Amicus Curiae Brief — Fed. Energy Regulatory Comm'n v. Elec. Power Supply Ass'n, 135 S. Ct. 2049 (2015) (No. 14-840)

. ne ~~

FILED

= 2013

IN THE OFFICE OF THE CLES”

Supreme Court of the Anited States

Nos. 14-840, 14-841

FEDERAL ENERGY REGULATORY COMMISSION,

Petitioner,

v.

ELECTRIC POWER SUPPLY ASSOCIATION, ET AL.,

Respondents.

ENERNOC, INC., ET AL.,

Petitioners,

v.

ELECTRIC POWER SUPPLY ASSOCIATION, ET AL.,

Respondents.

On Writs of Certiorari to the

United States Court of Appeals

for the District of Columbia Circuit

BRIEF OF ROBERT L. BORLICK, JOSEPH

BOWRING, JAMES BUSHNELL, AND 19 OTHER

LEADING ECONOMISTS AS AMICI CURIAE IN

SUPPORT OF RESPONDENTS

JOHN LEE SHEPHERD, JR. JEFFREY A. LAMKEN

SKADDEN, ARPS, SLATE, Counsel of Record

MEAGHER & FLOM LLP MOLOLAMKEN LLP

1440 New York Ave., NW The Watergate, Suite 660

Washington, D.C. 20005 600 New “oe Ave., NW

(202) 371-7338 W C. 20037

(202)

Counsel for Amici Curiae

WILSON-EPES PRINTING Co., INC. — (202) 789-0096 — WasnincTon, D.C. 20002

QUESTIONS PRESENTED

The questions presented for the Court’s review in this

case are:

1. Whether the Federal Power Act, 16 U.S.C. §§791a

et seq., authorizes the Federal Energy Regulatory Com-

mission (FERC) to induce retail reductions in electricity

consumption and to recoup those payments through ad-

justments to wholesale rates.

2. Whether FERC’s rule establishing the amount

paid to induce those reductions in electricity consumption

is arbitrary and capricious or otherwise unsupported by

reasoned decisionmaking.

This brief of amici curiae addresses only the second

question presented.

(i)

TABLE OF CONTENTS

I. FERC’s Rule Imposes Inefficient

Distortions Without Reasoned

A. The Need for Proper Price Signals

To Regulate Supply and Demand in

Electricity Markets .....................cccec-esce00es

1. The Critical Role of Price in

2. The Wholesale-Retail Disconnect

3. Aligning Wholesale and Retail

) Vantasutatee Subsidy That

1. Paying LMP Without Accounting

for Input Cost Is Inconsistent

with Basic Economics ...................-+.-+-

2. FERC’s Rule Creates Irrational

Incentives To Cease or Defer

Socially Valuable Production.............

3. The Rule Offers Arbitrary and

Uneconomic Rewards Based on

the Meter’s Location. .............c0ccs00-.

(iii)

iv

TABLE OF CONTENTS—Continued

II. Efficient Demand-Response Mechanisms

Replicate the Market .................c:sssceceseseereees

III. Contrary Arguments Are Unfounded..........

A. FERC’s Own Arguments Show

That Paying Full LMP Effects

an Unjustified Subsidy ...............ccs-0000s0-.

B. Reliance on the Net-Benefits Test

Underscores the Irrationality of

FERC’s Approach.............csc-scsceeceseesreseees

C. Petitioner and Its Amici Seek

Subsidies Not Supported by the

(| TAEeseeeere e e

Addendum — List of Amici CUuriae...........ccccccceeeeee

v

TABLE OF AUTHORITIES .

Page(s)

CASES

Delaware Dep't of Natural Res. v. EPA,

No. 13-1093, slip op. (D.C. Cir. May 1,

it hcshjeineinantinnisnashpianaeaaiaaidaiitiinseinmnsees 20

FPC v. Conway Corp., 426 U.S. 271 (1976)......... 24

FPC v. Hope Natural Gas Co.,

8 30

Morgan Stanley Capital Grp. Inc. v. Pub.

Util. Dist. No. 1, 554 U.S. 527 (2008) .............. 30

Nat'l Soc’y of Prof'l Eng’rs v. United

States, 435 U.S. 679 (1978) .........cccccccsee sceeseerees 7

New York v. FERC, 535 U.S. 1 (2002) ................. 7

Permian Basin Area Rate Cases,

BOO U.S. TET (19GB) ..ccccececcevesscscssseccsccsosscscceccsosses 30

Sacramento Mun. Util. Dist. v. FERC,

616 F.3d 520 (D.C. Cir. 2010) 20.0.0... ccccceeeseeees 8

SEC v. Chenery Corp., 332 U.S. 194 (1947)......... 31

ADMINISTRATIVE MATERIALS

kf ee 10

Cal. Indep. Sys. Operator Corp., 116

FERC 961,274 (2006) .........cc-cccecessesecsessersereesenes 8

Md. Pub. Serv. Comm’n v. PJM

Interconnection, L.L.C., 127 FERC

ice alicmreernntccams 8

Milford Power Co., LLC, 119 FERC

a 29

PJM Indus. Customer Coal. v. PJM

Interconnection L.L.C., 121 FERC

eB cccentcsscesseccssannesemnnsennnecnmamenaocsaces 22, 23

vi

TABLE OF AUTHORITIES—Continued

Page(s)

PJM Interconnection, L.L.C., 99 FERC

, SENT NR Sao ee 22

OTHER AUTHORITIES

Borenstein, Time-Varying Retail

Electricity Prices: Theory and

Practice, in E Deregulation:

Choices and Challenges 317 (Griffin &

Pe hs i attitiecarnieinirentnsnenniemantmerernccee 8,9

Borlick, Paying for Demand-Side Response

at the Wholesale Level: The Small

Consumers’ Perspective, 24 Elec. J. 8

Borlick, Pricing Negawatts, 148 Pub.

Utils. Fort. 14 (2010)...................0000- 9 12,21, 26

Chao, Price-Responsive Demand

Management for a Smart Grid World,

ee 29

Cicchetti & Hogan, Including Unbundled

Demand-Side Options in Electric

Utility Bidding Programs, Pub. Utils.

i REET 30

Electric Energy Market Competition Task

Force, Report to Congress on

Competition in Wholesale and Retail

Markets for Electric Energy (2007) ..............+- 7

Faruqui & Palmer, Dynamic Pricing and

Its Discontents, 34 Regulation 16 (2011)........ 23

Griffin & Puller, /ntroduction to Electricity

Deregulation: Choices & Challenges

(Griffin & Puller eds., 2005) ...............ccccccceeseeees 4

vii

TABLE OF AUTHORITIES—Continued

Page(s)

Hilke & Wise, Who Turned Out the Lights?

Competition and California’s Power

Crisis, 15 Antitrust 76 (2001) ..............ccseeeeeee 9

Hogan, Providing Efficient Incentives for

Demand Response (Oct. 29, 2009),

http-//elibrary ferc.gov/idmws/common/

OpenNat.asp?fileID = 12228734

SERRE SC ee 9, 21, 33

Newell, DR Distortion, 148 Pub. Utils.

| See 19, 21, 33

IN THE

Supreme Court of the Anited States

Nos. 14-840, 14-841

FEDERAL ENERGY REGULATORY COMMISSION,

Petitioner,

V.

ELECTRIC POWER SUPPLY ASSOCIATION, ET AL.,

Respondents.

ENERNOC, INC., ET AL.,

Petitioners,

Vv.

ELECTRIC POWER SUPPLY ASSOCIATION, ET AL.,

Respondents.

On Writs of Certiorari to the

United States Court of Appeals

for the District of Columbia Circuit

BRIEF OF ROBERT L. BORLICK, JOSEPH

BOWRING, JAMES BUSHNELL, AND 19 OTHER

LEADING ECONOMISTS AS AMICI CURIAE IN

SUPPORT OF RESPONDENTS

INTEREST OF AMICI CURIAE

Amici curiae (listed in the Addendum) are leading

economists and educators who have designed, studied,

and written about the electricity markets affected by the

2

decisions under review here.’ FERC’s Final Rule estab-

lishes the rate paid for reducing purchases of electricity

(called “demand response”). FERC now requires market

participants to pay the full “locational marginal price” of

electricity that is not consumed, treating non-consump-

tion as equivalent to selling energy one owns or gener-

ates. See U.S. Br. 45-61.

Although amici’s views may diverge on other issues,

they agree that FERC’s decision cannot be reconciled

with the economic principles FERC has repeatedly en-

dorsed. FERC’s pricing methodology overcompensates

reduced consumption by compensating demand response

as if parties that reduce consumption sell electricity they

generate, or resell into the market energy they pur-

chased. But demand-responders do neither: They de-

cline to buy electricity at a particular rate; they do not

provide electricity they own to others. Requiring that

’ Pursuant to this Court’s Rule 37.6, counsel for amici curiae states

that no counsel for a party authored this brief in whole or part or

made a monetary contribution to the preparation and submission of

this brief. Counsel’s fees and expenses incurred to prepare this brief

were paid by American Electric Power Service Corp., the Independ-

ent Power Producers of New York, Inc. and the New England Power

Generators Association, Inc. No other person or entity made a mon-

etary contribution intended to fund the preparation or submission of

this brief. Robert Borlick, William Hogan, and Roy Shanker were

compensated experts in FERC proceedings below. James Bushnell,

Seott Harvey, and Benjamin Hobbs were compensated as members

of the Market Surveillance Committee of the California Independent

System Operator, Inc., which filed comments before FERC. Paul

Centolella was a Commissioner on the Ohio Public Utilities Commis-

sion and, as part of his official duties, participated in submitting

comments before FERC. All parties consented to the filing of this

brief. Copies of the letters granting consent have been filed with the

Clerk.

3

demand-responders be paid as if they contributed and

sold their own electricity into the market creates exces-

sive incentives to curtail demand. It induces the cessa-

tion of productive activity even when that harms social

welfare. And it creates bizarre incentives by compensat-

ing those who operate their own (typically dirtier and

more costly) generation “behind the meter” for self-

consumption more generously than generators who sup-

ply electricity to the interstate grid for everyone’s use.

FERC’s methodology harms social welfare and the long-

term health of electricity markets. If FERC has legiti-

mate reasons for imposing those costs, it never articulat-

ed them.

SUMMARY OF ARGUMENT

I.A. FERC has regulated wholesale electricity mar-

kets to promote efficient competition, mandating the use

of a locational marginal price (LMP) that incorporates all

costs of delivering the electricity necessary to meet real-

time demand at each location. However, because end-use

consumers in state-regulated retail markets often pay

fixed rates, their energy consumption may be relatively

inelastic, regardless of real-time changes in wholesale

prices.

To make retail consumption more responsive to whole-

sale price changes, FERC mandated “demand-response”

payments of full LMP to customers for not consuming

electricity. Although FERC attempted to justify pay-

ment of full LMP on efficiency grounds, its analysis de-

fies the economic principles FERC purported to follow.

B. Under FERC’s rule, everyone actually supplying

electricity to market at LMP must bear the cost of ob-

taining the electricity they sell. Generators must pro-

duce energy, and traders must buy it, to sell it at LMP.

But not demand-responders. They are paid full LMP, as

4

if they “sold” electricity into the market, even though

they do not own (or buy) electricity and supply it to mar-

ket. They simply decline to consume electricity.

Basic math shows that paying full LMP encourages

purchasers to forgo electricity consumption and produc-

tive activity even when it harms overall social welfare.

FERC previously employed that math to conclude that

paying full LMP would be an unjustified subsidy for de-

mand response, as did the Federal Trade Commission in

these proceedings.

FERC’s rule also compensates users more handsome-

ly for generating electricity solely for their own “behind-

the-meter” use, discriminating against generators that

offer electricity into the grid. That favors less efficient

self-supply over public generation—even when self-

supply costs more than prevailing LMP FERC never

attempted to justify that result.

II. FERC offered no economically plausible rationale

for rejecting the more efficient alternative of LMP-

minus-G (where G is the full retail price). FERC had

previously approved that workable formula for compen-

sating demand-responders.

III. FERC relies on the “net-benefits test,” which re-

quires RTOs to perform a complex calculation using “his-

torical data” and an indeterminate list of factors. The

necessity of superimposing a net-benefits test demon-

strates that FERC’s pricing rule is flawed. If paying full

LMP produced appropriate pricing signals, FERC would

not need to superimpose another test as a precondition

for demand response.

By its terms, moreover, the test merely precludes de-

mand-response payments unless those payments lower

the total amount consumers pay for electricity. But arti-

+)

ficially lowering prices can impede efficiency and deter

necessary investment in new facilities. This Court and

FERC have recognized that pricing must look to the in-

terests of the market as a whole, including generators

and customers. Any economist seeking to maximize ben-

efits to society will measure total social welfare, not just

the consumer side of the ledger. FERC never justified

its exclusive focus on lower short-term prices and point-

edly ignored the harm to social welfare its pricing

scheme imposes. Regulators sometimes depart from ef-

ficiency to pursue other goals. But damaging efficiency

while purporting to pursue it is arbitrary and capricious

rather than reasoned decisionmaking.

ARGUMENT

FERC’s Final Rule defies the basic economic princi-

ples upon which FERC established organized electricity

markets. Declaring that it “is not limited to textbook

economic analysis,” FERC urged that it “may account for

the practical realities of how [electricity] markets oper-

ate.” Pet. App. 89a-90a. But FERC’s decision defies

both economic theory and the practical realities alike.

FERC’s formula for compensating decisions to reduce

retail consumption exceeds what a competitive market

would provide; systematically provides greater incentives

to reduce consumption than to generate energy for useful

purposes; and creates perverse incentives that injure

consumers in the long run. Notwithstanding the many

post-hoc rationalizations now asserted for that rule,

FERC gave no rationale that makes theoretical or prac-

tical economic sense.

Amici find FERC’s order disturbing. FERC purports

to promote efficiency but instead undermines efficiency

by creating obvious subsidies while denying that any

subsidy exists. Regulators may sometimes depart from

6

efficiency to pursue other legitimate goals. Here, howev-

er, FERC departed from efficiency while purporting to

pursue it, without acknowledging the costs of that depar-

ture.

I. FERC’s RULE IMPOSES INEFFICIENT DISTORTIONS

WITHOUT REASONED JUSTIFICATION

There is a well-acknowledged disconnect between

wholesale and retail energy markets. Wholesale prices

for electricity, which are set in competitive, organized

markets, reflect minute-to-minute fluctuations in demand

and supply. But the retail rates paid by electricity con-

sumers are often fixed in advance; purchasers pay a flat

rate for each additional megawatt consumed regardless

of real-time changes in supply and demand. As a result,

real-time price signals are not transmitted to retail elec-

tricity consumers, whose demand remains highly inelas-

tic to fluctuations in wholesale prices.

Here, FERC attempted to give retail customers incen-

tives to curtail usage in response to wholesale price

changes. But FERC created incentives that depart dra-

matically from proper competitive outcomes and any sen-

sible notion of economic efficiency. By overcompensating

reductions in demand, FERC discourages socially valua-

ble, productive activity, without acknowledging that con-

sequence. It likewise compensates self-generation more

handsomely than generation for the public’s benefit. At

bottom, FERC has chosen to subsidize demand response

without offering economic reasoning for that subsidy—or

the corresponding burden on generators and small con-

sumers who must fund it.

7

A. The Need for Proper Price Signals To Regulate

Supply and Demand in Electricity Markets

1. The Critical Role of Price in Energy Markets

“Price is the ‘central nervous system of the economy.’”

Nat'l Soc’y of Prof’l Eng’rs v. United States, 435 U.S.

679, 692 (1978). “[Tyhe influence of prices on the behavior

of consumers and producers is crucial for how a market

economy allocates scarce resources.” Mankiw, Princi-

ples of Economics 7 (5th ed. 2008). In efficient markets,

rising prices signal the need to increase production (and

trigger decreased consumption); falling prices signal the

need to decrease production (and trigger greater con-

sumption). The resulting buyer and seller conduct “natu-

rally move{s] markets toward the equilibrium of supply

and demand.” /d. at 77.

Those bedrock principles apply to electricity markets.

On the demand side, “[{p}rice changes signal to customers

in wholesale and retail markets that they should change

their decisions about how much and when to consume

electric power.” Electric Energy Market Competition

Task Force, Report to Congress on Competition in

Wholesale and Retail Markets for Electric Energy 47

(2007). Suppliers respond as well. “If the cost of increas-

ing [electricity] production is small, a relatively small

price increase may * * * encourage producers to increase

production in response to increased demand.” /d. at 50.

But if production costs are high, “suppliers will not in-

crease production unless the price increases enough to

cover the higher costs.” [bid.

FERC has recognized the importance of price signals

to balance supply and demand. It restructured organized

wholesale electricity markets to ensure accurate price

signals. See New York v. FERC, 535 US. 1, 45 (2002).

And it has used price signals by establishing auctions for

8

long-term supply. See, eg., Md. Pub. Serv. Comm'n v.

PJM Interconnection, L.L.C., 127 FERC 961,274 (2009).

2. The Wholesale-Retail Disconnect on Pricing

Although electricity markets are subject to the princi-

ples of supply and demand, there is a gulf between

wholesale pricing (regulated by FERC) and retail pricing

(regulated by States). That separation impedes the criti-

cal signaling function of price—an impediment FERC’s

rule here was supposed to overcome.

Wholesale rates are generally efficient. Most whole-

sale electricity in orgenised markets is priced according

to “locational marginal pricing,” or “LMP.” Under LMP,

“prices are designed to reflect the least-cost of meeting

an incremental megawatt-hour of demand at each loca-

tion on the grid, and thus prices vary based on location

and time.” Sacramento Mun. Util. Dist. v. FERC, 616

F.3d 520, 524 (D.C. Cir. 2010). LMP incorporates each

cost component for delivering electricity to its destina-

tion—the cost of generating it, the cost of transmitting it

from the generator, and “line losses” (energy losses that

occur during transport). /bid) LMP “communicate[s]

the true market value of electricity at each location”;

“create[s] financial incentives to dispatch the lowest cost

energy”; and “encourage{s] transmission and generation

investment at appropriate locations.” Cal Indep. Sys.

Operator Corp., 116 FERC 961,274 at 62,126 (2006). And

LMP is dynamic: Because it “reflect{[s] the supply-

demand interaction,” it “varies constantly,” with prices

changing in minute-by-minute increments. Borenstein,

Time-Varying Retail Electricity Prices: Theory and

Practice, in Electricity Deregulation: Choices and Chal-

lenges 317, 317 (Griffin & Puller eds., 2005). Wholesale

market participants thus have an incentive to respond to

9

increased prices, triggered by peak-demand or supply

shortfalls, by reducing purchases.

By contrast, retail rates typically do not reflect real-

time price fluctuations. Borenstein, supra, at 317. Ra-

ther, the retail price “typically is constant for months at a

time.” Ibid. As a result, moment-to-moment or day-to-

day wholesale price signals are not transmitted to most

consumers. The wholesale price during a peak-usage pe-

riod thus might increase dramatically, but retail consum-

ers paying fixed rates have no economic incentive to de-

crease consumption in response. See Pet. App. 96a-98a.

Consumer demand is thus rendered “inelastic” with re-

spect to wholesale price. The lack of incentives to reduce

demand during peak periods was one of the “most obvi-

ous culprit{s]” for the 2000-2001 California energy crisis.

Hilke & Wise, Who Turned Out the Lights? Competition

and California’s Power Crisis, 15 Antitrust 76, 76 (2001).

One way to ensure appropriate incentives is dynamic

pricing—charging retail customers the “relevant real-

time or day-ahead LMP” of the electricity they consume.

Hogan, Providing Efficient Incentives for Demand Re-

sponse 12 (Oct. 29, 2009) (“Whitepaper”), http://elibra-

ry ferc.gov/idmws/common/OpenNat.asp?fileID = 12228734

(Attach. A). Exposing consumers to market-driven, real-

time price fluctuations requires consumers to consider

the real-time cost of consumption—encouraging con-

sumption during low-demand periods, and discouraging it

during peak-demand periods. Borlick, Pricing Nega-

watts, 148 Pub. Utils. Fort. 14, 16 (2010). But real-time

pricing is not always feasible, because of customer or leg-

islative resistance and the costs of real-time metering

equipment. Griffin & Puller, /ntroduction to Electricity

Deregulation: Choices & Challenges 26 (Griffin & Puller

10

eds., 2005). As a result, wholesale and retail prices rarely

converge in real time.

8. Aligning Wholesale and Retail Incentives

Through Demand-Response Payments

To overcome the fact that many retail consumers do

not receive real-time price signals that would inform con-

sumption decisions, see J.A.293-294, regulators have at-

tempted to recreate otherwise missing market incentives

through “demand-response” policies. “Demand response

means a reduction in the consumption of electric energy

by customers from their expected consumption in re-

sponse to an increase in the price of electric energy” or,

more commonly, “incentive payments designed to induce

lower consumption of electric energy.” 18 C.F.R.

§ 35.28(b)(4).

Properly structured incentives can increase social wel-

fare. Excessive incentive payments for demand re-

sponse, however, are problematic. They encourage busi-

nesses not to produce when it would be more efficient—

and socially valuable—for them to continue production.

They encourage large consumers to self-supply, even

when their self-generated electricity is more costly to so-

ciety. And they deter and distort investment in genera-

tion by artificially suppressing consumption and prices.

The key demand-response question thus asks how

much to pay a retail customer to forgo purchasing elec-

tricity at a fixed price when wholesale prices climb. The

obvious answer is to replicate market forces by asking

what a fixed-price customer would earn by purchasing

energy at retail and reselling it at the prevailing whole-

sale market price (7e., LMP). In that case, the custom-

er’s profit would be the difference between the price at

which it sold electricity, and the contract price at which it

could buy electricity. That difference provides an incen-

ll

tive to forgo consumption only when doing so is more

profitable than using the electricity to produce goods and

services (~e., when non-production is more efficient than

production).

Here, FERC instead required that retail customers be

paid the market price for wholesale electricity—the full

LMP—without subtracting the price the customer would

have paid to buy it. See Pet. App. 100a-101a, 210a. The

customer is thus compensated as if it acquired or gener-

ated electricity for free and sold electricity it owned to

others in the market. As explained below, that result is

neither justified nor justifiable.

B. FERC’s Pricing Produces an Unacknowledged

Subsidy That Harms Social Welfare

Throughout the rulemaking, FERC claimed that pay-

ing full LMP for demand response was justified by eco-

nomic efficiency. FERC announced at the outset that

“paying demand response resources the LMP will com-

pensate those resources in a manner that reflects the

marginal value of the resource.” Pet. App. 91a; accord

J.A.35. Doing so, FERC stated, “will improve the com-

petitiveness of the organized wholesale energy markets

and * * * help to ensure that energy prices in those mar-

kets are just and reasonable.” J.A.35.

FERC maintained that focus in its Final Rule. It ex-

plained that pursuing these goals will “balance supply

and demand,” Pet. App. 96a, “caus{ing] wholesale and re-

tail prices to converge on a price level reflecting de-

mand’s ability to respond to the marginal cost of energy,”

Pet. App. 216a. While purporting to pursue economic ef-

ficiency, however, FERC adopted an approach that dis-

torts rather than reflects market forces and fundamental

economic principles, creating implicit subsidies FERC

cannot explicitly justify.

12

1. Paying LMP Without Accounting for Input

Cost Is Inconsistent with Basic Economics

Any demand-response mechanism must account for

the costs the consumer avoids by not purchasing electric-

ity. In a competitive market, a consumer who receives

electricity at a favorable rate might resell that electricity

instead of using it. But the customer must buy electricity

to resell it. The consumer who forgoes using electricity,

and makes that electricity available to others for a price,

thus receives the difference between the price at which

he can sell electricity and the price he paid for it.

Here, FERC required demand-responders to be paid

the full value of electricity they decline to purchase with-

out accounting for what they would have paid for it—they

get full LMP. That pays demand-responders as if they

actually sell electricity to the market, having somehow

generated or acquired it for free. But that erroneously

assumes that not buying electricity is the same as pro-

ducing it. A company that does not use electricity can in

some loose sense be said to make that unused electricity

available to others. But to make electricity available to

others, it needs another input: It needs someone to pro-

duce the electricity that it has a right to use, does not use,

and thereby leaves available for others. FERC’s meth-

odology ignores that essential input.

For that reason, FERC erred in assuming that not us-

ing one megawatt-hour of electricity (the so-called “neg-

awatt-hour”) is economically equivalent to producing one

megawatt-hour of electricity. See U.S. Br. 49-50. That

“assumption is wrong.” Borlick, Pricing Negawatts, su-

pra, at 14. “The characteristics of a megawatt and a

‘negawatt’ are different, both in terms of physics and in

economic impact.” Pet. App. 160a (Moeller, Comm’r, dis-

senting). In terms of physics, a megawatt exists and can

13

be measured; a “negawatt” is an imputed quantity calcu-

lated by comparing what the customer might have used

to what it actually used. In terms of economics, a mega-

watt and a negawatt are not equivalent, because a nega-

watt does not include the costs the consumer avoided by

not purchasing electricity. That is, the negawatt is not

the costless generation of electricity. It is the non-use of

energy that someone else generated, and built expensive

facilities to create. See C.A. App. 483 (Hogan, Demand

Response Pricing in Organized Wholesale Markets).

The fallacy of FERC’s approach becomes evident if

one imagines that this case concerned automobiles rather

than energy. If a particular automobile is in high de-

mand but short supply, the wholesale price might sky-

rocket to, for example, $50,000 per car (yielding profits

for the manufacturer). A car dealer who has obligated

himself to sell those cars to certain consumers at $40,000

might be willing to pay those consumers not to buy a car

(this might be called a “naughtomobile”). But no one

would say the dealer should pay each potential buyer

$50,000—the full cost of a new car—for merely declining

to purchase one at $40,000. The dealer would at most of-

fer the potential buyer $10,000—the amount the dealer

would lose if the consumer insisted on making the pur-

chase. FERC’s approach gives the customer a windfall.

He gets paid $50,000, as if he himself manufactured and

sold an additional car.

Compare that to the treatment of a consumer who has

already bought a car for $40,000, but has not taken deliv-

ery. No one would object to his reselling that car at the

market rate of $50,000. He sells that car, which he paid

for and owns, to others for the same $50,000 the manu-

facturer gets for selling an identical car. But then the

consumer’s profit is $10,000, not $50,000. Under FERC’s

14

approach, however, that seller gets /ess profit for bring-

ing an actual car to market than would someone who

simply declined to buy one in the first instance. FERC

never explains why the market price for selling an actual

product into the market (that the seller owns or made)

would be the same price the market pays others who of-

fer nothing they own or paid for but who simply decline

to make purchases.”

Even if a customer were deemed to “supply” some-

thing he never purchased or generated, FERC’s rule still

makes no sense. In energy markets, diverse generation

sources compete: Coal-powered plants compete with hy-

droelectric dams, which compete with gas turbines. If

FERC decreed that coal-powered plants need not pay for

inputs needed to generate electricity—for example, that

coal mines must provide coal for free—coal-powered

generators receiving LMP would become more profita-

ble. The same would be true if FERC ordered that coal-

plants could take electricity from other generators, and

resell it at LMP, without paying the original generator.

In both situations, coal-plants would be paid for power

that, by decree, they acquire for little or nothing. LMP

would fall, as electricity sold by coal-plants—subsidized

by free inputs—displace unsubsidized energy sources.

That might (initially) benefit consumers. But it distorts

the market in favor of one source that may not be the

most efficient. Allowing lower-cost subsidized power

* This is no mere hypothetical: Companies that profit by buying in

day-ahead markets and reselling at LMP in real-time markets would

now do better by not buying at all and receiving LMP. See, e.g., Mo-

tion to Intervene and Comments of the California Department of

Water Resources State Water Project at 4, FERC Dkt. No. RM10-

17-000 (May 13, 2010).

15

drives down prices for other, potentially more efficient

sources, deterring investment in power plants that might

be socially desirable.

FERC’s approach is almost indistinguishable from

that hypothetical coal-plant subsidy. It directs that all

resources that putatively make electricity “available” be

paid LMP Under that rule, everyone actually supplying

electricity (nuclear plants, solar generators, and even

traders importing energy from adjacent markets) must

bear the cost of obtaining the electricity they make avail-

able. But not demand-response providers. They are

deemed to have “sold” electricity into the market even

though they have not generated or purchased any.

FERC thus subsidizes demand response at the expense

of other suppliers and the markets’ long-term health.

Pet. App. 170a (Moeller, Comm’r, dissenting).

It does not matter, as FERC urges (at 53-54), that

demand-response providers incur some costs (e.g., costs

of shutting down factories, shifting production times, or

organizing demand-responders), or that competing gen-

erators incur some costs in ramping down to stop provid-

ing electricity. The point is that FERC’s order gives de-

mand-responders one critical input for free—they effec-

tively sell electricity they do not own without bearing the

cost of generating or acquiring it. If the hypothetical coal

order excused coal-plants from paying for only some of

their inputs, it still would be a subsidy and it still would

distort the market. The result is no different if one sub-

stitutes the word “demand-responders” for “coal power-

plant.”

Thus, even if demand response and generatior were

said to provide the same “marginal value,” Pet. App. 91a,

FERC’s order allows one to obtain an input—the elec-

tricity it purports to make available—without paying for

16

it. There is a simple way to eliminate the subsidy. De-

mand-responders can be required to “pay” for the elec-

tricity they purport to offer. Demand-responders can

start with a payment of LMP for the electricity they of-

fer; but FERC must deduct the retail rate demand-

responders would pay for the electricity if they were to

buy it before reselling it at LMP.

2. FERC’s Rule Creates Irrational Incentives To

Cease or Defer Socially Valuable Production

FERC’s approach produces irrational incentives that

FERC never acknowledges. Basic math shows that pay-

ing demand-responders full LMP encourages productive

facilities to cease operations even when that damages so-

cial welfare.

An example makes that clear. If a widget-maker must

spend $.30 for electricity at retail to produce each widget,

and it must spend $.10 per widget in other costs, the

widget-maker ordinarily will produce widgets whenever

the market price for widgets exceeds $.40. If the market

price is $1.00, the widget-maker will make widgets at a

$.60 profit. Society gains handsomely: The widget-

maker converts $.40 worth of inputs into a product con-

sumers value at $1 (or more).

Nonetheless, society is better off if the widget-maker

stops production once LMP for electricity reaches $.91.

At that point, the real cost of making a widget (labor and

materials at $.10 per widget plus the $.91 per-widget

electricity cost, totaling $1.01) exceeds the $1.00 market

price. But a demand-response payment of LMP will in-

duce the widget-maker to stop production earlier—when

LMP reaches $.61. At that point, the widget-maker

earns more by doing nothing: It avoids $.30 in electricity

costs (at retail) and $.10 in production costs per widget

(assuming production costs fall to zero when production

17

ceases). And it collects $.61 in demand-response pay-

ments, more than the net profit of $.60 it would earn from

making widgets.

Society, however, is worse off. Society would benefit

from production—it gets widgets worth $1 from inputs

costing $.71 ($.61 in electricity and $.10 in other costs).

Yet FERC’s rule would cause the widget-maker to stop

production. See J.A.193-194 (Hogan, Implications for

Consumers of the NOPR’s Proposal To Pay the LMP for

All Demand Response). The pro-competition, consumer-

protective Federal Trade Commission made precisely

that point: Paying full LMP would “give the wrong in-

centives” by “creatiing] situations in which a demand-

response provider would find it more profitable to sell its

power rights (z.¢., provide demand response) than to con-

sume that power, even though the value to society of con-

suming that power exceeds the power’s cost to society.”

J.A.282-283.5

The suggestion that widget-makers may defer produc-

tion (rather than ceasing production), Priv. Pet. Br. 54, is

no answer. Overcompensating demand response encour-

ages curtailed consumption even when the social cost—

whether of shifting or ceasing production—exceeds the

social benefit. Changing the subject cannot change the

calculation.

* If LMP declines because of demand response, the calculations be-

come more complex, but the result is unchanged: Demand-

responders curtail consumption even when that harms social welfare.

The same is true for behind-the-meter generation, discussed pp. 18-

20, infra. See J.A.193-194 (Hogan, Implications, supra).

18

3. The Rule Offers Arbitrary and Uneconomic

Rewards Based on the Meter’s Location

A proper pricing rule should provide the same incen-

tive—to consumers and generators alike—to invest in

generation facilities. In other words, the incentives

“should be the same with respect to placing the genera-

tor on the consumer side of the [electricity] meter versus

[the grid] side of the meter.” C.A. App. 484 (Hogan, De-

mand Response Pricing, supra). But FERC arbitrarily

compensates users more handsomely for generating elec-

tricity solely for their own behind-the-meter use, discrim-

inating against plants that offer electricity into the grid.

Returning to the widget-maker: Assume he can avoid

drawing power from the public grid by building or oper-

ating his own generator. In an efficient market, he would

self-supply only when it costs less than purchasing elec-

tricity. Thus, if self-generation costs $.60 per widget, he

should not self-supply until LMP exceeds $.60. But un-

der FERC’s approach, he would self-supply sooner. If

his retail electricity rate is $.30 per widget, the widget-

maker will self-supply when LMP hits $.31 per widget.

He spends $.60 running his own generator instead of pay-

ing $.30 retail, which would result in a net loss of $.30 per

widget. But he then collects a demand-response payment

of LMP, or $.31 per widget, for not buying the electricity

from the grid. So he ends up $.01 ahead for each widget.

Thus, even though the economic cost of self-generating

electricity is almost double the grid price—$.60 per

widget rather than LMP of $.31—the widget-maker will

self-supply. And, under FERC’s rule, that self-supply

would appear as a “reduction in electricity consumption”

even though the demand-response payment “has merely

moved the consumption somewhere where it is not visible

to the RTO.” J.A.198 (Hogan, /mplications, supra).

19

The absurdity of that result crystalizes if one com-

pares what would happen if the widget-maker offered the

output of his generator into the market, rather than us-

ing it himself. If (once again) self-generation costs $.60

per widget, and LMP is $.31 per widget, the widget mak-

er cannot profit by selling self-generated electricity: It

costs him $.60 to make something he can sell for only

$.31. Plus he must still pay for any electricity he draws

from the grid for his own use (at the retail rate of $.30

assumed above). But if he moves the generator behind

the meter and keeps it for his own use, he does better.

He gets paid $.31 per widget in demand-response pay-

ments for not using electricity from the grid (instead

generating it himself); and he saves the $.30 per widget

he would have paid for the electricity he now generates

for himself. Thus, even though total consumption in each

example is identical, the result changes depending on

who generates electricity and where it is metered or

“seen.” Paying full LMP thus provides a subsidy for self-

supply. Indeed, “whether full LMP constitutes a subsidy

isn’t a serious economic question.” Newell, DR Distor-

tion, 148 Pub. Utils. Fort. 36, 41 (2010).

That behind-the-meter problem also provides “the:

seeds of [the Final Rule’s] own demise.” J.A.200 (Ho-

gan, Implications, supra). Inefficient demand-response

payments “must be recouped somewhere.” /d. at 199.

Under FERC’s approach, “every dollar of benefit gained

by the wholesale buyers is expropriated” in the short-

term “from the generators.” Borlick, Paying for De-

mand-Side Response at the Wholesale Level: The Small

Consumers’ Perspective, 24 Elec. J. 8, 16 (2011). That

windfall for demand-responders will, in turn, “induce

others to leave the [grid] system” and go behind-the-

meter, creating an unsustainable inefficiency spiral.

20

J.A.200 (Hogan, Implications, supra). And that spiral

will disproportionately affect “residential and small

commercial consumers whose operations are not of suffi-

cient size and scope” to warrant behind-the-meter self-

generation, but who will ultimately pay the price for un-

derinvestment in common generation units. /d. at 203.

Rather than address that concern, FERC “assum/[ed]

that consumers would not respond to these incentives to

move generators behind the meter or operate inefficient

backup equipment.” J.A.205. But large, industrial con-

sumers often have that option, mostly through “diesel

units that lack emissions controls.” J.A.1149 n.58 (Com-

petitive Power Supplier Ass’ns). FERC agrees that de-

mand-responders will self-supply using “standby genera-

tor{s),” touting that as a “cost” they incur. U.S. Br. 54.

And FERC offered no “evidence to support [a contrary]

assumption.” J.A.205 (Hogan, /mplications, supra).

Favoring behind-the-meter generation is thus “nothing

less than a subsidy for dirty’"—and more costly—

“generating resources.” Delaware Dep't of Natural Res.

v. EPA, No. 13-1093, slip op. at 23 (D.C. Cir. May 1, 2015)

(quotation marks omitted). And the disparity in compen-

sation based on the meter’s location demonstrates the

defect in FERC’s approach: It facially discriminates, and

distorts the market, by favoring behind-the-meter ac-

tions, including demand response, over in-front-of-the-

meter activity like generation for public use.

Il. EFFICIENT DEMAND-RESPONSE MECHANISMS REP-

LICATE THE MARKET

In an efficient market, real-time pricing would require

consumers to pay for their marginal consumption. See

C.A. App. 482 (Hogan, Demand Response Pricing, su-

pra). If our widget-maker were required to pay LMP for

his electricity, he would continue making widgets until

21

LMP brought his total costs—for energy and other in-

puts—above the market price for widgets. He would not

need demand-response payments to encourage him to

decrease consumption when wholesale prices rise; he

would do so whenever increasing input prices drive prof-

its to zero.

Absent real-time pricing, however, demand response

can provide a second-best solution—if one avoids creat-

ing subsidies. “Payments based on the LMP minus retail

rate structure eliminate these subsidies.” C.A. App. 1143

(Dr. Shanker). That mechanism is known as “LMP-

minus-G,” where “G” is the full retail rate the consumer

would have paid for electricity. That methodology gives

demand-response providers incentives that parallel those

in competitive, real-time electricity markets. See C.A.

App. 1139 (Shanker, supra); Newell, supra, at 37.

For example, if a consumer has a contract right to

purchase electricity at a particular rate, it might pur-

chase electricity and, in turn, sell it to others at LMP. If

our widget-maker had a retail contract allowing him to

purchase electricity at $.30 per unit, and LMP rose to

$.91 per unit, he could stop production, exercise the im-

plicit option to buy electricity for $.30, and sell the elec-

tricity at $.91, yielding a profit of $.61 per unit. “The net

transaction for the customer would be the LMP minus

the fixed price of the contract.” Hogan, Whitepaper, su-

pra, at 13.

LMP-minus-G mimics that result for demand re-

sponse. It recognizes that “[e]conomic demand response

isn’t a sale of energy; rather, it’s a sale of a call option on

energy.” Borlick, Pricing Negawatts, supra, at 14. Like

other call options, its value is measured by the value of

the underlying good, less the cost of exercising the option

(the strike price)—here, the retail rate for electricity. If

22

the retail price is lower than LMP, buyers can sell for a

profit (LMP less the retail rate). Failing to subtract the

retail rate, however, allows the consumer to sell its elec-

tricity at full rates without exercising the option and thus

without buying a key input. That induces demand-

responders to sell load reductions, or self-supply, even

when it is more efficient for them to purchase electricity

from the grid.

FERC previously recognized as much. See PJM In-

terconnection, L.L.C., 99 FERC 961,227 at 61,941 (2002).

Indeed, FERC refused to mandate payment of full LMP

because it would create an excessive incentive to curtail

demand. See PJM Indus. Customer Coal. v. PJM Inter-

connection, L.L.C., 121 FERC 961,315 at P26 & n.20

(2007). The RTO, it explained, already “reimburses [de-

mand-responders} for the difference between the whole-

sale rate and their retail rate,” ie., LMP-minus-G. /d. at

P26. “That payment provides customers under retail

rates with the same economic incentive to curtail load as

if they were paying the wholesale rate itself.” bid

FERC did the math to prove that conclusion: It showed

that LMP-minus-G gives demand-responders the same

incentives as efficient wholesale rates—and why paying

full LMP creates excessive incentives.‘ FERC thus la-

* FERC stated: “The financial benefit [of reduced consumption] is

comprised of two components: (i) the retail rate that is avoided by

not consuming, and (ii) the payment received under the program,

which is equal to the difference between the LMP and the retail

rate.” PJM Indua., 121 FERC 161,315 at P26 n.19. Thus, “if the

LMP is $600/MWh and the retail rate is $50/MWh, the customer who

would receive a payment under the program of $550/MWh (ie., $600-

$50), for a total financial benefit of $600/MWh, which is the value of

the LMP.” /bid. Paying LMP, by contrast, would produce excessive

23

beled the payment of full LMP a “subsidy” that “could

lead [consumers] to curtail cost-effective production.” /d.

at P26 n.20. Today, FERC identifies no error in its for-

mer math or economic analysis. Instead, it imposes pre-

cisely the subsidy it previously rejected.

The irony is that, by overcompensating demand re-

sponse, FERC may doom truly efficient real-time pric-

ing. The “technical barrier to dynamic pricing”—

availability of time-sensitive metering for retail users—

“should be lifted in the next five to 10 years.” Faruqui &

Palmer, Dynamic Pricing and Its Discontents, 34 Regu-

lation 16, 17 (2011). But overcompensated demand-

responders have little incentive to change to a more effi-

cient model that eliminates their windfall subsidies.

Ill. CONTRARY ARGUMENTS ARE UNFOUNDED

Petitioners and supporting amici offer several eco-

nomic justifications for FERC’s full-LMP rule. None has

merit. FERC did not rely on many of them. And many

are naked requests for subsidies FERC never admitted

providing, in never-justified amounts, to overcome barri-

ers FERC never found.

A. FERC’s Own Arguments Show That Paying

Full LMP Effects an Unjustified Subsidy

FERC and its amici urge that FERC’s pricing mech-

anism is efficient, while LMP-minus-G is not. But that

consists mostly of assertion—not analysis.

incentives: “{TJhe financial benefit for reducing consumption would

exceed the LMP. That is, with the subsidy payment, the customer

would receive a payment of $600/MWh (ie., the LMP) and in addi-

tion would avoid the $50/MWh retail rate, for a total financial benefit

of $650.” Ibid.

24

1. Critically, there is much FERC and its amici do

not say. They nowhere deny that FERC’s pricing regime

favors those who self-supply behind the meter over those

who supply electricity to the grid for all. See pp. 18-20,

supra. As the only economists to submit a brief below,

we made precisely that point. C.A. Amicus Br. of Lead-

ing Economists 20-23. We made that point before

FERC. See, e.g., J.A.196-205; see also pp. 2, 18-20, su-

pra. No one offers an answer; none exists. Likewise, no

economist seriously addresses the skewed incentives that

paying LMP produces. See pp. 16-17, supra.°

Ine vad, FERC relies almost exclusively on a submis-

sion by Dr. Alfred Kahn, see Pet. App. 97a-98a, 101a,

214a, 217a-218a, which posits that demand response “is in

all essential respects economically equivalent to supply

response” (i.e., increased generation) and should be com-

pensated accordingly. Pet. App. 69a-70a. As FERC

characterizes that theory, “reduction in demand provides

the same benefits to a wholesale system as an equivalent

increase in supply” and should “be compensated equiva-

lently.” U.S. Br. 47-48, 52-53 (invoking Dr. Kahn).

* At times, FERC seems to embrace its refusal to address these

costs. The “Final Rule does not attempt to measure what would

have happened in a retail program,” FERC declares, because the

sole focus was “the net price effect of paying the demand response

resources the LMP in the wholesale market.” Pet. App. 238a (em-

phasis added); see Cicchetti Br. 30; ECDRP Br. 26. But neither eco-

nomics nor precedent allows FERC to wreak havoc on retail incen-

tives—ignoring damage to total social welfare—by myopically focus-

ing on FERC-regulated wholesale markets in isolation. FPC v.

Conway Corp., 426 U.S. 271, 272-278 (1976). FERC must consider

“the entire factual context in which the proposed wholesale rate will

function,” which “will naturally include those related to nonjurisdic-

tional transactions.” Jd. at 280 (quotation marks omitted).

25

The late Dr. Kahn was one of the great economists of

his time. Nonetheless, the arguments now imputed to

him do not withstand economic scrutiny, and he never

had or took the opportunity to address the critique of his

initial analysis. The cited submission ignores the fact

that retail customers who curtail consumption in ex-

change for demand-response payments receive two bene-

fits. First, they benefit by avoiding the cost of purchas-

ing electricity (a benefit generators do not receive). And

they receive demand-response payments as well. When

determining the compensation necessary to induce ap-

propriate conduct, the effect of both benefits must be

considered—as FERC previously recognized. See pp. 22

& n.4, supra. Professor Kahn’s submission ignores the

cost avoided by not buying electricity at retail. Genera-

ters, moreover, surrender electricity they own so others

may use it, receiving full LMP for it. Traders buy elec-

tricity on the market, pay for it, and sell that paid-for

electricity at full LMP. But demand-responders do not

offer electricity they own. They surrender the right to

buy and use electricity generated by others. Yet FERC

insists they be paid the same rate as someone who sells

electricity he bought, paid for, and owns.

No right-thinking economist would urge the market

should pay a consumer who gives up a car he bought and

owms the same price that would be paid to a consumer

who never paid for or acquired a car but merely surren-

ders the unexercised option of buying it. FERC’s order

does just that. That difference makes the overcompensa-

tion—and the subsidy—clear. Because a generator must

produce electricity, its profit is not LMP. It is LMP mi-

nus the costs incurred to deliver power, including all in-

puts to its generation. As a result, “[tJhere is no need to

deduct costs from the LMP payments made to genera-

26

tors because,” when they receive LMP, “they earn the

net amount equal to LMP minus” the cost of creating the

power they sell. C.A. App. 651 (NEPGA).

2. The government responds (at 54) that, because

FERC “does not ordinarily take into account the costs of

production when paying generators LMP, it would not

take into account a demand-response provider’s costs ei-

ther.” See also Priv. Pet. Br. 52-53; ECDRP Br. 21. If

demand-responders were selling electricity they owned,

as generators do, that would make sense. But demand-

responders do not sell electricity they own. They sur-

render only an option to buy electricity. By compensat-

ing demand-responders as if they owned electricity and

were reselling it, FERC makes them owners of electrici-

ty for free. It is no different than if FERC decreed that

coal-powered plants may sell electricity at full rates

without paying for certain inputs. See pp. 14-16, supra.

It is incongruous for a regulator like FERC to order the

creation of a subsidy by giving one set of suppliers an in-

put for free—treating them as owning electricity they did

not buy—and then defend its order by saying it should

never take costs into account. By doing so, FERC subsi-

dizes demand response, forcing the system to “dispatch

excessive and inefficient amounts” instead of generated

electricity. Borlick, Pricing Negawatts, supra, at 17.

3. FERC’s purported concerns about “practical diffi-

culties” are unfounded (even FERC conceded LMP-

minus-G was “perhaps feasible,” Pet. App. 102a). Not

one ISO or RTO—the entities that would implement this

mechanism—suggested that LMP-minus-G would be

problematic. PJM and the Midcontinent ISO already

offset avoided retail costs with little practical difficulty.

J.A.1173 n.117 (Competitive Power Supplier Ass’n).

FERC cited the New York PUC’s assertion that sub-

27

tracting retail rates would be “an administrative burden”

that could engender “confusion.” Pet. App. 76a; see U.S.

Br. 58-59. But the New York ISO, which would imple-

ment it, supported LMP-minus-G because it “avoids the

need for [FERC’s] complicated and contentious net-

benefits test and cost allocation rules.” C.A. App. 1153.

B. Reliance on the Net-Benefits Test Underscores

the Irrationality of FERC’s Approach

Ultimately, virtually every defense of FERC’s orders

falls back on FERC’s “net-benefits” test. That test re-

quires each RTO or ISO to use “historical data,” and an

indeterminate list of factors bearing on supply conditions

(e.g., fuel prices and generator unit availability), to calcu-

late the point at which consumer benefits from reduced

LMP exceed the cost of paying LMP to demand-response

providers. Pet. App. 56a-57a.

1. FERC’s need to overlay a net-benefits test repre-

sents an admission that its basic pricing methodology is

flawed. If paying full LMP produced appropriate pricing

signals, FERC would not need to superimpose another

test as a precondition for demand response. As the Fed-

eral Trade Commission explained, “there is no need for a

net benefits test so long as FERC utilizes efficient prices

in compensating demand response providers”; “[tlhe

proposal to implement a net benefits test *** arises

*** only if FERC sets inefficiently high compensation

levels for demand response.” C.A. App. 1180.

FERC ignores the practical problems with its net-

benefits test. That test assumes an omniscient regulator

who can accurately predict when demand response is ap-

propriate, rather than allowing rational price signals to

achieve that result. To replicate the effects of price sig-

nals, regulators would have to apply the net-benefits test

for each proposed demand-response provider for each

28

transaction. It is far from clear that could be accom-

plished accurately. Thus, a “clear majority of the wit-

nesses (representing a spectrum of interests that includ-

ed demand response advocates, economists, generators,

and the RTOs and ISOs)” opposed the test, characteriz-

ing it as exceedingly difficult to administer. Pet. App.

164a (Moeller, Comm’r, dissenting). And FERC’s reli-

ance on unsubstantiated “practical difficulties” to reject

LMP-minus-G, while simultaneously imposing a complex

and difficult net-benefits test to mitigate the impact of its

erroneous rule, is not reasoned decisionmaking.

2. Even if perfectly administered, FERC’s net-

benefits test is fundamentally flawed. By its terms, the

test precludes demand-response payments unless those

payments lower the amount consumers pay for electrici-

ty. See U.S. Br. 55-56; Priv. Pet. Br. 43; Cicchetti Br. 29-

30. FERC acknowledges that “[t]he cost allocation

methodology required in the Final Rule is based upon the

benefits of demand response to wholesale load”—e.,

customers. Pet. App. 248a (emphasis added); see id. at

234a (“net benefits test” met when demand-response

“payment is a cost-effective purchase from the custom-

er’s standpoint,” ie., “[wJhen reductions in LMP from

implementing demand response results in a reduction in

the total amount consumers pay * * * greater than the

money spent acquiring those demand-response resources

at LMP”) (emphasis added). The net-benefits test thus

focuses exclusively on benefits to buyers, without looking

to the impact on industry or society.

That erroneously assumes that lower prices are al-

ways beneficial. Manipulating markets to force prices

below competitive rates may seem desirable for consum-

ers in the short-run. But FERC has repeatedly held that

artificially suppressed prices “deter investment in new

29

generation” and harm consumers in the long-run by

“broadly hinder[ing] market development and perfor-

mance.” Milford Power Co., LLC, 119 FERC 961,167 at

P31 & nn.48-51 (2007) (listing cases). The net-benefits

test, by myopically focusing solely on whether prices are

lower rather than whether they are efficient, ignores that

wisdom.

Perhaps more important, the net-benefits test ignores

half the societal benefits of economic activity. Any econ-

omist examining pricing and production to maximize

benefits to society will measure total social welfare. So-

cial welfare consists of the sum of the consumer surplus

(the difference between how much the consumer values a

good and what he pays for it), plus the producer surplus

(the difference between the cost of production and the

price). Mankiw, Principles of Macroeconomics 148 (5th

ed. 2009). Social welfare—societal benefit—is maximized

when that sum is maximized. [bid.

But FERC’s test looks only to consumer welfare—

whether consumers end up paying less. See, e.g., Pet.

App. 234a-235a (“When demand response produces a suf-

ficient reduction in LMP to cover the increased billing

costs imposed on remaining customers, it is beneficial to

customers; when the reduction does not cover costs, the

demand response is not beneficial.”); p. 28, supra. By do-

ing that, FERC approves demand-response payments

even where it lowers total social welfare (such as where

the consumer benefit is merely a wealth transfer from

producers). See Chao, Price-Responsive Demand Man-

agement for a Smart Grid World, 23 Elec. J. 7, 17 (2010).

Dr. Kolstad, while purpc:tiny to support the order, thus

concedes that “the test focuses exclusively on consumers’

expenditures” and thus “cannot guarantee that compen-

sating demand response at LMP optimizes overall social

30

welfare.” Kolstad Br. 17-18. That test thus contradicts

both FERC’s understanding of its mandate, see pp. 7-8,

22, supra, and this Court’s as well: For more than half a

century, this Court has emphasized that “FERC must

choose a method that entails an appropriate ‘balancing of

investor and consumer interests.” Morgan Stanley

Capital Grp. Inc. v. Pub. Util. Dist. No. 1, 554 U.S. 527,

532 (2008) (quoting FPC v. Hope Natural Gas Co., 320

U.S. 591, 608 (1944)) (emphasis added); accord Permian

Basin Area Rate Cases, 390 U.S. 747, 776 (1968).

Because the net-benefits test looks at only half of that

balance—the consumer interest in lower prices—

Professor Cicchetti’s brief studiously avoids addressing

total social welfare. Instead, he states that “economic

efficiency [is] improved” in various situations “because

buyers would pay less” despite the cost of “compen-

sat{ing] providers” for demand response. Br. 25-26 (em-

phasis added). The net-benefits test, he asserts, “prohib-

its the dispatch of DRR that is not cost-effective from a

buyer’s perspective.” Jd. at 27 (emphasis added). But

efficiency is not the same as lower prices. Total social

welfare is not the same as buyer welfare. And “balanc-

ing” the interests of generators and consumers is not the

same as ensuring consumers always pay less.

Professor Cicchetti’s publications (unlike his brief)

recognize that. Professor Cicchetti cites (at 2) a paper he

co-authored with Dr. William Hogan, amicus here. That

paper concludes that demand-response payments “equal

to the utility’s marginal cost” of buying from the grid at

LMP “would obviously be overkill. The conserver would

be paid twice for the same decision, once when his/her bill

was reduced [by the amount he conserved], and again

when receiving a subsidy equal to avoided [wholesale]

cost.” Cicchetti & Hogan, Including Unbundled De-

31

mand-Side Options in Electric Utility Bidding Pro-

grams, Pub. Utils. Fort., June 8, 1989, at 9, 10. Critiqu-

ing the payment of full LMP, he urged that the “incentive

given the [demand-responder] is wrong. He is paid

twice: once by giving him the dollars for the conservation

*** again by providing a free service. This misplaced

incentive could result in inefficient over-investment in

conservation.” Jd. at 11; see pp. 16-17, supra.®

Dr. Cicchetti may now believe that social goals justify

inefficient pricing that lowers total welfare. But he never

purports to pursue a goal other than efficient pricing.

The same thing is true of FERC. Its stated goal was to

“improve the competitiveness of the organized wholesale

energy markets,” J.A.35, so as to “balance supply and

demand,” Pet. App. 96a. With respect to that goal, Dr.

Cicchetti’s scholarship, not his brief, is correct.

* Dr. Cicchetti now offers a theory FERC never relied on, precluding

its consideration here. See SEC v. Chenery Corp., 332 U.S. 194, 196

(1947). Nonetheless, his new theory shows why LMP-minus-G, not

full LMP, produces efficient results. Because utilities pay end-users

for the right to limit their consumption at key moments—allowing

the utility to limit its purchases—itilities incur some costs. But that

does not speak to the rate utilities should be paid. Where utilities

are saving money by not buying expensive wholesale electricity for

customers who would pay lesser retail rates, they save LMP simply

by reducing their purchases. Dr. Cicchetti would have them receive

a demand-response payment of LMP on top of that, making the ben-

efit twice LMP, less forgone retail revenue. Moreover, Dr. Cicchetti

concedes that “large industrial end-users” (or demand-response ag-

gregators) who sell demand response directly into wholesale mar-

kets “are an exception to” his critique. Cicchetti Br. 33 n.21. But Dr.

Cicchetti does not explain why those suppliers should receive double

payments. Moreover, if industrial end-users would receive only

LMP-minus-G in an efficient market, the same should be true for

other demand-responders.

32

3. FERC and its amici make various efforts to shore

up FERC’s leaky economics, but they fail for similar rea-

sons. For example, Professor Cicchetti and others assert

that demand response “should enter the wholesale mar-

ket whenever its marginal cost is less than the marginal

cost of supplying electricity.” Cicchetti Br. 21; see EC-

DRP Br. 22; U.S. Br. 48; Priv. Pet. Br. 47. True. But Dr.

Cicchetti (and the others) ignore how FERC has distort-

ed the concept of marginal cost. If demand-responders

and generators were both bringing electricity to the

market so others could use it, they each would bear the

marginal cost of obtaining or generating that electricity;

they would account for that cost when deciding when to

enter the market. Under FERC’s order, generators and

traders do that—they are not relieved of paying the costs

of the electricity they generate (or buy) and then sell.

But demand-responders receive full price for electricity

they never buy. The marginal cost of demand response is

thus effectively subsidized by FERC’s decree that they

be treated as if they are selling electricity they own when

they never buy or pay for it.

Those subsidies for non-consumption significantly re-

duce demand and lower prices. But “every dollar of ben-

efit gained by the wholesale buyers is expropriated” in

the short-term “from the generators.” Borlick, Paying

for Demand-Side Response, supra, at 16. “[MJost of the

parties advocating such a net benefits calculation explicit-

ly base it on the ‘benefits’ of such price suppression.”

C.A. App. 1141-1142 (Shanker, supra). If there are non-

economic, policy reasons for artificially suppressing pric-

es and establishing a one-sided test that ignores total so-

cial welfare, regulators should acknowledge them and

balance them against the resulting costs. Here, FERC

did not do that. “[Ijn the long run,” the burden of its fail-

33

ure will not befall generators, RTOs or ISOs, or FERC.

The costs will eventually be passed along to “residential

and small business consumers,” which cannot provide

demand response, but end up paying demand-responders

to displace otherwise more efficient generation. Borlick,

Paying for Demand-Side Response, supra, at 8. That “is

inefficient for the economy as a whole, distorts produc-

tion and consumption decisions, and raises prices in the

long run.” C.A. App. 1142 (Shanker, supra); see Newell,

supra, at 38.

C. Petitioner and Its Amici Seek Subsidies Not

Supported by the Record

Finally, some parties urge that overcompensation is

necessary to overcome putative “barriers to full demand

response participation,” Priv. Pet. Br. 46-47, or to cover

“the cost of technology” and “infrastructure” “necessary

to participate in wholesale markets,” U.S. Br. 53-54, 55.

Those are, at bottom, requests for a subsidy. Nothing in

the record establishes the extent of any putative “barri-

ers.” There is no finding that paying full LMP is cali-

brated to overcome those “barriers.” And there is no

analysis suggesting that demand-response providers face

more significant “barriers” than new generators, who

must invest in multi-year, multi-billion-dollar plants.

Undoubtedly, FERC effected a subsidy for demand re-

sponse. It just refused to admit or justify it. That is ar-

bitrary and capricious."

’ Paying too much for demand response likewise is not calibrated to

address any particular externality. See Borlick, Paying for De-

mand-Side Response, supra, at 11. It “dof{es) nothing to induce elec-

tricity suppliers to reduce the environmental externalities they im-

pose on others.” Hogan, Whitepaper, supra, at 7. It merely induces

generators to supply less energy—period. And because it encour-

34

The government also suggests (at 53-54) that FERC

decided that higher payments were warranted to offset

certain demand-responder costs, such as the cost of “ob-

taining power from another source” (e.g., “standby gen-

erators”) or the costs of “going without the foregone

power in its production of goods.” See also Priv. Pet. Br.

55 (overcoming increased labor costs).

Demand-responders do encounter costs when they

curtail demand. See U.S. Br. 53-54; Priv. Pet. Br. 52-53;

ECDRP Br. 22; Kolstad Br. 11. But they should consider

those costs when deciding whether to curtail consump-

tion. If those costs exceed the available payment for de-

mand response, demand-responders should continue

production (or should continue using electricity from the

grid rather than switching to self-generation). Only if

total benefits exceed total costs will curtailing production

or shifting to self-generation benefit society as a whole.

The problem with FERC’s position is that it establishes

an incentive to move to behind-the-meter, stand-by gen-

eration even when that costs more than electricity at

market rates, or to cease or defer production when doing

so damages total social welfare. See pp. 16-20, supra.

FERC’s failure to confront those distortions dooms its

order.

CONCLUSION

The judgment of the court of appeals should be af-

firmed.

ages unregulated behind-the-meter generation, overcompensating

demand reduction may increase pollution.

Respectfully submitted.

JOHN LEE SHEPHERD, JR. JEFFREY A. LAMKEN

SKADDEN, ARPS, SLATE, Counsel of Record

MEAGHER & FLOMLLP MOLOLAMKEN LLP

1440 New York Ave.,N.W. The W: Suite 660

Washington, D.C. 20005 § 600 New Hampshire Ave., N.W.

(202) 371-7338 Washington, D.C. 20037

(202) 556-2000

jlamken@mololamken.com

Counsel for Amici Curiae

September 2015

ADDENDUM

la

ADDENDUM

LIST OF AMICI CURIAE'

Robert L. Borlick is a Managing Director at Borlick

Associates. He has more than 35 years’ experience in the

electric power industry, and recently served as a con-

sultant to the Demand Response Working Group of the

Midcontinent Independent System Operator. He has

held a number of positions at the United States Depart-

ment of Energy, including Chief of the Economic and

Financial Analysis Branch and Director of the Electric

Power Analysis Division.

Joseph Bowring is the President of Monitoring Ana-

lytics, LLC, and the Independent Market Monitor for

PJM Interconnection. He is the former Chief Economist

for the New Jersey Department of the Public Advocate’s

Division of Rate Counsel. He has taught economics as a

member of the faculty at Bucknell University and Villa-

nova University.

James Bushnell is a Professor in the Department of

Economics at the University of California, Davis. He is

also a Research Fellow at the National Bureau of Econ-

omic Research and a member of the Market Surveillance

Committee of the California Independent System Oper-

ator. He serves as an Associate Editor of the journal

Operations Research.

Paul A. Centolella is President of Paul Centolella &

Associates, LLC. He is a member of the Secretary of

Energy’s Electricity Advisory Committee, the National

Institute of Science and Technology's Smart Grid

Advisory Committee, and the National Academy of

Sciences Committee on Adoption of Advanced Energy

' Institutional affiliations are listed for identification purposes only.

2a

Efficiency & Clean Energy. He previously served as a

Commissioner of the Public Utilities Commission of Ohio.

Hung-po Chao is President of Energy Trading Ana-

lytics, LLC. Previously, he served as Director of Market

Monitoring and then Director of Market Strategy at ISO

New England. Before that, he held various research

management positions at the Electric Power Research

Institute, and was a Consulting Professor at Stanford

University.

Ahmad Farugqui is a Principal at The Brattle Group.

Formerly, he was Vice President of CRA International.

He has edited or co-edited four books and authored or co-

authored more than a hundred papers on energy effi-

ciency, demand response, tariff design, and load fore-

casting. His work has been cited in The Economist, The

New York Times, USA Today, The Wall Street Journal,

and The Washington Post, and he has appeared on Fox

News and National Public Radio.

Michael Giberson is Associate Professor of Practice,

in the area of Energy, Economics, and Law, at the Rawls

College of Business, Texas Tech University. Before that,

he was a consultant at Potomac Economics, Ltd. He has

worked with the Critical Infrastructure Protection Pro-

ject and the Interdisciplinary Center for Economic Sci-

ence at George Mason University.

Dinos Gonatas is a Principal at CPG Advisors. He

has worked extensively in the area of electric transmis-

sion for wind power. Previously, he was a business devel-

opment manager at Oxford Instruments, and a licensing

officer at the Massachusetts Institute of Technology. He

holds six U.S. patents in the area of synthetic fuel tech-

nologies for coal and gas conversion.

3a

Scott Harvey is a Consultant at FTI Consulting, Inc.

He is currently a member of the California Independent

System Operator’s market surveillance committee. He

has consulted with the New York Independent System

Operator on market design and performance issues since

its inception.

Benjamin F. Hobbs is the Theodore K. and Kay W.

Schad Professor of Environmental Management at the

Johns Hopkins University and Director of the Johns

Hopkins Environment, Energy, Sustainability & Health

Institute. He currently serves as the Chairman of the

Market Surveillance Committee of the California Inde-

pendent System Operator.

William W. Hogan is the Raymond Plank Professor

of Global Energy Policy at the John F. Kennedy School

of Government, Harvard University, and Research Di-

rector of the Harvard Electricity Policy Group. He has

been actively engaged in the design and improvement of

competitive electricity markets in many regions of the

United States, as well as around the world. In the past,

he was President of the International Association for

Energy Economics.

Joseph P. Kalt is the Ford Foundation Professor

(Emeritus) of International Political Economy at the

John F. Kennedy School of Government, Harvard Uni-

versity. He is also Co-Director of the Harvard Project on

American Indian Economic Development, and has pub-

lished widely in the area of natural resources economics

and policy.

Robert J. Michaels is Professor of Economics at the

California State University, Fullerton; Senior Fellow at

the Institute for Energy Research; Adjunct Scholar at

the Cato Institute; and an independent consultant. His

4a

research on electricity markets frequently appears in

academic and industry media. He has testified before

state regulatory commissions, FERC, and Committees of

Congress.

Shmuel S. Oren is the Earl J. Isaac Professor in the

Science and Analysis of Decision Making at the De-

partment of Industrial Engineering and Operations Re-

search, University of California at Berkeley. He is a for-

mer member of the California Independent System

Operator Market Surveillance Committee, and served as

Senior Advisor to the Market Oversight Division of the

Public Utility Commission of Texas. He has consulted on

market design issues with ISO New England, Pacific Gas

& Electricity, and other companies.

David B. Patton is President of Potomac Economics.

He currently serves as the Independent Market Monitor

for Midcontinent Independent System Operator, the

New York Independent System Operator, [SO New Eng-

land, and the Energy Reliability Council of Texas. He is

also a former Senior Economist at FERC, where he

worked on issues such as transmission pricing.

Craig Pirrong is Professor of Finance and Director of

the Global Energy Management Institute at the Bauer

College of Business, University of Houston. In the past,

he has consulted for the New York Mercantile Exchange,

the Chicago Board of Trade, the Chicago Stock Ex-

change, and several electric utilities.

Susan Liese Pope is Managing Director at FTI Con-

sulting, Inc. She has more than 20 years’ experience in

the electric power sector, and has worked with the New

York Independent System Operator, PJM Interconnec-

tion, the California Independent System Operator, the

Midcontinent Independent System Operator, and other

5a

RTOs to design rules for the operation of wholesale elec-

tricity markets.

Larry E. Ruff is an independent consultani. He was

formerly Senior Vice President at National Economic

Research Associates. He has advised governments and

utilities in the United States, Canada, Europe, the Asia-

Pacific region, and Latin America on competitive restruc-

turing of electric and gas utilities.

Richard Schmalensee is the Howard W. Johnson

Professor of Management, Emeritus; Professor of Ap-

plied Economics, Emeritus; and Dean, Emeritus at the

Sloan School of Management, Massachusetts Institute of

Technology. He is a fellow of the American Academy of

Arts and Sciences. He is a former member of the Presi-

dent’s Council of Economic Advisors.

Roy J. Shanker is an independent consultant in ener-

gy markets with over 40 years’ experience. He has been

actively involved in the design, development, and conduct

of the PJM and New York Independent System Operator

markets since approximately 1995 when design efforts

began. He also has had engagements involving the

design and ongoing development of the Midcontinent

Independent System Operator and ISO New England.

He has been an invited speaker and frequently testifies

before FERC.

Vernon L. Smith is the George L. Argyros Endowed

Chair in Finance and Economics at Chapman University

and Professor of Economics and Law at the Dale E.

Fowler School of Law at Chapman University. He was

awarded the Nobel Memorial Prize in Economic Sciences

in 2002. He is a founding member and President of the

International Foundation for Research in Experimental

Economics and a Senior Fellow at the Cato Institute. He

6a

served as a consultant on the privatization of electric

power in Australia and New Zealand and has partici-

pated in numerous private and public discussions of ener-

gy deregulation in the United States.

Richard D. Tabors is President at Tabors Caramanis

Rudkevich, and was previously President and Principal

at Across the Charles. In the past, he was a senior lec-

turer in technology and policy at the Massachusetts In-

stitute of Technology. He received his Ph.D. from Syra-

cuse University and holds an honorary D.Sc. from the

University of Strathclyde, Glasgow, Scotland, where he

was a visiting professor.