PCC Issues Paper

No. 05, Series of 2020

PAPER
ISSUES

Competition in
Philippine Markets:
A Scoping Study of the
Manufacturing Sector
Erlinda Medalla, Francis Mark Quimba,
and Maureen Ane Rosellon
Competition in Philippine Markets: A Scoping Study
of the Manufacturing Sector
Erlinda Medalla, Francis Mark Quimba, and Maureen
Ane Rosellon

Published by:

Philippine Competition Commission

25/F Vertis North Corporate Center 1
North Avenue, Quezon City 1105

PCC Issues Papers aim to examine the structure, conduct, and

performance of select industries to better inform and guide PCC’s
advocacy and enforcement initiatives. The opinions, findings,
conclusions, and recommendations expressed in these studies are
those of the author(s) and do not necessarily reflect the views of the
Commission. This work is protected by copyright and should be cited
accordingly.

The views reflected in this paper shall not in any way restrict or confine
the ability of the PCC to carry out its duties and functions, as set out in
the Philippine Competition Act. PCC reserves the right, when examining
any alleged anti-competitive activity that may come to its attention,
to carry out its own market definition exercise and/or competition
assessment, in a manner which may deviate or differ from the views
expressed in this paper.

 [email protected] |  www.facebook.com/CompetitionPH |  www.twitter.com/CompetitionPH |  www.phcc.gov.ph

 CONTENTS
I. INTRODUCTION 1
II. REVIEW OF LITERATURE: RELEVANT FINDINGS FROM PAST STUDIES 2
Past Study by Aldaba (2008): summary of main findings 2
Relevant findings from past studies including other sectors 10

III. APPROACH AND METHODOLOGY 12

Structure 13
Conduct 16
Performance 17
SCP relationships 20

IV. EMPIRICAL FINDINGS: SCOPING COMPETITION IN THE PHILIPPINE

MANUFACTURING SECTOR 22
Brief Overview of Recent Manufacturing Performance 22
Findings on Concentration Ratios and the Herfindahl Index 25
Findings on Price-Cost Margins in Manufacturing 33
Findings on the relationships between Structure-Conduct-Performance 35

V. EXTERNAL FACTORS AFFECTING COMPETITION: SOME CRITICAL

GOVERNMENT POLICIES AND REGULATIONS 39
VI. CONCLUDING REMARKS: MATRIX FOR PRIORITIZATION 41
VII. REFERENCES 48
List of Tables
Table 1. Market Structure, Barriers to Entry, and Competition (As of 2008)........................... 3
Table 2. Four-firm Concentration Ratios in the Philippine Manufacturing Industry ............... 6
Table 3. Distribution of firms according to DRC/SER ratios...................................................... 8
Table 4. Simple Price Cost Margins in the Philippine Manufacturing Industry....................... 8
Table 5. Manufacturing correlation matrix: concentration ratio and price-cost margin,
1988, 1994, 1995 ........................................................................................................................... 9
Table 6. Manufacturing Sector Performance ............................................................................23
Table 7. 4-firm Concentration Ratio summarized at 2-digit PSIC level..................................26
Table 8. HHI summarized at 2-digit PSIC level .........................................................................27
Table 9. 4-firm Concentration Ratios and Number of Establishments, during and post-
trade reform periods ...................................................................................................................28
Table 10. Number of Establishments by 2-digit PSIC manufacturing sector, 2006-2014 ...29
Table 11. 4CR and HHI for Selected Industries, at 4-digit PSIC Code ...................................31
Table 12. Adjusted Price-Cost Margin summarized at the two-digit level ............................33
Table 13. PCM during and post-trade reform periods............................................................34
Table 14. Correlation between concentration indices and price-cost margin .....................35
Table 15. Correlation between concentration indices and price cost margin: excluding
‘outliers’ .........................................................................................................................................36
Table 16. Regression results including other explanatory variables ...................................37
Table 17. Regression results on the relationship of concentration on productivity ............38
Table 18. Prioritization matrix .....................................................................................................42
Table 19. PSIC sectors with highest priority, Category A (HHI > 2500, and APCM> 15%) 43
Table 20. PSIC sectors in other categories (B1 and C1) with Value-added share exceeding
1% ..................................................................................................................................................45
Table 21. Suggested Priority List ...............................................................................................46

List of Figures
Figure 1. The Structure-Conduct-Performance Framework ...................................................13
I. INTRODUCTION

The Philippines has been undertaking major structural reforms characterized by

liberalization, deregulation, and privatization, particularly during the three decades after
the Marcos regime. A key reform measure was trade liberalization which brought down
tariffs from an average of around 40 percent (arising from a structure with tariff peaks of
over 100 percent) before reforms (in the early 1980s) to an average of 7 percent (with
around 80 percent of tariff lines imposed duties of 5 percent or below) after a series of tariff
reforms by 2000s. Trade liberalization also eased quantitative restrictions and import
controls—from around a third of commodity classifications to less than 3 % by the end of the
significant trade reforms (Medalla 1998).

The primary rationale for trade reforms has been the recognition of the huge costs of
protection without the benefits that it was supposed to have produced—industrial
development, sustained growth, and increased consumer welfare. Instead, protectionist
trade policies have widely promoted rent-seeking, high prices, limited consumer choice,
lack of innovation, and aging capacities. A liberal trade regime was seen to remove the
market distortions from protectionist policies and promote a more dynamic economy in the
long-run, with better allocation of resources to more productive sectors, innovation, and
greater firm efficiency needed in a competitive environment.

While trade policy is the first layer of competition policy that can discipline the market, its
cross-border impact is not enough to fully counter the hold of existing domestic companies
with a dominant position that could arise from many factors such as imperfect substitution
between foreign and local goods, the presence of structural barriers to entry of firms, the
local distribution channels coopted by domestic suppliers, and the non-tariff barriers that
still abound. Thus, the lack of effective policy for domestic competition (behind the border)
minimizes the gains from trade liberation. In other words, competition policy is a necessary
complementary measure to trade liberalization. The synergies between trade and
competition policies cannot be overemphasized. (See Bartok and Miroudot 2008, and
Aldaba 2005 and 2008 for example).

Republic Act 10667, also known as the Philippine Competition Act (RA 10667 or PCA), in
2015 is thus a landmark legislation that has long been overdue. It is a much-needed
reform, primarily because of the new developments arising from new technologies that
have revolutionized the way business is done and intensified international
production sharing and (increasingly interlinked and broadened) supply chains.
Possibly more than the past trade reforms, it could have a substantial potential impact
on productivity, innovation, and even equity (as it levels the playing field for small and
medium enterprises or SMEs).

The Philippine Competition Commission (PCC) is an independent quasi-judicial

body mandated to implement the PCA. The task of the Commission and other related
agencies to implement competition policy and law is enormous. First and foremost, there

1
is a clear need for a greater understanding of the state of markets and competition
issues in the different sectors of the economy. The PCC has thus embarked on funding
crucial studies and developing a work program on key sectors such as
telecommunications, shipping, among others.

In particular, the PCC has commissioned this research project to do a scoping study on the
Manufacturing Sector, which aims to assess the state of competition, identify problem
areas, and formulate a prioritization matrix that would aid in achieving the objectives of
competition law and policy. The scoping study would, at least, serve as the basis for the
PCC in its advocacy initiatives and its selection of sectors for in-depth market studies. Also,
it could serve as a benchmark for future studies that would gauge the impact of competition
act and policy reforms that have been implemented.

Further, this paper reviews what was done in the past to assess the state of competition in
Philippine manufacturing. Specifically, this paper leverages on the study done by Aldaba
(2008) in the manufacturing sector for the Philippines. Findings from relevant studies that
look at other industries are also presented. As to approach and methodology, this study
adopts the underlying framework of the Structure-Conduct-Performance (SCP) Model, and
estimates similar relevant indicators as Aldaba (2008). This section also formulates
regression models that are tested to help understand the interaction among the three SCP
elements and how they impact the state of competition. The succeeding section provides
the empirical findings and analysis, which would aid in the assessment of competition and
identify parameters and guides for prioritization. A separate section looks briefly at major
external factors, specifically, selected government policies with significant impacts on
competition in the manufacturing sector. This section, however, would be more exploratory
and illustrative in nature, to understand their implications on the task of the PCC in
promoting competition. Finally, the last section on the conclusion and recommendation
suggests a prioritization matrix that could be used by the PCC.

II. REVIEW OF LITERATURE: RELEVANT FINDINGS FROM

PAST STUDIES
This section covers the major studies done on competition in manufacturing and other past
studies that could be relevant to this scoping study on the state of competition in the
manufacturing sector.

Past Study by Aldaba (2008): summary of main findings

Aldaba (2008) provides the most recent and comprehensive work on assessing competition
in Philippine markets. A summary of her main findings is presented below. Table 1 provides
an overview of market structure, relevant government regulator, barriers to entry in major
sectors and subsectors of the Philippine economy as of the time of the writing of the study.
Despite trade liberalization, market structures could encourage monopolistic or

2
oligopolistic behavior and barriers to entry. Such could be detrimental to the state of
competition in these markets.

Table 1. Market Structure, Barriers to Entry, and Competition (as of 2008)

Barriers to entry
Market Gov’t
Economic Sectors Structural & Behavioral Studies
Structure Regulator
regulatory
Agriculture
Rice Importation: NFA Import license AGILE (2000);
Monopoly Intal and Garcia
Trading: Cartel (2005); Reeder
Oligopoly (2000); Mendoza
and Rosegrant
Corn Importation: NFA Import license Cartel
(1995)
Monopoly
Trading:
Oligopoly
Sugar Oligopoly SRA Tariff quota Cartel Borrel etl (1994);
Philexport (1998),
Tolentino (1999)
Poultry & chicken Tariff quota
Bananas (for Oligopsony Abuse of Digal (2007)
export) market power
Pineapplies (for Oligopsony Abuse of
export) market power
Manufacturing Aldaba (2007,
2005, 2003 and
2002a); L. de Dios
(1993); Imbat &
Tanlapco (1993),
E. de dios (1986);
Lindsey (1977)
Motorcycles & Oligopoly Large capital Pineda (1994)
parts requirements
Economies of
Scale
Meat & dairy Oligopoly Tariff quotas: L. de Dios (1994a)
processing live swine Large
capital
requirements
Product
differentiation
Sunk costs
Appliance Oligopoly Large capital Lapid (1994)
requirements
Product
differentiation
Economies of
scale
Technology
acquisition
Access to

3
 Barriers to entry
Market Gov’t
Economic Sectors Structural & Behavioral Studies
Structure Regulator
regulatory
distribution
channels

Packaging (glass- Oligopoly Large capital Medillo (1994)

based) requirements
Economies of
scale
Flat glass Monopoly Safeguard Medillo (1994)
measure; Large
capital
requirements
Skill intensive
Economies of
scale
Synthetic resin Oligopoly Large capital Banzon (1994)
requirements
Economies of
scale
Agricultural Competitive Trabajo (1994)
machinery
Shipbuilding & Oligopoly Large capital
repair requirements
Automotive Oligopoly Large capital Aldaba (1997,
requirements 2000b, 2008b)
Economies of
scale Strong
parts supply
base
Downstream oil Oligopoly DOJ-DOE Large capital Cartel Salas (2002);
task requirements Galang & Solleza
force to Extensive retail (2001); Cabalu et
oversee network al (2001); Fabella
competitio & Aldaba (2004)
n
Pharmaceutical Oligopoly BFAD License & Cartel Lecciones (2004)
drugs Monopoly for regulates registration Gift giving Lao (1999)
patent entry Patents, practices by
holders Intellectual drug firms to
Property Law promote
Intensive expensive
advertising drugs to
physicians &
pharmacists
Cement Oligopoly Large capital Cartel Aldaba (2007,
requirements 20002b),
Lamberte, E. de
Dios et al (1992)
Services
Electricity Generation: ERC regulatory Llanto &
NPC, IPPs capacity & Patalinghug

4
 Barriers to entry
Market Gov’t
Economic Sectors Structural & Behavioral Studies
Structure Regulator
regulatory
independence (2004); Fabella &
of ERC Aldaba (2004);
Transmission: ERC Same Tuano (2001)
TRANSCO
Monopoly
Distribution: ERC Same Abuse of
MERALCO market
Monopoly Power, cross-
ownership of
distribution
and
generation
firms
Water Monopoly MWSS-RO Independence Fabella (2006);
of RO Santos (2003);
Llanto (2002);
Solon and
Pamintuan (2000)
Wholesale & Duenas-Caparas
Retail (2005)
Department Competitive:
stores & SM &
Supermarkets Robinson's
are the 2
biggest
players
Drug stores Oligopoly: Economies of
Mercury Drug, scale & scope
the dominant Customer
player goodwill &
loyalty
Supplier
network
Tele- Oligopoly NTC Congressional PLDT delaying Llanto &
communications Franchise interconnecti Patalinghug
Network on PLDT & (2004); Salazar
industry Smart merger (2007); Abrenica
Regulatory (2000); Aldaba
capacity & (2000a); Serafica
independence (1998a & b)
of NTC
Ports Monopoly PPAs Complex policy, Llanto, E. Basilio,
regulatory, & & L. Basilio
institutional (2005); PDP
framework Australia/Meyrick
Conflicting roles and Associates
of PPA (2005)
Water transport Oligopoly MARINA Cabotage law Mergers; Austria (2003);
Regulatory cartel & Llanto, E. Basilio
capacity & market and L. Basilio
independence sharing (2005)
of MARINA

5
 Barriers to entry
Market Gov’t
Economic Sectors Structural & Behavioral Studies
Structure Regulator
regulatory
Air transport Oligopoly CAB Congressional Mergers; Austria (2002);
Major routes: Franchise , regulatory Forsyth et al
Duopoly Cabotage law, capture (2004); Aldaba
Minor routes: Subsidies given (2005c); Lim
Monopoly only to PAL , (2004)
Regulatory
capacity &
independence
of CAB
Banking Oligopoly: BSP Pasadilla & Milo
Institutions competitive (2004); Milo
behavior (2001); Manzano
& Neri (2001);
Montinola &
Moreno (2001);
Lamberte, M. &
C. Manlangit
(2005)
Source: Table 13 of Aldaba 2008

Aldaba (2008) computed the 4-firm concentration ratios for the manufacturing sectors. Her
results are presented in Table 2. On average, concentration ratios rose for the period 1988-
1998, following the major trade reforms in the mid-1980s, which continued up to the end
of the 1990s. This increase in concentration ratio is not an inconsistent response to
competition from a move to a more open trade regime, as firms restructure to be able to
compete, and as inefficient firms are driven out.

Table 2. Four-firm Concentration Ratios in the Philippine Manufacturing Industry

Concentration Ratios Number of establishments

Sectors
1988 1994 1995 1998 1988 1994 1995 1998
High (above (70%)
Petroleum Refineries 100 100 100 99.93 4 4 4 5
Professional and Scientific 100 100 99.97 97.41 14 13 20 80
Tobacco 96.64 99.56 99.41 99.50 25 21 22 21
Nonferrous Metal Products 99.67 99.28 98.57 97.76 35 34 40 35
Glass and Glass Products 96.33 90.58 92.05 95.43 35 53 46 66
Industrial Chemical 90.14 87.52 84.65 86.49 112 171 197 375
Transport Equipment 80.98 86.2 84.4 77.67 230 264 265 364
Pottery, China and Earthen 92.82 86.05 93.74 d 59 68 61 -
Food Processing 79.51 81.37 81.74 a 915 751 717 -
Iron and Steel 84.18 80.64 70.55 79.43 128 191 201 505
machinery except Electrical 63.59 77.47 79.43 94.90 556 464 460 888
Petroleum and Coal Products 81.1 77.00 87.4 100 16 14 16 13
Fabricated Metal Products 73.45 74.48 74.32 78.24 469 555 550 975
Other Chemicals 66.37 75.64 69.09 80.92 300 288 295 397

6
 Concentration Ratios Number of establishments
Sectors
1988 1994 1995 1998 1988 1994 1995 1998
Rubber Products 79.15 73.5 73.66 90.33 137 187 181 136
Other Nonmetallic Mineral 68.92 71.31 74.54 90.03 353 304 253 701
Paper and Paper Products 78.97 71.23 70.4 78.14 167 215 206 335
Miscellaneous Manufacture 70.87 70.62 76.76 92.77 342 312 309 310
Textiles 64.12 64.14 72.37 72.84 549 537 508 586
Food Manufacturing 63.48 69.74 77.92 86.94 2003 1879 1798 3919
Beverages 48.19 70.08 63.43 73.51 91 86 88 129
Electrical Machinery 64.8 69.36 63.73 72.42 217 271 310 448
Leather and Leather Products 57.7 63.89 64.02 73.47 120 84 85 595
Wood and Cork Products 40.5 55.47 65.35 76.32 683 401 354 584
Printing and Publishing 42.13 47.26 51.08 82.08 636 637 636 988
Plastic Products 49.41 40.75 50.87 70.09 300 377 365 490
Moderate (40 to 69%)
Metal Furniture 80.88 79.49 62.67 b 36 34 35 -
Cement 45.3 48.3 45.37 68.22 17 18 18 20
Leather Footwear 30.33 41.7 55.0 c 425 384 373 -
Furniture 19.51 40.91 41.64 62.54 678 497 439 68
Low (below 39%)
Wearing Apparel ex Footwear 34.7 31.69 26.52 23.57 1556 1512 1521 2025
Total Manufacturing 70.88 73.63 73.64 80.55 11208 10726 10373 15674
Source of basic data: National Statistics Office, 1988 and 1994 Census of Establishments and 1995 and 1998 Annual
Survey of Establishments. The concentration ratios refer to the ratio of census value added by four largest firms to
total in each five-digit PSIC sector. The concentration ratios given above are weighted averages for 3-digit PSIC.
a
combined food manufacturing and food processing;
b
combined metal furniture and furniture; c combined leather footwear and leather products;
d
combined pottery, china and other nonmetallic products See: Aldaba (2007).
Source: Table 14 of Aldaba 2008

Indeed, Medalla (1998) finds that, on average, the efficiency of firms improved during trade
reforms. See Table 3. More establishments, both in terms of number and value of
production, became efficient in saving/earning foreign exchange.1 As the table shows, for
example, the share of productive firms in terms of production value rose from around 19
percent in 1983 to 44 percent in 1992 decline in the standard deviation of the DRC/SER
ratios indicating that resources are better allocated. According to Medalla (1998), a wide
variation implies room for movement of resources from high domestic resource cost sectors
to lower domestic resource cost sectors.

1 Clearly, the lower the DRC/SER ratio is (desirably not more than 1), the more efficient it is, especially from the
point of view of society.

7
 Table 3. Distribution of firms according to DRC/SER ratios

Notes: DRC/SER ratio is the domestic resource cost per unit (social) value of foreign exchange earned/saved (earned
for exporters and saved for local producers of import substitutes)

Source: Medalla (1998)

The increase in concentration ratios should not be ignored despite the efficiency of the
market outcome because of the potential for collusion and abuse of market power with
high firm concentration. Hence, there is a crucial need for competition policy that prevents
the potential for abuse, and ensure that firms play fairly (within the competition law).

Aldaba (2008) also computed price cost margins (PCM) to look more closely at what these
results could indicate in terms of possible monopoly rents. See Table 4.

Table 4. Simple Price Cost Margins in the Philippine Manufacturing Industry

1972- 1972- 1976- 1981- 1986- 1991- 1996-

Industry Sector
98 75 80 85 90 95 98
High (50 to 69%)
Cement 0.65 0.59 0.66 0.67 0.65 0.65
Beverages 0.53 0.56 0.46 0.51 0.56 0.57 0.55
Glass and Glass
Products 0.52 0.48 0.48 0.51 0.55 0.54 0.58
Moderate (20 to
49%)
Tobacco 0.47 0.44 0.40 0.31 0.50 0.57 0.66
Other Non-metallic
mineral 0.43 0.64 0.42 0.33 0.42 0.43 0.36
Other Chemicals 0.37 0.37 0.35 0.30 0.35 0.46 0.44
Paper and Paper
Products 0.36 0.35 0.38 0.36 0.36 0.34 0.36
Industrial Chemicals 0.35 0.40 0.33 0.37 0.32 0.38 0.32
Rubber Products 0.28 0.25 0.26 0.30 0.26 0.31 0.29
Food Manufacturing 0.28 0.34 0.24 0.23 0.29 0.28 0.37
Textiles 0.27 0.25 0.23 0.30 0.25 0.27 0.30
Iron and Steel 0.26 0.26 0.22 0.35 0.21 0.26 0.25
Plastic Products 0.25 0.25 0.22 0.26 0.20 0.29 0.32
Electrical Machinery 0.25 0.29 0.21 0.25 0.21 0.24 0.34
Wood and Cork
Products 0.26 0.33 0.22 0.30 0.24 0.25 0.22

8
 1972- 1972- 1976- 1981- 1986- 1991- 1996-
Industry Sector
98 75 80 85 90 95 98
Furniture except Metal 0.22 0.21 0.18 0.24 0.21 0.23 0.27
Nonferrous Metal
Products 0.21 0.37 0.29 0.17 0.11 0.14 0.19
Petroleum and Coal
Products 0.21 0.32 0.17 0.22 0.20 0.24 0.13
Miscellaneous
Manufacture 0.20 0.22 0.12 0.30 0.12 0.22 0.27
Low (19% and
below)
Fabricated Metal
Products 0.17 0.23 0.12 0.16 0.12 0.21 0.21
Printing and
Publishing 0.16 0.26 0.08 0.07 0.13 0.17 0.36
Leather and Leather
Products 0.16 0.14 0.16 0.22 0.10 0.15 0.20
Transport Equipment 0.14 0.11 0.17 0.12 0.05 0.16 0.28
Machinery except
Electrical 0.11 0.20 -0.14 0.18 0.13 0.17 0.17
Average 0.30 0.31 0.18 0.26 0.23 0.29 0.34
Standard deviation 0.14 0.14 0.32 0.17 0.20 0.15 0.14
Source: Table 15 of Aldaba (2008)

The results show that PCMs, despite the increasing concentration ratios, are still mostly
moderate to low. The declining correlation coefficient between concentration relation (CR)
and PCM (Medalla, 2003) supports the view that the observed increase in concentration
ratio during the period of extensive trade reforms was likely less indicative of increased
monopoly power in general. See Table 5.

Table 5. Manufacturing correlation matrix: concentration ratio and price-cost

margin, 1988, 1994, 1995
PCM 1988 PCM 1994 PCM 1995

CR 1988 0.4223 0.1854 0.12

CR 1994 0.272 0.1759

CR 1995 0.1423
Source: Obtained from Table 5 of Medalla 2003

Considering the trade reforms, improvements in efficiency, generally low to moderate

PCMs, and the declining positive correlation between PCM and concentration ratio, the
estimated CRs seem under control. Whether these results remain valid is something that
this study would explore. The indicators used by Aldaba (2008), where feasible, are thus
updated.

9
Relevant findings from past studies including other sectors

There are sectors, alleged of anti-competitive behavior (e.g. abuse of dominance) and other
unfair trade practices, that have been the subject of research. A study by Abad et al. (2012)
briefly reviewed the literature (e.g., Aldaba 2000, 2008, 2010, 2011; Lamberte et al. 1992)
on some of these sectors; in particular, cement, telecommunications, energy, agriculture,
food, and consumer products.

The cement industry is one of those industries alleged to have collusion among firms. An
analysis by Aldaba (2010) found that the industry is highly concentrated, with three firms
controlling almost 90% of the market. In the 1970s, during the protectionist regulations, a
few producers dominated the market, and a pricing pattern was observed from them
(Aldaba 2000; Lamberte et al. 1992). The issue of collusion waned when the industry was
liberalized and deregulated in the 1980s. However, a trend in pricing behavior recurred in
the late 1990s.

Price fixing and market allocation were two manifestations being connected to a cartel
(explicit or tacit) in the cement industry (Abad et al. 2012). There were observed price
increases at times of excess supply and weakened demand during the economic slowdown,
such as during the 1997 and 2008 financial crises, or when fuel prices and power rates
changed substantially (Aldaba 2000, 2010). Aldaba (2000) also found that despite having
different cost structures, there is low variation in pricing and that changes/increases in
prices happen in a rather ‘harmonious fashion’. On the other hand, market allocation
(production quotas and geographical division of the regional markets) was alleged to have
been discussed in meetings held among cement firms (Lamberte et al. 1992).

In the past, the government conducted investigations on the cartel in the cement industry.
However, as there was no clear and comprehensive competition law at that time that the
investigation did not produce substantial findings (Abad et al. 2012). But with the
enactment of the PCA and creation of the PCC, complaints on cement cartel are again being
probed2 but with guidance of a competition law.

Meanwhile, despite the liberalization and deregulation of the telecommunications sector,

there are still cases of complaints and allegations of unfair business or anti-competitive
practices. PLDT owns the domestic backbone system and can influence the speed and the
terms and conditions for interconnection and for revenue-sharing agreements, which is
perceived as disadvantageous, especially for new players (Aldaba 2008, 2011). PLDT also
expanded its coverage in the market with its merger with Smart Communications, one of
the top mobile network/ telecommunications companies. Globe, another
telecommunications company, and Smart were the top and competing companies in the
sector until a third up-and-coming player entered the market, Sun Cellular (owned by
Digital Telecommunications Philippines or Digitel) in 2003.

2 Bongquin, C. and P. Quintos. “Cartel, price fixing fight gets boost as anti-trust law takes effect. ABS-CBN News
Online. 02 August 2017.

10
It was also in 2003 when Globe and Smart filed separate complaints before the National
Telecommunications Commission (NTC), charging Sun Cellular with predatory pricing
when it offered unlimited call and text messaging (Abad et al. 2012). The two big firms
petitioned to implement fixed call rates and prevent the newcomer, Sun Cellular, from
charging much lower rates. The NTC ruled in favor of Sun Cellular. This incident intensified
the competition as the two big firms offered attractive and competitive packages to the
market.

Aside from predatory pricing, Sun Cellular had been accused of misleading or false
advertisement. In 2009, Sun Cellular-owner Digitel filed a complaint before the NTC
objecting to the false advertisements circulated by Red Mobile, another
telecommunications company. The ad had given the impression that Sun Cellular had lower
coverage (cellular sites) than it has.3

Mergers can lead to the lessening of competition. When PLDT acquired Digitel (including
Sun Cellular) in 2011, industry observers assessed that PLDT would control a majority
(about 70%) of the mobile network market. While it has benefits of improving and
enhancing the efficiency of services, the merger could lead to increased market power.
Aldaba (2011) recommended that the government ensure market contestability and
regulate business practices that could restrict competition. Meanwhile, a more recent case
was the acquisition by industry competitors PLDT Inc. and Globe of San Miguel
Corporation’s telecommunications business, which was believed to have hindered the
entry of a third player. According to reports, the case has been pending because PLDT
objects the investigation by the PCC.

Like the telecommunications sector, the energy sector has also been regulated and
liberalized (in 2001). However, Aldaba (2008) finds that the early stages of deregulation
lacked clear rules and regulatory framework, including regulations related to access rules
for transmission and distribution (e.g., dispatching order) and pricing system that would
allow consumers to share inefficiency gains.

There are claims that unfair business practices, particularly vertical agreement and unfair
dealing affecting price, take place in the energy sector. Meralco received accusations of
buying power from affiliated independent power producers (IPPs) when lower prices were
available at the National Power Corporation (NPC). This business practice is said to impact
on consumers who eventually pay higher prices as they subsidize the high-cost firms
(presumed to be inefficient) (SEPO 2009). Nonetheless, it was reported that the PCC is
investigating the alleged collusion in power rates.4

Similar to the cement industry, the agricultural sector, particularly rice, corn, and sugar
sectors, have been accused of operating in cartels, which was manifested by the pricing
behavior in these sectors. Low farm-gate and high retail prices were attributed to cartels in

3 TeleGeography.com, “Digitel calls foul over PLDT’s ‘unfair’ ad-based service Red Mobile.” 11 June 2009.
4 Cahiles-Magkilat, B. “PCC includes DTI DAO in cement ‘cartel’ probe. Manila Bulletin Online. 02 August 2017.

11
rice and corn sectors and high local prices of sugar to ‘integrated sugar mandates’
that control mining, refining, and marketing. Thus, the PCC commissioned
Dr. Roehlano Briones to conduct further studies on the rice and sugar industry.5

Moreover, a cartel in the garlic sector was also suspected in the mid-2010s, as prices
increased unusually (even more than doubled) when there was no supply shortage. The
Office for Competition of the Department of Justice investigated the case, and in a 2014
report, described the modus operandi and identified cartel operators. However, no case
was filed at that time.6 This garlic cartel is now one of the cases being investigated by the
PCC.7

Further, in the food sector, a case of obstruction of competition and unfair trade practice
was filed before the DTI in 2009 by a food manufacturing company, CDO-Foodsphere Inc,
against a large canned tuna company, Century Pacific Group (CPG). CDO-Foodsphere
accused CPG of blocking the entry of its new product, the corned tuna, into the market, and
for convincing retailers to enter into voluntary loyalty programs that hold off competitors.8

These illustrate that there could be continuing cases of anti-competitive conduct in the
absence of a working competition law.

III. APPROACH AND METHODOLOGY

The underlying framework commonly used for analyzing the state of competition in markets
is the Structure-Conduct-Performance (SCP) model. (See Figure 1). This model was used by
Aldaba (2008)9 and will also be the underlying framework for this scoping study.

5 The published studies are accessible at the resources section of the PCC website (https://www.phcc.gov.ph/).
6 Uy, J. “DA to suspend 55 garlic importers”. Philippine Daily Inquirer Online. 08 August 2017.
7 Bongquin, C. and P. Quintos, op. cit.
8 Manila Bulletin Newspaper Online, “DTI probes complaint on alleged unfair trade practices in canned
tuna.”13 November 2009.
9 The SCP is also the starting point of Porter’s five forces of competitive position, and industrial organization.

12
Figure 1. The Structure-Conduct-Performance Framework

Structure Conduct Performance

concentration of advertising, research and Industry

producers (number of development,
sellers) growth
diversification of products
barriers to entry/exit, in price trends
general pricing and volume
Firm
large capital requirements capacity change: whether
to expand or contract profitability
capacity utilization
entry/exit, divestment, productivity
sunk capital mergers/acquisition, enter
into legal contracts
economies of scale/scope
cost control and hiring
Import competition schemes

External Factors

Demographics Government regulations

Technology Industry Specific – Regulation of monopolies/ public utilities

Horizontal – Business permits/licenses; trade and investment policy…

The SCP model highlights the roles of the industry structure (S), firm conduct (C), and
performance (P) that contribute to the state of competition in any market.

Structure

The structure of the particular market is the primary indicator of competition one would
need to assess. In simplest terms, competition exists when there are rival firms, not acting
in collusion, to supply the market. Thus, in assessing the state of competition of markets,
the first general indicator to look at is the fundamental element of the market structure—the
degree of market concentration. How many players are competing in the market, in the
first place? For this, a significant task of the study is to estimate two commonly used
indicators: firm-concentration ratio (CR) and the Herfindahl-Hirschman Index (HHI).

The HHI is computed as the sum of the squared market shares of all suppliers in the market.
The inverse of this ‘raw’ HHI is interpreted as the ‘effective’ number of competitors. Thus,
for example, if there are four firms, all with equal market shares, the HHI is equal to 4 *
0.0652 = .25. The inverse of this is 4 ( that is, equal to 1/.25), the number of rival firms. In
addition, the higher the HHI, the higher the concentration ratio, and the lower the ‘effective’

13
number of rival firms. The ‘raw’ HHI (sum of squares) is multiplied by 10,000 to come up
with the conventional HHI. In our example. HHI is 2500. We do the same for this study.

The CR is computed as the market share (whether as share in value-added or share in sales)
of top firms (e. g., the top 4 firms, CR4).

These are not perfect indicators, as would be elaborated on later. Also, considering which
thresholds to apply could be arbitrary. For example, in the case of HHI, a standard guideline
used (e.g., by the US and the EU on mergers) for HHI is:

HHI of below 1000 is considered ‘unconcentrated’

HHI of between 1000 and 1800, as ‘moderately concentrated’
HHI of above 1800, as ‘highly concentrated’

Using these thresholds indicates that there should be more than four competing firms
supplying the market for it to be considered as not concentrated.

This study proposes to use more ‘lenient’ thresholds, given the much smaller market, and
investments in the case of the Philippines. Our proposed classification for the study is:

HHI of below 1500 is considered ‘unconcentrated’

HHI of between 1500 and 2500, as ‘moderately concentrated’
HHI of above 2500, as ‘highly concentrated’

In the case of CR4, one could use the threshold of 70% for highly concentrated, between
40% and 70% as moderate, and below 40% as low concentration as done by Aldaba (2008).

It is not just the degree of concentration that matters in assessing the structure of the market
and the state of competition. High concentration is a necessary condition for the existence
of market power. Still, it is not a sufficient condition to indicate market power and the state
(or lack) of competition. In the end, what is essential is market contestability, the ease of
entry/exit of firms in a particular market. In this regard, even if one finds some market
concentration, one would need to examine further other factors affecting supply,
particularly, the presence of barriers to entry and the nature of these barriers. It would also
help to understand the nature of the supply and value chain, and where the potential for
competition squeeze can occur.

Hence, one looks at the concentration (number of sellers) as the first indicator of potential
market power (and lack of competition). However, this should be supplemented by an
assessment of barriers to entry conditions, and other supply and demand conditions.

For example, is there competition from imports (can imports come in freely)? If so, high firm
concentration would be less indicative of market power. Is the good mainly exported? As
such, the market is much broader than what firm concentration ratios (as estimated)
capture. As such, it would be useful to supplement concentration estimates with indicators

14
such as the import penetration10 or export ratio, and the tariff rate and existence of non-
tariff barriers (NTB). In addition to imports, are there close local substitutes? How much do
firms engage in product differentiation?

Aside from these market supply and demand conditions, are there government regulations
that impede the entry of firms? Are there significant factors inherent in the nature of the
industry that could affect the entry of firms? Some sectors could be characterized by
economies of scale, huge capital requirements, and sunk capital. In such cases, there could
be inherent asymmetries between new entrants and incumbent firms, where there could be
costs that had to be borne by entrants but not by incumbents. For example, sunk capital
could either make it easier for a firm to erect barriers to entry, or make it difficult for it to
exit the market. Such a case is typical among industries with large capital requirements and
economies of scale: similar to excess capacity, which could be an indicator of an entry
barrier. In this case, capacity utilization is a useful indicator over time.

How elastic is the demand for the product? If demand is inelastic, the firm would have
greater market power as it will be able to induce a more considerable price increase when
limiting supply (the consumer is at greater mercy of the supplier). On the other hand, if
demand is elastic, limiting supply leads to a commensurate decline in firm revenues, and
the firm would thus have limited market power. The rate of growth of demand for the
product also makes it more difficult to sustain market power. (See World Bank and OECD,
1998 for a more in-depth discussion of barriers to entry).

In sum, various elements of market structure that impact market contestability need to be
examined. These include, among others:

• concentration of producers (number of sellers)

• barriers to entry/exit, in general
• large capital requirements
• capacity utilization (excess capacity could be an entry barrier)
• sunk capital (providing asymmetry between incumbent and potential entrants)
• economies of scale/scope (usually associated with sunk costs)
• import competition
• product homogeneity
• product differentiation
• elasticity of demand
• rate of growth

10 For more in-depth studies, estimating concentration ratios should include supply of imports in total domestic
supply (not just total local supply produced by local incumbent firms). This was attempted in the study but major
differences in industry classification (PSIC) and commodity classification used for imports and lack of data did
not make it possible for an overall manufacturing scoping study. What was done instead is to use imports as
one of the explanatory variables, as would be seen in the later section of the study.

15
Conduct

What do firms do to compete with other firms? Different firms have varying strategies. The
question is, are these within legal means (that is, not in violation of competition law)? Or do
they compete by striving to be more innovative and efficient? Some examples of what a
firm does to keep or enhance its market share or to maximize its profits are:
• advertising,
• research and development,
• diversification of products
• pricing and volume (which, in the presence of competition, should be close to
marginal cost)
• capacity change: whether to expand or contract
• entry/exit, divestment, mergers/acquisition, enter into legal contracts
• cost control and hiring schemes
• process and product innovation

What the PCC should watch out for are signs of anti-competitive behavior. These are the
prohibited acts under the law, which could include predatory pricing behavior, collusion
and cartel behavior, or unlawful refusal to deal. These are often difficult to spot and verify.
After finding a high concentration in the market, as earlier mentioned, more supplementary
indicators are needed, such as significant structural barriers to entry (including
government-photo-policy induced).

An example of an indicator that could reveal the anti-competitive conduct of a firm is the
presence of excess capacity. The incumbent firm/s might invest in excess capacity to deter
entry. It holds excess ability in reserve and threatens to use it if a new player would have
plans to enter. It launches a price war, thereby rendering entry unprofitable. The criticism
here is that the threat to utilize capacity post entry may not be credible because it will likely
not be profit-maximizing to increase output. Can the incumbent credibly maintain or
improve output post-entry? The question is thus whether or not an incumbent can credibly
threaten to produce the limited output post entry.

Another argument that an incumbent might use excess capacity as an entry deterrent is to
consider the excess capacity as sunk costs. It could become a strategic approach of the
incumbent, as sunk costs would provide it a cost advantage by reducing its variable costs
versus new firms post entry.

Such is an illustration of the complexity of proving anti-competitive conduct. Even the case
of the PLDT/Globe Telecoms acquisition is not clear-cut. Nonetheless, this is a clear case
for the investigation by the PCC because of the claim to a necessary resource (bandwidth)
which could limit the entry of rival firms. Whether the deal would lower consumer welfare
is not conclusive, as regulations and conditions could be imposed to make sure that overall
welfare is enhanced.

16
Performance

Finally, in assessing the state of competition in particular markets, there is a need to look at
firm performance, which is usually measured by productivity, firm growth, and profitability.
For a competition policy body, the question is whether the firms are enjoying “monopoly
rents” or “abnormal” profits (that is, earnings over and above the ‘normal’ returns to capital
that results from competition). Is the profitability performance of a firm an indicator of
efficiency or market power?

The second primary task of the study is to examine the performance of industries in the
manufacturing sector that could indicate the existence of monopoly rents. Towards this
end, a commonly used indicator is the (Lerner’s) Price-Cost margin (PCM).

The Price-Cost Margin is supposed to capture how much the market price (P) deviates from
marginal cost (MC). In a perfectly competitive market, P = MC and (P–MC)/P reflects market
power (how much the monopolist can control price and maximize profits over and above
the competitive level).

Hence, we want PCM to reflect a deviation from marginal costs. The problem is that
Marginal Costs are not observable from the PSA data set, which is the source of data we
need to come up with an assessment measure covering all manufacturing. Although PSA
Census/Survey provides establishment data, at best, this would represent average costs
and output values. It has been one of the criticisms about using the accounting PCM data.
One suggestion is using estimates of marginal costs from econometrically derived
coefficients (assuming a Cobb-Douglas production function). Aldaba (2008) did this and
found, as most previous studies did, that results remain the same. Hence this study uses
PCM from PSA’s annual data.

Another critical question is the comparability of using PCM as such over different sectors.
This arises mainly from differences in capital intensity and length of production cycle across
industries. Price-cost margins that allow viability of the firm/sector would depend on how
short (or long) the production cycle is and how much capital is leveraged. Higher PCM, for
example, is needed for industries with longer production cycles and higher capital
intensities. As such, some allowance should be given to this difference across sectors.
Hence, this study looks at the potential “abnormal” profits of the firm, industry, or sector to
use as a basis for the estimate of monopoly power, instead of directly looking at the wedge
between price and marginal cost (the PCM).

For simplicity, we use a one-period analysis (and simple inputs and output). Let us denote
the following (annual) variables as follows:

VO = value of output
RM = raw materials used
w = wage rate
L = labor used

17
 d = depreciation rate
R = rental payments, interest and amortization
K = capital used.

The profit rate, ∏/K, (or return to equity) accruing to the activity is derived as:

∏ / K = (VO – RM – wL – dK – R)/K

Ideally, under perfect competition, excess profits (abnormal profits) are zero. Above that is
rent. Specifically, if r denotes normal profits, ∏/K – r would be excess profits (or an indicator
of monopoly rent).

∏/K - r = (VO – RM – wL - dK - R)/K – r. Or Equation 1

∏ - rK = VO – RM – wL – dK - R - rK.

“Normal” rate of return, r, should equal the social rate of discount. We could assume, for
now, for r to be 10% (Medalla 2014). This means that ∏ - rK is measured using census of
manufacturing data as follows:

∏ - .10K = VO – RM – wL – dK - R - .10K
= Sales – Cost of Goods Sold - .10 K

where Sales = VO;

Cost of goods sold includes RM, fuels, electricity, Labor, overhead costs.

The “normal” rate of return should be the same across sectors in the ideal world, where
resources could move freely. Also, in assessing the state of competition and monopoly
power across industries in manufacturing, it is not necessary to come up with absolute
measures. Hence, we could simplify the estimation further by dropping the last term to
come up with relative measures across sectors (and over time, as r would be relatively stable
over the medium term).

Hence this brings us back to ∏ = VO – RM – wL – dK - R, as a relative measure of ‘monopoly’

rents.

Note that PCM = (VO – RM – wL)/VO. Hence, dividing ∏ by (VO), we have:

∏/(VO) = PCM – (dK + R)/VO. Equation 2

Aldaba (2008) refers to this as an adjusted PCM. Hence, this study estimates the adjusted
PCM or APCM.

Using APCM, however, does not adequately address the earlier question about how much
monopoly power is affected by capital intensity (or magnitude of capital requirement). Of
course, capital intensity, in theory, should not affect the competitive outcome of P = MC, a
marginal cost. In practice, however, businesses look at long-run average costs and long-run

18
returns to capital. Indeed, zero profits, even under perfect competition, do not mean zero
returns to capital. Instead, capital should not earn more than the cost of money, which, from
the society’s longer-run perspective, is the social rate of discount. Again, this is what we
consider the ‘normal’ return to capital.11

A possible adjustment to PCM that follows from the above discussion can be derived from
Equation 1. As previously mentioned, we can use 10 percent as the ‘normal’ rate of return,
which is the most recent estimate of the social discount rate (Medalla 2015).

If r= 10%, Equation 1 becomes:

∏/K - .10 = (VO – RM – wL - dK - R)/K – .10

There is monopoly rent if this is greater than 0.

That is:

(VO – RM – wL - dK - R)/K – .10 > 0

Or
(VO – RM – wL - dK - R)/K > .10.

This suggests a modified PCM, MPCM which uses K as a divisor, instead of VO (in contrast
with APCM used by Aldaba).

MPCM = (VO – RM – wL - dK - R)/K.

If MPCM is consistently above 10 percent (over a while), there is a significant indication that
the firm is enjoying monopoly rents.12

It is assumed that the capital is not sunk cost. Note that, if indeed, part or all of the capital is
sunk cost, MPCM will be underestimated. To illustrate, if we deduct sunk costs from K, and
assume that this is close to 100% of K, the denominator will be close to zero and MCPM will
be very high, which indicates monopoly power. It is consistent with the situation (earlier
mentioned) that sunk capital could provide a barrier to entry arising from the asymmetry
between incumbents and potential entrants.

This paper attempts to calculate both APCM and MPCM. There are more readily available
data from PSA, e. g. sales and cost of goods sold (which include overhead costs like
depreciation, fuel costs, electricity, etc.) to estimate APCM. For MPCM, an estimate of K
(which ideally should be the replacement cost of capital) is needed. However, it poses a

11 The discussion about having ‘normal’ profits disregards differences in levels of risks across different types of
investments. For this study, results are annual averages, as it uses PSA data, even at the disaggregated four-
digit manufacturing PSIC level. This is enough for a general scoping study.
12 It is important to see trends over time, rather than just looking at a single period, to arrive at a more robust
conclusion.

19
more difficult constraint. Two possible statistics that can be obtained from the PSA census
or survey to estimate K are (1) the book value of fixed assets (BVA) and (2) depreciation.
BVA is the acquisition value of assets (capital) less depreciation. The problem is that the
PSA survey does not provide the acquisition values,13 nor the age and expected life of the
asset/s. Capital expenditures could occur in different periods (but statistics provided would
only be for the current year), while the annual depreciation reported is the total for all capital
assets. Given these limitations, there are two stages of difficulties in coming up with the
estimate of K. First, the acquisition value cannot be directly derived from reported BVA and
depreciation for the year. And two, even if we are then able (using reasonable assumptions)
to estimate the acquisition value, this should be adjusted for inflation (and possibly deflated
by the difference in productivity between ‘old’ and ‘new’ capital). Unfortunately, given the
limited time and resources, this study is not able to come up with enough information to
make the necessary adjustments (either in BVA or depreciation) to estimate K. Hopefully,
future efforts could produce better results.

SCP relationships

The discussion above also suggests that one needs to go further and understand better the
relationships between the market structure, conduct, and performance of firms. The most
evident direction of the relationship in the SCP model is that the market structure affects
the behavior and performance of firms and the overall industry. However, the course of
effects could be two-way. Conduct could affect the structure. A firm could choose to divest
or merge with another (Conduct) for efficiency reasons (Performance) and thus influence
the structure of the market. Performance could also affect conduct. For example, the higher
the profits (Performance), the higher the firm could allocate to advertising or R&D (Conduct)
and this would then again impact on the market structure and performance of the firm.

Outside these interactions in the SCP, there would also be external shocks that can affect
these SCP elements. A major external blow is government policy or regulation itself. In the
analysis, this is a crucial aspect to consider, mainly as it affects the entry and exit of firms
and thus the market structure and market power.

There is no one size fits all configuration. Whether one element is more important than the
other, or which is the primary factor to consider could vary. Economists also differ about
which should be given emphasis. Does market structure largely determine market conduct?
This is the traditional “structure performance hypothesis.” Or is the market structure that is
possibly characterized by high concentration an efficient market outcome? This is the
“efficient structure” hypothesis.

The PCA implicitly recognizes this. In fact, various clauses of the PCA provide for the
exemptions and exceptions, which are mainly based on the impact on efficiency and

13 The PSA Census/Survey provides only capital expenditures for the year.

20
consumer welfare. In this regard, the PCC focuses on the conduct of firms. When firms
invest in R&D activities, product and process innovation to keep their competitiveness and
market share, there is evidence of pro-competitive impacts and pressure. The PCC’s task is
to watch out for anti-competitive firm conduct14 that leads to harmful, ‘unfairly gained’
profits that counter innovation and efficiency.

To recapitulate, structure, conduct, and performance are interrelated. The market structure
would have an impact on the conduct of firms, such as strategies, cost minimization, and
measures taken, to name a few. Market structure could also make it easier for firms to
collude if the market is highly concentrated. Inversely, conduct could also impact the
market structure, as firms try to keep or enhance their respective market shares. They could
do this within legal means, like advertising and R&D. Or they could use non-legal measures
listed as prohibited acts under the competition law. Either way, it is the task of a
Competition Authority to determine if the firm’s conduct, whether it enabled or was
enabled by the market structure, violates the competition law.

Furthermore, high profits could have been enabled by excellent and efficient strategy or
anti-competitive conduct or, in turn, may have enabled firms to conduct individual acts or
strategies. Nonetheless, compliance with the competition law has to be checked.

Estimates of concentration (CR4 or HHI) and PCM would be the starting point of this general
assessment and scoping of competition in the manufacturing sector. Since the results could
indicate the state of competition as well as its impact on the performance and market
power, it would be interesting to identify these relationships empirically.

Past studies show mixed results. Some find a weak relationship between the structural
variables and performance (Salinger, 1984). The same is likely to be the case as well for this
study. Nonetheless, the regression results would still be useful and insightful either way. Is
there a significant correlation between these two variables? A positive correlation between
concentration and PCM generally suggests uncompetitive markets and a high potential for
abuse of dominance. It is the expected result of a structural hypothesis of the SCP model.
Again, even then, this would need to be supported by further analysis of barriers to entry
and other performance indicators. For example, the conclusion could be strengthened by
findings on lack of innovation or a decline in productivity. On the other hand, a negative or
even an absence of correlation weakens the case for the presence of abuse of market
power.15

Aside from APCM, the study attempts to get some measure (estimate) of total factor
productivity (TFP), or at least, labor productivity. What happens to productivity would reveal
a lot about the impact of market structure on performance. If the higher concentration is

14 Whether by unfairly shutting out and/or colluding with rival firms, or by exploiting its dominant position.

15 Similarly, if we find no general correlation, are there particular sectors with consistently high concentration and
PCM over time? What are they? And what are possible explanation. Is it the type of industry? Government
policy?

21
accompanied by increased productivity, the opposite, or no correlation at all will show how
important it would be to guard against monopolies.

Hence, possible regressions to be done would include:

(1) ones that look at the relationship between concentration and price-cost margins, and
(2) those that look at factors that could explain or correlate with concentration.

These are discussed further in the presentation of findings and results.

Given that this paper is a scoping study of the manufacturing sector, it aims to provide a
bird’s-eye view of the state of competition in manufacturing and trends related to it, and
suggest a tool and some guidelines for prioritization. For this purpose, the study employs
the most useful comprehensive data set collected by the PSA on the Census and Survey of
manufacturing establishments, at the four-digit manufacturing PSIC (Philippine Standard
Industrial Classification).16 Whether the level of classification of the sector represents or
defines the market is not considered in the use of such data sets from PSA. Identifying
relevant markets requires more in-depth studies of the specific sectors, subject to
investigation of the Commission.

IV. EMPIRICAL FINDINGS: SCOPING COMPETITION IN THE

PHILIPPINE MANUFACTURING SECTOR

Brief Overview of Recent Manufacturing Performance

First, it is useful to keep in mind the general context of the results: the performance of
the manufacturing sector.

The manufacturing sector constitutes about 68% of the industrial sector and 23% of
GDP (2016). It employs around 8% of employed persons in all industries. The
industry experienced sluggish growth in the past two decades but has demonstrated
growth in the last five years. It has even surpassed the annual increase in the services
sector, particularly in 2013 (at 10.3%) and 2014 (at 8%), and average growth in the
period 2012-2016, with manufacturing growing at 7.3% and the services sector at 6.9%.

Labor productivity has also been increasing in manufacturing. The sector’s

labor productivity (computed as GVA per employed worker) grew by about 24% from
2012 to 2016. In the same period, the industry registered an average annual growth rate
of 5.3%, which is higher than that of services (3.6%) and agriculture (3.2%).

16 This four-digit manufacturing PSIC is five-digit from the point of view of the census of the whole economy which
covers services, agriculture, forestry fishing and mining as well.

22
Sectoral GVA distribution in manufacturing in 2012-2016 revealed some of the significant
contributors to GDP. These include food manufactures, which posted an average share of
8.2%; radio, television and communication equipment and apparatus, 3.9%; chemical and
chemical products, 2.6%; furniture and fixtures, 1.1%; and beverage industries, 1.0%.

However, in 2016, the top contributors to manufacturing GVA did not contribute
significantly to the 7% growth in production (some of them even posted negative growth).
The major contributors to growth include office, accounting and computing machinery,
which grew by 43%; basic metal industries, 40.5%; machinery and equipment except
electrical, 24.9%; transport equipment, 24.4%; rubber and plastic products, 24.4%; and
wood, bamboo, cane and rattan articles, 18.5%.

Recent data also indicate that GVA has been increasing in some of the subsectors such as
footwear and leather and leather products; basic metal; electrical machinery and
apparatus; transport equipment; and food manufactures. (See Table 6)

Table 6. Manufacturing Sector Performance

2012 2013 2014 2015 2016

Gross value added (in million pesos, at
2000 constant prices) 1,395,711 1,538,912 1,666,514 1,760,989 1,884,320
GVA growth rate (%) 5.4 10.3 8.3 5.7 7.0
Share to GDP (%) 22.1 22.8 23.3 23.2 23.2
Employment (in thousand workers) 3,112 3,159 3,212 3,209 3,390
Share to Total (%) 8.3 8.3 8.3 8.3 8.3

Labor Productivity (in thousand pesos per

worker) 448.5 487.2 518.8 548.8 555.8

Manufacturing Value Added (in million

pesos, at 2000 constant prices)
Food manufactures 531,704 554,984 593,577 603,249 652,709
Beverage industries 60,303 58,632 73,080 72,375 79,341
Tobacco manufactures 4,675 4,349 4,307 5,480 5,854
Textile manufactures 30,102 26,435 30,428 32,384 29,737
Wearing apparel 39,554 33,330 31,994 31,258 31,332
Footwear and leather and leather
products 6,269 6,993 7,137 7,478 8,110
Wood, bamboo, cane and rattan articles 14,316 13,316 13,567 17,366 20,572
Paper and paper products 13,592 12,708 13,437 15,392 16,401
Publishing and printing 8,509 8,225 15,308 17,916 18,791
Petroleum and other fuel products 48,790 43,266 49,683 49,035 49,689
Chemical & chemical products 95,267 184,363 191,229 220,902 242,814
Rubber and plastic products 22,516 23,208 24,561 25,398 31,596
Non-metallic mineral products 38,010 41,392 39,637 43,362 41,976
Basic metal industries 20,983 31,348 33,218 35,290 49,587
Fabricated metal products 13,961 14,063 20,335 21,994 21,986

23
 2012 2013 2014 2015 2016
Machinery and equipment except
electrical 20,271 21,426 26,568 31,424 39,245
Office, accounting and computing
machinery 20,940 20,936 23,638 20,342 29,090
Electrical machinery and apparatus 35,749 33,405 34,476 37,373 42,035
Radio, television and communication
equipment and apparatus 238,396 262,166 276,537 311,241 305,489
Transport equipment 33,285 26,845 28,867 31,301 38,943
Furniture and fixtures 53,346 77,078 94,741 90,378 89,898
Miscellaneous manufactures 45,176 40,444 40,189 40,050 39,124
Manufacturing Value Added (% Share to GDP)*
Food manufactures 8.4 8.2 8.3 7.9 8.0
Beverage industries 1.0 0.9 1.0 1.0 1.0
Tobacco manufactures 0.1 0.1 0.1 0.1 0.1
Textile manufactures 0.5 0.4 0.4 0.4 0.4
Wearing apparel 0.6 0.5 0.4 0.4 0.4
Footwear and leather and leather
products 0.1 0.1 0.1 0.1 0.1
Wood, bamboo, cane and rattan articles 0.2 0.2 0.2 0.2 0.3
Paper and paper products 0.2 0.2 0.2 0.2 0.2
Publishing and printing 0.1 0.1 0.2 0.2 0.2
Petroleum and other fuel products 0.8 0.6 0.7 0.6 0.6
Chemical & chemical products 1.5 2.7 2.7 2.9 3.0
Rubber and plastic products 0.4 0.3 0.3 0.3 0.4
Non-metallic mineral products 0.6 0.6 0.6 0.6 0.5
Basic metal industries 0.3 0.5 0.5 0.5 0.6
Fabricated metal products 0.2 0.2 0.3 0.3 0.3
Machinery and equipment except
electrical 0.3 0.3 0.4 0.4 0.5
Office, accounting and computing
machinery 0.3 0.3 0.3 0.3 0.4
Electrical machinery and apparatus 0.6 0.5 0.5 0.5 0.5
Radio, television and communication
equipment and apparatus 3.8 3.9 3.9 4.1 3.8
Transport equipment 0.5 0.4 0.4 0.4 0.5
Furniture and fixtures 0.8 1.1 1.3 1.2 1.1
Miscellaneous manufactures 0.7 0.6 0.6 0.5 0.5
Source: PSA
* Used data in 2000 constant prices (from www.industry.gov.ph)
Note: Labor productivity was computed using data from PSA.

The manufacturing sector performance has thus been nothing short of outstanding during
the last six years, a departure from its decades of lackluster performance in the past.

24
Findings on Concentration Ratios and the Herfindahl Index

Following Aldaba (2008), 4-firm concentration ratios were calculated for industries (four-
digit manufacturing PSIC) and then summarized at the two-digit PSIC level (weighted
average using share in total manufacturing value-added). In 2006, the manufacturing sector
in the Philippines had, on average, a 4-firm concentration ratio of 68 percent. While this
value increased slightly to 72.7 percent in 2008, it had settled to 65.4 percent in 2014 (see
Table 6). These figures indicate that improvements in the overt measure of the degree of
competition in the manufacturing sector did occur, albeit limited and with some losses over
time.

Historically, only printing and reproduction of recorded media and manufacture of rubber
and plastic products have 4CR less than 40 percent. In 2014, aside from the sectors
mentioned earlier, paper and paper products manufacturing also has 4CR below 40
percent.

One of the criticisms about the use of the 4CR is that it fails to consider the market shares
of all the firms in a given industry. The Herfindahl-Hirschman Index (HHI) of market
concentration may be used to address this issue. HHI measures market concentration in the
form of the sum of the squared market shares of all companies in the industry (equation 2).

𝐻𝐻𝐼 = ∑𝑁 2
𝑖=1 𝑆𝑖 (equation 3)

By using the square of the market shares, the more significant weight is given to firms with
larger market shares.

HHI value ranges from near zero, which reflects perfect competition, up to 10000, as in the
case of a pure monopoly.17 An industry or market is non-concentrated if the HHI is less than
1500, while the market is moderately concentrated if the value of the index ranges between
1500 and 2500. The value of HHI above 2500 represents a highly concentrated industry.
Table 6 presents a summary of the HHI at the 2-digit PSIC level. Similar to the findings of
Table 6, Table 7 shows that Total Manufacturing is highly concentrated with HHI values
ranging from 2000 to 3000. In 2014, Total Manufacturing had an HHI value of 2541.

The HHI index confirms the consistently highly-concentrated sectors identified by 4CR:
Tobacco products, Leather and related products, Coke and refined petroleum products,
Machinery and equipment, n.e.c., and Manufacture of other transport equipment.

17 Note that the inverse of HHI (times 10000) can be interpreted as the effective number of rival firms (players).
HHI = 1000, for example is like having 10 rival firms or industry players, and HHI = 10000, one firm or
monopoly.

25
Table 7. 4-firm Concentration Ratio summarized at 2-digit PSIC level

PSIC 2-
Description 2006 2008 2010 2012 2014
digit
High concentration (80% and above)
C19 Coke & refined petroleum prods 100 99.9 99.9 99.8 99.5
C12 Tobacco products 99.2 99.1 95.8 99.9 99.6
C30 Manufacture of other transport eqpt 92.3 93.1 97.7 97 93.7
C28 Machinery and equipment, n.e.c. 93.2 87 79.4 86.7 81
C15 Leather and related products 80.3 85.6 83.7 77.5 84.3
C16 Wood and cork products, etc. 59.1 82.5 83.2 88.9 87.2
C27 Electrical equipment 68.9 72 87.8 83.6 77.3
C32 Other manufacturing 76.9 73.7 72.9 81.5 76.2
C33 Repair/installation of mach & eqpt 82 57.5 78.6 75.4 67.1
C21 Pharmaceutical & pharmaceutical preps 70.2 77.7 72.1 65.4 72.7
C20 Chemical and chemical products 62.9 71.5 73.6 70.5 68.4
C24 Basic Metals 79.6 66.8 79.6 60.2 58.6
Moderate Concentration (40 - 69%)
C10 Manufacturing of food products 67 70.8 69.6 65.9 62.8
C13 Textiles 73.9 62.4 65.8 67.6 54.1
C11 Beverages 63.8 54.9 62.1 59.3 66.9
C23 Non-metallic mineral products 55.5 56.1 64.2 62.8 68.1
C29 Vehicles, trailers, semi-trailers 61.1 66.3 56.7 39.2 81.3
C26 Computer, Electronic & electrical prods 63.9 59.6 55.9 63.1 47.9
C31 Manufacture of furniture 43.9 48.2 53.7 56.3 57.3
Fabricated metal prods expt mach &
C25 46.9 44.4 51.5 47.3 48.4
eqpt
C17 Paper and paper products 45.7 47.8 45.3 42.6 37
Low concentration (40% and below)
C14 Wearing apparel 32.8 36 44.7 42.9 46.1
Printing/reproduction of recorded
C18 27.7 27.2 49 28.5 33.7
media
C22 Rubber and plastic products 26.5 25 26.5 24.3 21.6
C Total Manufacturing 68 72.7 72 63.8 65.4
Note: Concentration ratios were calculated at the 5-digit PSIC level and then summarized at the 2-digit PSIC using GVA as weights.
Source: CPBI 2006; ASPBI 2008, ASPBI 2010, ASPBI 2014; CPBI 2012; Philippine Statistics Authority

26
The 4CR indicated only three unconcentrated industries in 2014. While the HHI identified
more under the same classification in the same year, including Textiles, Wearing apparel,
Paper and paper products, Printing and reproduction of recorded media, Rubber and
plastic products, Fabricated metal products expt machinery and equipment, Electrical
equipment, Manufacture of furniture.

Table 8. HHI summarized at 2-digit PSIC level

PSIC Description 2006 2008 2010 2012 2014

High concentration, HHI > 2500
C12 Tobacco products 4,479 4,641 3,243 9,082 7,181

C19 Coke and refined petroleum products 5,111 5,092 5,106 5,110 4,696
C30 Manufacture of other transport eqpt 4,693 4,836 4,488 4,580 5,474
C15 Leather and related products 4,773 5,976 5,158 3,894 3,332
C28 Machinery and equipment, n.e.c. 6,310 4,924 2,612 4,480 3,409
C24 Basic Metals 4,822 3,503 4,834 2,528 2,179

C16 Wood and cork products, etc. 1,720 3,571 3,473 3,980 3,706

C27 Electrical equipment 2,125 3,408 4,233 3,300 2,634

C32 Other manufacturing 2,408 3,057 2,982 3,277 2,741

C21 Pharmaceutical & pharmaceutical preps 1,954 3,273 3,011 2,549 3,259

C10 Manufacturing of food products 3,415 3,314 2,338 2,274 1,859

C23 Non-metallic mineral products 1,454 1,633 2,275 3,055 4,507

C33 Repair/installation of mach & eqpt 3,923 989 2,900 2,483 2,270
Moderate Concentration, 1500 < HHI < 2500
C20 Chemical and chemical products 1,735 2,524 3,169 2,789 1,992

C29 Vehicles, trailers, semi-trailers 1,610 1,879 1,597 731 4,218

C13 Textiles 2,929 1,417 1,743 1,941 1,125
C11 Beverages 1,562 1,411 1,474 1,547 1,951
Low concentration, HHI < 1500
C26 Computer, Electronic & electrical prods 1,804 1,513 1,265 1,586 953

C25 Fabricated metal products expt mach & eqpt 1,277 1,228 1,584 1,295 1,317

C31 Manufacture of furniture 813 964 1,148 1,598 1,291

C22 Rubber and plastic products 682 827 1,050 917 811
C17 Paper and paper products 831 847 899 733 572
C14 Wearing apparel 528 580 836 776 920

C18 Printing/reproduction of recorded media 393 414 1,027 427 517

C Total Manufacturing 2,695 3,088 2,828 2,078 2,541
Note: Concentration ratios were calculated at the 5-digit PSIC level and then summarized at the 2-digit PSIC using GVA as weights.
Source: CPBI 2006; ASPBI 2008, ASPBI 2010, ASPBI 2014; CPBI 2012; Philippine Statistics Authority

27
Compared with the results of Aldaba (2008), which cover the country’s substantial trade
reform period, the manufacturing sector appears to have become less concentrated.
However, as noted earlier, the average concentration ratio is still higher than 70 percent for
specific years. The reduction, however, is significant and has fallen below the ‘high
concentration’ threshold by 2014. It is worth noting that this trend coincided with a period
of remarkably high growth in GDP and high growth in the number of establishments. (see
Table 8 and Table 9)

Table 9. 4-firm Concentration Ratios and Number of Establishments, during and post-
trade reform periods

Year CR4 # of establishments

Trade Reform period
1988 70.88 11208
1994 73.63 10726
1995 73.64 10373
1998 80.55 15674
Post Trade Reforms (pre-PCA)
2006 68 18,331
2008 72.7 15,868
2010 72 15,849
2012 63.8 25,038
2014 65.4 25,166
Source: Trade reform Years: Aldaba 2008
Post-trade reform, before the Philippine Competition Act: Authors' Computations

28
 Table 10. Number of Establishments by 2-digit PSIC manufacturing sector, 2006-
2014

We also examine vertical and horizontal sectors and compare the 4CR and HHI at the PSIC
4-digit level. In particular, we look at tobacco products, leather and related products, textile
and wearing apparel, and beverages (Table 11; complete listing in Annex 1).

Tobacco products have high concentrations based on the 4CR and HHI computations at
the 2-digit PSIC level. A disaggregation of the industry indicates that the manufacture of

29
cigarettes composes mostly of the establishments (around 40%) of the tobacco industry
and is classified to have high concentration based on 4CR and HHI in 2006-2014. The other
sectors registered high density but have fewer establishments. The manufacture of chewing
and smoking tobacco; and Curing and re-drying tobacco leaves had, on average five
establishments from 2006-2014 and was decreasing in number during the same period.
Meanwhile, the manufacture of cigarettes increased from 5 to 8 establishments.

The other sector with a high concentration ratio or index is the leather and related products.
Generally, the leather and associated industries, from the tanning of leather to the
manufacture of footwear, registered a decline in the number of establishments (except for
manufacture of products of leather and imitation leather), but still classified as moderate to
highly concentrated; though declining slightly over the years. The manufacture of products
of leather and imitation leather increased in terms of the number of establishments (83 to
96 from 2006 to 2014), still moderate to highly concentrated, but is showing less
concentration over the years, as indicated by the substantial decrease in HHI.

In the manufacture of shoes, leather shoes have the most number of establishments. The
industry remains moderately concentrated during this period, though the number of
establishments has decreased considerably over the years from about 160 in 2006 to 80 in
2014.

As for beverages, most establishments are engaged in the manufacture of drinking water.
The number of establishments under this industry has increased over the years, while
concentration declined from low high to moderate. Meanwhile, the manufacture of
sports/energy drink has been highly concentrated, consisting of 3 establishments in 2006
and declining to 1 in 2014. For other beverages, there is an increase from 1 in 2010 to 3 in
2014, and a considerable decrease in HHI during this period. The sports or energy drink
sector is an interesting and illustrative example of how nuances in the nature of the product
matter. While there are only a few large manufacturers, the product has many substitutes
that would limit their market power.

For textiles and wearing apparel, most industries are in garments manufacturing, custom
tailoring, and custom dressmaking. These industries have low to moderate concentration,
except for custom dressmaking, which is moderate to high. Along the industry chain,
mainly, weaving or preparation of textiles, there are relatively fewer establishments, market
concentration is moderate, and competition seems to be improving as indicated by
concentration values decreasing over the years.

30
Table 11. 4CR and HHI for Selected Industries, at 4-digit PSIC Code

31
Further analysis of the market structure could include the growth indicators, related markets
other than local, e. g. exports or competing imports. Is the market growing, stable, or
declining? One expects that there will be demand for new production capacity, and that
entry will be easier if market demand is growing. In contrast, sectors which are declining
over time will not attract new investments. If the firms are producing for exports, then a high
concentration ratio as computed would lose much of its meaning as an indicator of
monopoly power (at least within the local market. Similarly, if there is supply coming from
imports, CR4 and HHI measures would be misleading indicators of monopoly power (real
market share will, in effect, be lower if imports are taken into account). This consideration is
thus included in the regression analysis presented below.

32
Findings on Price-Cost Margins in Manufacturing

Table 12. Adjusted Price-Cost Margin summarized at the two-digit level

2-digit
Description 2006 2008 2010 2012 2014
PSIC
C10 Manufacturing of food products 0.214 0.153 0.048 -0.053 0.122
C11 Beverages 0.193 0.148 0.153 0.124 0.252
C12 Tobacco products 0.167 0.238 0.338 0.089 0.011
C13 Textiles -0.206 0.037 0.011 0.073 0.043
C14 Wearing apparel 0.090 0.093 0.060 -0.094 0.065
C15 Leather and related products 0.152 0.000 0.049 -0.062 0.090

C16 Wood and cork products, etc. 0.173 0.068 0.045 0.074 0.135

C17 Paper and paper products 0.111 0.126 -0.592 0.115 0.113
Printing and reproduction of recorded
C18 -0.064 0.077 0.003 0.071 0.115
media
C19 Coke & refined petroleum prods 0.104 0.233 0.217 0.149 0.146
C20 Chemical and chemical products 0.104 0.096 0.053 -0.408 0.053
C21 Pharmaceutical and pharmaceutical preps 0.087 -0.172 -0.108 0.003 0.036
C22 Rubber and plastic products 0.134 0.080 0.082 0.029 0.075
C23 Non-metallic mineral products -0.025 0.267 0.177 0.190 0.184
C24 Basic Metals 0.113 0.128 0.127 0.069 0.136
Fabricated metal products expt machinery
C25 -0.030 0.041 0.050 0.429 0.094
and equipment

C26 Computer, Electronic and electrical products 0.103 -0.208 0.091 0.138 0.082
C27 Electrical equipment 0.140 0.078 -0.009 0.105 -74.955
C28 Machinery and equipment, n.e.c. 0.152 0.138 0.098 0.073 0.101
C29 Vehicles, trailers, semi-trailers 0.112 -0.010 0.071 -0.191 0.108
C30 Manufacture of other transport equipment 0.132 0.076 0.146 -0.038 -11.785
C31 Manufacture of furniture -0.319 0.076 0.095 0.086 0.133
C32 Other manufacturing 0.148 0.063 0.106 0.062 0.102
Repair and installation of machinery and
C33 0.110 0.023 0.007 0.152 0.080
equipment
C Total Manufacturing 0.118 0.094 0.095 0.066 -2.107

Initial results indicate that sector PCMs from 2006 to 2014 as shown in Table 12 are mostly
low to moderate. Some sectors that have over 10% PCM and have reached reasonable
margins (over 20%) at some point between 2006 and 2014 are manufacture of food

33
products; tobacco; beverages; coke and refined petroleum products; and fabricated metal
products expt machinery and equipment. Beverages have had higher margins (from 19%
to 25%) while the other four sectors’ margins decreased, though still over 10% (except for
tobacco and fabricated metal products expt machinery and equipment). Most of the
sectors/products’ PCM decreased from 2006 to 2014, except for printing and reproduction
of recorded media; manufacture of furniture; and basic metals, whose margins increased
by about 4 to 6 percentage points.

Estimates appear very low, mainly driven down by the high negative PCMs of -75 and -11.8
for sectors C27 (electrical equipment) and C30 (manufacture of transport equipment),
respectively. A possible explanation is accelerated depreciation for these sectors. On the
whole, however, the overall downward trend could be a temporary outcome of rapid
growth in the number of establishments. 18

It would be useful and interesting to find out what happens if we remove the two sectors
with large negative APCMs. Doing such could be justified not only because these may be
considered to be outliers, but even more because these are sectors that are mainly for
exports. Table 13 presents what has happened to the average APCM during and post-trade
reform periods, including and excluding these observations. With or without these
observations, the general reduction in ACPM is evident between the two periods. The
range went down from 23-34 percent to around 12 percent.

Table 13. PCM during and post-trade reform periods

Year Adj PCM

Trade Reform Period
1981-85 0.26
1986-90 0.23
1991-95 0.29
1996-98 0.34
Post Trade Reforms (pre-PCA)
all Mfg C sectors Exc. C27, C30
2006 0.12 0.12
2008 0.09 0.12
2010 0.10 0.12
2012 0.07 0.23
2014 -2.11 0.13

18 Scheduling limitations in accessing PSA establishment data have been a major constraint in cleaning up the
data and sorting out methodological problems.
34
Findings on the relationships between Structure-Conduct-Performance

Regression results

As discussed in the earlier section, a positive correlation between concentration and PCM
could suggest the presence of uncompetitive markets and a high potential for abuse. This
result is expected in a structuralist hypothesis. Our initial findings on the regression
estimates, however, are the opposite, showing a negative and significant correlation.

Using the single-variable regression analysis, we tried to calculate the correlation between
measures of concentration and price-cost margin (Table 14). Ordinary least squares
estimates show that there is a negative correlation between APCM and measures of
concentration. The same regression specification was estimated using panel fixed effects
to control the time-invariant omitted variable bias. The correlations still reveal a negative
and significant relationship between concentration and price-cost margin. See Table 14 .

Table 14. Correlation between concentration indices and price-cost margin

(1) (2) (3) (4)

VARIABLES CR4 HHI CR4 HHI

APCM -0.000819*** -0.000178 -0.000593*** -0.000279***

(0.000199) (0.000130) (1.76e-05) (4.63e-05)

Constant 0.772*** 0.347*** 0.772*** 0.347***

(0.00747) (0.00911) (4.59e-06) (1.21e-05)

Observations 886 886 886 886

F statistics 16.99 1.871 1141 36.25

Number of id 188 188

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

At first glance, these results appear counter-intuitive, as we do not expect a correlation

between sectors with higher concentration and low price-cost margins. It therefore implies
that the market environment for the manufacturing sector is competitive. This means that
the high concentration is correlated with a low price-cost margin. Because even in highly
concentrated industries, the incumbent firms appear to be not enjoying ‘rents’ or abnormal

35
profits. However, further consideration of its overall condition and performance at the time,
this may not seem improbable and may explain this general assessment of the state of
competition in the manufacturing sector. It was a period of high, accelerating, and
sustained growth for the manufacturing sector in particular. Firms will tend to go where
returns are high.

Nonetheless, we considered the possibility that this could be a ‘misleading’ result of the
high negative APCM in two sectors noted earlier. It would be interesting to find out what
happens if we remove the observations showing large negative APCM (sectors C27 and
C30).

The relationship between concentration and PCM became positive, but has also become
insignificant. Table 15 shows that it is more in line with the usual expectation. The sign of
the correlation suggests that the state of the ‘domestic’ market is not as competitive as
indicated by the previous results, and that, indeed, the competition authority should look
at concentration ratios. However, the weak (insignificant) finding magnifies the need to
consider other factors as previously discussed.

Table 15. Correlation between concentration indices and price cost margin: excluding
‘outliers’

Panel FE Panel FE
(3) (4)
VARIABLES CR4 HHI
APCM -0.00291 0.00553
(0.0094) (0.0133)
Constant 0.772*** 0.346***
(0.00088) (0.00125)
Observations 884 884
R-squared 0 0
F statistics 0.0959 0.172
Number of id 188 188
Robust standard errors in parentheses
*** p<0.01

Still, it is not clear that the high negative observations should be considered outliers. These
could be brought about by many factors such as fluctuations in the export market where
the sectors in question belong, or capital intensity and accelerated depreciation. Perhaps,
the modified regression results, excluding C27 and C30, would be more reflective of the
state of competition, if one would consider only the domestic market. However, the entire
manufacturing sector including those producing for exports, the original results would
perhaps be more applicable.

36
Conducting further analysis by estimating other specifications suggests that there might be
other factors contributing to the degree of concentration in the industry. While we still
obtain a negative and significant correlation between CR4 and PCM, the relationship has
weakened for HHI, which is no longer significant. What needs to be explained further is the
relationship of other markets to concentration ratios. Table 16 [(3) and (4)] shows that
industry groups that have external markets tend to have higher degrees of concentration,
as reflected by a positive and significant correlation between exports of the industry and
the dependent variable. Similarly, the positive association between imports of similar
products and concentration ratios shows that the presence of foreign competition fosters
further concentration in the industry. This is consistent with our earlier explanation about
Aldaba’s initial finding of increasing concentration ratios during trade reforms. Foreign
competition drives out inefficient firms that could lead to a higher concentration.

In sum, there appears to be a significant negative correlation between the degree of

concentration and PCM. A possible interpretation is that in general, a low PCM provides a
disincentive to new entrants or drives out less efficient firms, and thus a higher degree of
concentration. Additionally, a higher PCM encourages more firms to engage in the market
and hence more players and lower degree of concentration. This is the opposite of what is
expected from a monopolistic structural outcome, and more in line with an efficient, more
or less competitive market outcome. Furthermore, these results are not entirely surprising,
given the steady growth of the manufacturing sector during the period in question.

Table 16. Regression results including other explanatory variables

37
Table 17. Regression results on the relationship of concentration on productivity

We tried to look for other variables/factors that could affect the market structure. Table 17
above presents the relationship between concentration and labor productivity. Models (1)-
(4) reflects the relationship between HHI and labor productivity while Models (5)-(8)
demonstrates the correlation between CR4 and labor productivity. The positive and
significant coefficient in the model reflects that increases in concentration measured by the
change in concentration index correlates with increases in labor productivity, regardless of
the measure of concentration (HHI or CR4). The relationship is positive and significant,
indicating that concentration may be associated with increased efficiency more than
increasing the cost. The link seems robust across models even after controlling for time-
invariant omitted variables (models 2, 4, 6, 8).

Another explanation for the unexpected results could be deduced from the significant
positive correlation found between imports and concentration, along with a negative
correlation between PCM and concentration. This suggests that indeed, imports place
competitive pressure on the market, and that the concentration ratios as defined, i.e., ‘local’
firm concentration could be misleading.19

19 Imports summarized at the ISIC level (4-digit) is incorporated in the regressions but incorporating it to the
calculation of the 4CR and HHI (calculated at the 5 digit industry group) is problematic. There are several factors:
38
While the results are encouraging, the presence of relatively high concentration in specific
sectors implies that the PCC should remain vigilant in preventing anti-competitive acts, and
safeguarding the competitive process.

V. EXTERNAL FACTORS AFFECTING COMPETITION: SOME

CRITICAL GOVERNMENT POLICIES AND REGULATIONS
The government makes policies and regulations for various (valid) reasons and objectives.
These could include security, safety, environmental safeguards, health and sanitary
reasons, and provision of public goods. Generally, the market fails when either the market,
on its own, would not advance the social objective, or even when the market, would
independently operate against these objectives.

Government regulations could be industry-specific. The most notable examples are those
dealing with ‘natural’ monopolies and public utilities. Often, these industry-specific
regulations would have multiple objectives, like, safety and affordability. There are also
horizontal government regulations, such as, business regulations that are related to ease of
doing business. These serve particular purposes, such as taxation, accountability, and
traceability. However, these would always have impacts on markets- demand, supply, cost
of goods, and resources. No matter how good the intentions are for these government
policies and regulations, there will always be unintended consequences among which

(1) ISIC lumps together industries which we have distinctly differentiated in the calculation of concentration ratios
using the PSIC. For example: Manufacture of virgin coconut oil, Manufacture of fish oil and other marine animal
oils etc. belong in one group (Manufacture of oils). Another example would be: manufacture of leather shoes,
manufacture of rubber shoes, etc. also belong in one group (Manufacture of footwear). Lumping them in one
category implies that the firms in this "industry group" compete with each other in terms of market share. It is
not clear whether these products are substitutes to each other and thus compete for a share in the market.
(2) Using the ISIC instead of the PSIC expands the industry groups by incorporating too many firms which will surely
affect the concentration ratios. This is evident in the example on shoes (footwear). There are 6 sub-industries in
footwear (Manufacture of leather shoes, Manufacture of rubber shoes, Manufacture of plastic shoes,
Manufacture of shoes made of textile materials with applied soles, Manufacture of wooden footwear and
accessories, Manufacture of footwear, n.e.c) and to use the ISIC to incorporate imports would result to
concentration ratios decreasing because of the increased number of establishments dividing the market.
(3) Related to number 2 would be the limitation to assess the impact of imports because the same concentration
ratios that were initially calculated (using the PSIC) could not be compared with the ISIC.
(4) Finally, the ISIC does not distinguish between horizontal relationships (Example: Manufacture of leather shoes,
Manufacture of rubber shoes, Manufacture of plastic shoes, Manufacture of shoes made of textile materials with
applied soles, Manufacture of wooden footwear and accessories, Manufacture of footwear, n.e.c) versus vertical
relationships (Example: Preparation and spinning of textile fibers, Weaving of textiles, Finishing of textiles
Preparation and finishing of textiles (integrated)). The horizontal relationships are competing with the market
while those in the vertical relationships are actually part of a value chain. This actually complicates the
interpretation of the concentration ratios that could be calculated using these.

39
would be on the competitive process. Regulations have thus been a significant subject of
concern for competition impacts. As such, the PCC has noteworthy studies on law and
competition.

Horizontal laws related to governing business, particularly regulations on starting a

business and its operations have a direct significance to the manufacturing sector. They
directly affect the ease of entry and exit of firms, which have primary impacts on
competition. In terms of policy recommendations, the general principle is streamlining
these regulations and removing those that are counter-productive. Reforms along these
lines would possibly have the most positive impact, not just on competition, but on the
efficiency and viability of the manufacturing sector. Again, this has been a subject of study
by the PCC, and a grave concern shared by many government agencies, like the
Department of Trade and Industry (DTI).

Perhaps a less unified view is in the area of industrial policy. Indeed, there could be some
limitation of competition in the chosen industrial strategy of the DTI, especially where there
are interventions in preferred sectors, such as granting incentives that promise long-run
benefits. The main rationale of industrial policy is to address ‘market failures’ that make the
sector, firm, and industry unable to realize its potential. The ‘exemptions and exclusions’
are incorporated in the PCA for this purpose. In this sense, there is no conflict between PCC
and the DTI mandates. However, it should not prevent PCC from coming up with a review
and recommendations that could point out the possible anti-competition impact that could
go against the overall economic objectives. In the same manner, the industrial policy should
recognize ‘competition’ policy as one of its pillars. This means, for example, sunset
provisions in its preferential incentives and cooperation with the PCC in promoting
competition.

An excellent example of where competition and industrial policy could go hand in hand is
in SME development policy. Competition policy and SME development policy are mutually
reinforcing. Competition policy is a prerequisite for SME development. First of all,
competition policy and law, in general, would lower transaction costs for SMEs, especially
in terms of the potential impact of competition policy on reducing the costs of infrastructure
services, including ICT, transportation, and logistics. At a more micro-level, it prevents
more prominent firms from abusing their ‘bigger’ market power arising from more
significant market shares. Addressing abusive conduct that is prohibited or sanctioned by
the PCA levels the playing field for SMEs. In addition, the PCC’s implementation of
competition policy and law would promote an environment of trust and mitigate some risks
of doing business provided that the rules are fairly implemented, and the SMEs ensure that
market players use only efficiency and innovation strategies to compete. This element of
creating trust in the market environment would have invaluable benefits in terms of
reducing the cost of doing business.

On the other hand, SME development could reinforce Competition Policy. The primary
element of SME policies is not just the provision of subsidies to address market failure and
additional distortion arising from firm size but promoting ease of entry and regulations, and
40
providing facilitating measures, for example, in border trade and creating the right
environment for start-ups and then growing the business. In the end, these start-ups and
viable, growing SMEs would create significant competitive pressure on incumbents. Even
just a potential threat that start-ups could quickly emerge could pose a competitive threat.
Growing SMEs would create even more threatening potential competitive pressure. Thus,
the SME Development Policy lays a good foundation for a competitive market.

These points are well summarized by Allan Fels (Fels 2001),20 the former Chairman of the
Australian Competition and Consumer Commission (ACCC), as follows.

“A competition regime needs to cooperate in conjunction with other government policies.

Inevitably, conflict between policies will arise, and it will, therefore, be necessary to
determine priorities based on an assessment of national interests.”

In this regard, it is good to note that monitoring anti-competitive behavior is indeed a

mutual agenda of DTI and PCC and that they are seeking ways for better coordination and
exchange of relevant information.

VI. CONCLUDING REMARKS: MATRIX FOR PRIORITIZATION

The PCC is not a regulator in the usual sense. The PCA lays down rules of fair play, and it is
an essential task of the PCC to remove or ease barriers to entry. In essence, it has a
facilitative role in running the economy. Aside from this facilitative role, it also has a vital
role as an advocate for regulatory reforms.

Medalla (2000) prioritized the review and reform of regulations as a crucial task of
competition policy, especially for developing countries like the Philippines, where
infrastructure regulations are of utmost importance. The regulations impact the price and
efficiency of backbone infrastructure affecting the costs across all sectors of the economy.
However, it is not just infrastructure regulations that impose unnecessary costs and
impediments. The overall Ease of Doing Business (EODB) has pervasive impacts on the
entry and exit of firms, which is essential in promoting competition. As such, among its
priorities, PCC should add support to other agencies, particularly DTI/BOI, in getting rid of
unnecessary, harmful regulations. Even if its findings may only be recommendatory, its
position on certain issues would bolster actions for regulatory reform, especially when it
has established enough track record and credibility in recommending well thought out
policies.

Nonetheless, its core function remains to be the implementation of the PCA, which focuses
on preventing harmful anti-competitive acts. Competition analysis and investigations are

20 Presented in the conference on The Future of Canadian Competition Policy in the 21st Century in 2001.

41
complex endeavors. Studies, information and education, and quasi-judicial functions would
be regular components of PCC functions, covering theoretically all sectors of the economy.
It is a huge task, especially for a new competition authority. This study hopes to aid in the
prioritization process by providing some benchmark and guidelines.

Step 1. Prioritization Matrix. Categorize sectors, according to concentration and price-cost

margin.

The first step in coming up with the priority list is categorizing sectors, according to the
degree of concentration, in this case based on HHI and the estimated Price-Cost Margin
(PCM). Where there is both high PCM and high concentration would be a greater
possibility of market power and abuse thereof.21 See Table 18.

Table 18. Prioritization matrix

Low APCM Moderate PCM High PCM

(<10%) (10-15%) (>15%)
1500 < HHI < 2500
Low to moderate O C1 B1
concentration
HHI > 2500
High concentration C2 B2 A

Sectors that are highly concentrated and with consistently high price-cost margin would be
the top priority (Category A) for investigation. These are the sectors where there is highest
possibility of monopoly power and abuse. The cut-off levels suggested here are arbitrary,
and would likely need adjustments. Also, concentration groups could be adjusted if
deemed more appropriate. For now, we use below 10% as low, between 10-15% as
moderate, and greater than 15% as high. The next categories are B1 and B2. The lines
between B1 and B2 are more blurred. The high concentration means greater opportunities
for collusion, even if it does not show in the estimate of APCM that could arise from
estimation errors, given the data limitation.

On the other hand, the high APCM could be an indicator of rents, and the estimated HHI
could be high enough. An example of the latter is the case of cement. It appears to have
only a moderate degree of concentration, but among the highest APCM.22 For those falling
in these categories, consideration would be based on reports of anti-competitive conduct,
and other important, transparent, and known factors affecting barriers to entry.

21 We tried fewer categories using four quadrants, having only two groupings of PCM- high and low. However,
we found this to be too broad.
22 The APCM estimates for cement over the period ranges from 22-28 percent. On the other hand the
moderate/low concentration ratio could have been underestimated, as geographic location (and possible
market segregation) is not considered.
42
Judging by the list of sectors alone, it appears that the majority of sectors are at least
moderately unconcentrated. In contrast, there is a much shorter list of industries falling in
the highest priority- Category A (HHI > 2500, and PCM>15%).23 See Table 19.

Table 19. PSIC sectors with highest priority, Category A (HHI > 2500, and APCM>
15%)

23 This very lopsided result should be expected, given the positive correlation between concentration and PCM.

43
Those sectors with 1500 < HH <2500 and APCM <10% are in Category O (see Annex 6).
These are treated similarly as those with HHI < 1500 (unconcentrated). These are
considered competitive markets, and thus not in the priority list. Industries falling in this
category with HHI < 1500 are presented in Annex 7.

Step 2. Importance in the economy

The benefits of enforcing competition related to a particular sector or firm would depend
on its importance in the economy. If the PCC is to maximize the benefits from its review and
investigation, it must consider essential market activities as indicated by a share in value-
added, forward linkages, or place in supply and value chain. Applicability and significance
to social objectives and data resources should also be considered. A good example would
be the importance of a sector (product) in terms of consumption patterns.

To illustrate, one can use the data from the Family Income and Expenditure Survey 2012
(FIES 2012) to calculate the total household expenditure on consumer goods and match
these with their corresponding industries in the PSIC. Annex 8 presents the industries with
the highest expenditure share. Expenditure categories that have a significant share in
consumer expenditure would be those that are related to the following:

1. key food items (rice, meat and meat products, fish and other food items like sugar,
bakery products);
2. water, drinking water, and soft drinks;
3. clothes and footwear (ready-made embroidered garments, women’s and girls’
garments);
4. manufacture of leather shoes; and
5. cigarettes.

Similarly, using the data from the 2012 Input-Output table (Annex 9), the industries with the
highest share in a firm’s expenditure were used to trim the list further. Essentially, the
intermediate products that have the largest share of intermediate demand belonging to
the following broad categories: electronics export industry (C2612), petroleum and fuel
inputs (C1920), chemical inputs (C2011-C2013; C2021-C2023), fertilizers (inputs to
construction steel (C2411), and cement (C2394).

Table 18, including sectors in Category A, could thus be trimmed down. For this study,
share in manufacturing value-added is used. The cut-off level for the percentage of value-
added share would depend on how much could be handled by the PCC. We could trim
down the list by including only those whose value-added in manufacturing is greater than
.02 percent. Thus, trimming down the number of 4-digit sectors to 15.

Many of the important sectors that have a high share in value-added will not be in Category
A. Hence, there may be higher priority sectors to be found in Categories B and C if we
consider its importance in terms of value-added share. Thus, we suggest including sectors
44
from the lower Categories B1 and B2 in the priority, moving them up to the priority levels
that have high value-added share. For example, we can prioritize those sectors in Category
B whose value-added contribution is more than 1%. Hence, providing additional sectors of
cement and meat processing. We can go further to Category C to include those with a high
share in Value-added (greater than 1%), adding two more PSIC sectors—sugar, and motor
vehicles parts and accessories.

Table 20. PSIC sectors in other categories (B1 and C1) with Value-added share
exceeding 1%

Share in
CR4
PSIC Code Description HHI APCM Mfg. value
in %
added
B 1 Category
C2394 Manufacture of cement 821 43.1 0.269 3.31
Production, processing and
C1012 preserving of meat and meat 723 42.9 0.183 1.18
products
C1 Category
Manufacture of parts and
C2930 458 29.7 0.120 4.16
accessories for motor vehicles
C1072 Manufacture of sugar 853 46.5 0.105 1.26

Next Steps is the analysis of entry barriers and other significant factors affecting the sector’s
state of competition.

The next steps would involve further analysis of barriers to entry. This could start with the
existing government policies. First and foremost is the trade regime environment within
which the sector operates. For example, if the firm is mainly producing for exports, this
would be a reasonable explanation for high CR4/HHI, hence these sectors should be
deleted from the priority list. This is indicated by significant export ratio for the sector; or
trade in the commodity is open enough to import competition, effectively constraining its
monopoly power. This could be indicated by significant import penetration ratio. Also,
some products may have many apparent close substitutes. As such, we could also delete
sector C1106 (Manufacture of sports and energy drink).

Another government policy to consider is the imposition of ‘sin’ tax. The ‘sin’ tax does not
only indicate the social valuation of the government on the product, but could also act as a
tax on monopoly. This would further trim down the list by two items—C1101 (distilled spirits)
and C1201 (cigarettes).

There would be other factors that could affect the list of priorities. For this exercise, we look
at two more: (1) the homogeneity, and (2) the place of the sector in the industrial policy of

45
the government. If the product is homogenous, the sector would be a strong candidate for
inclusion in the priority list. In contrast, heterogeneity would weaken the case for inclusion.
For the second factor, PCC could leave the matter to DTI, and remove it from the
prioritization list. Hence, for the new sectors suggested in Table 20, meat processing, is
heterogenous and could be removed from Category B1 ‘candidate’ for the priority list. In
the case of Category C1 ‘candidate’, the manufacture of parts and accessories for motor
vehicles could also be removed. The sector covers products under the Industrial program
of the DTI. Besides, the sector is heterogenous products, and has very low HHI. In sum, we
could only add cement and sugar in the priority list, on top of the Category A products
whose share in manufacturing value-added is greater than .02 %. From category A,
candidate for deletion as they are under DTI industrial priorities are sectors C2811 (engines
and turbines) and C3101 (furniture and fixtures).

The trimmed-down list is presented in Table 21 below. In sum, the first step was to prioritize
those sectors with high concentration and high PCM. Consequently, the list is trimmed
down, considering its importance in the economy. For this study, we use its value-added
contribution (.02 % of manufacturing value-added cut-off for Category A and 1% for
categories B and C).

The list is further trimmed down concerning how they relate to other government policies.

Table 21. Suggested Priority List

PSIC Number of APCM % GVA share in

Description HHI
Code establishments (in %) Manufacturing
Manufacture of refined
C1920 23 4686 15.1 1.373
petroleum products
Manufacture of fertilizers
C2012 31 2912 22.9 0.605
and nitrogen compounds
Manufacture of milk-
C1056 based infants' and dietetic 1 10000 16.6 0.416
foods
Manufacture of dairy
C1059 16 4495 16.5 0.370
products, n.e.c.
Manufacture of
C2821 agricultural and forestry 29 5390 24.8 0.130
machinery
Manufacture of pens and
C3291 4 6636 19.2 0.037
pencils of all kinds

46
 PSIC Number of APCM % GVA share in
Description HHI
Code establishments (in %) Manufacturing
Manufacture of refractory
C2391 17 8680 24.1 0.024
ceramic products
Manufacture of power-
C2818 2 9694 24.9 0.023
driven hand tools
C2394 Manufacture of cement 821 26.9 3.31
C1072 Manufacture of sugar 853 10.5 1.26

There would be ‘priority’ sectors that may not have been captured by this matrix and
prioritization steps. These would be the low hanging fruits that could provide a good
demonstration of how the PCC works. These would be cases of reported anti-competitive
conduct, which are relatively easy to confirm (cost of administration is low). The redress
need not be in the form of litigation, but utilizing the other provisions (administrative
measures) of the PCA.

The list could also be further shortened, as earlier suggested, if the sector is too
heterogeneous and many product substitutes are available, or by using a higher cut-off rate
in value-added.

In many cases, opening up markets by easing entry that government policy or regulation
itself impeded would be the ideal solution. Beyond this would be to examine industry
characteristics, as suggested in the discussion in the earlier sections. It includes defining
markets by looking at product substitutability, heterogeneity, geography, and looking at
other factors affecting SCP, such as capacity utilization, capital intensity, and technology
considerations, more rigorously.

Finally, the results and findings of this study could provide benchmarks for future evaluation
and assessment of the impact of the PCA.

47
VII. REFERENCES

Abad, A., K. Gonzales, M. Rosellon and J. Yap (2012). “Unfair Trade Practices in the
Philippines.” PIDS Discussion Paper Series No. 2012-39. Philippine Institute for
Development Studies, Makati.

Aldaba R.M. (2000). “Tacit Price Collusion in the Philippine Cement Industry?” PIDS
Discussion Paper No. 2000-13. Philippine Institute for Development Studies, Makati
City.

Aldaba R.M. (2005). The Impact of Market Reforms on Competition, Structure and
Performance of the Philippine Economy. PIDS Discussion Paper Series No. 2005-24.

Aldaba R.M.(2008).“Assessing Competition in Philippine Markets.” PIDS Discussion Paper

No.
2008-23. Philippine Institute for Development Studies, Makati City.

Aldaba R.M. (2010). “Why Cement Prices Remain High Despite Zero Tariffs”. PIDS
Discussion
Paper No. 2010-02. Philippine Institute for Development Studies, Makati City.

Aldaba R.M. (2011). “PLDT-Sun Acquisition: good or bad?” PIDS Discussion Paper No.
2011-
08. Philippine Institute for Development Studies, Makati City.

Bonquin, C., & Quintos, P. (2017, August 2). Cartel, price-fixing fight gets boost as anti-trust
law takes effect. Retrieved from https://news.abs-cbn.com/business/08/02/17/cartel-price-
fixing-fight-gets-boost-as-anti-trust-law-takes-effect

Bartók, Csilla & Miroudot, Sébastien. (2008). The Interaction amongst Trade, Investment
and Competition Policies. OECD, Trade Directorate, OECD Trade Policy Working Papers.

Cahiles-Magkilat, B. (2017). “PCC includes DTI DAO in cement ‘cartel’ probe.” Manila
Bulletin Online. 02 August 2017.

Fels, A. (2001). “Competition Policy: Governance Issues – What are the alternative
structures?” The Australian Experience. Australian Competition and Consumer
Commission.

Medalla, E (1998). Trade and Industrial Policy Beyond 2000: An Assessment of the
Philippine Economy. PIDS Discussion Paper Series No. 2008-05.

Abad, A.R.A et. al. (2002). Toward a National Competition Policy for the Philippines. PIDS
Book 2002-05.

Medalla, E. (2003). Philippine Competition Policy in Perspective. Perspective Series Paper

No. 4. PIDS. 2003

Abad, A.R.A et. al. (2014). Using the social rate of discount in evaluating public investments
in the Philippines. Policy Notes No. 14-02. PIDS.
48
Lamberte, M. B. et al. (1992). Barriers to Entry Study. Final Report, Volume I. United States
Agency for International Development (USAID).

World Bank and OECD. Annex 1. Barriers to Entry. A Framework for the Design and
Implementation of Competition Law and Policy. 1998

Manila Bulletin Newspaper Online (2009). “DTI probes complaint on alleged unfair trade
practices in canned tuna.” 13 November 2009.

TeleGeography.com, “Digitel calls foul over PLDT’s ‘unfair’ ad-based service Red Mobile.”
11 June 2009.

Uy, J. (2017). “DA to suspend 55 garlic importers.” Philippine Daily Inquirer Online. 08
August 2017.

49
ANNEX 1. Number of observations, HHI and CR4 at 5-digit PSIC level

50
51
52
53
54
55
56
57
 Annex 2. Determinants of Concentration Ratios

(1) (2) (3) (4)

VARIABLES CR4 HHI CR4 HHI

APCM -0.000716*** 0.000326 -0.000302*** 0.000162

(0.000272) (0.000261) (8.87e-05) (0.000134)
No. of establishments (growth
rate) 0.00152 0.0133 -0.0115** 0.000737
(0.00833) (0.0126) (0.00476) (0.00817)
Value added in the industry
(growth rate) 0.000408*** 0.000784*** -2.29e-05 0.000327***
(0.000104) (0.000197) (4.20e-05) (9.76e-05)
Exports of the industry
(ratio of total output) 0.0946*** 0.212*** 0.0432*** 0.134***
(0.0252) (0.0446) (0.00868) (0.0327)
Imports of similar products
(ratio of total output+imports-
exports) 0.129*** 0.279*** 0.0122 0.0837
(0.0375) (0.0672) (0.0496) (0.196)
R&D Expenditure (ratio of total
output) 3.902** 11.04*** 4.598** 8.059**
(1.743) (2.554) (2.040) (3.342)
Labor productivity 6.11e-07*** 6.53e-07** 1.17e-07 3.66e-07
(1.73e-07) (2.56e-07) (1.13e-07) (3.21e-07)
2010 year dummy 0.0217 0.0137 0.0307*** 0.0240
(0.0240) (0.0274) (0.00942) (0.0156)
2012 year dummy 0.0287 0.0257 0.0458*** 0.0430**
(0.0259) (0.0292) (0.0125) (0.0182)
2014 year dummy -0.00594 0.00993 -0.00732 0.00132
(0.0242) (0.0274) (0.0106) (0.0159)
Constant 0.750*** 0.305*** 0.748*** 0.308***
(0.0176) (0.0191) (0.00814) (0.0113)

Observations 693 693 693 693

R-squared 0.029 0.089 0.122 0.118
Number of id 184 184
Robust standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
The variables that were incorporated in the above regression results include: Year dummies with
2008 as the base year, labor productivity and indicators for RDE expenditure. The calculation of
the import concentration ratio is also modified with the denominator incorporating total imports
and total exports.

The results still indicate a negative correlation between APCM and measures of concentration,
particularly 4-firm concentration ratio.

58
Annex 3. PSIC sectors with highest priority, Category A (HHI > 2500, and APCM> 15%)

59
Annex 4. PSIC sectors with priority Category B

60
Rows in italics pertain to B2 (HHI>2500 and 10%<APCM<15%), otherwise B1(1500<HHI<2500 and 15%<APCM).

Annex 5. PSIC sectors with Category C

61
Rows in italics pertain to C2 (HHI>2500 and APCM<10%), otherwise C1(1500<HHI<2500 and
10%<APCM<15%).

62
Annex 6. PSIC sectors with Category O (HHI > 2500, and APCM> 15%)

63
Annex 7. Unconcentrated sectors (HHI<1500)

64
65
 Annex 8. Industries with the highest share in consumer expenditure

PSIC 4- % of Total
PSIC 4-digit Description Expenditure (FIES
digit
2012)

C1011 Slaughtering and meat packing 5.86

C1020 Processing and preserving of fish and fish products and other 5.39
C1030 Processing
seafoods and preserving of fruits and vegetables 3.76
C1041 Manufacture of virgin coconut oil 0.72
C1052 Manufacture of powdered milk (excep for infants) and condensed or 1.77
evaporated milk (filled, combined or reconstituted)
C1061 Rice/corn milling 9.02
C1062 Manufacture of grain and vegetable mill products, except rice and 0.67
C1071 Manufacture
corn of bakery products 1.82
C1072 Manufacture of sugar 0.64
C1073 Manufacture of cocoa, chocolate and sugar confectionery 0.45
C1074 Manufacture of macaroni, noodles, couscous and similar farinaceous 0.45
C1077 Coffee roasting and processing
products 0.95
C1079 Manufacture of other food product, n.e.c. 1.07
C1104 Manufacture of soft drinks 0.87
C1105 Manufacture of drinking water 0.42
C1201 Manufacture of cigarettes 0.86
C1412 Women's and girls' and babies' garment manufacturing 1.62
C1413 Ready-made embroidered garments manufacturing 1.69
C1521 Manufacture of leather shoes 0.75

66
Annex 9. Industries with the largest share of intermediate demand

PSIC Code Description IO % of Total

C1311- Preparation and spinning of textile fibers; weaving 1.02
Intermediate
C1621
1312 Manufacture of veneer
oftextilesWeaving sheets; manufacture of plywood,
of textiles 0.57
Demand
C1701 Manufacture of pulp, paper and paperboard 0.59
C1920 Manufacture of refined petroleum products 5.50
C1990 Manufacture of other fuel products 0.80
C2011 Manufacture of basic chemicals except fertilizers and 0.93
C2012 Manufacture
nitrogen of fertilizers and nitrogen compounds 0.56
C2013 Manufacture of plastics in primary forms and of synthetic 0.48
C2021 Manufacture of pesticides and other agro- chemical 0.47
C2022 Manufacture
products of paints, varnishes and similar coatings, 0.51
C2023 Manufacture
printing of soap and detergents, cleaning and polishing 0.57
C2029 Manufacture of other chemical products, n.e.c. 0.61
C2100 Manufacture of pharmaceuticals, medicinal chemicals and 0.65
C2394 Manufacture of cement 0.75
C2411 Operation of blast furnaces and steel making furnaces 2.09
C2421 Gold and other precious metal refining 0.44
C2511 Manufacture of structural metal products 0.44
C2612 Manufacture of semi-conductor devices and other 14.55
C2720 Manufacture
electronic of batteries and accumulators 0.58
C2817 Manufacture of office machinery and equipment 0.54

67
Contact Us

The Philippine Competition Commission is open

Mondays through Fridays, from 8:00 a.m. to 5:00
p.m. Submissions of notifications and complaints are
accepted during these hours.

 25/F Vertis North Corporate Center 1, North Avenue,

Quezon City 1105 Philippines

 www.phcc.gov.ph

 +632.8771.9722

 [email protected]