[Federal Register Volume 83, Number 156 (Monday, August 13, 2018)]
[Proposed Rules]
[Pages 39939-39957]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2018-17221]
=======================================================================
-----------------------------------------------------------------------
POSTAL REGULATORY COMMISSION
39 CFR Part 3015
[Docket No. RM2017-1; Order No. 4742]
Competitive Postal Products
AGENCY: Postal Regulatory Commission.
ACTION: Proposed rulemaking.
-----------------------------------------------------------------------
SUMMARY: The Commission is revising its previously proposed rules
related to the minimum amount that competitive products as a whole are
required to contribute to institutional costs annually, based on
comments received. The Commission invites public comment on the revised
proposed rules.
DATES: Comments are due: September 12, 2018.
ADDRESSES: Submit comments electronically via the Commission's Filing
Online system at http://www.prc.gov. Those who cannot submit comments
electronically should contact the person identified in the FOR FURTHER
INFORMATION CONTACT section by telephone for advice on filing
alternatives.
FOR FURTHER INFORMATION CONTACT: David A. Trissell, General Counsel, at
202-789-6820.
SUPPLEMENTARY INFORMATION:
Table of Contents
I. Introduction
II. Organization of Discussion
III. Background
IV. Proposed Modified Formula and Commission Analysis
V. Section 703(d) of the PAEA
VI. Administrative Actions
VII. Ordering Paragraphs
I. Introduction
On February 8, 2018, the Commission issued a Notice of Proposed
Rulemaking (Order No. 4402) proposing that a formula be used to
calculate the minimum amount that competitive products as a whole are
required to
[[Page 39940]]
annually contribute to institutional costs (i.e., the appropriate
share).\1\ Order No. 4402 was the result of the Commission's second
review of the appropriate share, conducted pursuant to 39 U.S.C.
3633(b) in order to determine whether the existing appropriate share
requirement of 5.5 percent should be retained, modified, or eliminated.
See 39 U.S.C. 3633(b); see also 39 CFR 3015.7(c). For the reasons
discussed below, the Commission proposes modifications to its formula-
based approach and related revisions to the proposed rules.
---------------------------------------------------------------------------
\1\ Notice of Proposed Rulemaking to Evaluate the Institutional
Cost Contribution Requirement for Competitive Products, February 8,
2018 (Order No. 4402). The Notice of Proposed Rulemaking to Evaluate
the Institutional Cost Contribution Requirement for Competitive
Products was published in the Federal Register on February 14, 2018.
See 83 FR 6758 (February 14, 2018).
---------------------------------------------------------------------------
II. Organization of Discussion
Section III of this Revised Notice of Proposed Rulemaking provides
an overview of 39 U.S.C. 3633 and a recap of the Commission's two
previous decisions concerning competitive products' appropriate share.
In addition, section III provides a synopsis of Order No. 4402,
including a brief summary of the formula-based approach previously
proposed by the Commission and that approach's compliance with the
elements set forth in 39 U.S.C. 3633(b). Section III also provides a
list of comments received in response to Order No. 4402.
In section IV, the Commission proposes modifications to Order No.
4402's formula-based approach. In conjunction with the proposed
modifications, the Commission discusses comments received in response
to Order No. 4402 that directly relate to a modification proposed in
this Order as well as several comments applicable to aspects of the
formula's calculation.\2\ As it did in Order No. 4402, the Commission
also analyzes its modified proposed formula pursuant to the
requirements of 39 U.S.C. 3633(b).
---------------------------------------------------------------------------
\2\ The Commission received a range of comments related to its
proposed formula-based approach and its analysis pursuant to the
elements of 39 U.S.C. 3633(b). The Commission has reviewed and
considered all comments received in response to Order No. 4402. For
the purposes of this Revised Notice of Proposed Rulemaking, the
Commission addresses those comments that relate to the formula
modifications the Commission is proposing in this Order. Comments
received in response to Order No. 4402 but not addressed in this
Order will be addressed in a subsequent order in this proceeding.
---------------------------------------------------------------------------
In section V, the Commission affirms its finding in Order No. 4402
pursuant to section 703(d) of the Postal Accountability and Enhancement
Act (PAEA).\3\
---------------------------------------------------------------------------
\3\ Uncodified section 703 of the PAEA, Public Law 109-435, 120
Stat. 3198 (2006), directs the Commission, when revising regulations
under 39 U.S.C. 3633, to consider subsequent events that affect the
continuing validity of a Federal Trade Commission (FTC) report that
analyzed the Postal Service's economic advantages and disadvantages
in the competitive product market when compared to private
competitors. See PAEA, 120 Stat. 3244; see also Federal Trade
Commission, Accounting for Laws that Apply Differently to the United
States Postal Service and its Private Competitors, December 2007
(FTC Report), available at: https://www.ftc.gov/sites/default/files/documents/reports/accounting-laws-apply-differently-united-states-postal-service-and-its-private-competitors-report/080116postal.pdf.
---------------------------------------------------------------------------
Section VI takes administrative steps to allow for comments on the
modifications to the proposed formula and related revisions to the
proposed rules by interested persons.
III. Background
A. Relevant Statutory Requirements
The PAEA requires that competitive products collectively cover what
the Commission determines to be an appropriate share of the Postal
Service's institutional costs. 39 U.S.C. 3633(a)(3).
The Commission is required to review the appropriate share
regulation at least every 5 years to determine if the contribution
requirement should be ``retained in its current form, modified, or
eliminated.'' See 39 U.S.C. 3633(b). In making such a determination,
the Commission is required to consider ``all relevant circumstances,
including the prevailing competitive conditions in the market, and the
degree to which any costs are uniquely or disproportionately associated
with any competitive products.'' 39 U.S.C. 3633(b). Thus, by its terms,
section 3633(b) establishes three separate elements that the Commission
must consider during each review: (1) The prevailing competitive
conditions in the market; (2) the degree to which any costs are
uniquely or disproportionately associated with competitive products;
and (3) all other relevant circumstances. See Order No. 4402 at 6.
B. Previous Commission Decisions
In promulgating its initial competitive product rules in Docket No.
RM2007-1, the Commission determined that basing competitive products'
minimum contribution on a percentage of total institutional costs was
easily understood and, in tying it to historic contribution at the
time, set the appropriate share at 5.5 percent.\4\
---------------------------------------------------------------------------
\4\ See Docket No. RM2007-1, Order Proposing Regulations to
Establish a System of Ratemaking, August 15, 2007, at 70 (Order No.
26); Docket No. RM2007-1, Order Establishing Ratemaking Regulations
for Market Dominant and Competitive Products, October 29, 2007, at
91, 138 (Order No. 43); see also Order No. 4402 at 6-7.
---------------------------------------------------------------------------
The Commission completed its first review of the appropriate share,
pursuant to 39 U.S.C. 3633(b), in Docket No. RM2012-3.\5\ After
considering the elements established by section 3633(b), the Commission
determined that the appropriate share should be retained at 5.5
percent. See generally Order No. 1449.
---------------------------------------------------------------------------
\5\ See Docket No. RM2012-3, Order Reviewing Competitive
Products' Appropriate Share Contribution to Institutional Costs,
August 23, 2012 (Order No. 1449); see also Order No. 4402 at 7-11.
---------------------------------------------------------------------------
C. Current Commission Review: Docket No. RM2017-1
1. Procedural History
On November 22, 2016, the Commission issued an Advance Notice of
Proposed Rulemaking, which established this docket as its second review
of the appropriate share pursuant to 39 U.S.C. 3633(b), appointed a
Public Representative, and provided interested persons with an
opportunity to comment.\6\ On February 8, 2018, after considering
initial and reply comments received, the Commission issued Order No.
4402, which responded to comments, presented a new formula-based
approach to setting the appropriate share, and provided another
opportunity for interested persons to submit comments. See generally
Order No. 4402.
---------------------------------------------------------------------------
\6\ Advance Notice of Proposed Rulemaking to Evaluate the
Institutional Cost Contribution Requirement for Competitive
Products, November 22, 2016 (Order No. 3624). The Advance Notice of
Proposed Rulemaking to Evaluate the Institutional Cost Contribution
Requirement for Competitive Products was published in the Federal
Register on November 29, 2016. See 81 FR 229 (November 29, 2016).
---------------------------------------------------------------------------
2. Order No. 4402
In Order No. 4402, the new formula-based approach proposed to set
the appropriate share through a dynamic formula, which would annually
update the appropriate share percentage based on market conditions. Id.
at 11-33.
a. Formula-Based Approach
The proposed formula-based approach used two components to annually
capture changes in the Postal Service's market power and in the overall
size of the competitive market: The Postal Service Lerner Index and the
Competitive Market Output. Id. at 15.
The purpose of the Postal Service Lerner Index was to measure the
Postal Service's market power within the competitive market. Id. at 16.
In Order No. 4402, the Commission noted that
[[Page 39941]]
market power is a competitor's ability to profitably set prices well
above costs with little chance that entry or expansion by other
competitors would erode such profits. Id. The Commission determined
that evaluating the Postal Service's market power allowed it to assess
whether competition was being preserved and whether the Postal Service
possessed any competitive advantage. Id.
The purpose of the second component of the proposed formula, the
Competitive Market Output, was to measure the overall size of the
competitive market. Id. at 22. The Commission proposed evaluating the
overall size of the market because doing so enabled the Postal
Service's market power to be placed into context relative to the market
as a whole. Id.
With the two components discussed above, the Commission proposed
calculating the appropriate share using the following formula: \7\
---------------------------------------------------------------------------
\7\ Id. at 29.
ASt+1 = ASt * (1 + %[Delta]LIt-1 +
%[Delta]CMOt-1)
I[fnof] t = 0 = FY 2007, AS = 5.5%
The Commission proposed measuring the year-over-year percentage change
in the Postal Service Lerner Index and Competitive Market Output,
weighting both components equally. Id. at 29-31. As proposed in Order
No. 4402, the formula's calculation was recursive with the Commission
proposing to begin the calculation in FY 2007, using an initial
appropriate share value of 5.5 percent. Id. at 31-32. The Commission
proposed adjusting the appropriate share annually by using the formula
to calculate the appropriate share for the upcoming fiscal year. Id. at
30. The appropriate share for each upcoming fiscal year would be
reported in the Commission's Annual Compliance Determination (ACD). Id.
b. Compliance With Statutory Requirements
As part of Order No. 4402, the Commission examined how its proposed
formula-based approach complied with section 3633(b) and accounted for
the requirements of that section: (1) The prevailing competitive
conditions in the market; (2) whether any costs are uniquely or
disproportionately associated with any competitive products; and (3)
other relevant circumstances. 39 U.S.C. 3633(b); Order No. 4402 at 34-
53. For prevailing competitive conditions and other relevant
circumstances, the Commission addressed the ways the proposed formula
captured the prevailing competitive conditions and other relevant
circumstances described in previous Commission decisions concerning the
appropriate share. Id. at 34-40, 45-51. In addition, the Commission
found that all costs uniquely or disproportionately associated with
competitive products were already attributed to those products under
the Commission's costing methodology.\8\
---------------------------------------------------------------------------
\8\ Order No. 4402 at 43-45. The Commission's analysis of ``the
degree to which any costs are uniquely or disproportionately
associated with any competitive products'' relied on current costing
methodologies approved in Docket No. RM2016-2. Id. at 40-45; see
Docket No. RM2016-2, Order Concerning United Parcel Service, Inc.'s
Proposed Changes to Postal Service Costing Methodologies (UPS
Proposals One, Two, and Three), September 9, 2016 (Order No. 3506).
UPS challenged the Commission's costing methodologies approved in
Order No. 3506 in the United States Court of Appeals for the
District of Columbia Circuit. See Petition for Review, United Parcel
Serv., Inc. v. Postal Reg. Comm'n, No. 16-1354 (D.C. Cir. filed Oct.
7, 2016). The Court issued its opinion on May 22, 2018. See United
Parcel Serv., Inc. v. Postal Reg. Comm'n, 890 F.3d 1053 (D.C. Cir.
2018) (UPS). In its opinion, the Court denied UPS's Petition for
Review and found that the Commission exercised reasonable judgment
in ``settling on a cost-attribution methodology that implements its
statutory mandate and falls well within the scope of its
considerable discretion.'' Id. at 1069. UPS petitioned for rehearing
en banc, which was denied by the United States Court of Appeals for
the District of Columbia Circuit. See Petition for Rehearing En
Banc, United Parcel Serv., Inc. v. Postal Reg, Comm'n, No. 16-1354
(D.C. Cir. filed July 6, 2018), denied per curiam, No. 16-1354 (D.C.
Cir. filed July 27, 2018).
---------------------------------------------------------------------------
c. Comments in Response to Order No. 4402
The Postal Service, the Public Representative, Amazon.com Services,
Inc. (Amazon), the Greeting Card Association (GCA), the Parcel Shippers
Association, Pitney Bowes Inc., United Parcel Service, Inc. (UPS),
Robert J. Shapiro, and the American Consumer Institute Center for
Citizen Research filed comments in response to Order No. 4402.\9\ In
addition, representatives for the Public Representative and UPS filed
declarations supporting comments on Order No. 4402.\10\
---------------------------------------------------------------------------
\9\ Comments of the United States Postal Service in Response to
Order No. 4402, April 16, 2018 (Postal Service Comments); Public
Representative Comments in Response to Notice of Proposed
Rulemaking, April 16, 2018 (PR Comments); Comments of Amazon.com
Services, Inc. on Order No. 4402, April 16, 2018 (Amazon Comments);
Comments of the Greeting Card Association, April 16, 2018 (GCA
Comments); Comments of the Parcel Shippers Association, April 16,
2018; Comments of Pitney Bowes Inc., April 16, 2018; Initial
Comments of United Parcel Service, Inc. on Notice of Proposed
Rulemaking to Evaluate the Institutional Cost Contribution
Requirement for Competitive Products, April 16, 2018 (UPS Comments);
Declaration of Robert J. Shapiro, April 16, 2018; Comments of
American Consumer Institute Center for Citizen Research Regarding
Docket No. RM2017-1 Submitted to the Postal Regulatory Commission,
April 16, 2018.
\10\ Declaration of Soiliou Daw Namoro for the Public
Representative, April 16, 2018 (Namoro Decl.); Declaration of J.
Gregory Sidak on Behalf of United Parcel Service, April 16, 2018
(Sidak Decl.). Soiliou Daw Namoro filed in support of the Public
Representative, and J. Gregory Sidak filed in support of UPS.
---------------------------------------------------------------------------
IV. Proposed Modified Formula and Commission Analysis
As noted above, in this Revised Notice of Proposed Rulemaking, the
Commission is proposing modifications to both the Postal Service Lerner
Index and the Competitive Market Output previously presented in Order
No. 4402. As discussed in more detail below, these proposed
modifications are made in response to comments received in response to
Order No. 4402. The Commission proposes modifications to the Postal
Service Lerner Index in order to address concerns related to the
aggregation of data used in its calculation, provide a better measure
of Postal Service market power, and more clearly distinguish the
Commission's component from a traditional Lerner index. The Commission
proposes modifications to the Competitive Market Output in order to
more explicitly incorporate Postal Service market share.
A. Modified Formula-Based Approach
In this section, the Commission reviews pertinent portions of Order
No. 4402, examines relevant comments, describes its proposed
modifications to both components, and discusses the resulting formula.
1. Modification to Postal Service Lerner Index
a. Order No. 4402
The Postal Service Lerner Index component was designed to gauge the
Postal Service's market power in the competitive market. Order No. 4402
at 15-16. The Commission determined that evaluating the Postal
Service's market power enables it to assess whether competition is
being preserved and whether the Postal Service possesses a competitive
advantage in the competitive market. Id. at 16. A Lerner index
quantitatively assesses market power for a given firm by measuring the
difference between the price charged by the firm for a particular
product and the marginal cost incurred by the firm in producing that
product. Id. at 17. In general, the further a firm is able to price its
product above marginal cost, the more market power the firm possesses.
Id.
In Order No. 4402, the Commission used a traditional Lerner index
as a
[[Page 39942]]
starting point and proposed to develop a measure of market power
specific to the Postal Service using Postal Service data. The
Commission noted that the Postal Service is a multi-product firm, with
each product having its own unique marginal cost and associated set of
prices. Id. Therefore, in order to develop a measure that would be
applicable to competitive products as a whole, the Commission proposed
using average competitive product marginal cost and average competitive
product price to calculate what it referred to as the Postal Service
Lerner Index. Id.
The Commission determined that marginal cost data for the Postal
Service's competitive products could be obtained from the Postal
Service's Cost and Revenue Analysis (CRA) report, which is submitted to
the Commission annually as part of the Postal Service's Annual
Compliance Report (ACR).\11\ The Commission uses the CRA report as an
input to the Postal Service Product Finances analysis (PFA), which is
produced each year as part of the Commission's ACD.\12\ Order No. 4402
at 18. The CRA report calculates marginal costs using volume-variable
costs, which are the costs of specific Postal Service operations that
vary with respect to relevant cost drivers. Id. The volume-variable
costs are then distributed to individual Postal Service products. Id.
Dividing the total volume-variable costs of a product by the product's
total volume results in unit volume-variable costs, which are
equivalent to marginal costs. Id. The Commission, therefore, proposed
to divide the sum of all competitive product volume-variable costs in
the PFA by the sum of all competitive product volume in order to
calculate the aggregate competitive product unit volume-variable cost.
Id. This number is equivalent to the average marginal cost for all
competitive products.
---------------------------------------------------------------------------
\11\ Order No. 4402 at 18; see 39 U.S.C. 3652.
\12\ See 39 U.S.C. 3653.
---------------------------------------------------------------------------
The Commission determined that the price variable could be obtained
using average revenue-per-piece, which incorporates all of the prices
for all of the Postal Service's competitive products. Id. The PFA
presents revenue data by product. Id. at 18-19. The Commission proposed
dividing the sum of all competitive product revenue by the sum of all
competitive product volume in order to calculate competitive product
average revenue-per-piece. Id. at 19. This number is equivalent to the
average price for all competitive products.
Using the two variables described above, the Commission developed
its proposed Postal Service Lerner Index, which consisted of the
following formula: \13\
---------------------------------------------------------------------------
\13\ Id.
[GRAPHIC] [TIFF OMITTED] TP13AU18.034
b. Comments
Multiple commenters address the proposed Postal Service Lerner
Index. Some of these commenters allege that the Postal Service Lerner
Index suffers from a number of defects resulting from the aggregation
of data. Specifically, UPS and Sidak assert that it is improper to
calculate the Postal Service Lerner Index using an average of the
marginal costs for each of the Postal Service's competitive products.
UPS Comments at 32; Sidak Decl. at 24-26. They contend that because the
Postal Service is a multi-product firm with different cost
characteristics for each of its products, averaging costs across
different products is misleading. Id. Sidak maintains that even if the
aggregate Postal Service Lerner Index is positive, the Lerner index for
an individual product could still be negative, which could enable the
Postal Service to engage in below-cost pricing for individual products.
Sidak Decl. at 24. Sidak states that, for a multi-product firm,
economists typically develop separate Lerner indices for each product.
Id.
UPS asserts that averaging product costs together could result in
distortions and instability in the Postal Service Lerner Index
following any future reclassifications of market dominant products as
competitive or any future changes within the competitive product mail
mix. UPS Comments at 32-33. UPS maintains that such changes would
result in the composition of products within the Postal Service Lerner
Index shifting for reasons unrelated to changes in market conditions.
Id. For example, if a market dominant product had its own Lerner index
with a value lower than the Postal Service Lerner Index (which is the
aggregate of all competitive products), and that market dominant
product were to be reclassified as a competitive product, then its
addition to the Postal Service Lerner Index would reduce the Postal
Service Lerner Index's overall value.
With regard to the Commission's proposed use of average revenue,
UPS and Sidak argue that it is improper to calculate the Postal Service
Lerner Index using average revenue as a measure of price. UPS Comments
at 33; Sidak Decl. at 28-31. Sidak asserts that average revenue is an
inaccurate measure of price for a firm that engages in price
discrimination, as he states the Postal Service does through its
offering of negotiated service agreements (NSAs).\14\ Under these
circumstances, he notes that as the quantity of a good that is sold
increases, the price of a marginal unit of that good will decrease more
quickly than average revenue will decrease.\15\ Sidak concludes that
average revenue can overstate price, and a Lerner index built on such
data can overstate the difference between price and marginal costs,
thereby serving as an inaccurate measure of market power.\16\
---------------------------------------------------------------------------
\14\ Sidak Decl. at 30. Price discrimination is a form of
nonlinear pricing where the same good is sold at different prices.
See Jeffrey Church & Roger Ware, Industrial Organization: A
Strategic Approach 157 (2000) (Church & Ware), available at: https://works.bepress.com/jeffrey_church/23/. The Postal Service regularly
enters into NSAs, which are contractual agreements between the
Postal Service and specific mailers providing for customized prices
and classifications in exchange for volume commitments by the
mailer.
\15\ Id. The Commission provides a simple example to explain
Sidak's concern. If the Postal Service were to sell 100 parcel
deliveries at $5 each to retail consumers, and then sell 200 parcel
deliveries at $3 each to a particular mailer pursuant to an NSA,
then the price of a marginal unit of parcel delivery would be $3
(because marginal price is defined as the price of the last unit
sold), but the average revenue for all 300 units sold would be
$3.67.
\16\ Id. Sidak does not argue that revenue in general is
inappropriate as a measure of price--only that average revenue is an
inappropriate measure of price because the Postal Service offers
NSAs. Id. at 28-31. Sidak does not suggest an alternative measure of
price to be used in this case.
---------------------------------------------------------------------------
c. Commission Analysis and Proposed Modification
After considering the comments received, the Commission proposes to
replace the Postal Service Lerner Index with an alternate measurement
the Commission labels as the Competitive Contribution Margin. The
Competitive Contribution Margin has two primary differences when
compared to the Postal Service Lerner Index: (1) It uses total
[[Page 39943]]
competitive product values rather than average competitive product
values; and (2) it uses competitive product attributable costs instead
of competitive product volume-variable costs. The formula for
calculating the Competitive Contribution Margin is as follows:
[GRAPHIC] [TIFF OMITTED] TP13AU18.035
This modification presents several benefits. First, it addresses an
apparent misunderstanding with the mathematical functioning of the
Postal Service Lerner Index as initially proposed by the Commission.
With regard to UPS's and Sidak's assertions that the Postal Service
Lerner Index inappropriately uses average revenue in place of price,
Namoro's declaration demonstrates that the use of averages has no
actual effect on the calculation. See Namoro Decl. at 6-7.
The Postal Service Lerner Index, as initially proposed by the
Commission, used revenue-per-piece (i.e., average revenue) and unit
volume-variable cost (i.e., average cost). Revenue-per-piece is
calculated by dividing total competitive product revenue by total
competitive product volume, and unit volume-variable cost is calculated
by dividing total competitive product volume-variable cost by total
competitive product volume.
Because every term is divided by volume, the volume terms cancel
each other out, which is mathematically demonstrated as follows:
[GRAPHIC] [TIFF OMITTED] TP13AU18.036
The final construction of the Postal Service Lerner Index shown
above is mathematically equivalent to the Postal Service Lerner Index
as originally proposed in Order No. 4402, but does not use averaging.
See id.; see also Order No. 4402 at 19. As demonstrated above,
averaging is immaterial to the calculation of this component. For that
reason, the Commission proposes to omit averaging and to use total
revenue for all competitive products in its modified component. Because
this modification does not affect what the component measures, the
modified component will continue to measure the market power of the
Postal Service's competitive products as a whole. At the same time, the
Commission recognizes that using total amounts departs somewhat from a
traditional calculation of a Lerner index, which is typically
calculated using unit cost and unit price.\17\ Therefore, the
Commission proposes to refer to the modified component as the
Competitive Contribution Margin to distinguish it from a traditional
Lerner index.
---------------------------------------------------------------------------
\17\ A traditional Lerner index is defined by the ratio of price
minus marginal cost to price. See Church & Ware at 31-36.
---------------------------------------------------------------------------
The second major benefit of this modification is that by using
total attributable costs, it more accurately reflects competitive
product costs than the Postal Service Lerner Index. The Postal Service
Lerner Index only included volume-variable costs, whereas the
Competitive Contribution Margin uses attributable costs, which include
volume-variable costs, product-specific costs, and inframarginal costs
calculated as part of each competitive product's incremental costs.\18\
In addition, by incorporating the inframarginal costs of competitive
products collectively, the Competitive Contribution Margin also
reflects costs which are not caused by any one competitive product, but
by competitive products as a whole. Reflecting all costs caused by
competitive products mitigates the risk of overstating the Postal
Service's market power in the competitive market because the
modification allows the component to more accurately measure the
relationship between cost and price.
---------------------------------------------------------------------------
\18\ See Order No. 3506 at 60 (directing Postal Service to begin
basing attributable costs for competitive products on incremental
costs, which include a portion of inframarginal costs).
---------------------------------------------------------------------------
The third benefit of this proposed modification is that it better
reflects modern economic literature on the subject of measuring market
power. As Sidak notes, ``[e]conomists routinely use the ratio of
`operating profits net of depreciation, provisions and an estimated
financial cost of capital [to] sales' as a proxy for a firm's Lerner
[i]ndex.'' \19\ Sidak estimates UPS's and FedEx's Lerner index values
for FY 2017 using each firm's operating profit-to-revenue ratio. Sidak
Decl. at 47. The Competitive Contribution Margin follows the same
calculation outlined in the economic literature cited to by Sidak,
determining the ratio of operating profit to revenue.\20\ This measure
is frequently referred to in economic literature as the price-cost
margin.
---------------------------------------------------------------------------
\19\ Sidak Decl. at 47, Figure 4 (citing Philippe Aghion et al.,
Competition and Innovation: An Inverted-U Relationship, 120 Q.J.
Econ. 701, 704 (2005); Frederick H. deB. Harris, Structure and
Price-Cost Performance Under Endogenous Profit Risk, 35 J. Indus.
Econ. 35, 43 (1986)).
\20\ The difference between total competitive product revenue
and total competitive product attributable costs constitutes the
profit derived from competitive products. Dividing this difference
by total competitive product revenue results in the profit-to-
revenue ratio that Sidak uses.
---------------------------------------------------------------------------
With regard to UPS's and Sidak's concerns that an index which
aggregates
[[Page 39944]]
total costs across multiple competitive products could be used to mask
below-cost pricing for individual competitive products, the Commission
finds that such a situation is, as a practical matter, highly unlikely
to occur. First, because the PAEA allows the Postal Service to retain
earnings, the Postal Service is incentivized to maximize profits on
competitive products. To price below-cost for individual competitive
products would be economically disadvantageous for the Postal Service.
As the Commission noted in Order No. 4402, a firm pricing below
marginal cost should suspend production in the short run, and if cost
or market characteristics do not change, exit the industry in the long
run. Order No. 4402 at 36 n.63. Second, an individual competitive
product that was priced below cost would violate 39 U.S.C. 3633(a)(2),
which requires each competitive product to recover its attributable
costs. See 39 U.S.C. 3633(a)(2). Such violations are addressed annually
in the ACD, with the Commission having authority to order appropriate
remedies.\21\
---------------------------------------------------------------------------
\21\ See, e.g., Docket No. ACR2007, Annual Compliance
Determination, March 27, 2008, at 112-13; Docket No. ACR2008, Annual
Compliance Determination, March 30, 2009, at 86-89; Docket No.
ACR2009, Annual Compliance Determination, March 29, 2010, at 117;
Docket No. ACR2010, Annual Compliance Determination, March 29, 2011,
at 139-40; Docket No. ACR2011, Annual Compliance Determination,
March 28, 2012, at 156-63; Docket No. ACR2012, Annual Compliance
Determination, March 28, 2013, at 162-72; Docket No. ACR2013, Annual
Compliance Determination, March 27, 2014, at 79-91; Docket No.
ACR2014, Annual Compliance Determination, March 27, 2015, at 72-82;
Docket No. ACR2015, Annual Compliance Determination, March 28, 2016,
at 79-92; Docket No. ACR2016, Annual Compliance Determination, March
28, 2017, at 80-88; Docket No. ACR2017, Annual Compliance
Determination, March 29, 2018, at 82-92 (FY 2017 ACD).
---------------------------------------------------------------------------
With respect to UPS's concern that the effects of future product
reclassifications or competitive product mail mix changes could result
in distortions, the Commission finds that although such a change would
alter the inputs to the calculation, the Competitive Contribution
Margin would accurately reflect the Postal Service's market power in
the expanded (or contracted) market that resulted from the change. For
example, if a market dominant product were to be re-classified as
competitive, the addition of that product to the competitive mail mix
would change both competitive products' total attributable costs and
total revenue. However, because the Competitive Contribution Margin is
calculated by subtracting total attributable costs from total revenue,
and dividing that number by total revenue, the result would continue to
indicate how much market power the Postal Service possessed after the
transfer.
Table IV-1 provides a comparison of annual changes in the
Competitive Contribution Margin and the Postal Service Lerner Index.
Table IV-1--Comparison of Competitive Contribution Margin and Postal Service Lerner Index
----------------------------------------------------------------------------------------------------------------
Percentage
Competitive change in Percentage change
Fiscal year Contribution Competitive Postal Service in Postal Service
Margin Contribution Lerner Index Lerner Index
Margin
----------------------------------------------------------------------------------------------------------------
FY 2007................................... 0.226 N/A 0.228 N/A
FY 2008................................... 0.213 -5.9 0.217 -5.1
FY 2009................................... 0.241 13.4 0.251 15.9
FY 2010................................... 0.279 15.7 0.298 18.6
FY 2011................................... 0.257 -7.9 0.276 -7.3
FY 2012................................... 0.266 3.7 0.275 -0.3
FY 2013................................... 0.281 5.5 0.290 5.4
FY 2014................................... 0.282 0.4 0.292 0.8
FY 2015................................... 0.275 -2.6 0.284 -2.7
FY 2016................................... 0.325 18.1 0.332 16.6
FY 2017................................... 0.329 1.3 0.356 7.5
----------------------------------------------------------------------------------------------------------------
As shown in Table IV-1, the growth and decline in the two measures
is generally consistent. Two divergences warrant discussion: FY 2012,
when the Postal Service Lerner Index declined while Competitive
Contribution Margin grew; and FY 2017, when the difference between the
Postal Service Lerner Index and Competitive Contribution Margin was
more than 6 percentage points.
As noted above, the Competitive Contribution Margin uses
attributable costs while the Postal Service Lerner Index uses only
volume-variable costs.\22\ In a given fiscal year, if the percentage
growth in attributable costs was greater than the percentage growth in
volume-variable costs, the Competitive Contribution Margin would grow
less than the Postal Service Lerner Index. If the percentage growth in
attributable costs was less than the percentage growth in volume-
variable costs, the Competitive Contribution Margin would grow more
than the Postal Service Lerner Index. Between FY 2011 and FY 2012,
volume-variable costs increased by 27 percent, while attributable costs
increased by 25 percent.\23\ Thus, the Competitive Contribution Margin
grew in FY 2012, while the Postal Service Lerner Index decreased.
---------------------------------------------------------------------------
\22\ For FY 2007 through FY 2016, attributable costs were
calculated as the sum of volume-variable costs and product-specific
fixed costs.
\23\ The smaller increase in attributable costs was caused by a
decrease in product-specific fixed costs of 42 percent. This
decrease in product-specific fixed costs was primarily driven by a
decrease in competitive product advertising costs.
---------------------------------------------------------------------------
In FY 2017, the Commission included a portion of inframarginal
costs in the calculation of attributable costs for the first time,
which increased the overall level of cost attribution.\24\ This
resulted in attributable costs growing 11 percent from FY 2016 to FY
2017, while volume-variable costs (which were not affected by this
methodological change) grew only 8 percent during the same period. This
produced an inverse situation to that which occurred in FY 2012--
because the growth in attributable costs was greater than volume-
variable costs, the Competitive Contribution Margin grew less than the
Postal Service Lerner Index.
---------------------------------------------------------------------------
\24\ See Order No. 3506 at 60.
---------------------------------------------------------------------------
These differences reflect how the Competitive Contribution Margin
more accurately measures the Postal Service's market power for
competitive products. Because the Competitive Contribution Margin
measures all costs caused by competitive products, including those that
cannot be attributed to any one
[[Page 39945]]
competitive product specifically, the Competitive Contribution Margin
provides a more complete view of the Postal Service's market power. For
that reason, the Commission proposes to replace the Postal Service
Lerner Index with the Competitive Contribution Margin in its revised
formula.
2. Modification to Competitive Market Output
a. Order No. 4402
The second component of the formula initially proposed by the
Commission was the Competitive Market Output, which was designed to
measure the overall size of the competitive market. Order No. 4402 at
22. The Commission proposed that evaluating the overall size of the
market provided context for assessing the prevailing competitive
conditions in the market and the Postal Service's market power. Id. The
Commission stated that the appropriate share requirement should balance
encouraging the Postal Service to increase competitive products'
contribution to institutional costs when the market is growing with the
need to adjust competitive products' pricing in the event of a market
decline. Id.
The Commission determined that the relevant market consisted of two
groups: The Postal Service's competitive products and ``similar
products'' offered by the Postal Service's competitors. Id. The
Commission proposed using revenue, rather than volume, to measure the
size of the overall market. Id. at 23. This was because revenue data
for all competitors were available and directly comparable, whereas
volume data were not uniformly available and would require frequent
adjustments. Id.
The Commission proposed obtaining the necessary revenue data for
the Postal Service's competitive products from the PFA, which the
Commission produces every year as part of its ACD. Id. The Commission
proposed obtaining the necessary revenue data for the Postal Service's
competitors from two surveys conducted by the United States Census
Bureau: The Quarterly Services Survey (QSS) and the Services Annual
Survey (SAS). Id. The methodology for collecting and aggregating these
data was described in Order No. 4402. Id. at 22-29.
Using the foregoing information, the Commission developed its
proposed Competitive Market Output measure, which consisted of the
following formula: \25\
---------------------------------------------------------------------------
\25\ See Order No. 4402 at 23.
Competitive Market Output = RevenueUSPS + RevenueC&M \26\
---------------------------------------------------------------------------
\26\ ``C&M'' stands for ``Couriers and Messengers,'' the name of
the relevant dataset for the Postal Service's competitors within the
Census Bureau data. See id. at 24.
---------------------------------------------------------------------------
b. Comments
Multiple commenters address the proposed Competitive Market Output
component. These comments can be broadly grouped into six different
areas.
First, the Public Representative and his declarant, Namoro, both
express concern that the Competitive Market Output component, as
proposed, disproportionately incorporates competitor revenue. Namoro
Decl. at 10-11; PR Comments at 5-6. Namoro explains that this is due to
the fact that not all competitor revenue within Competitive Market
Output is weighted by market share. Namoro Decl. at 10-11. As a result,
the Public Representative and Namoro assert that coordinated price
increases by the Postal Service's competitors could cause the required
appropriate share to increase, regardless of other market conditions.
Id. at 11; PR Comments at 5-6.
Second, several commenters note that the Competitive Market Output
as proposed does not incorporate the Postal Service's market share.
Sidak observes that the Competitive Market Output will not reflect
changes in market share; it will simply show the size of the overall
market. Sidak Decl. at 49-51. Namoro likewise posits that the
Competitive Market Output as proposed implicitly and incorrectly
assumes that ``the Postal Service's specific gains or losses from total
market expansion or market contraction are irrelevant to the
computation of the appropriate share[ ] . . . .'' Namoro Decl. at 3.
UPS argues that the appropriate share should take into account how much
the Postal Service's competitive products are growing within the
context of the overall market. UPS Comments at 35. The Postal Service
asserts that under the formula as proposed, the appropriate share would
not decrease if the Postal Service were to lose market share but the
measured Competitive Market Output did not also decrease. Postal
Service Comments at 20. The Postal Service states that a circumstance
where it loses market share without the Competitive Market Output
similarly decreasing is not merely theoretical. Id. If the Postal
Service's competitors were to begin competing more aggressively or
shippers and non-traditional competitors were to expand their delivery
operations, then the Competitive Market Output (which measures the
total size of the package delivery market) might remain the same even
as the Postal Service's individual share of the market decreased. Id.
at 20-21.
Third, UPS asserts that there is no economic basis for linking the
size of the overall competitive market (measured by revenue) with the
question of what the appropriate share should be. UPS Comments at 34.
UPS states this is because ``[n]either the Commission nor the Postal
Service ha[s] the ability to control what prices are charged by other
participants in the market,'' and considering market size alone ``does
not account for the possibility of customers making in-house
deliveries, which would not impact overall market volume but would
decrease [the Competitive Market Output] nonetheless.'' Id. at 34-35.
The Postal Service also notes this issue. It states that both the
Competitive Market Output and the appropriate share could increase
without necessarily reflecting additional market opportunities, for the
Postal Service or any other package delivery company, if there were to
be a market change towards greater self-delivery of packages by
shippers themselves. Postal Service Comments at 21.
Fourth, UPS and Sidak both criticize the Competitive Market Output
for measuring output in terms of revenue, as opposed to volume. UPS
Comments at 35; Sidak Decl. at 36-38. Sidak asserts that ``a firm's
costs are more directly a function of its unit volume than of its
revenue.'' Sidak Decl. at 36. Furthermore, Sidak maintains that
``[m]easuring output on the basis of revenue can fail to capture market
growth if competitive pressure decreases prices more rapidly than unit
volume increases, or if growth in volume is driven by below-cost
pricing.'' Id. Sidak notes that measuring industry output by unit
volume would be consistent with the approach taken by other regulatory
agencies. Id. at 36-38.
Fifth, the Postal Service criticizes the Competitive Market Output
for failing to take into account inflation, considering that the
Competitive Market Output constitutes an absolute measure of market
size by revenue, denominated in current dollars. Postal Service
Comments at 21. By presenting growth rates in the Competitive Market
Output based on revenues expressed in nominal dollars, rather than
constant dollars adjusted for inflation, the Postal Service maintains
that the Competitive Market Output includes purely inflationary
increases in revenue, demand, and market power. Id. The Postal Service
[[Page 39946]]
also asserts that if the Competitive Market Output were to grow more
slowly than inflation, the Competitive Market Output growth may not
accurately reflect growth in the Postal Service's ability to increase
competitive products' contribution to institutional costs because, in
such a situation, institutional costs (which are also subject to
inflation) would be increasing faster in real terms than the Postal
Service's competitive revenue. Id. at 21-22.
Sixth, the Postal Service asserts that the Competitive Market
Output fails to take into account differentiation between the Postal
Service's and its competitors' respective product offerings, which can
impact the ability of competitive products to contribute to
institutional costs.\27\
---------------------------------------------------------------------------
\27\ Postal Service Comments at 16. Although the Postal Service
does not explain this particular argument in detail, it appears to
suggest that to the extent the Postal Service's and its competitors'
products are not perfect substitutes for each other, those products
will not be in direct competition, and arguably should not be
considered part of the same market. Therefore, to the extent that
the Competitive Market Output includes such products in the same
market, it could be said to overstate the size of the market.
---------------------------------------------------------------------------
c. Commission Analysis and Proposed Modification
After considering the comments received, the Commission proposes to
replace the Competitive Market Output with an alternate measurement the
Commission labels the Competitive Growth Differential. Unlike the
Competitive Market Output, which sought to determine overall market
size, the Competitive Growth Differential assesses the growth or
decline of the Postal Service's market position from year-to-year. It
explicitly incorporates the Postal Service's market share and accounts
for inflation and whether market growth is structural or caused by
coordinated pricing by competitors. It is calculated using the
following equation:
Competitive Growth Differential = Market ShareUSPS *
(%[Delta]RevenueUSPS - %[Delta]RevenueC&M)
The Competitive Growth Differential is calculated by subtracting
the year-over-year percentage change in competitors' revenue from the
year-over-year percentage change in the Postal Service's competitive
product revenue to determine the Postal Service's growth relative to
that of its competitors, and multiplying the result by the Postal
Service's market share. The Postal Service's market share is determined
by dividing the Postal Service's total competitive product revenue by
the sum of the Postal Service's total competitive product revenue and
total competitor revenue, as depicted in the following formula:
[GRAPHIC] [TIFF OMITTED] TP13AU18.037
As with the Competitive Market Output, the Competitive Growth
Differential is measured using revenue, rather than volume. As
explained in Order No. 4402, the Commission selects revenue data
because volume data would need to be adjusted for intra-industry
transactions, while revenue data can be used directly, without
adjustment.\28\ Additionally, revenue data are also available for all
firms in the relevant market through publicly available sources,
whereas volume data for the Postal Service's competitors are not
publicly available. Id.
---------------------------------------------------------------------------
\28\ See Order No. 4402 at 23. An example of an intra-industry
transaction is a Postal Service competitor transporting a package
from a sender in California to a recipient's destination delivery
unit (i.e., the Postal Service facility where mail carriers depart
for local mail delivery) in New York. The Postal Service would then
deliver the package to the recipient (i.e., last-mile delivery).
---------------------------------------------------------------------------
As with the Competitive Market Output, revenue data for the Postal
Service are obtained from the PFA, and revenue data for the Postal
Service's competitors are obtained from Census Bureau data--
specifically the QSS and SAS survey data. Unlike the Competitive Market
Output, revenue data under the Competitive Growth Differential are
adjusted for inflation, using the Consumer Price Index for All Urban
Consumers (CPI-U) as the deflator.\29\ CPI-U data are obtained from the
Bureau of Labor Statistics (BLS).\30\ The Commission indexes the CPI-U
data to FY 2007; that is, FY 2007 constitutes the base year for any
inflation adjustment. This aligns the CPI-U data with the beginning
year for the Commission's proposed formula.\31\
---------------------------------------------------------------------------
\29\ The CPI-U is a measure of the average change over time in
the prices paid by urban consumers for a market basket of consumer
goods and services. See Bureau of Labor Statistics, Consumer Price
Index, Frequently Asked Questions, available at: https://www.bls.gov/cpi/questions-and-answers.htm.
\30\ Bureau of Labor Statistics, Consumer Price Index--All Urban
Consumers (Series ID CUUR0000SA0),'' available at: https://data.bls.gov/timeseries/CUUR0000SA0.
\31\ See Order No. 4402 at 32. For additional discussion of the
beginning year of the Commission's formula, see section IV.A.3.c,
infra.
---------------------------------------------------------------------------
The Competitive Growth Differential better reflects the Postal
Service's position in the overall competitive market and addresses the
concerns raised by commenters discussed above. First, the change to the
Competitive Growth Differential eliminates the disproportionate
inclusion of competitor revenue from the component's underlying
equation. To illustrate this, the Commission starts with the formula
for calculating the year-over-year percentage change in Competitive
Market Output (which was an input into the formula as initially
proposed in Order No. 4402): \32\
---------------------------------------------------------------------------
\32\ This equation and all equations in this section are
calculated for t for simplicity of demonstration, while the input
(i.e., when using the formula to determine the appropriate share) is
calculated for t-1.
[GRAPHIC] [TIFF OMITTED] TP13AU18.038
[[Page 39947]]
Although not explicitly depicted in the formula, both the change in
Postal Service revenue and the change in competitor revenue are
weighted by their respective market shares. This is because an
aggregate rate of growth is not equivalent to the sum of individual
rates of growth.\33\ The formula is therefore mathematically equivalent
to the following:
---------------------------------------------------------------------------
\33\ A simple example can be used to demonstrate why this is the
case. Consider an entity with two products, one generating revenue
of $100,000 in FY 2017 and $105,000 in FY 2018 (a 5-percent year-
over-year increase) and the other generating revenue of $50,000 in
FY 2017 and $55,000 in FY 2018 (a 10-percent year-over-year
increase). If the entity were trying to calculate the aggregate rate
of revenue growth, it would be incorrect to add the individual rates
of growth (i.e., 5 percent for the first product and 10 percent for
the second product = 15 percent total). Instead, the entity would
calculate each product's share of total revenue (i.e., $100,000/
$150,000 = 66 percent for the first product and $50,000/$150,000 =
34 percent for the second product), and then multiply each product's
share of total revenue by the percentage revenue change (i.e., 66
percent * 5 percent = 3.3 percent for the first product, and 34
percent * 10 percent = 3.4 percent for the second product). The
final step would be to add the two numbers to calculate the
aggregate rate of revenue growth for the entity (i.e., 3.3 percent +
3.4 percent = 6.7 percent).
%[Delta]Competitive Market Output
= (Market ShareUSPS * %[Delta]RevenueUSPS + ((1 - Market ShareUSPS)
* (%[Delta]Revenue)C&M) \34\
---------------------------------------------------------------------------
\34\ For a rigorous demonstration of this transformation, see
Namoro Decl. at 11-13, reproduced in Library Reference PRC-LR-
RM2017-1/2.
Weighting by market share is necessary in order to incorporate the
relative contribution of each source of revenue growth to the overall
growth. As Library Reference PRC-LR-RM2017-1/2 illustrates, the year-
over-year percentage change in the Competitive Market Output is
equivalent to the year-over-year percentage change in the Postal
Service's revenue, weighted by the Postal Service's market share, plus
the year-over-year percentage change in competitors' revenue, weighted
by competitors' market share.\35\ In order to demonstrate how this
equation over-incorporates competitor revenue, it is helpful to state
its terms differently. The terms of the equation can be mathematically
rewritten as follows:
---------------------------------------------------------------------------
\35\ Competitors' market share is determined by calculating 1 -
Market ShareUSPS. This constitutes the residual left over after the
Postal Service's market share has been determined.
%[Delta]Competitive Market Output
= ((Market ShareUSPS) * (%[Delta]RevenueUSPS - %[Delta]RevenueC&M))
+ (%[Delta]RevenueC&M) \36\
---------------------------------------------------------------------------
\36\ This formula is the result of a three-step transformation
from the formula directly above it. The three-step transformation is
demonstrated in detail in Library Reference PRC-LR-RM2017-1/2.
This construction of the Competitive Market Output growth rate
equation is mathematically equivalent to the previous construction and
demonstrates that growth in Competitive Market Output constitutes the
sum of two terms: The market share weighted difference in revenue
growth between the Postal Service and its competitors; and the
unweighted growth in competitor revenue. It is this second term (+
(%[Delta]RevenueC&M)) that results in the disproportionate
incorporation of competitor revenue because the growth in competitor
revenue is not weighted by market share. The Competitive Growth
Differential removes the second term, thereby resolving the problem of
disproportionate incorporation of competitor revenue.\37\ Eliminating
the disproportionate incorporation of competitor revenue by adopting
the Competitive Growth Differential addresses the concerns raised by
the Public Representative and Namoro that competitors' pricing
decisions alone could influence the appropriate share.
---------------------------------------------------------------------------
\37\ The Commission notes that this adjustment was identified as
a possible solution by Namoro in his declaration. See Namoro Decl.
at 17 n. 12.
---------------------------------------------------------------------------
This modification also changes the nature of the component from a
measure of overall market size to a measure of the Postal Service's
market position because the modification captures the change in the
size of the Postal Service's competitive business relative to that of
the Postal Service's competitors.
Additionally, the Competitive Growth Differential directly
incorporates the Postal Service's market share into the appropriate
share calculation, which addresses comments that the Competitive Market
Output failed to consider the Postal Service's market share.\38\ The
Competitive Growth Differential directly incorporates the Postal
Service's market share as a weight. This ensures that any change in the
appropriate share due to changes in the Competitive Growth Differential
are not solely driven by growth in the overall market but are also
reflective of whether those changes give the Postal Service greater (or
reduced) market share. This is important because if both the Postal
Service's and its competitors' respective revenues increase but the
Postal Service's market share remains the same, the Postal Service's
relative position in the market may not have changed. With the
Competitive Growth Differential, the Commission's proposed formula will
now reflect this. Similarly, the change from the Competitive Market
Output to the Competitive Growth Differential will prevent the scenario
identified by the Postal Service in which, despite the Postal Service
having lost market share, the appropriate share requirement may not
decrease due to the size of the overall market remaining unchanged.
---------------------------------------------------------------------------
\38\ The Commission found in Order No. 4402 that market share
was indirectly incorporated into the Competitive Market Output
because any large shift in revenue share between the Postal Service
and its competitors would be reflected in the Competitive Market
Output. Order No. 4402 at 38-39. Market share is also indirectly
incorporated into the Competitive Market Output because determining
growth rates for the Competitive Market Output implicitly requires a
determination of the Postal Service's market share, as demonstrated
in Library Reference PRC-LR-RM2017-1/2.
---------------------------------------------------------------------------
With regard to UPS's assertion that there is no economic basis for
linking the size of the overall competitive market to the appropriate
share, the Commission reiterates its explanation in Order No. 4402 that
evaluating the overall size of the market provides context for
assessing prevailing competitive conditions. See id. at 22. The size of
the market serves as an indicator of how healthy the market is, both
when the market is considered in isolation and when the market is
considered relative to the broader economy. Evaluating the overall size
of the market is also necessary to determine the relative shares of the
competitors in it. For these reasons, it remains appropriate to
consider the overall size of the competitive market, as well as the
Postal Service's position in the market, as relevant to the appropriate
share.
As discussed above, the Competitive Growth Differential tracks
changes in the market more accurately than the Competitive Market
Output. It accomplishes this by using real revenue growth instead of
nominal revenue growth. The Commission agrees with the Postal Service's
suggestion that taking into account inflation will improve this
component of the formula. Without such an adjustment, the formula could
interpret inflationary changes in the market as market growth.
Relatedly, with regard to UPS's and Sidak's criticisms of this
component for measuring output in terms of revenue, it is true that
there are circumstances in which using revenue as a measure of output
could be misleading, such as when a firm is attempting to strategically
price its products at a low level in order to gain market share.
However, because the Competitive Growth Differential accounts for
inflation, those circumstances do not apply here. Even if the Postal
Service or its competitors were to engage in strategic pricing in order
to gain market share, causing revenue to diverge from volume, as long
as revenue is measured in real terms, the Competitive Growth
[[Page 39948]]
Differential would accurately reflect the Postal Service's relative
position in the market.\39\
---------------------------------------------------------------------------
\39\ With regard to Sidak's assertion that measuring industry
output by volume would be more consistent with practice in other
agencies, the Commission notes that the use of revenue to determine
output is consistent with the methodology employed by agencies such
as the United States Department of Commerce, which uses revenue as
an initial measure of output when calculating Gross Domestic Product
(GDP). GDP is the total expenditure on the economy's output of goods
and services. See N. Gregory Mankiw, Macroeconomics 18, 27 (7th ed.
2010). For information on the use of revenue in calculating GDP, see
Bureau of Economic Analysis, Concepts and Methods of the U.S.
National Income and Product Accounts, November 2017, at 4-9, 5-30,
available at: https://www.bea.gov/national/pdf/all-chapters.pdf.
---------------------------------------------------------------------------
The Postal Service's concern that this component fails to directly
consider product differentiation is mitigated by the overarching
similarities between the Postal Service's and its competitors'
products. Furthermore, product differentiation would be reflected in
the Competitive Growth Differential because changes in product
differentiation will affect the relative growth in revenue for the
Postal Service compared to its competitors. This is because if the
Postal Service's and its competitors' products became less and less
interchangeable to the point that they were occupying different markets
with different characteristics, those products' growth rates would be
likely to diverge, resulting in greater changes in the Competitive
Growth Differential. In addition, such differentiation would be
reflected by larger increases in the Competitive Contribution Margin
because that index measures the market power of the Postal Service; and
to the extent that the Postal Service has fewer competitors, it will
have greater market power. Further, if differentiation between the
Postal Service's and its competitors' products were to occur such that
the products were no longer considered to constitute the same market,
the 5-year review of the appropriate share mandated by 39 U.S.C.
3633(b) would allow the Commission to examine whether the data obtained
from Census Bureau continues to be an appropriate measure of
competitors' revenue.\40\
---------------------------------------------------------------------------
\40\ Should a change be necessary in advance of the 5-year
review, the Commission is also permitted to revise its regulations
when circumstances warrant. See 39 U.S.C. 3633(a); Order No. 1449 at
13.
---------------------------------------------------------------------------
The Competitive Market Output and Competitive Growth Differential
results for each fiscal year since the PAEA was enacted are reported in
Table IV-2 below.
Table IV-2--Comparison of Annual Changes in Competitive Market Output
Growth and Competitive Growth Differential \41\
------------------------------------------------------------------------
Competitive
Competitive growth
Fiscal year market output differential
growth (%) (%)
------------------------------------------------------------------------
FY 2007................................. N/A N/A
FY 2008................................. -1.5 0.7
FY 2009................................. -13.9 1.2
FY 2010................................. -0.8 0.9
FY 2011................................. 5.3 -0.2
FY 2012................................. 6.4 2.7
FY 2013................................. 5.0 2.5
FY 2014................................. 4.7 1.2
FY 2015................................. 6.5 0.2
FY 2016................................. 5.1 1.4
FY 2017................................. 6.3 1.1
------------------------------------------------------------------------
The Competitive Growth Differential values differ substantially
from the Competitive Market Output values because they measure
different things: The Competitive Market Output measures absolute
growth in the market, whereas the Competitive Growth Differential
measures the Postal Service's growth relative to that of its
competitors.
---------------------------------------------------------------------------
\41\ Because the Competitive Growth Differential evaluates
relative growth rather than absolute growth, it is inappropriate to
include the absolute Competitive Market Output values in this table.
No corresponding absolute Competitive Growth Differential values
exist.
---------------------------------------------------------------------------
For example, in FY 2008, FY 2009, and FY 2010, the Competitive
Market Output decreased and the Competitive Growth Differential
increased. This occurred because the Postal Service maintained (and in
some years, increased) its competitive product output despite a global
financial crisis, both through NSAs and the reclassification of certain
market dominant products as competitive. As such, the Postal Service
was able to improve its market position relative to its competitors,
even as the overall market declined. In FY 2011, the Competitive Growth
Differential was negative because the Postal Service's competitive
revenue displayed no material growth, while competitor revenue, and
hence the overall market, grew. This demonstrates that the Competitive
Growth Differential reflects the source of the growth in the market in
ways that the Competitive Market Output did not. Subsequent fiscal
years reflect similar differences, with the Competitive Growth
Differential better reflecting the Postal Service's market position in
the overall competitive market than the Competitive Market Output
would.
In the next section, the Commission discusses the formula proposed
in Order No. 4402, as well as specific comments received related to the
operation of the formula. The Commission then describes how the two
modified components, the Competitive Contribution Margin and the
Competitive Growth Differential, are incorporated into the Commission's
proposed formula to calculate the appropriate share.
3. Resulting Formula
a. Order No. 4402
In Order No. 4402, the Commission proposed calculating the
appropriate share using the following formula: \42\
---------------------------------------------------------------------------
\42\ Order No. 4402 at 29.
ATt+1 = ASt * (1 + %[Delta]LIt-1 +
%[Delta]CMOt-1)
If t = 0 = FY 2007, AS = 5.5%
[[Page 39949]]
Where,
AS = Appropriate Share \43\
---------------------------------------------------------------------------
\43\ This figure would be expressed as a percentage and rounded
to one decimal place for simplicity and consistency with the
Commission's past practice of expressing an appropriate share using
one decimal place. Id. at 29 n.52.
---------------------------------------------------------------------------
LI = Postal Service Lerner Index
CMO = Competitive Market Output
t = Fiscal Year
As noted above, under the previously proposed formula, the
Commission would have calculated the year-over-year percentage changes
for both the Postal Service Lerner Index and Competitive Market Output
components. Id. at 31; see section III.C.2.a, supra. In order to
calculate an upcoming fiscal year's appropriate share percentage
(ASt+1), the formula multiplied the sum of the percentage
changes in the Postal Service Lerner Index and the Competitive Market
Output from the previous fiscal year \44\ (1 + %[Delta]LIt-1
+ %[Delta]CMOt-1) by the current fiscal year's appropriate
share (ASt). Order No. 4402 at 30. In addition, both components were
given equal weight in the calculation in order to balance changes in
the competitive market with changes in the Postal Service's market
power. Id. at 29-30.
---------------------------------------------------------------------------
\44\ As noted in Order No. 4402, the ``1 +'' is a necessary
mathematical concept for any percentage change formula in order to
incorporate the pre-existing value being changed. Id. at 30 n.54;
see Jagdish Arya & Robin Lardner, Mathematical Analysis for Business
and Economics 202-03 (2d ed. 1985).
---------------------------------------------------------------------------
In order to calculate the appropriate share for the current fiscal
year, the Commission needed to determine the beginning appropriate
share percentage (AS) and the beginning fiscal year (t). The Commission
proposed to begin the calculation in FY 2007, when the PAEA was
enacted, and set the initial appropriate share value at 5.5 percent,
which was the appropriate share initially set by the Commission. Id. at
32. Both beginning values were chosen to allow for incorporation of the
changes in the competitive market in the years since the PAEA's
enactment. Id. Using FY 2007 and the 5.5-percent appropriate share as
the beginning point of the formula's calculation, the Commission used
the cumulative formula results from FY 2008 through FY 2018 in order to
reach FY 2019's proposed appropriate share (10.8 percent). Id. at 33.
In Order No. 4402, the Commission proposed adjusting the
appropriate share annually by using the formula to calculate the
appropriate share for the upcoming fiscal year. Id. at 30. Due to the
timing of when all necessary data were available, the Commission
proposed that the appropriate share would be reported as part of the
Commission's ACD issued each year in March and would take effect at the
beginning of the following fiscal year on October 1. Id.
b. Comments Concerning Beginning Appropriate Share, Beginning Fiscal
Year, and the Weighting of Components
In response to Order No. 4402, the Commission received comments
from several parties concerning the beginning appropriate share,
beginning fiscal year, and the weighting of the two components of the
formula. As these comments relate directly to the modified formula as
well as the previously proposed formula, the Commission discusses the
comments received on those three topics in this section.
i. Beginning Appropriate Share
UPS contends that using 5.5 percent as the beginning appropriate
share percentage is ``irrational'' because the initial 5.5 percent
appropriate share was an ``intentionally low'' figure and was based on
different analysis. UPS Comments at 36. UPS states that the initial 5.5
percent was set based on factors, such as small Postal Service market
share and the risk of setting appropriate share too high, and was
intended to provide flexibility to the Postal Service. Id. UPS
maintains ``[t]hese concerns have no bearing today.'' Id.
In the Order No. 4402, the Commission proposed that the appropriate
share be modified to better reflect the modern competitive market that
had exhibited changes since the Commission's last appropriate share
review and the PAEA's enactment. Order No. 4402 at 12. UPS interprets
this as Commission recognition that the 5.5-percent appropriate share
level is ``too low given current market conditions'' and thus questions
its use as the beginning value for the Commission's calculation of the
appropriate share. UPS Comments at 37. UPS contends that if the
Commission is increasing the appropriate share from 5.5 percent to
better reflect current market conditions, the beginning value of the
appropriate share calculation should not be 5.5 percent and instead
should reflect current market conditions. Id. For these reasons, UPS
recommends the Commission use the average revenue share of Postal
Service competitive products over the last 3 fiscal years (26.6
percent) as the beginning value of the appropriate share (AS). Id. at
39-40.
ii. Beginning Fiscal Year
UPS and the Postal Service address the beginning fiscal year used
in the proposed formula in their comments. In recommending the
Commission use 26.6 percent as the beginning value of the appropriate
share, UPS notes that percentage should be considered ``in the
Commission's formula for 2018 and onwards,'' which implies that UPS is
recommending the Commission change the beginning fiscal year (t) to FY
2018. Id. at 40.
The Postal Service recommends that the Commission eliminate or
reduce the appropriate share. Postal Service Comments at 3-8. However,
if the Commission retains the formula, the Postal Service alternatively
recommends that the Commission change the formula's beginning fiscal
year (t)to FY 2017. Id. at 23-24. The Postal Service contends there is
``no basis for applying the new formula beginning in FY 2007 and
continuing forward on a cumulative basis.'' Id. at 23.
In Order No. 4402, the Commission stated that the formula's
calculation, beginning in FY 2007, would be recursive in order to
capture the cumulative effects of changes in prevailing competitive
conditions in the market on the appropriate share. Order No. 4402 at
31-32. The Postal Service states that the current prevailing
competitive conditions are already captured by the proposed formula's
two components and do not need to be captured by beginning the
formula's calculation in FY 2007. Postal Service Comments at 23-24. In
addition, the Postal Service notes that the formula produces a
hypothetical appropriate share for each fiscal year between FY 2007 and
FY 2017, and that the use of those figures is ``inappropriate'' and
``arbitrary'' because the actual appropriate share for those same
fiscal years are known.\45\ For these reasons, the Postal Service
maintains that the beginning fiscal year (t)``should be FY 2017, the
most recent year in which the appropriate share requirement was a fixed
5.5 percent,'' or in the alternative, FY 2012, the most recent time the
Commission reviewed the appropriate share. Postal Service Comments at
23.
---------------------------------------------------------------------------
\45\ Id. at 24. The ``hypothetical'' appropriate shares the
Postal Service references can be found in Order No. 4402 at 33,
Table IV-6, column ``Appropriate Share for the Following Year
(ASt+1).''
---------------------------------------------------------------------------
iii. Weighting of the Components
Related to the Commission's equal weighting of both components,
Sidak asserts that the Commission's decision is an arbitrary one. Sidak
Decl. at 39. He maintains the Commission provides no
[[Page 39950]]
reasonable explanation for the equal weighting of the components. Id.
Sidak contends that the Commission failed to evaluate whether the two
components are endogenous, whether a correlation exists between the two
components and attributable costs, or how the formula would evolve
under alternative weights. Id. He suggests the Commission should have
``conduct[ed] some research and analysis to find the correct ratio'' of
the two components. Id.
c. Commission Analysis and Modified Formula
After consideration of the comments received, the Commission elects
to maintain Order No. 4402's approach to the beginning appropriate
share, the beginning fiscal year, and the weighting of components. In
this section, the Commission initially discusses the modified formula's
configuration and then provides its analysis of the commenters'
recommendations.
Based on the proposed modifications to both components discussed in
sections III.A.1 and III.A.2, supra, the Commission proposes to
calculate the appropriate share using the following modified formula:
ASt+1 = AS * (1 + %[Delta]CCMt-1)
If t = 0 = FY 2007, AS = 5.5%
Where:
AS = Appropriate Share \46\
---------------------------------------------------------------------------
\46\ This figure continues to be expressed as a percentage and
rounded to one decimal place for simplicity and consistency with the
Commission's past practice of expressing an appropriate share using
one decimal place.
---------------------------------------------------------------------------
CCM = Competitive Contribution Margin
CGD = Competitive Growth Differential
t = Fiscal Year
Procedurally, the Commission proposes that the appropriate share be
adjusted annually through the same process as proposed in Order No.
4402. Under that process, the appropriate share would be adjusted
annually by using the formula to calculate the minimum appropriate
share for the upcoming fiscal year.\47\ The Commission also retains
that the new appropriate share level for the upcoming fiscal year would
be reported as part of the Commission's ACD.\48\
---------------------------------------------------------------------------
\47\ In response to Order No. 4402, GCA requested the Commission
confirm that, despite the use of its formula-based approach, the
appropriate share continues to act as a minimum contribution level
or floor, to be exceeded, if possible. GCA Comments at 1-2. As noted
in Order No. 4402, ``the Commission has and continues to view the
appropriate share as a minimum requirement.'' Order No. 4402 at 81;
see id. at 6 (citing Order No. 26 at 72). The Commission continues
to view the appropriate share as a minimum requirement. The minimum
requirement nature of the appropriate share is embodied in the
proposed rule itself, which states ``. . . the appropriate share of
institutional costs to be recovered from competitive products
collectively, at a minimum, will be calculated using the following
formula. . . .'' See Order No. 4402, Attachment A at 1.
\48\ See Order No. 4402 at 30. It is important to note that, as
recently as its FY 2017 ACD, the Commission has stated the
appropriate share requirement of 39 U.S.C. 3633(a)(3) applies to the
Postal Service annually. See FY 2017 ACD at 92-93. Thus, to comply
with 39 U.S.C. 3633(a)(3), the Postal Service's competitive products
must collectively cover the Commission-determined appropriate share
of institutional costs as set forth in 39 CFR 3015.7(c) in each
fiscal year. See id. Although the Postal Service may exceed this
minimum contribution level, any contribution that exceeds the
minimum level cannot be used as a form of ``prepayment'' for future
fiscal years. See id.
---------------------------------------------------------------------------
In order to calculate an upcoming fiscal year's appropriate share
percentage (ASt+1), the modified formula multiplies the sum
of the Competitive Growth Differential and the percentage change in the
Competitive Contribution Margin, (1 + %[Delta]CCMt-1 +
CGDt-1),\49\ by the current fiscal year's appropriate share
(ASt). The modified formula continues to be recursive in nature in
order to incorporate year-over-year changes in the competitive market.
See Order No. 4402 at 31.
---------------------------------------------------------------------------
\49\ See n.44, supra.
---------------------------------------------------------------------------
Thus, as an example of how the modified formula functions, if the
following conditions hold:
Current year appropriate share is 5.5 percent
(ASt+1)
Competitive Contribution Margin grew by 6 percent in the
prior year (%[Delta]CCMt-1)
Competitive Growth Differential \50\ was 0.4 percent when:
---------------------------------------------------------------------------
\50\ As discussed above, the Competitive Growth Differential is
calculated as follows: Market ShareUSPS * (%[Delta]RevenuesUSPS -
%[Delta]RevenuesC&M). See section IV.2.c, supra.
--Postal Service revenue grew 5 percent in the prior year
(%[Delta]RevenueUSPS)
--Competitor revenue grew 3 percent in the prior year
(%[Delta]RevenueC&M)
--Postal Service market share was 20 percent (ShareUSPS)
Then the appropriate share for the next year is calculated as follows:
Appropriate Share = 0.055* (1 + 0.06 + (0.2 *(0.05 - 0.03))) = 0.059 or
5.9%
Under this scenario, the next year's appropriate share would be 5.9
percent. As noted above, this result will be the starting point for
calculating the appropriate share for the following year.
Using 5.9 percent as the starting point for calculating the
appropriate share for the following year (ASt=1), if the following
conditions hold:
Competitive Contribution Margin declined by 1 percent in
the prior year (%[Delta]CCMt-1)
Competitive Growth Differential was 2.2 percent, when:
--Postal Service revenue grew 6 percent in the prior year
(%[Delta]RevenueUSPS)
--Competitor revenue declined 4 percent in the prior year
(%[Delta]RevenueC&M)
--Postal Service market share was 22 percent (ShareUSPS)
Then the appropriate share for the next year is calculated as follows:
Appropriate Share = 0.059 * (1 - 0.01 + (0.22 * (0.06 - (-0.04))))
= 0.06 or 6.0%
Under this scenario, the next year's appropriate share would be 6.0
percent and would become the starting point for calculating the
appropriate share for the next year.
As it relates to comments received concerning the beginning
appropriate share and beginning fiscal year, the Commission finds that
it is appropriate to use 5.5 percent as the beginning appropriate share
and FY 2007 as the beginning fiscal year when calculating the modified
formula. Those beginning values allow the resulting appropriate share
to capture the impact of market fluctuations on the appropriate share
over time and moving forward.
The Commission's selection of 5.5 percent as the beginning
appropriate share does not imply that the Commission believes the
initial 5.5 percent set in Docket No. RM2007-1 was ``too low'' or
``inadequate'' as UPS suggests. See UPS Comments at 37. To the
contrary, the initial 5.5 percent appropriate share was reasonably
based on historical contribution. Order No. 4402 at 7. However, since
the PAEA's enactment, the Postal Service, competitors, and market
conditions have changed, and the goal of the formula-based approach is
to better capture these changes both historically and moving forward.
As a result, UPS's proposed use of Postal Service competitive products'
revenue share would be inappropriate because it does not appropriately
reflect market conditions in FY 2007 and subsequent years. In addition,
the use of revenue share to begin the calculation of the formula is
improper for the reasons discussed by the Commission in Order No. 4402
when it rejected using Postal Service competitive products' revenue
share to set the appropriate share. See Order No. 4402 at 82. Postal
Service competitive products' share of revenue is not reflective of
market conditions, the elements of 39 U.S.C. 3633(b), and Commission
precedent. Id. As discussed in Order No. 4402, competitive products'
share of revenue is driven in large part by market dominant revenue,
which has been declining due to a decline in demand for market dominant
products. Id. As a result of declining market dominant demand and
revenue, the competitive revenue share has
[[Page 39951]]
increased and is likely to continue to increase. However, this increase
in revenue share has little do with the criteria of 39 U.S.C. 3633(b)
that drive the determination of the appropriate share. As a result, use
of revenue share would be inappropriate because such use would allow
the appropriate share to be substantially impacted by factors unrelated
to the prevailing market conditions and other relevant circumstances
required pursuant to 39 U.S.C. 3633(b).
Additionally, it would be inappropriate to begin the formula's
calculation in FYs 2012, 2017, or 2018, as the Postal Service and UPS
respectively suggest. Calculating the appropriate share beginning in
any fiscal year other than FY 2007 would result in the Commission
disregarding the cumulative impact that changes in market have had on
the initial 5.5 percent appropriate share in the years since the PAEA's
enactment. The proposed formula's calculation incorporates the changes
from those fiscal years, a necessary action to better capture the
impact that changes in market conditions have had on the appropriate
share.
As noted above, the Postal Service makes two specific critiques
regarding the use of FY 2007 as the beginning fiscal year. The Postal
Service contends that the two components themselves reflect current
prevailing competitive conditions, leaving no reason to begin the
formula's calculation in FY 2007 in order to capture historical market
changes. Although it is true both components capture changes in
prevailing competitive conditions in the market,\51\ the beginning
fiscal year serves a different purpose. The components, as applied
through the formula, capture market changes, including prevailing
competitive conditions, over a single fiscal year. However, they do not
capture the prevailing competitive conditions in the market as they
have evolved since the PAEA's enactment. As the Commission explained in
Order No. 4402, it is appropriate to set FY 2007 as the beginning year
for the formula because the prevailing competitive conditions in the
market, as well as other relevant circumstances, have changed since FY
2007. Order No. 4402 at 32. By using FY 2007 as the beginning year, the
proposed formula allows the appropriate share to reflect the cumulative
effect of developments in competitive market conditions since the
PAEA's enactment.
---------------------------------------------------------------------------
\51\ The components, as applied through the formula, also
capture other relevant circumstances. See section IV.B, infra.
---------------------------------------------------------------------------
Additionally, the Postal Service maintains that it is inappropriate
and arbitrary to assign ``hypothetical'' values that represent the
appropriate share dating back to FY 2007 when the actual appropriate
share for those fiscal years are known. Postal Service Comments at 24.
The Commission acknowledges that the actual appropriate share \52\ is
known for prior fiscal years and clarifies that its approach does not
purport to change the actual values for any prior fiscal year. However,
as explained above, the Commission finds that the formula should ensure
the appropriate share reflects the market conditions as they have
evolved since the PAEA's enactment. As a result, it is neither
inappropriate nor arbitrary for the Commission to use these values to
determine the impact that market changes have had on the appropriate
share. The formula's calculation is purposefully and appropriately
cumulative in order to determine this impact.
---------------------------------------------------------------------------
\52\ In using the term ``actual appropriate share'' the
Commission is referring to the fact that, since its regulations in
Docket No. RM2007-1 became final, as required by the PAEA, the
appropriate share has remained at 5.5 percent. See supra at 4 n.4.
---------------------------------------------------------------------------
As it relates to comments received concerning the weighting of the
two components of the formula, the Commission finds that it is
appropriate from both a legal and economic perspective to weight the
components equally. First, from a legal perspective, the Commission's
decision to weight both components equally is appropriate because it is
based on the required consideration of the statutory criteria set forth
in 39 U.S.C. 3633(b). The Commission notes that the modified components
measure two discrete concepts. As described in sections IV.A.1 and
IV.A.2, supra, the Competitive Contribution Margin measures the Postal
Service's absolute market power; that is, its own ability to raise
prices above costs, whereas the Competitive Growth Differential
measures the Postal Service's market position relative to its
competitors. These concepts measure different aspects of the
competitive market, as the Competitive Contribution Margin considers
the Postal Service's market power with respect to consumers and the
Competitive Growth Differential measures the Postal Service's market
position with respect to competitors. Both modified components play
critical and equal roles in supporting the formula's ability to capture
the criteria set forth in 39 U.S.C. 3633(b). For example--as it relates
to capturing prevailing competitive conditions in the market--the
Competitive Contribution Margin provides insight into potential Postal
Service competitive advantage; the Competitive Growth Differential
reflects any changes in Postal Service market share; and both are
equally necessary in order to capture various changes to the market and
competitors. See section IV.B.1, infra. Additionally, both modified
components play a role in capturing each of the other relevant
circumstances the Commission considers. See section IV.B.3, infra.
Given that neither component is more significant than the other in
capturing the criteria set forth in 39 U.S.C. 3633(b), the Commission
finds it is appropriate to weight the components equally.
Second, from an economic perspective, the Commission's decision to
weight both components equally is appropriate. Although Sidak maintains
that ``from an economic perspective'' the Commission failed to offer a
reasonable explanation for the formula's configuration and suggests
that weights be assigned at the component level, Sidak's criticism is
problematic for two reasons. See Sidak Decl. at 39. First, the
assignment of weights at the component level, without unique economic
justification, is inconsistent with economic practice. Typically,
weighting is applied in survey analyses to correct imperfections in
surveys or in regression analyses to normalize errors.\53\ In those
instances, a unique weight is applied to each variable, for each
observation, using a function or a formula.\54\ Sidak seems to suggest
weights be assigned as follows:
---------------------------------------------------------------------------
\53\ See Jeffrey M. Wooldridge, Introductory Econometrics: A
Modern Approach 280-94 (5th ed. 2013) (Wooldridge); see also Sharon
L. Lohr, Sampling: Design and Analysis 225-29 (1999).
\54\ Wooldridge at 280-94.
Weighted Competitive Contribution Margin = Weight * %[Delta]CCM
Weighted Competitive Growth Differential = Weight * (Market
ShareUSPS,t-1 * (%[Delta]RevenueUSPS - %[Delta]RevenueC&M))
However, statistically, a more accurate assignment of weights would be
as follows:
[[Page 39952]]
[GRAPHIC] [TIFF OMITTED] TP13AU18.039
The Commission finds that assigning unique weights to each variable in
the context of the proposed formula would be inappropriate without an
economic rationale for each weight (e.g., to correct imperfections
(survey analysis) or to normalize errors (regression analysis)).\55\
Sidak does not propose an economic rationale for assigning any
particular set of weights, and the Commission has not separately
identified any. Without an economic rationale or justification, the
application of unique weights to each variable would be artificial and
thus inappropriate. Id.
---------------------------------------------------------------------------
\55\ Id. at 280-94.
---------------------------------------------------------------------------
Second, it would be problematic to assign weights at the component
level because both the Competitive Contribution Margin and the
Competitive Growth Differential rely in part on a shared input, the
Postal Service's competitive product revenue. See Order No. 4402 at 18-
19, 23; see also sections IV.A.1.c and IV.A.2.c, supra. For this
reason, the components are not independent and are considered
economically related.\56\ Due to the relatedness of variables (i.e.,
(Revenue) from the Competitive Contribution Margin and
(%[Delta]RevenueUSPS) from the Competitive Growth Differential), if
unique weights are assigned to the two components, the effect on those
components and the formula's calculation would be disproportionate. To
weight the components in a formula of this type would be inconsistent
with statistical practice and would diminish the accuracy of the
formula by changing how the components interact with each other.\57\
---------------------------------------------------------------------------
\56\ Related terms are commonly used in econometric models. See
Wooldridge at 198-200.
\57\ See id. at 280-94.
---------------------------------------------------------------------------
Table IV-3 below illustrates the calculation using the Commission's
revised proposed formula starting with an appropriate share of 5.5
percent in FY 2007.
---------------------------------------------------------------------------
\58\ Source: Library Reference PRC-LR-RM2017-1/2.
Table IV-3--Calculation of Appropriate Share, FY 2007-FY 2019 \58\
----------------------------------------------------------------------------------------------------------------
Percentage
change in Competitive
Appropriate Competitive Growth Appropriate
share for the Contribution Differential for share for the
Fiscal year current year (AS) Margin for the the prior year following year
(%) prior year (CGD-1) (%) (AS + 1) (%)
(%[Delta]CCM) (%)
----------------------------------------------------------------------------------------------------------------
FY 2007............................. 5.5 N/A N/A 5.5
FY 2008............................. 5.5 0.0 0.0 5.5
FY 2009............................. 5.5 -5.9 0.7 5.2
FY 2010............................. 5.2 13.4 1.2 6.0
FY 2011............................. 6.0 15.7 0.9 7.0
FY 2012............................. 7.0 -7.9 -0.2 6.4
FY 2013............................. 6.4 3.7 2.7 6.8
FY 2014............................. 6.8 5.5 2.5 7.3
FY 2015............................. 7.3 0.4 1.2 7.4
FY 2016............................. 7.4 -2.6 0.2 7.2
FY 2017............................. 7.2 18.1 1.4 8.6
FY 2018............................. 8.6 1.3 1.1 8.8
----------------------------------------------------------------------------------------------------------------
The proposed revised formula and each resulting appropriate share
percentage reflect trends in the market. For example, Table IV-3 shows
that the appropriate share would have decreased from FY 2009 to FY 2010
under the
[[Page 39953]]
proposed modified formula (comparing the second column with the last
column of the FY 2009 row). This decrease would have occurred in
response to a decline in the Postal Service's market power in FY 2008
(as measured by the Competitive Contribution Margin shown in the third
column of the FY 2009 row) largely due to the global financial crisis.
Although there was an increase in the Competitive Growth Differential
in FY 2008 (as shown in the fourth column of the FY 2009 row), it would
not have offset the decline in the Competitive Contribution Margin. The
appropriate share would have also decreased from FY 2012 to FY 2013
(comparing the second column with the last column of the FY 2012 row),
again in response to a decline in the Postal Service's market power (as
measured by the Competitive Contribution Margin shown in the third
column of the FY 2012 row). In this case, the decline was due to
changes in the mail mix that caused competitive products' revenue to
increase less than attributable costs. Beginning with FY 2014's
appropriate share, the appropriate share would have steadily increased
as the Postal Service expanded its market power and market position. As
a result, the appropriate share for FY 2019 (as indicated in the
bottom-right cell in Table IV-3) would be 8.8 percent under the
Commission's modified formula.
B. Analysis Pursuant to 39 U.S.C. 3633(b)
As it did in Order No. 4402, in this section, the Commission
explains how its modified formula captures the prevailing competitive
conditions in the market and other relevant circumstances as required
by 39 U.S.C. 3633(b). Additionally, the Commission addresses the
remaining element of section 3633(b)--whether any costs are uniquely or
disproportionately associated with Postal Service competitive products.
1. Prevailing Competitive Conditions in the Market
a. Order No. 4402
In Order No. 4402, to assess the prevailing competitive conditions
in the market, the Commission considered whether there was any evidence
of Postal Service competitive advantage; whether there had been any
changes in Postal Service market share; and whether there had been any
changes in the package delivery market or to competitors since the
Commission's last appropriate share review.\59\
---------------------------------------------------------------------------
\59\ Order No. 4402 at 34-40. The Commission also mentioned a
purely qualitative factor previously considered as a market
condition--whether there was any evidence of antitrust actions filed
against the Postal Service. Id. at 34 n.60. The Commission found
that that factor could not be explicitly captured through the
proposed quantitative formula. Id. However, the Commission did
determine antitrust actions were implicitly captured by the
previously proposed formula because changes in the Postal Service's
market power could offer insight into whether the Postal Service was
engaging in the kinds of anticompetitive behavior that would
underlie an antitrust action. See id. Because the Competitive
Contribution Margin continues to measure the Postal Service's market
power, the Commission finds that the modified formula implicitly
captures antitrust actions for the same reasons described in Order
No. 4402.
---------------------------------------------------------------------------
The Commission identified and discussed changes in market
conditions that had occurred since its last appropriate share review
and determined that its formula-based approach captured these
considerations. Order No. 4402 at 34-40. For example, the Commission
found that the Postal Service Lerner Index would reflect any Postal
Service competitive advantage because the more market power the Postal
Service possesses, the larger the Postal Service Lerner Index would be.
Id. at 35. The Commission also determined that the formula would
capture any evidence of predatory pricing because, should the Postal
Service ever engage in predatory pricing, the Postal Service Lerner
Index value would be negative. Id. at 36-37. In addition, the
Commission found that the formula captured Postal Service and
competitor market share by revenue mainly through the Competitive
Market Output. Id. at 38-39. Finally, the Commission found that changes
in the market including overall growth, entry and exit of firms, and
innovation would be observed in both the Postal Service Lerner Index
and Competitive Market Output. Id. at 39-40.
b. Modified Formula's Compliance With Section 3633(b)
Despite modifications to the previously proposed components, the
modified formula captures the prevailing competitive conditions in the
market. First, similar to the Postal Service Lerner Index, the
Competitive Contribution Margin provides insight into whether the
Postal Service possesses a competitive advantage.\60\ The higher the
Competitive Contribution Margin, the more market power the Postal
Service possesses. Any large increases in the Competitive Contribution
Margin may indicate a competitive advantage under certain
circumstances. Just as with the Postal Service Lerner Index, the
Competitive Contribution Margin also indicates whether the Postal
Service is engaging in predatory pricing because the resulting
Competitive Contribution Margin would be negative. If the Postal
Service were engaging in predatory pricing, its attributable costs
would be greater than its revenue, and, as calculated in the
Competitive Contribution Margin, the difference between them would be
less than zero, resulting in a negative value. Figure IV-1 below
displays the Competitive Contribution Margin from FY 2007 to FY 2017.
---------------------------------------------------------------------------
\60\ As discussed in Order No. 4402, the Commission also uses
its analysis required by section 703(d) to assess whether Postal
Service competitive products have a competitive advantage. See Order
No. 4402 at 35, 54-68. The Commission clarifies that a section
703(d) analysis is the primary way the Commission assesses whether
Postal Service competitive products have a competitive advantage due
to differences in the application of federal and state laws to the
Postal Service compared to competitors. The Commission notes that it
also uses other factors (e.g., large increases in market power or
evidence of Postal Service predatory pricing) to assess whether the
Postal Service has a competitive advantage.
---------------------------------------------------------------------------
[[Page 39954]]
[GRAPHIC] [TIFF OMITTED] TP13AU18.040
As shown in Figure IV-1, the Competitive Contribution Margin has
never been negative. As a result, the Commission continues to find no
evidence of Postal Service predatory pricing. The Commission maintains
that the use of the Competitive Contribution Margin in its modified
formula will provide an ongoing indication of whether the Postal
Service is engaging in predatory pricing.
---------------------------------------------------------------------------
\61\ Source: Library Reference PRC-LR-RM2017-1/2.
---------------------------------------------------------------------------
Second, the change in the Postal Service's market share by revenue
would be reflected in the Competitive Growth Differential even more so
than the Competitive Market Output component of the previously proposed
formula. Unlike the Competitive Market Output, which reflected market
share in its composition, the Competitive Growth Differential directly
incorporates Postal Service market share into the calculation of the
appropriate share, as discussed in section IV.A.2.c, supra. If the
Postal Service's market share were to grow from an increase in revenue,
the Competitive Growth Differential would increase, thereby increasing
the appropriate share if all other factors were to remain constant. If
the Postal Service's market share were to decline from a decrease in
revenue, the Competitive Growth Differential would decrease, thereby
decreasing the appropriate share if all other factors were to remain
constant. Additionally, similar to the Postal Service Lerner Index, any
growth or decline in the Postal Service's market share caused by shifts
in demand or pricing strategies would be reflected in the Competitive
Contribution Margin because such shifts would affect the Postal
Service's ability to price above costs and therefore its market power.
See Order No. 4402 at 39.
Finally, changes in the market and to competitors, such as overall
market growth, firm entry or exit from the market and innovation, are
reflected by both of the modified components. For example,\62\ if a
firm enters the market and generates new business, competitor revenue
relative to the Postal Service's revenue would increase, thereby
decreasing the Competitive Growth Differential. Alternatively, if a
firm enters the market and takes business from the Postal Service--
whether through pricing or innovation--the Postal Service would have to
price closer to marginal cost to remain competitive, thereby reducing
the Competitive Contribution Margin. However, if a firm exits the
market and the business it used to generate is lost, it could cause a
decrease in competitor revenue and an increase the Postal Service's
market share, thereby increasing the Competitive Growth Differential.
These various examples illustrate the modified formula's ability to
capture overall changes, including expansion or retraction in the
competitive market.
---------------------------------------------------------------------------
\62\ Each example assumes all other factors remain constant.
---------------------------------------------------------------------------
2. Unique or Disproportionate Costs
As previously noted, the second element of section 3633(b) is that
the Commission must consider ``the degree to which any costs are
uniquely or disproportionately associated with any competitive
products.'' See 39 U.S.C. 3633(b); see section III.A, supra. The
analysis of this second element differs from the other elements in
section 3633(b) because the Commission's consideration of the second
element is unrelated to the Commission's formula-based approach.
For that reason, in Order No. 4402, the Commission's discussion of
whether any costs are uniquely or disproportionately associated with
any competitive product relied on its current costing methodologies.
See Order No. 4402 at 43-45. The Commission's current costing
methodology attributes all reliably identifiable, causally related
costs that can be traced to individual products to those products and
was recently upheld
[[Page 39955]]
by the D.C. Circuit.\63\ The requirement that cost attribution must be
based on reliably identified causal relationships comes from the PAEA.
Order No. 4402 at 43 (citing 39 U.S.C. 3622(c)(2)). The Commission
noted that ``[b]y definition, costs identified as institutional are
those that cannot be causally linked to any specific product'' and
found that there were no costs uniquely associated or
disproportionately associated with any competitive products that were
not already attributed to competitive products under the Commission's
methodology. Id. at 43-44.
---------------------------------------------------------------------------
\63\ Id.; see generally UPS, 890 F.3d 1053.
---------------------------------------------------------------------------
The Commission's discussion on whether any costs were uniquely
associated or disproportionately associated with any competitive
products elicited multiple comments.\64\ However, as this Revised
Notice of Proposed Rulemaking is concentrated on modifications to its
proposed formula-based approach, the Commission will address the
comments related to ``the degree to which any costs are uniquely or
disproportionately associated with any competitive products'' in a
subsequent order.
---------------------------------------------------------------------------
\64\ See, e.g., Amazon Comments at 8-11; Postal Service Comments
at 4-5, 13, 16, 26-28; Sidak Decl. at 53-55.
---------------------------------------------------------------------------
3. Other Relevant Circumstances
a. Order No. 4402
In its assessment of other relevant circumstances in Order No.
4402, the Commission considered the effects of: (1) Products which have
been transferred from the market dominant product list to the
competitive product list since the Commission's last review of the
appropriate share; (2) changes to the mail mix (i.e., the relative
proportions of individual mail products' volumes within the overall
postal system) since the last review of the appropriate share; (3)
uncertainties in the marketplace; and (4) the risks associated with
setting the appropriate share either too high or too low. Order No.
4402 at 45-53. The Commission identified and discussed changes in these
relevant circumstances and determined that all were reflected in its
proposed formula-based approach. Id.
First, the Commission identified product transfers since its last
review of the appropriate share and determined that they were reflected
in the previously proposed formula because the transferred products'
revenue was automatically included in the Postal Service's portion of
the Competitive Market Output, and the transferred products' revenue-
per-piece and unit volume-variable cost were incorporated into the
composition of the Postal Service Lerner Index. Id. at 46.
Second, the Commission noted that the Postal Service has
experienced mail mix changes since the Commission's last review of the
appropriate share, as market dominant volumes have continued to decline
and competitive volumes have continued to increase. Id. at 46-49. The
Commission determined that the formula's Competitive Market Output
component incorporated changes in the Postal Service's mail mix by
including revenue that the Postal Service received from any increase in
competitive product volume. Id. at 48-49. Likewise, the Postal Service
Lerner Index would reflect the growth or decline of competitive
products with varying degrees of profitability. Id.
Third, with regard to market uncertainties, the Commission
explained that ``shifts in market demand or macroeconomic conditions
would be reflected in the appropriate share determination through
changes in the Postal Service Lerner Index and Competitive Market
Output.'' Id. at 49. The Commission also noted that the last 5 years
have been a time of significant innovation and development in the
delivery industry, and that it is important for the Commission's
proposed formula-based approach to be able to incorporate such changes.
Id. For potential competitor innovation or changes in e-commerce, the
Commission explained that both would be reflected in the Competitive
Market Output because competitor revenue would change as their
innovations succeeded or failed. Id. The Commission also noted it was
possible for competitor innovation to affect the Postal Service Lerner
Index should it cause the Postal Service to alter its pricing of
competitive products. Id. at 49-50.
Finally, the Commission has consistently recognized that there are
risks inherent in setting the appropriate share either too high or too
low. Id. at 50-51; see also Order No. 1449 at 12. If the appropriate
share were set too high, the Postal Service would be forced to raise
its prices to non-competitive levels. Order No. 4402 at 50. If the
appropriate share were set too low, the Postal Service might be
incentivized to discount its prices in order to gain market share. Id.
at 50. The Commission found that its proposed formula should limit
increases in the appropriate share to no higher than appropriate to
account for the Postal Service's growth in market power and the growth
of the market as a whole. Id. With regard to the risk of the
appropriate share being set too low, the Commission noted that price
discounting on the scale necessary to gain market share would come at
the expense of the Postal Service's overall profitability. Id. at 50-
51. The Commission therefore concluded that the Postal Service
possesses little incentive to engage in such behavior. Id. at 51.
b. Modified Formula's Compliance With Section 3633(b)
Despite changes to the previously proposed components, with the
Competitive Contribution Margin and the Competitive Growth
Differential, the modified formula captures other relevant
circumstances. First, the modified formula continues to capture changes
caused by Postal Service product transfers to the competitive product
list. When a product is transferred from the market dominant to the
competitive product list, the modified formula continues to incorporate
it directly through the Competitive Growth Differential because the
modified component continues to include the transferred product's
revenue as part of the Postal Service's revenue. The effect of product
transfers would also be reflected in changes in Postal Service market
share because market share is calculated using, in part, Postal Service
revenue, which would include the revenue of any transferred product. In
addition, the transferred product's attributable costs and revenue are
incorporated into the Competitive Contribution Margin. Any change in
the Competitive Contribution Margin resulting from a transfer reflects
the Postal Service's market power in the expanded competitive market,
as discussed above. See section IV.A.1.c, supra.
Second, as it relates to changes in the mail mix, the Commission
noted in Order No. 4402 that mail mix changes occur as demand for
postal products shifts. Order No. 4402 at 46. Most recently, Postal
Service market dominant product demand has decreased, while demand for
its competitive products has increased. Id. at 46-48. The modified
formula captures these mail mix changes as the Competitive Growth
Differential reflects the revenue the Postal Service receives from any
increase in competitive product volume. The Competitive Contribution
Margin, similar to the Postal Service Lerner Index, would reflect the
growth or decline of very profitable or less profitable competitive
products. See id. at 48-49.
Third, regarding market uncertainties, the modified formula
captures changes in market demand or other macroeconomic conditions
through
[[Page 39956]]
changes in either of the modified components. For example, if demand in
the market declines, because of a recession or other conditions, there
may be downward pressure on prices in the market. This occurrence may
cause the Postal Service to reduce its prices in order to preserve
volume, reducing the Completive Contribution Margin. Other competitors
may reduce prices as well, resulting in changes to the market overall;
an occurrence that would be reflected in the Competitive Growth
Differential.
The Commission also finds that its modified formula should capture
efforts to innovate or changes in e-commerce, accomplishing the same
objective as the previously proposed formula. The Competitive Growth
Differential captures these changes as they affect the Postal Service's
position in the market. For example, if competitors in the aggregate
were to successfully innovate and generate more revenue relative to the
Postal Service, the Competitive Growth Differential would decrease if
all other factors were to remain constant. If the Postal Service were
to successfully innovate and generate more revenue relative to its
competitors, the Competitive Growth Differential would increase if all
other factors were to remain constant.
Finally, in terms of the risk involved with setting the appropriate
share too high, the Commission finds that this risk is addressed by the
modified formula, just as it was by the previously proposed formula.
The modified formula continues to limit increases in the appropriate
share to no higher than appropriate to account for the Postal Service's
growth in market power and for growth in the Postal Service's market
position. In terms of the risks involved in setting the appropriate
share too low and allowing the Postal Service to gain market share by
discounting prices, the Commission continues to find that this risk is
minimal. As noted in Order No. 4402, the Postal Service has little
incentive to discount prices in order to gain market share because
discounting prices to gain market share would decrease the Postal
Service's profitability at a time when it continues to face financial
challenges.\65\
---------------------------------------------------------------------------
\65\ See Order No. 4402 at 50-51. The modified formula continues
to be calculated with a time lag that further discourages price
discounting by the Postal Service because the negative consequences
would appear before the benefits. See id. at 51.
---------------------------------------------------------------------------
V. Section 703(d) of the PAEA
As discussed in Order No. 4402,\66\ in order to determine whether
Postal Service competitive products enjoyed advantages over private
carriers, Congress directed the FTC to prepare a report identifying
federal and state laws that apply differently to the Postal Service's
competitive products than similar products offered by private
competitors and to account for the net economic effect resulting from
such differences.\67\ Additionally, section 703(d) directs the
Commission, when revising regulations under 39 U.S.C. 3633, to consider
subsequent events that may affect the continuing validity of the FTC's
net economic effect finding.\68\
---------------------------------------------------------------------------
\66\ See id. at 54-58.
\67\ See PAEA, 120 Stat. 3244; see also S. Rep. No. 108-318 at
29 (2004); PAEA section 703(a) and (b). Section 703 was not codified
and is reproduced in the notes of 39 U.S.C.A. 3633. See also FTC
Report.
\68\ PAEA section 703(d).
---------------------------------------------------------------------------
Order No. 4402 presented the first proposed revision to a
regulation issued under 39 U.S.C. 3633 since the PAEA's enactment. The
Commission provided its analysis pursuant to section 703(d) in Order
No. 4402. Order No. 4402 at 54-68. In that analysis, the Commission
discussed the FTC Report and its findings, defined the scope of its
review pursuant to section 703(d), and performed the required analysis
based on the statute. Id. The comments received in response to Order
No. 4402 have not identified any subsequent events pursuant to the
Commission's interpretation of section 703(d) that were not addressed
in Order No. 4402 or that have subsequently occurred.\69\ The
Commission also has not identified any subsequent events that would
affect its section 703(d) analysis in Order No. 4402. As such, the
Commission affirms its finding in Order No. 4402 that the FTC's
conclusion that the Postal Service operates at a net economic
disadvantage continues to be valid.
---------------------------------------------------------------------------
\69\ The Commission's discussion on the FTC Report and section
703 elicited multiple comments. See, e.g., UPS Comments at 22-26;
Sidak Decl. at 6, 9-15, 52-53. However, as this Revised Notice of
Proposed Rulemaking is concentrated on modifications to the proposed
formula-based approach, the Commission will address the comments
received on the FTC Report and section 703(d) in a subsequent order.
---------------------------------------------------------------------------
VI. Administrative Actions
Additional information concerning this rulemaking may be accessed
via the Commission's website at http://www.prc.gov. Interested persons
may submit comments on the modified formula-based approach and related
revisions to proposed rules \70\ no later than 30 days after the date
of publication of this Revised Notice of Proposed Rulemaking in the
Federal Register. Pursuant to 39 U.S.C. 505, Kenneth R. Moeller
continues to be designated as an officer of the Commission (Public
Representative) to represent the interests of the general public in
this proceeding.
---------------------------------------------------------------------------
\70\ The Commission makes one revision to proposed Sec.
3015.7(c)(1). The Commission replaces the formula proposed in Order
No. 4402 with the formula proposed in this Revised Notice of
Proposed Rulemaking. The proposed rules are set forth below the
signature of this Order.
---------------------------------------------------------------------------
The Regulatory Flexibility Act requires federal agencies, in
promulgating rules, to consider the impact of those rules on small
entities. See 5 U.S.C. 601, et seq. (1980). If the proposed or final
rules will not, if promulgated, have a significant economic impact on a
substantial number of small entities, the head of the agency may
certify that the initial and final regulatory flexibility analysis
requirements of 5 U.S.C. 603 and 604 do not apply. See 5 U.S.C. 605(b).
In the context of this rulemaking, the Commission's primary
responsibility is in the regulatory oversight of the United States
Postal Service. The rules that are the subject of this rulemaking have
a regulatory impact on the Postal Service, but do not impose any
regulatory obligation upon any other entity. Based on these findings,
the Chairman of the Commission certifies that the rules that are the
subject of this rulemaking will not have a significant economic impact
on a substantial number of small entities. Therefore, pursuant to 5
U.S.C. 605(b), this rulemaking is exempt from the initial and final
regulatory flexibility analysis requirements of 5 U.S.C. 603 and 604.
VII. Ordering Paragraphs
It is ordered:
1. Interested persons may submit comments no later than 30 days
from the date of the publication of this notice in the Federal
Register.
2. Pursuant to 39 U.S.C. 505, Kenneth R. Moeller continues to be
appointed to serve as the Public Representative in this proceeding.
3. The Secretary shall arrange for publication of this Order in the
Federal Register.
By the Commission.
Stacy L. Ruble,
Secretary.
List of Subjects for 39 CFR Part 3015
Administrative practice and procedure.
For the reasons stated in the preamble, the Commission proposes to
amend chapter III of title 39 of the Code of Federal Regulations as
follows:
PART 3015--REGULATION OF RATES FOR COMPETITIVE PRODUCTS
0
1. The authority citation for part 3015 continues to read as follows:
[[Page 39957]]
Authority: 39 U.S.C. 503; 3633.
0
2. Amend Sec. 3015.7 by revising paragraph (c) to read as follows:
Sec. 3015.7 Standard for compliance.
* * * * *
(c)(1) Annually, on a fiscal year basis, the appropriate share of
institutional costs to be recovered from competitive products
collectively, at a minimum, will be calculated using the following
formula:
ASt+1 = ASt * (1 + %[Delta]CCMt-1 +
CGDt-1)
Where,
AS = Appropriate Share, expressed as a percentage and rounded to one
decimal place
CCM = Competitive Contribution Margin
CGD = Competitive Growth Differential
t = Fiscal Year
If t = 0 = FY 2007, AS = 5.5 percent
(2) The Commission shall, as part of each Annual Compliance
Determination, calculate and report competitive products' appropriate
share for the upcoming fiscal year using the formula set forth in
paragraph (c)(1) of this section.
[FR Doc. 2018-17221 Filed 8-10-18; 8:45 am]
BILLING CODE 7710-FW-P