990) The variation of estimated impacts from five regional

input-output models. Int Reg Sci Rev 13:119–139.

Crihfield JB, Campbell HS Jr (1991) Evaluating alternative regional planning models. Growth and Change

22:1–16.

Crihfield JB, Campbell HS Jr (1992) Evaluating alternative regional planning models: reply. Growth and

Change 23:521–530.

Grimes DR, Fulton GA, Bonardelli MA (1992) Evaluating alternative regional planning models: a

comment. Growth and Change 23:516–520.

Radtke H, Detering S, Brokken R (1985) A comparison of economic impact estimates for changes in the

federal grazing fee: secondary vs primary data I/O models. Western J Agricultural Econ 10:382–390.

REMI (2019) Data sources and estimation procedures, https://www.remi.com/wpcontent/uploads/2019/08/Data-Sources-and-Estimation-Procedures.pdf .

Richman DS, Schwer RK (1993) A systematic comparison of the REMI and IMPLAN models: the case of

southern Nevada. Rev Reg Stud 23:143–161.

Rickman DS, Schwer RK (1994) Multiplier comparisons of the IMPLAN and REMI models across versions:

illuminating black boxes. Environ Plan A26.

Stevens BH, Treyz GI, Ehrlich DJ, Bower JR (1983) A new technique for the construction of non-survey

regional input-output models. Int Reg Sci Rev 8:271–286.

Stevens BH, Treyz GI, Lahr ML (1989) On the comparative accuracy of RPC estimating techniques. In:

Miller RE, Polenske KR, Rose AZ (eds) Frontier of input-output analysis. Oxford University Press, New

York Oxford.

Treyz GI, Richman DS, Shao G (1992) The REMI economic-demographic forecasting and simulation

model. Int Reg Sci Rev 14:221–253.

Treyz, GI (1993). Regional Economic Modeling: A Systematic Approach to Economic Forecasting and

Policy Analysis. New York, NY: Springer Netherlands.

39

App.180

Westmoreland Mining, LLC (2019). Personal communication, July 23, 2019, summarized as follows:

Current headcount employment – 365

Total compensation to employees - $44M annually

Total annual mine production – 8.5 million tons

Employment and compensation of outside contractors – 4 employees, $15M/year

Selected vendor and supplier purchases -- $8M/year electricity, $10M/year diesel

Taxes paid:

Montana Severance Tax – 23M /year

Rosebud County Gross Proceeds Tax - $8M/year

Federal Royalties - $13M/year

Bond amount for remediation - $165M

Note: Employment and other measures used in the analysis incorporate reductions from these

amounts to account for the impact of the closure of Colstrip SES Units 1 and 2 in the baseline

scenario.

40

App.181

A

PPENDIX 1:

TABLES

APPENDIX

1: REMI

REMI TABLES

41

41

App.182

Forecast

&ŽƌĞĐĂƐƚ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

Individuals

/ŶĚŝǀŝĚƵĂůƐ

Individuals

/ŶĚŝǀŝĚƵĂůƐ

Individuals

/ŶĚŝǀŝĚƵĂůƐ

Millions

of Fixed (2019) Dollars

DŝůůŝŽŶƐŽĨ&ŝdžĞĚ;ϮϬϭϵͿŽůůĂƌƐ

Millions

of Fixed (2019) Dollars

DŝůůŝŽŶƐŽĨ&ŝdžĞĚ;ϮϬϭϵͿŽůůĂƌƐ

Millions

of Fixed (2019) Dollars

DŝůůŝŽŶƐŽĨ&ŝdžĞĚ;ϮϬϭϵͿŽůůĂƌƐ

Millions

of Fixed (2019) Dollars

DŝůůŝŽŶƐŽĨ&ŝdžĞĚ;ϮϬϭϵͿŽůůĂƌƐ

Millions

of Fixed (2019) Dollars

DŝůůŝŽŶƐŽĨ&ŝdžĞĚ;ϮϬϭϵͿŽůůĂƌƐ

Millions

of Fixed (2012) Dollars

DŝůůŝŽŶƐŽĨ&ŝdžĞĚ;ϮϬϭϮͿŽůůĂƌƐ

Thousands

of Fixed (2012) Dollars

dŚŽƵƐĂŶĚƐŽĨ&ŝdžĞĚ;ϮϬϭϮͿŽůůĂƌƐ

2012=100

(Nation)

ϮϬϭϮсϭϬϬ;EĂƚŝŽŶͿ

Population

WŽƉƵůĂƚŝŽŶ

Labor

Force

>ĂďŽƌ&ŽƌĐĞ

Gross

Domestic Product

'ƌŽƐƐŽŵĞƐƚŝĐWƌŽĚƵĐƚ

Output

KƵƚƉƵƚ

Value-Added

sĂůƵĞͲĚĚĞĚ

Personal

Income

WĞƌƐŽŶĂů/ŶĐŽŵĞ

Disposable

Personal Income

ŝƐƉŽƐĂďůĞWĞƌƐŽŶĂů/ŶĐŽŵĞ

Real

Disposable Personal Income

ZĞĂůŝƐƉŽƐĂďůĞWĞƌƐŽŶĂů/ŶĐŽŵĞ

Real

Disposable Personal Income per Capita

ZĞĂůŝƐƉŽƐĂďůĞWĞƌƐŽŶĂů/ŶĐŽŵĞƉĞƌĂƉŝƚĂ

PCE-Price

Index

WͲWƌŝĐĞ/ŶĚĞdž

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

0.000:

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

0.000:

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

0.000:

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

0.000:

Ϭ͘ϬϬϬ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

0.000!

Ϭ͘ϬϬϬ

id

Adjusted

ZĞƐŝĚĞŶĐĞĚũƵƐƚĞĚŵƉůŽLJŵĞŶƚ

ploy

2018

ϮϬϭϴ

Westmoreland

Shutdown Postland2

tĞƐƚŵŽƌĞůĂŶĚ^ŚƵƚĚŽǁŶWŽƐƚϭĂŶĚϮ

Private

Non-Farm Employment

WƌŝǀĂƚĞEŽŶͲ&ĂƌŵŵƉůŽLJŵĞŶƚ

Units

hŶŝƚƐ

Comparison

Type

ŽŵƉĂƌŝƐŽŶdLJƉĞ

Differences

ŝĨĨĞƌĞŶĐĞƐ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Category

ĂƚĞŐŽƌLJ

Region

ZĞŐŝŽŶ

Total

Employment

dŽƚĂůŵƉůŽLJŵĞŶƚ

Montana

DŽŶƚĂŶĂ

Summary

^ƵŵŵĂƌLJ

Comparison

Forecast

ŽŵƉĂƌŝƐŽŶ&ŽƌĞĐĂƐƚ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

0.000:

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

0.000:

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

0.000:

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

0.000:

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

0.000:

Ϭ͘ϬϬϬ

0.000!

Ϭ͘ϬϬϬ

-0.003

ͲϬ͘ϬϬϯ

-0.049

ͲϬ͘Ϭϰϵ

-73.795

Ͳϳϯ͘ϳϵϱ

-81.125

Ͳϴϭ͘ϭϮϱ

-95.626

Ͳϵϱ͘ϲϮϲ

-199.704

Ͳϭϵϵ͘ϳϬϰ

325.891

ͲϯϮϱ͘ϴϵϭ

-199.704

Ͳϭϵϵ͘ϳϬϰ

-427.680

ͲϰϮϳ͘ϲϴϬ

-568.216

Ͳϱϲϴ͘Ϯϭϲ

-1702.460

ͲϭϳϬϮ͘ϰϲϬ

-1176.778

Ͳϭϭϳϲ͘ϳϳϴ

-1716.455

Ͳϭϳϭϲ͘ϰϱϱ

2020

ϮϬϮϬ

App.1838

App.183

2019

ϮϬϭϵ

Standard

Regional Control

^ƚĂŶĚĂƌĚZĞŐŝŽŶĂůŽŶƚƌŽů

-0.024|

ͲϬ͘ϬϮϰ

-0.038

ͲϬ͘Ϭϯϴ

-78.340

Ͳϳϴ͘ϯϰϬ

-86.121

Ͳϴϲ͘ϭϮϭ

-101.421

ͲϭϬϭ͘ϰϮϭ

-213.063

ͲϮϭϯ͘Ϭϲϯ

347.753

Ͳϯϰϳ͘ϳϱϯ

-213.063

ͲϮϭϯ͘Ϭϲϯ

-706,292

ͲϳϬϲ͘ϮϵϮ

-997.782

Ͳϵϵϳ͘ϳϴϮ

-1824,223

ͲϭϴϮϰ͘ϮϮϯ

-1249.764

ͲϭϮϰϵ͘ϳϲϰ

-1851.973

Ͳϭϴϱϭ͘ϵϳϯ

2021

ϮϬϮϭ

Year

zĞĂƌ

2022

ϮϬϮϮ

-0.027

ͲϬ͘ϬϮϳ

-0.031

ͲϬ͘Ϭϯϭ

-84.545

Ͳϴϰ͘ϱϰϱ

-92.942

ͲϵϮ͘ϵϰϮ

-109.150

ͲϭϬϵ͘ϭϱϬ

-217.918

ͲϮϭϳ͘ϵϭϴ

-355.220

Ͳϯϱϱ͘ϮϮϬ

-217.918

ͲϮϭϳ͘ϵϭϴ

-908.732

ͲϵϬϴ͘ϳϯϮ

-1337.341

Ͳϭϯϯϳ͘ϯϰϭ

-1867.377

Ͳϭϴϲϳ͘ϯϳϳ

-1259.691

ͲϭϮϱϵ͘ϲϵϭ

-1892.618

ͲϭϴϵϮ͘ϲϭϴ

2023

ϮϬϮϯ

-0.029

ͲϬ͘ϬϮϵ

-0.024

ͲϬ͘ϬϮϰ

-87.697

Ͳϴϳ͘ϲϵϳ

-96.407

Ͳϵϲ͘ϰϬϳ

-113.036

Ͳϭϭϯ͘Ϭϯϲ

-217.036

ͲϮϭϳ͘Ϭϯϲ

-353.373

Ͳϯϱϯ͘ϯϳϯ

-217.036

ͲϮϭϳ͘Ϭϯϲ

-1050.001

ͲϭϬϱϬ͘ϬϬϭ

-1591.959

Ͳϭϱϵϭ͘ϵϱϵ

-1851.420

Ͳϭϴϱϭ͘ϰϮϬ

-1225.317

ͲϭϮϮϱ͘ϯϭϳ

-1871.783

Ͳϭϴϳϭ͘ϳϴϯ

2024

ϮϬϮϰ

-0.029

ͲϬ͘ϬϮϵ

-0.018

ͲϬ͘Ϭϭϴ

-89.188

Ͳϴϵ͘ϭϴϴ

-98.046

Ͳϵϴ͘Ϭϰϲ

-114.801

Ͳϭϭϰ͘ϴϬϭ

-213.282

ͲϮϭϯ͘ϮϴϮ

-346.844

Ͳϯϰϲ͘ϴϰϰ

-213.282

ͲϮϭϯ͘ϮϴϮ

-1136.611

Ͳϭϭϯϲ͘ϲϭϭ

-1777.074

Ͳϭϳϳϳ͘Ϭϳϰ

-1805.309

ͲϭϴϬϱ͘ϯϬϵ

-1169.818

Ͳϭϭϲϵ͘ϴϭϴ

-1819.331

Ͳϭϴϭϵ͘ϯϯϭ

2025

ϮϬϮϱ

-0.029

ͲϬ͘ϬϮϵ

-0.013

ͲϬ͘Ϭϭϯ

-89.711

Ͳϴϵ͘ϳϭϭ

-98.622

Ͳϵϴ͘ϲϮϮ

-115.356

Ͳϭϭϱ͘ϯϱϲ

-208.439

ͲϮϬϴ͘ϰϯϵ

-338.521

Ͳϯϯϴ͘ϱϮϭ

-208.439

ͲϮϬϴ͘ϰϯϵ

-1184.804'

Ͳϭϭϴϰ͘ϴϬϰ

-1909.481

ͲϭϵϬϵ͘ϰϴϭ

-1746.773

Ͳϭϳϰϲ͘ϳϳϯ

-1107.546

ͲϭϭϬϳ͘ϱϰϲ

-1754.040

Ͳϭϳϱϰ͘ϬϰϬ

2026

ϮϬϮϲ

-0.029

ͲϬ͘ϬϮϵ

-0.009

ͲϬ͘ϬϬϵ

-88.954

Ͳϴϴ͘ϵϱϰ

-97.789

Ͳϵϳ͘ϳϴϵ

-115.261

Ͳϭϭϱ͘Ϯϲϭ

-203.468

ͲϮϬϯ͘ϰϲϴ

-329.970

ͲϯϮϵ͘ϵϳϬ

-203.468

ͲϮϬϯ͘ϰϲϴ

-1207.785

ͲϭϮϬϳ͘ϳϴϱ

-2001.623

ͲϮϬϬϭ͘ϲϮϯ

-1686.573

Ͳϭϲϴϲ͘ϱϳϯ

-1046.949

ͲϭϬϰϲ͘ϵϰϵ

-1687.595

Ͳϭϲϴϳ͘ϱϵϱ

2027

ϮϬϮϳ

-0.028

ͲϬ͘ϬϮϴ

-0.005

ͲϬ͘ϬϬϱ

-88.087

Ͳϴϴ͘Ϭϴϳ

-96.836

Ͳϵϲ͘ϴϯϲ

-114.896

Ͳϭϭϰ͘ϴϵϲ

-198.946

Ͳϭϵϴ͘ϵϰϲ

-322.350

ͲϯϮϮ͘ϯϱϬ

-198.946

Ͳϭϵϴ͘ϵϰϲ

-1213.279

ͲϭϮϭϯ͘Ϯϳϵ

-2062.205

ͲϮϬϲϮ͘ϮϬϱ

-1630.498

ͲϭϲϯϬ͘ϰϵϴ

-992.925

ͲϵϵϮ͘ϵϮϱ

-1625.836

ͲϭϲϮϱ͘ϴϯϲ

2028

ϮϬϮϴ

-0.028

ͲϬ͘ϬϮϴ

-0.003

ͲϬ͘ϬϬϯ

-88.156

Ͳϴϴ͘ϭϱϲ

-96.912

Ͳϵϲ͘ϵϭϮ

-114.721

Ͳϭϭϰ͘ϳϮϭ

-195.221

Ͳϭϵϱ͘ϮϮϭ

-316.116

Ͳϯϭϲ͘ϭϭϲ

-195.221

Ͳϭϵϱ͘ϮϮϭ

-1208.059

ͲϭϮϬϴ͘Ϭϱϵ

-2101.659

ͲϮϭϬϭ͘ϲϱϵ

-1583.703

Ͳϭϱϴϯ͘ϳϬϯ

-948.781

Ͳϵϰϴ͘ϳϴϭ

-1573.938

Ͳϭϱϳϯ͘ϵϯϴ

2029

ϮϬϮϵ

-0.027

ͲϬ͘ϬϮϳ

-0.001

ͲϬ͘ϬϬϭ

-88.234'

Ͳϴϴ͘Ϯϯϰ

-96.998

Ͳϵϲ͘ϵϵϴ

-114.681

Ͳϭϭϰ͘ϲϴϭ

-192.308

ͲϭϵϮ͘ϯϬϴ

-311.303

Ͳϯϭϭ͘ϯϬϯ

-192.308

ͲϭϵϮ͘ϯϬϴ

-1197.573

Ͳϭϭϵϳ͘ϱϳϯ

-2125.971

ͲϮϭϮϱ͘ϵϳϭ

-1544.985

Ͳϭϱϰϰ͘ϵϴϱ

-913.432

Ͳϵϭϯ͘ϰϯϮ

-1531.020

Ͳϭϱϯϭ͘ϬϮϬ

2030

ϮϬϯϬ

-0.027

ͲϬ͘ϬϮϳ

0.000:

Ϭ͘ϬϬϬ

-88.355

Ͳϴϴ͘ϯϱϱ

-97.131

Ͳϵϳ͘ϭϯϭ

-114.702

Ͳϭϭϰ͘ϳϬϮ

-190.283

ͲϭϵϬ͘Ϯϴϯ

-308.026

ͲϯϬϴ͘ϬϮϲ

-190.283

ͲϭϵϬ͘Ϯϴϯ

-1183.176

Ͳϭϭϴϯ͘ϭϳϲ

-2139.229

ͲϮϭϯϵ͘ϮϮϵ

-1513.331

Ͳϭϱϭϯ͘ϯϯϭ

-886.312

Ͳϴϴϲ͘ϯϭϮ

-1496.067

Ͳϭϰϵϲ͘Ϭϲϳ

Category

ĂƚĞŐŽƌLJ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Support

activities for mining

^ƵƉƉŽƌƚĂĐƚŝǀŝƚŝĞƐĨŽƌŵŝŶŝŶŐ

Utilities

hƚŝůŝƚŝĞƐ

Construction

ŽŶƐƚƌƵĐƚŝŽŶ

Wood

product manufacturing

tŽŽĚƉƌŽĚƵĐƚŵĂŶƵĨĂĐƚƵƌŝŶŐ

Nonmetallic

mineral product manufacturing

EŽŶŵĞƚĂůůŝĐŵŝŶĞƌĂůƉƌŽĚƵĐƚŵĂŶƵĨĂĐƚƵƌŝŶŐ

Primary

metal manufacturing

WƌŝŵĂƌLJŵĞƚĂůŵĂŶƵĨĂĐƚƵƌŝŶŐ

metal product

&ĂďƌŝĐĂƚĞĚŵĞƚĂůƉƌŽĚƵĐƚŵĂŶƵĨĂĐƚƵƌŝŶŐ

Machinery

manufacturing

DĂĐŚŝŶĞƌLJŵĂŶƵĨĂĐƚƵƌŝŶŐ

Computer

and electronic product manufacturing

ŽŵƉƵƚĞƌĂŶĚĞůĞĐƚƌŽŶŝĐƉƌŽĚƵĐƚŵĂŶƵĨĂĐƚƵƌŝŶŐ

and

ng

ůĞĐƚƌŝĐĂůĞƋƵŝƉŵĞŶƚ͕ĂƉƉůŝĂŶĐĞ͕ĂŶĚĐŽŵƉŽŶĞŶƚŵĂŶƵĨĂĐƚƵƌŝŶŐ

Motor

vehicles, bodies and trailers, and parts manufacturing

DŽƚŽƌǀĞŚŝĐůĞƐ͕ďŽĚŝĞƐĂŶĚƚƌĂŝůĞƌƐ͕ĂŶĚƉĂƌƚƐŵĂŶƵĨĂĐƚƵƌŝŶŐ

Other

i

i

i

KƚŚĞƌƚƌĂŶƐƉŽƌƚĂƚŝŽŶĞƋƵŝƉŵĞŶƚŵĂŶƵĨĂĐƚƵƌŝŶŐ

Furniture

and related product manufacturing

&ƵƌŶŝƚƵƌĞĂŶĚƌĞůĂƚĞĚƉƌŽĚƵĐƚŵĂŶƵĨĂĐƚƵƌŝŶŐ

Miscellaneous

manufacturing

DŝƐĐĞůůĂŶĞŽƵƐŵĂŶƵĨĂĐƚƵƌŝŶŐ

Food

manufacturing

&ŽŽĚŵĂŶƵĨĂĐƚƵƌŝŶŐ

Beverage

and tobacco product manufacturing

ĞǀĞƌĂŐĞĂŶĚƚŽďĂĐĐŽƉƌŽĚƵĐƚŵĂŶƵĨĂĐƚƵƌŝŶŐ

Textile

mills; Textile product mills

dĞdžƚŝůĞŵŝůůƐ͖dĞdžƚŝůĞƉƌŽĚƵĐƚŵŝůůƐ

Apparel

manufacturing; Leather and allied product manufacturing

ƉƉĂƌĞůŵĂŶƵĨĂĐƚƵƌŝŶŐ͖>ĞĂƚŚĞƌĂŶĚĂůůŝĞĚƉƌŽĚƵĐƚŵĂŶƵĨĂĐƚƵƌŝŶŐ

Paper

manufacturing

WĂƉĞƌŵĂŶƵĨĂĐƚƵƌŝŶŐ

Printing

and related support acti’

S

WƌŝŶƚŝŶŐĂŶĚƌĞůĂƚĞĚƐƵƉƉŽƌƚĂĐƚŝǀŝƚŝĞƐ

Petroleum

and coal products manufacturing

WĞƚƌŽůĞƵŵĂŶĚĐŽĂůƉƌŽĚƵĐƚƐŵĂŶƵĨĂĐƚƵƌŝŶŐ

Chemical

manufacturing

ŚĞŵŝĐĂůŵĂŶƵĨĂĐƚƵƌŝŶŐ

Plastics

and rubber products manufacturing

WůĂƐƚŝĐƐĂŶĚƌƵďďĞƌƉƌŽĚƵĐƚƐŵĂŶƵĨĂĐƚƵƌŝŶŐ

Wholesale

trade

tŚŽůĞƐĂůĞƚƌĂĚĞ

Retail

trade

ZĞƚĂŝůƚƌĂĚĞ

Air

transportation

ŝƌƚƌĂŶƐƉŽƌƚĂƚŝŽŶ

Rail

transportation

ZĂŝůƚƌĂŶƐƉŽƌƚĂƚŝŽŶ

Water

transportation

tĂƚĞƌƚƌĂŶƐƉŽƌƚĂƚŝŽŶ

Truck

transportation

dƌƵĐŬƚƌĂŶƐƉŽƌƚĂƚŝŽŶ

Couriers

and messengers

ŽƵƌŝĞƌƐĂŶĚŵĞƐƐĞŶŐĞƌƐ

Transit

and ground passenger transportation

dƌĂŶƐŝƚĂŶĚŐƌŽƵŶĚƉĂƐƐĞŶŐĞƌƚƌĂŶƐƉŽƌƚĂƚŝŽŶ

Pipeline

transportation

WŝƉĞůŝŶĞƚƌĂŶƐƉŽƌƚĂƚŝŽŶ

Scenic

and sightseeing transportation; Support activities for transportation

^ĐĞŶŝĐĂŶĚƐŝŐŚƚƐĞĞŝŶŐƚƌĂŶƐƉŽƌƚĂƚŝŽŶ͖^ƵƉƉŽƌƚĂĐƚŝǀŝƚŝĞƐĨŽƌƚƌĂŶƐƉŽƌƚĂƚŝŽŶ

Warehousing

and storage

tĂƌĞŚŽƵƐŝŶŐĂŶĚƐƚŽƌĂŐĞ

Publishing

industries, except Internet

WƵďůŝƐŚŝŶŐŝŶĚƵƐƚƌŝĞƐ͕ĞdžĐĞƉƚ/ŶƚĞƌŶĞƚ

Motion

picture and sound recording industries

DŽƚŝŽŶƉŝĐƚƵƌĞĂŶĚƐŽƵŶĚƌĞĐŽƌĚŝŶŐŝŶĚƵƐƚƌŝĞƐ

Data

processing, hosting, and related services; Other information services

ĂƚĂƉƌŽĐĞƐƐŝŶŐ͕ŚŽƐƚŝŶŐ͕ĂŶĚƌĞůĂƚĞĚƐĞƌǀŝĐĞƐ͖KƚŚĞƌŝŶĨŽƌŵĂƚŝŽŶƐĞƌǀŝĐĞƐ

Broadcasting,

except Internet

ƌŽĂĚĐĂƐƚŝŶŐ͕ĞdžĐĞƉƚ/ŶƚĞƌŶĞƚ

Telecommunicat

ns

dĞůĞĐŽŵŵƵŶŝĐĂƚŝŽŶƐ

Monetary

authorities - central bank; Credit intermediation and related acti

ies

DŽŶĞƚĂƌLJĂƵƚŚŽƌŝƚŝĞƐͲĐĞŶƚƌĂůďĂŶŬ͖ƌĞĚŝƚŝŶƚĞƌŵĞĚŝĂƚŝŽŶĂŶĚƌĞůĂƚĞĚĂĐƚŝǀŝƚŝĞƐ

other i

Funds, trusts, other financial vehicles

^ĞĐƵƌŝƚŝĞƐ͕ĐŽŵŵŽĚŝƚLJĐŽŶƚƌĂĐƚƐ͕ŽƚŚĞƌŝŶǀĞƐƚŵĞŶƚƐ͖&ƵŶĚƐ͕ƚƌƵƐƚƐ͕ŽƚŚĞƌĨŝŶĂŶĐŝĂůǀĞŚŝĐůĞƐ

Insurance

carriers and related acti

/ŶƐƵƌĂŶĐĞĐĂƌƌŝĞƌƐĂŶĚƌĞůĂƚĞĚĂĐƚŝǀŝƚŝĞƐ

Real

estate

ZĞĂůĞƐƚĂƚĞ

Rental

and lea: g services; Lessors of nonfinancial intangible assets

ZĞŶƚĂůĂŶĚůĞĂƐŝŶŐƐĞƌǀŝĐĞƐ͖>ĞƐƐŽƌƐŽĨŶŽŶĨŝŶĂŶĐŝĂůŝŶƚĂŶŐŝďůĞĂƐƐĞƚƐ

Professional,

scientific, and technical services

WƌŽĨĞƐƐŝŽŶĂů͕ƐĐŝĞŶƚŝĨŝĐ͕ĂŶĚƚĞĐŚŶŝĐĂůƐĞƌǀŝĐĞƐ

Management

of companies and enterprises

DĂŶĂŐĞŵĞŶƚŽĨĐŽŵƉĂŶŝĞƐĂŶĚĞŶƚĞƌƉƌŝƐĞƐ

Administrative

and support services

ĚŵŝŶŝƐƚƌĂƚŝǀĞĂŶĚƐƵƉƉŽƌƚƐĞƌǀŝĐĞƐ

App.184

App.184

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Mining

(except oil and gas)

DŝŶŝŶŐ;ĞdžĐĞƉƚŽŝůĂŶĚŐĂƐͿ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

0.000

Ϭ͘ϬϬϬ

0.000

Ϭ͘ϬϬϬ

Oil

and gas extraction

KŝůĂŶĚŐĂƐĞdžƚƌĂĐƚŝŽŶ

0.000

Ϭ͘ϬϬϬ

2019

ϮϬϭϵ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

2018

ϮϬϭϴ

Forecast

&ŽƌĞĐĂƐƚ

Westmoreland

Shutdown Postland2

tĞƐƚŵŽƌĞůĂŶĚ^ŚƵƚĚŽǁŶWŽƐƚϭĂŶĚϮ

0.000

Ϭ͘ϬϬϬ

Comparison

Type

ŽŵƉĂƌŝƐŽŶdLJƉĞ

Differences

ŝĨĨĞƌĞŶĐĞƐ

Support

activities for agriculture and forestry

^ƵƉƉŽƌƚĂĐƚŝǀŝƚŝĞƐĨŽƌĂŐƌŝĐƵůƚƵƌĞĂŶĚĨŽƌĞƐƚƌLJ

Units

hŶŝƚƐ

Employment

ŵƉůŽLJŵĞŶƚ

Individuals

(Jobs)

/ŶĚŝǀŝĚƵĂůƐ;:ŽďƐͿ

Industries

/ŶĚƵƐƚƌŝĞƐ

Region

ZĞŐŝŽŶ

Forestry

and Logging; Fishing, hunting and trapping

&ŽƌĞƐƚƌLJĂŶĚ>ŽŐŐŝŶŐ͖&ŝƐŚŝŶŐ͕ŚƵŶƚŝŶŐĂŶĚƚƌĂƉƉŝŶŐ

Montana

DŽŶƚĂŶĂ

Employment

by Industry

ŵƉůŽLJŵĞŶƚďLJ/ŶĚƵƐƚƌLJ

Comparison

Forecast

ŽŵƉĂƌŝƐŽŶ&ŽƌĞĐĂƐƚ

2020

ϮϬϮϬ

-25.980

ͲϮϱ͘ϵϴϬ

-0.032

ͲϬ͘ϬϯϮ

-33.489

Ͳϯϯ͘ϰϴϵ

-8.202

Ͳϴ͘ϮϬϮ

-46.357

Ͳϰϲ͘ϯϱϳ

-1,159

Ͳϭ͘ϭϱϵ

-4.124

Ͳϰ͘ϭϮϰ

-0.659

ͲϬ͘ϲϱϵ

-0.645

ͲϬ͘ϲϰϱ

-0,.228

ͲϬ͘ϮϮϴ

-0.335

ͲϬ͘ϯϯϱ

-0.487

ͲϬ͘ϰϴϳ

-0.020

ͲϬ͘ϬϮϬ

-0.838

ͲϬ͘ϴϯϴ

-1.333

Ͳϭ͘ϯϯϯ

-0.418

ͲϬ͘ϰϭϴ

-6.468

Ͳϲ͘ϰϲϴ

-0.657

ͲϬ͘ϲϱϳ

-9.375

Ͳϵ͘ϯϳϱ

-0.011

ͲϬ͘Ϭϭϭ

5.505

ϱ͘ϱϬϱ

-0.707

ͲϬ͘ϳϬϳ

-150.633}

ͲϭϱϬ͘ϲϯϯ

-25.743]

ͲϮϱ͘ϳϰϯ

-0.461

ͲϬ͘ϰϲϭ

-0.109

ͲϬ͘ϭϬϵ

-0.579

ͲϬ͘ϱϳϵ

-0.234

ͲϬ͘Ϯϯϰ

-0.008

ͲϬ͘ϬϬϴ

-0.432

ͲϬ͘ϰϯϮ

-0.286

ͲϬ͘Ϯϴϲ

-1.756

Ͳϭ͘ϳϱϲ

-2.041

ͲϮ͘Ϭϰϭ

-0.318

ͲϬ͘ϯϭϴ

-1.003

Ͳϭ͘ϬϬϯ

-0.019

ͲϬ͘Ϭϭϵ

-0.172

ͲϬ͘ϭϳϮ

-0.001

ͲϬ͘ϬϬϭ

-0.201

ͲϬ͘ϮϬϭ

0.224

ͲϬ͘ϮϮϰ

-0.839

ͲϬ͘ϴϯϵ

-0.047

ͲϬ͘Ϭϰϳ

-0.556

ͲϬ͘ϱϱϲ

-1.491

Ͳϭ͘ϰϵϭ

-256.458}

ͲϮϱϲ͘ϰϱϴ

5,104

Ͳϱ͘ϭϬϰ

-4.164

Ͳϰ͘ϭϲϰ

-378.541|

Ͳϯϳϴ͘ϱϰϭ

-1.039

Ͳϭ͘Ϭϯϵ

-0.240

ͲϬ͘ϮϰϬ

-26.893

ͲϮϲ͘ϴϵϯ

0.028

Ϭ͘ϬϮϴ

-36.880

Ͳϯϲ͘ϴϴϬ

-8.337

Ͳϴ͘ϯϯϳ

-50.033

ͲϱϬ͘Ϭϯϯ

-1.047

Ͳϭ͘Ϭϰϳ

-3.792

Ͳϯ͘ϳϵϮ

0.142

Ϭ͘ϭϰϮ

-0.506

ͲϬ͘ϱϬϲ

-0.237

ͲϬ͘Ϯϯϳ

-0.353

ͲϬ͘ϯϱϯ

-0.478

ͲϬ͘ϰϳϴ

0.040

Ϭ͘ϬϰϬ

-0.832

ͲϬ͘ϴϯϮ

-1.148

Ͳϭ͘ϭϰϴ

-0.164

ͲϬ͘ϭϲϰ

-6.566

Ͳϲ͘ϱϲϲ

-0.467

ͲϬ͘ϰϲϳ

-9.663

Ͳϵ͘ϲϲϯ

-0.010

ͲϬ͘ϬϭϬ

5.686

ϱ͘ϲϴϲ

-0.696

ͲϬ͘ϲϵϲ

-159.442

Ͳϭϱϵ͘ϰϰϮ

-26.137

ͲϮϲ͘ϭϯϳ

-0.478

ͲϬ͘ϰϳϴ

-0.094

ͲϬ͘Ϭϵϰ

-0.515

ͲϬ͘ϱϭϱ

-0.207

ͲϬ͘ϮϬϳ

-0.008

ͲϬ͘ϬϬϴ

-0.362

ͲϬ͘ϯϲϮ

-0.271

ͲϬ͘Ϯϳϭ

-2.026

ͲϮ͘ϬϮϲ

-2.118

ͲϮ͘ϭϭϴ

-0.257

ͲϬ͘Ϯϱϳ

-1.060

Ͳϭ͘ϬϲϬ

-0.019

ͲϬ͘Ϭϭϵ

-0.181

ͲϬ͘ϭϴϭ

0.002

Ϭ͘ϬϬϮ

-0.157

ͲϬ͘ϭϱϳ

-0.226

ͲϬ͘ϮϮϲ

-0.880

ͲϬ͘ϴϴϬ

-0.047

ͲϬ͘Ϭϰϳ

-0.577

ͲϬ͘ϱϳϳ

-1.646

Ͳϭ͘ϲϰϲ

-310.170

ͲϯϭϬ͘ϭϳϬ

-4.468

Ͳϰ͘ϰϲϴ

-4.301

Ͳϰ͘ϯϬϭ

-377.334

Ͳϯϳϳ͘ϯϯϰ

-0.265

ͲϬ͘Ϯϲϱ

-0.024

ͲϬ͘ϬϮϰ

-0.210

ͲϬ͘ϮϭϬ

2021

ϮϬϮϭ

Year

zĞĂƌ

-0.270

ͲϬ͘ϮϳϬ

Standard

Re

nal Control

^ƚĂŶĚĂƌĚZĞŐŝŽŶĂůŽŶƚƌŽů

1.174

ϭ͘ϭϳϰ

0.418

Ϭ͘ϰϭϴ

-0.072

ͲϬ͘ϬϳϮ

2023

ϮϬϮϯ

1.730

ϭ͘ϳϯϬ

0.601

Ϭ͘ϲϬϭ

-0.009

ͲϬ͘ϬϬϵ

2024

ϮϬϮϰ

-27.249

ͲϮϳ͘Ϯϰϵ

0.086

Ϭ͘Ϭϴϲ

-38.132

Ͳϯϴ͘ϭϯϮ

-8.152

Ͳϴ͘ϭϱϮ

-53.676

Ͳϱϯ͘ϲϳϲ

-0.906

ͲϬ͘ϵϬϲ

-3.384

Ͳϯ͘ϯϴϰ

0.888

Ϭ͘ϴϴϴ

-0.370

ͲϬ͘ϯϳϬ

-0.240

ͲϬ͘ϮϰϬ

-0.364

ͲϬ͘ϯϲϰ

-0.465

ͲϬ͘ϰϲϱ

0.095

Ϭ͘Ϭϵϱ

-0.808

ͲϬ͘ϴϬϴ

-26.726

ͲϮϲ͘ϳϮϲ

0.136

Ϭ͘ϭϯϲ

-37.945

Ͳϯϳ͘ϵϰϱ

-7.820

Ͳϳ͘ϴϮϬ

-53.851

Ͳϱϯ͘ϴϱϭ

-0.749

ͲϬ͘ϳϰϵ

-2.943

ͲϮ͘ϵϰϯ

1.536

ϭ͘ϱϯϲ

-0.247

ͲϬ͘Ϯϰϳ

-0.236

ͲϬ͘Ϯϯϲ

-0.367

ͲϬ͘ϯϲϳ

-0.438

ͲϬ͘ϰϯϴ

0.142

Ϭ͘ϭϰϮ

-0.769

ͲϬ͘ϳϲϵ

0.288

Ϭ͘Ϯϴϴ

-0.733

ͲϬ͘ϳϯϯ

0.077

Ϭ͘Ϭϳϳ

-6.307

Ͳϲ͘ϯϬϳ

-0.059

ͲϬ͘Ϭϱϵ

-8.820

Ͳϴ͘ϴϮϬ

-0.009

ͲϬ͘ϬϬϵ

5.970

ϱ͘ϵϳϬ

-0.624

ͲϬ͘ϲϮϰ

-164.088

Ͳϭϲϰ͘Ϭϴϴ

-24.482

ͲϮϰ͘ϰϴϮ

-0.455

ͲϬ͘ϰϱϱ

-0.060

ͲϬ͘ϬϲϬ

-0.351

ͲϬ͘ϯϱϭ

-0.148

ͲϬ͘ϭϰϴ

-0.008

ͲϬ͘ϬϬϴ

-0.239

ͲϬ͘Ϯϯϵ

-0.238

ͲϬ͘Ϯϯϴ

-2.383

ͲϮ͘ϯϴϯ

-2.105

ͲϮ͘ϭϬϱ

-0.119

ͲϬ͘ϭϭϵ

-1.066

Ͳϭ͘Ϭϲϲ

-0.016

ͲϬ͘Ϭϭϲ

-0.175

ͲϬ͘ϭϳϱ

0.007

Ϭ͘ϬϬϳ

-0.031

ͲϬ͘Ϭϯϭ

-0,194

ͲϬ͘ϭϵϰ

-0.782

ͲϬ͘ϳϴϮ

-0.040

ͲϬ͘ϬϰϬ

-0.472

ͲϬ͘ϰϳϮ

-1.467

Ͳϭ͘ϰϲϳ

-300.673

ͲϯϬϬ͘ϲϳϯ

-3.602

Ͳϯ͘ϲϬϮ

-3.428

Ͳϯ͘ϰϮϴ

-0.944

ͲϬ͘ϵϰϰ

-6.516

Ͳϲ͘ϱϭϲ

-0.263

ͲϬ͘Ϯϲϯ

-9.394

Ͳϵ͘ϯϵϰ

-0.010

ͲϬ͘ϬϭϬ

5.846

ϱ͘ϴϰϲ

-0.669

ͲϬ͘ϲϲϵ

-164.580

Ͳϭϲϰ͘ϱϴϬ

-25.653

ͲϮϱ͘ϲϱϯ

-0.474

ͲϬ͘ϰϳϰ

-0.078

ͲϬ͘Ϭϳϴ

-0.435

ͲϬ͘ϰϯϱ

-0.179

ͲϬ͘ϭϳϵ

-0.008

ͲϬ͘ϬϬϴ

-0.301

ͲϬ͘ϯϬϭ

-0.257

ͲϬ͘Ϯϱϳ

-2.245

ͲϮ͘Ϯϰϱ

-2.143

ͲϮ͘ϭϰϯ

-0.191

ͲϬ͘ϭϵϭ

-1.087

Ͳϭ͘Ϭϴϳ

-0.018

ͲϬ͘Ϭϭϴ

-0.181

ͲϬ͘ϭϴϭ

0.005

Ϭ͘ϬϬϱ

-0.097

ͲϬ͘Ϭϵϳ

-0.214

ͲϬ͘Ϯϭϰ

-0.853

ͲϬ͘ϴϱϯ

-0.045

ͲϬ͘Ϭϰϱ

-0.541

ͲϬ͘ϱϰϭ

-1.616

Ͳϭ͘ϲϭϲ

-318.215

Ͳϯϭϴ͘Ϯϭϱ

-4.031

Ͳϰ͘Ϭϯϭ

-3.971

Ͳϯ͘ϵϳϭ

-25.791

ͲϮϱ͘ϳϵϭ

0.176

Ϭ͘ϭϳϲ

-37.025

Ͳϯϳ͘ϬϮϱ

-7.430

Ͳϳ͘ϰϯϬ

-52.199

ͲϱϮ͘ϭϵϵ

-0.602

ͲϬ͘ϲϬϮ

-2.535

ͲϮ͘ϱϯϱ

2.057

Ϯ͘Ϭϱϳ

-0.148

ͲϬ͘ϭϰϴ

-0.230

ͲϬ͘ϮϯϬ

-0.366

ͲϬ͘ϯϲϲ

-0.407

ͲϬ͘ϰϬϳ

0.179

Ϭ͘ϭϳϵ

-0.726

ͲϬ͘ϳϮϲ

-0.539

ͲϬ͘ϱϯϵ

0.465

Ϭ͘ϰϲϱ

-6.040

Ͳϲ͘ϬϰϬ

0.122

Ϭ͘ϭϮϮ

-8.133

Ͳϴ͘ϭϯϯ

-0,009

ͲϬ͘ϬϬϵ

6.042

ϲ͘ϬϰϮ

-0.576

ͲϬ͘ϱϳϲ

-160.762

ͲϭϲϬ͘ϳϲϮ

-23.056

ͲϮϯ͘Ϭϱϲ

-0.430

ͲϬ͘ϰϯϬ

-0.044

ͲϬ͘Ϭϰϰ

-0.274

ͲϬ͘Ϯϳϰ

-0.121

ͲϬ͘ϭϮϭ

-0.007

ͲϬ͘ϬϬϳ

-0.187

ͲϬ͘ϭϴϳ

-0.219

ͲϬ͘Ϯϭϵ

-2.470

ͲϮ͘ϰϳϬ

-2.041

ͲϮ͘Ϭϰϭ

-0.053

ͲϬ͘Ϭϱϯ

-1.028

Ͳϭ͘ϬϮϴ

-0.014

ͲϬ͘Ϭϭϰ

-0.167

ͲϬ͘ϭϲϳ

0.010

Ϭ͘ϬϭϬ

0.028

Ϭ͘ϬϮϴ

-0.174

ͲϬ͘ϭϳϰ

-0.695

ͲϬ͘ϲϵϱ

-0.035

ͲϬ͘Ϭϯϱ

-0.392

ͲϬ͘ϯϵϮ

-1,268

Ͳϭ͘Ϯϲϴ

-270.693

ͲϮϳϬ͘ϲϵϯ

3,225

Ͳϯ͘ϮϮϱ

-2.814

ͲϮ͘ϴϭϰ

-365.958

Ͳϯϲϱ͘ϵϱϴ -355,.856

Ͳϯϱϱ͘ϴϱϲ -346.316

Ͳϯϰϲ͘ϯϭϲ

0.497

Ϭ͘ϰϵϳ

0.204

Ϭ͘ϮϬϰ

-0.142

ͲϬ͘ϭϰϮ

2022

ϮϬϮϮ

-24.735

ͲϮϰ͘ϳϯϱ

0.205

Ϭ͘ϮϬϱ

-35.823

Ͳϯϱ͘ϴϮϯ

-7.036

Ͳϳ͘Ϭϯϲ

-49.785

Ͳϰϵ͘ϳϴϱ

-0.482

ͲϬ͘ϰϴϮ

-2.189

ͲϮ͘ϭϴϵ

2.444

Ϯ͘ϰϰϰ

-0.078

ͲϬ͘Ϭϳϴ

0.224

ͲϬ͘ϮϮϰ

-0.363

ͲϬ͘ϯϲϯ

-0.376

ͲϬ͘ϯϳϲ

0.206

Ϭ͘ϮϬϲ

-0.685

ͲϬ͘ϲϴϱ

-0.375

ͲϬ͘ϯϳϱ

0.600

Ϭ͘ϲϬϬ

-5.765

Ͳϱ͘ϳϲϱ

0.270

Ϭ͘ϮϳϬ

-7.445

Ͳϳ͘ϰϰϱ

-0.008

ͲϬ͘ϬϬϴ

6.055

ϲ͘Ϭϱϱ

-0.528

ͲϬ͘ϱϮϴ

-156.172

Ͳϭϱϲ͘ϭϳϮ

-21.619

ͲϮϭ͘ϲϭϵ

-0.403

ͲϬ͘ϰϬϯ

-0.029

ͲϬ͘ϬϮϵ

-0.207

ͲϬ͘ϮϬϳ

-0,098

ͲϬ͘Ϭϵϴ

-0.007

ͲϬ͘ϬϬϳ

-0.145

ͲϬ͘ϭϰϱ

-0.202

ͲϬ͘ϮϬϮ

-2,519

ͲϮ͘ϱϭϵ

-1.973

Ͳϭ͘ϵϳϯ

0.002

Ϭ͘ϬϬϮ

-0.982

ͲϬ͘ϵϴϮ

-0.012

ͲϬ͘ϬϭϮ

-0.157

ͲϬ͘ϭϱϳ

0.011

Ϭ͘Ϭϭϭ

0.077

Ϭ͘Ϭϳϳ

-0.155

ͲϬ͘ϭϱϱ

-0.607

ͲϬ͘ϲϬϳ

-0.030

ͲϬ͘ϬϯϬ

-0.314

ͲϬ͘ϯϭϰ

-1,061

Ͳϭ͘Ϭϲϭ

-236.844

ͲϮϯϲ͘ϴϰϰ

-2.924

ͲϮ͘ϵϮϰ

-2.205

ͲϮ͘ϮϬϱ

-337.089

Ͳϯϯϳ͘Ϭϴϵ

2.146

Ϯ͘ϭϰϲ

0.744

Ϭ͘ϳϰϰ

0.042

Ϭ͘ϬϰϮ

2025

ϮϬϮϱ

-23.724

ͲϮϯ͘ϳϮϰ

0.223

Ϭ͘ϮϮϯ

-34.630

Ͳϯϰ͘ϲϯϬ

-6.675

Ͳϲ͘ϲϳϱ

-47.231

Ͳϰϳ͘Ϯϯϭ

-0.392

ͲϬ͘ϯϵϮ

-1.911

Ͳϭ͘ϵϭϭ

2.708

Ϯ͘ϳϬϴ

-0.034

ͲϬ͘Ϭϯϰ

-0,.219

ͲϬ͘Ϯϭϵ

-0.361

ͲϬ͘ϯϲϭ

-0.348

ͲϬ͘ϯϰϴ

0.224|

Ϭ͘ϮϮϰ

-0.649

ͲϬ͘ϲϰϵ

-0.245

ͲϬ͘Ϯϰϱ

0.690

Ϭ͘ϲϵϬ

-5.509

Ͳϱ͘ϱϬϵ

0.383

Ϭ͘ϯϴϯ

-6.817

Ͳϲ͘ϴϭϳ

-0.007

ͲϬ͘ϬϬϳ

6.018

ϲ͘Ϭϭϴ

-0.485

ͲϬ͘ϰϴϱ

-150.980

ͲϭϱϬ͘ϵϴϬ

-20.295

ͲϮϬ͘Ϯϵϱ

-0.378

ͲϬ͘ϯϳϴ

-0.017

ͲϬ͘Ϭϭϳ

-0.153

ͲϬ͘ϭϱϯ

-0.081

ͲϬ͘Ϭϴϭ

-0.007

ͲϬ͘ϬϬϳ

-0.112

ͲϬ͘ϭϭϮ

-0.186

ͲϬ͘ϭϴϲ

-2,538

ͲϮ͘ϱϯϴ

-1.909

Ͳϭ͘ϵϬϵ

0.045

Ϭ͘Ϭϰϱ

-

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.

A word about cookies

We need a few to keep you signed in and the library working. The rest help us see which pages people use and where they get stuck. They stay off unless you say yes.

Emergency Application — Westmoreland Mining Holdings LLC, et al., Applicants v. Environmental Protection Agency, et al. | Frix