competitive industrial performance report 2018

COMPETITIVE INDUSTRIAL PERFORMANCEREPORT 2018

Biennial CIP report, edition 2018

ACKNOWLEDGEMENTS

The Competitive Industrial performance (CIP) Report 2018 was prepared by Nelson Correa andThomas Nice under the supervision of Shyam Upadhyaya, Chief Statistician of UNIDO. ValentinTodorov, Senior Management Information Officer, provided data support for the updated 2018version of the CIP database. Several other colleagues of Statistics Division including Romana Bauer,Anja Boukhari, Kateryna Gumeniuk, Martin Haitzmann, Juergen Muth and Romana Ransmayrcontributed to the report. Our deepest gratitude is also due to Niki Rodousakis, who providedediting assistance and improved the report’s language and style. Finally, Marielena Ayala, NicolaCantore, Petra Kynclova and Rita Liang provided insightful comments that greatly enhanced thequality of this report.

Copyright c© 2019 United Nations Industrial Development Organization

The designations employed and the presentation of material in this report do not imply the ex-pression of any opinion whatsoever on the part of the Secretariat of the United Nations IndustrialDevelopment Organization (UNIDO) concerning the legal status of any country, territory, city orarea or of its authorities, or concerning the delimitation of its frontiers or boundaries.Terms such as “developed”, “industrialized” and “developing” are intended for statistical conve-nience and do not necessarily express a judgment about the state reached by a particular country orarea in the development process.Material in this publication may be freely quoted or reprinted, but acknowledgement is requested.

First printing, March 2019

Contents

I CIP report 2018: Chapter 1

1 The CIP Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1.1 Introduction 91.1.1 Competitiveness and industrial development . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.1.2 International embeddedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121.1.3 Technological absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

1.2 Measuring competitiveness 151.2.1 Composition of the CIP Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151.2.2 Capacity to produce and export . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161.2.3 Technological deepening and upgrading . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161.2.4 World impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

1.3 Sustainable Development Goal 9 181.3.1 Measuring SDG progress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181.3.2 SDG 9.2.1 Manufacturing value added as a proportion of GDP and per capita 181.3.3 SDG 9.2.2 Manufacturing employment as a proportion of total employment . 181.3.4 SDG 9.3.1 Proportion of small-scale industries in total industry value added . . 191.3.5 SDG 9.3.2 Proportion of small-scale industries with a loan or line of credit . . . . 201.3.6 SDG 9.4.1 CO2 emissions per unit of value added . . . . . . . . . . . . . . . . . . . . . . . 201.3.7 SDG 9.B.1 Proportion of MHT industry value added in total value added . . . . . 201.3.8 Countries’ progress in achieving SDG 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

1.4 How to use the CIP Index 211.4.1 The CIP index as analytical tool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

II CIP report 2018: Chapter 2

2 Highlights of the CIP report 2018 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2.1 Basic facts about the CIP 2018 edition 272.1.1 Country coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272.1.2 Data sources and compilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272.1.3 The CIP ranking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

2.2 By development stage 342.2.1 Main findings by development country group . . . . . . . . . . . . . . . . . . . . . . . . . 34

2.3 By geographical region 372.3.1 Main findings by geographical region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372.3.2 East Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372.3.3 Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412.3.4 Latin America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 452.3.5 Middle East and North Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482.3.6 North America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 512.3.7 South and South East Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542.3.8 Sub-Saharan Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 572.3.9 Other Asia and Pacific . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

2.4 By indicator 612.4.1 Main findings by indicator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 612.4.2 Dimension 1: Capacity to produce and export manufactured goods . . . . . . . 612.4.3 Dimension 2: Technological deepening and upgrading . . . . . . . . . . . . . . . . . 632.4.4 Dimension 3: World impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

III CIP report 2018: Chapter 3

3 CO2-adjusted CIP Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

3.1 Introduction to the CO2-adjusted CIP Index 693.1.1 General considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

3.2 Emission efficiency and carbon footprint 723.2.1 The global distribution of production and emissions . . . . . . . . . . . . . . . . . . . . . 723.2.2 Emission intensity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 723.2.3 The manufacturing carbon footprint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 733.2.4 Linking emission efficiency and the carbon footprint . . . . . . . . . . . . . . . . . . . . 763.2.5 Changes in emission efficiency and carbon footprint over time . . . . . . . . . . . . 77

3.3 Calculating the CO2-adjusted CIP Index 823.3.1 A non-linear adjustment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

3.4 CIP adjustment 843.4.1 Adjustment in ranks and scores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 843.4.2 By regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

IV Concluding remarks

Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

I

1 The CIP Index . . . . . . . . . . . . . . . . . . . . . . . . 91.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.2 Measuring competitiveness . . . . . . . . . . . . . . 151.3 Sustainable Development Goal 9 . . . . . . . . . . 181.4 How to use the CIP Index . . . . . . . . . . . . . . . . 21

CIP report 2018: Chapter 1

1. The CIP Index

1.1 Introduction1.1.1 Competitiveness and industrial development

The 2018 CIP Report assesses and benchmarksindustrial competitiveness across economies,providing valuable information on the strengthsand of weaknesses in national manufacturing sec-tors. This information is crucial to policymakers,as competitive industries drive the process ofstructural change, which development dependson. By promoting competitiveness, it is possibleto maximize economic efficiency in the alloca-tion of scarce resources while generating greaterprosperity for the population.

Yet recent political and social developmentsseem to contradict the foundations on which in-dustrial competitiveness is built. For example,global trade liberalization has historically con-tributed to an increase in the mobility of goodsand services among countries, with internationalcompetition being a main source of greater eco-nomic efficiency. In recent years, however, tradefrictions have emerged in North America, Eu-rope and Asia, with major economic powers try-ing to boost their domestic industries and gettingan edge on their direct competitors with protec-tionist policies. The established global trade or-der is being questioned, which has put the debate

on competiveness and industrial development onthe agenda again.

Despite its widespread use, the term com-petitiveness still causes some confusion. Theterm competitiveness is used in very differentcontexts, ranging from political campaigns, in-dustrial policy plans, academic discussions toeconomic debates. This has resulted in a blur-ring of the term competitiveness over time.

The term competitiveness seems to bestraightforward, yet it is difficult to explicitlydefine. One important reason lies in the fact thatcompetitiveness has different meanings at thefirm and at the country level. At the firm level,competitiveness refers to the ability of firms tocompete, i.e. their capacity to sell their prod-ucts in domestic or global markets. The impli-cation is that one firm’s gain comes at its com-petitor’s loss. However, competition betweencountries in international markets is not a zero-sum game: all countries can compete and at thesame time—as alleged by David Ricardo around200 years ago (Ricardo, 1817)—benefit frominternational trade.

The concept of competitiveness at the coun-

10 Chapter 1. The CIP Index

try level seems to be associated with the follow-ing three elements: international trade, economicperformance and overall prosperity. One of themost widely recognized definitions of compet-itiveness is that developed by the World Eco-nomic Forum (WEF). It defines competitiveness“as the set of institutions, policies, and factorsthat determine the level of productivity of aneconomy, which in turn sets the level of prosper-ity that the economy can achieve” (WEF, 2017,page 11).

This definition emphasizes the clear link be-tween economic performance (measured in thiscase as productivity) and overall prosperity, butdoes not mention international trade. Etymologi-cally, competitiveness derives from “compete”,and there is no competition between countrieswithout international trade. In this context, PaulKrugman claims that the use of the term compet-itiveness without including international trademakes very little sense because prosperity—thatis, the rise in living standards—would be almostentirely determined by the rate of productivitygrowth. According to Krugman: “for an econ-omy with very little international trade, ‘com-petitiveness’ would turn out to be a funny wayof saying ‘productivity’ and would have nothingto do with international competition” (Krugman,1994, page 32).

When shifting the scope of analysis to indus-trial competitiveness—which is the focus of thisreport—we redirect our attention to a country’scapacity to increase its presence in internationaland domestic markets whilst developing indus-trial sectors and activities with a higher valueadded and technological content, the purpose be-ing the improvement of the population’s overallprosperity (UNIDO, 2013).1

When we shift our scope of analysis to in-dustrial competitiveness, we must also changeour focus from economic performance and in-ternational trade flows to industrial performanceand manufacturing trade. Manufactured goodsrepresent around 75 per cent of total merchan-dise trade (UNCTAD, 2018), with the other 25

per cent composed of primary products or com-modities. The main distinction between thesetwo groups of goods is that the determinants forcompetition tend to differ. For example, whilecommodities tend to face strong price competi-tion, its role tends to be much weaker for manu-factured goods, where technology plays a moredominant role.

Malik and Temple assert that manufacturedgoods are less likely to be affected by price fluc-tuations than commodities, and as a result oftheir higher value added, often yield more bene-fits for those that produce them (Malik and Tem-ple, 2009). Based on this notion, an increasein industrial competitiveness implies that thecountry is exporting manufactured goods—asopposed to commodities—and consequently hasa wider margin of benefits, which in turn has ahigher impact on the country’s overall economicperformance and prosperity.

An increase in industrial competitivenesscan contribute to a country’s overall prosperityin many different ways. For example, it canencourage more investment from national andinternational firms. It increases a sector’s re-silience to external shocks, including surges incommodity prices or economy-wide recessionsWEF, 2017). Competitiveness is decisive if acountry’s industrial sector is to flourish, and itdetermines the pace and quality of the country’sstructural change as its economy develops aswell as the extent to which these changes willcontribute to society’s wellbeing. The industrialsector’s contribution to prosperity depends onits capacity to produce manufactured goods, toexchange those goods in global markets and tospecialize in complex production processes.

Yet the benefits of greater competitivenessin manufacturing are not limited to the devel-opment of the country’s industrial sector or itseconomic growth. Greater industrial competi-tiveness translates into economic, social and en-vironmental benefits, in addition to technologicalprogress. For example, changes in the structureof a country’s economy towards a stronger man-

1The term prosperity here is defined as in the Sustainable Development Goals publication; it is the extent to which“all human beings can enjoy prosperous and fulfilling lives and that economic, social and technological progress occursin harmony with nature” (United Nations, 2015, page 5). Thus, it entails all three dimensions of sustainable development:the economic, social and environmental. In other words, overall prosperity is much more than the rise in living standards,measured by the rate of productivity growth or income per capita.

1.1 Introduction 11

ufacturing sector have historically been accom-panied by social progress. This includes greatergender equality, higher levels of education, re-duced poverty and better health (UNIDO, 2014).The environmental impact of industrializationsignificantly affects a society’s living standardsand should therefore be considered in all strate-gies and polices aimed at increasing a country’soverall prosperity.

While the aforementioned benefits are notcomponents of economic growth or productiv-

ity measures per se, they contribute to an im-proved and more sustainable future. There isa strong link between industrial competitive-ness and the Sustainable Development Goals(SDGs). Greater industrial competitivenessraises an economy’s likelihood of succeeding inachieving the SDG targets, particularly SDG 9to “Build resilient infrastructure, promote inclu-sive and sustainable industrialization and fosterinnovation”.2

Box: 1.1 Competitiveness, Inclusive and Sustainable Industrialization and the SDGs

UNIDO’s mandate is to promote and acceler-ate Inclusive and Sustainable Industrial Devel-opment (ISID). The underlying objective is toimprove the living standards of the entire popula-tion in every country through industrial progress,while protecting the environment. ISID impliesthat no one is left behind and that all parts of so-ciety are to benefit from industrial progress, thusproviding countries the means to tackle criticalsocial and humanitarian needs.Industrial sector development drives both com-petitiveness and ISID, its impact reaching farbeyond manufacturing to stimulate economic,social and environmental progress. The rele-vance of ISID and industrial competitiveness areexplicitly recognized in SDG 9, which aims to“Build resilient infrastructure, promote inclusiveand sustainable industrialization and foster inno-

vation”.While the link between ISID and SDG 9 is ex-pressly outlined in its title, industrial compet-itiveness and SDG 9 merge at a deeper level.Countries must keep track of the progress theymake in each SDG, which requires the setting oftargets and identification of indicators for eachSDG. Six targets and their corresponding indica-tors have been proposed for SDG 9 to measurethe progress made in different components ofSDG 9. Three of these indicators are included inthe CIP Index and focus on a country’s 1) pro-duction capabilities, 2) technological deepeningand 3) environmental damage from industrialproduction. The link between SDG 9 and theCIP Index is discussed in more detail in Section1.3: “Sustainable Development Goals”.

2Section 1.3 demonstrates that many SDG9 targets overlap with the indicators used to create the CIP Index.


1.1.2 International embeddedness

International trade is a key element of industrialcompetitiveness. As the capacity of countriesto increase their presence in international mar-kets rises, the potential impact on their industrialdevelopment and prosperity also grows.

Economic history has shown how countries— at different paces — have opened up theireconomies to exchange their goods based on thenotion that selling goods they can produce (ef-ficiently) and buying those they cannot produce(efficiently) generates economic benefits for theireconomy. This notion is not new – it is based onthe principle of comparative advantages devel-oped by David Ricardo in 1817 to explain whycountries benefit from international trade.

Although countries have been exchanginggoods for centuries, trade flows have intensifiedin recent decades. Figure 1.1 presents the globaltrend since 1990. It shows the surge of manufac-turing exports, particularly after 2001, and com-pares it with the development of manufacturingvalue added. The gap between these two trendsdemonstrates how profound the integration pro-cess has been during this period; it has been char-acterized by disruptive technological shocks thathave changed the manufacturing business modelfrom huge industrial plants able to work inde-pendently and built for mass production to thefragmentation of the production process into sev-eral units that work together and are integratedin a global production chain.

Global manufacturing value added (MVA)was 2.5 greater in 2015 than in 1990, yet thevalue of manufacturing exports increased by afactor of 4.5 over the same period.3 The growthrate of manufacturing exports was particularly

high between 2000 until the onset of the fi-nancial crisis in 2008. Manufacturing exportswere affected considerably more than overallMVA throughout the financial crisis. The pe-riod from 1990 to 2015 was generally character-ized by a rapid increase in the globalization ofmarkets: frictions such as industrial tariffs andother barriers to trade declined, while technolog-ical progress facilitated international goods trade(McMillan and Rodrik, 2011).

Opening up to new markets and consumersallowed firms to realize economies of scale.Economies of scale increased the benefits of pro-ducing greater quantities due to falling averagecosts or increasing bargaining power. However,opening up the economy also meant increasedcompetition from abroad, forcing manufactur-ing firms to innovate and specialize in niches inwhich they had a comparative advantage (OECD,2008). Firms that could not adequately respondto these challenges faced serious difficulties stay-ing in business.

While it is widely agreed that internationaltrade has contributed positively to industrial de-velopment, some industries have had negativeexperiences, i.e. although international competi-tion has boosted some local industries, it harmedothers. It is, however, unquestionable that theintegration of global markets has had a profoundimpact on countries’ economic system.

The data presented in Figure 1.2 suggestthat countries with the highest MVA growthrates tend to also register the highest increasein their manufacturing exports. Countries witha weak manufacturing export performance arealso likely to show poor industrial performance.

3MVA growth is usually calculated in real prices. Current USD prices were used here to make a comparison ofmanufactured exports possible, as exports are valued in current USD prices. Please note that even if both series are atcurrent prices, their comparability is still ambiguous as manufactured exports prices (Free On Board (FOB), accordingto UN Trade Statistics) should include the transaction value of goods and the value of services rendered to deliver thegoods to the border of the exporting country, while manufacturing value added should include the value of materials andsupplies for production and the cost of services received. Therefore, the value of exports does not reflect the value addedin a particular country. Simply spreading individual production across value chains in multiple countries—while keepingtotal production constant—would therefore also lead to an increase in the value of exports. Additionally, heterogeneityin country sizes may produce confusing data. Countries with large domestic markets are likely to have a lower exportshare as they are less focused on foreign demand (UNIDO, 2018b).

1.1 Introduction 13

Figure 1.1: World manufacturing exports and MVA, 1990-2015 (Index, 1990=100)Source: UNIDO, 2018a

Figure 1.2: Annual growth rates of MVA and manufacturing exports, 1990-2016 (%)Note: 148 countries used in the sample, with two countries removed as they presented outlying values. Source: UNIDO,

2018a

https://stat.unido.org/




The increase of international trade flowsreaches beyond the growth of the manufactur-ing sector. Opening up to international tradeoften translates into stronger integration of man-ufacturing firms in global value chains, whichin turn leads to an inflow of foreign direct in-vestment and know-how. Knowledge transfersand spillover effects are the result of greater in-teraction with foreign firms. These effects arenot exclusive to the manufacturing sector andcan occur as technologies are diffused, workersmove between firms or production processes areimitated.

Consequently, some economists claim thattoday’s economic and industrial developmentdepends much more on a country’s ability to ex-port goods abroad while importing capital, tech-nology and know-how than it does on local in-dustrial and economic policies designed in au-tarky.4 There is a clear trend showing that thoseeconomies that have registered long episodes ofeconomic growth over the world average tend toalso report total export and manufactured exportgrowth that are higher than the world average.This trend is presented in Figure 1.3.

Figure 1.3 presents the average annualgrowth rates of both total exports and manufac-turing exports for the world’s fastest growingeconomies from 1990 to 2016. High economicgrowth rates are, on the whole, accompanied by

annual increases in manufacturing exports thatlie well above the global average. For example,in China, Myanmar, Mozambique, Viet Nam andUganda, the average annual growth rate in man-ufacturing exports was above 13 per cent – morethan twice the world average. While total ex-ports in these countries also grew much morerapidly than the world average, the differenceis not as significant as it is for manufacturingexports. The only exceptions are Qatar, wheremanufacturing exports grew roughly proportion-ally with GDP, and Laos, where GDP grew morethan manufacturing exports.

Economic growth is linked to export-orientation in the manufacturing sector, particu-larly in fast-growing countries. This link wascrucial to the industrialization of East Asiancountries and regions such as Hong Kong SAR,Taiwan ROC, Singapore and the Republic of Ko-rea. In these countries and regions, domesticfirms initially received state support to operatein global markets until they became competitive(McMillan and Rodrik, 2011). As projected bydevelopment economists a long time ago, large-scale diversification of the economy from agri-cultural production and commodities to manu-facturing entailed considerable productivity in-creases as the workforce shifted to the manufac-turing sector (Lewis, 1954; Kaldor, 1966).

1.1.3 Technological absorption

The link between manufacturing export perfor-mance and economic growth reaches beyond thestatistical correlation. A positive correlation usu-ally reflects countries that have succeeded in in-creasing their exports considerably by takingadvantage of the economies of scale in key man-ufacturing industries, particularly technology-intensive industries with high value added. Thus,successful industrialization processes are notonly characterized by the expansion of manu-factured exports but also by profound structural

change towards technology-intensive industries.At the same time, production diversifies awayfrom agriculture and commodities to manufac-turing activities and services (Haraguchi and Re-zonja, 2011).

This has been the case for China and othersuccessful East Asian countries. Figure 1.4presents the share of medium- and high-tech(MHT) manufacturing exports in total manu-facturing exports for selected countries. In theAsian countries and regions previously men-

4Industrial and competitiveness policies continue to be crucial for industrial development, as they redefine thecountry’s comparative advantage and its attractiveness in terms of foreign direct investment (FDI). Education andinfrastructure are key to attracting FDI, for transferring technology and knowledge and having an export-friendlyenvironment.

1.2 Measuring competitiveness 15

Figure 1.3: Average annual growth rates of GDP, total exports and manufacturing exports, 8 fastestgrowing economies in terms of GDP, 1990-2016 (%)

Source: UNIDO, 2018a

tioned, the share of MHT manufacturing exportswas considerably higher in 2016 than in 1990,ranging from 37 per cent in Hong Kong SAR tonearly 78 per cent in Singapore. A number offactors determine a country’s ability to absorbadvanced technologies and innovations, and thusincreases its competitiveness and export perfor-mance. Factors of production must be available,for example, through sufficient capital on effi-cient capital markets or mobile labour that cantransition to innovative firms. Government poli-cies also play an important role in improving aneconomy’s capacity to transition to MHT indus-tries. The quality of the education and trainingsystem determines whether workers are able tocontribute to technologically advanced produc-tion processes. Likewise, infrastructure mustbe in place to enable the introduction of MHTtechnology.

Openness to trade and the resulting diffusionof new technologies thus contribute to develop-

ment differently, depending on each country’scharacteristics. Moreover, there might be (short-term) trade-offs to this form of technology acqui-sition, particularly when foreign technology re-places local knowledge in the form of embodiedknowledge in capital goods. For example, whenworkers are replaced with machines and otherforms of capital, this may reduce the domesticcomparative advantage in labour-intensive indus-tries, thereby reducing the potential for tradewith other countries (Rodrik, 2018). Addition-ally, development experts have argued that whencountries are not prepared to absorb and retaina skilled workforce, to create new knowledgeand achieve technological catch up with moreadvanced countries, openness to trade may resultin a locked-in situation in which the developingcountry ends up being no more than a sourceof low-wage labour and production for moreadvanced economies Dosi, Pavitt, and Soete,1990).

1.2 Measuring competitiveness1.2.1 Composition of the CIP Index



Figure 1.4: Medium- and high-tech manufacturing exports, selected countries, 1990-2016Source: UNIDO, 2018a

The 2018 CIP Index assesses and benchmarksthe industrial competitiveness of 150 countries.It provides an indication whether a country’smanufacturing sector contributes to its develop-ment. The CIP Index measures how successfula country’s industries are at producing and sell-ing their goods on domestic and foreign markets,and consequently how much they contribute tostructural change and development.

The CIP Index covers three main dimensions,each consisting of two indicators. These dimen-sions are: i) the capacity to produce and exportmanufactured goods, ii) technological deepen-ing and upgrading, and iii) world impact. Thehigher the scores in any of the three dimensions,the higher the country’s industrial competitive-ness and its CIP Index. Figure 1.5 illustrates theconfiguration of the CIP Index.5

1.2.2 Capacity to produce and export

The CIP Index’s first dimension provides a com-parable measure of countries’ manufacturingproduction, either for local or for foreign con-sumption. Manufacturing value added per capita,MVApc, allows for country comparisons inde-pendent of their size. This indicator is closelylinked to a country’s stage of development andis expected to change throughout the process ofstructural change.

In a globalized economy, a country’s capac-ity for production must be accompanied by anability to export manufactured goods. Manufac-turing industries unable to specialize and inte-

grate in global value chains are unlikely to becompetitive and will face limitations in termsof demand for their products which is in directrelation to the size of their economies. Manufac-turing exports per capita, MXpc, is an additionalindicator reflecting the ability to realize compar-ative advantage in specific industries. Per capitamanufacturing exports allows country compar-isons irrespective of country size. However, thisis just a simple indicator that does not fully cap-ture all important elements in a country’s exportperformance.

1.2.3 Technological deepening and upgrading

5Definitions and conceptual descriptions in this chapter are elaborations based on the Industrial Development Report2012/13 and subsequent CIP reports (UNIDO, 2013).


1.2 Measuring competitiveness 17

Figure 1.5: Composition of the CIP IndexSource: UNIDO, 2013

The types of goods a country’s manufacturingsector produces are also relevant to measure com-petitiveness. Technological spillovers and lowvulnerability to price shocks implies that themore technology-intensive the goods being pro-duced are, the higher the expected benefits fromproducing (and further exporting) them.

The CIP Index accounts for technologicaldeepening and upgrading with two compositeindices. First, the degree of industrialization in-tensity, INDint, serves to estimate the complex-ity of production processes. This consists of theshare of medium- and high-tech (MHT) MVAin total MVA (MHVAsh) and the share of MVAin total MVA, the GDP (MVAsh). And second,the export quality, MQual, which measure thequality of the integration process in the country’smanufacturing sector.

A greater share of MHT production in totalmanufacturing production denotes an economywith a high level of productivity, innovative ac-

tivity and technological progress. In turn, thegreater the complexity of the manufacturing pro-cesses in certain industries, the higher the likeli-hood of knowledge spillovers across industriesand sectors. In this respect, successful structuralchange entails a transition from low-technology,labour-intensive activities to MHT activities, aswas the case of the four East Asian countries andregions mentioned in the previous section.

It is important to distinguish between thecomposition of a country’s total manufactur-ing production and of its manufacturing exports.MHT goods face competition from foreign mar-kets, in particular, and the characteristics of ex-ports are thus a further indication of industries’competitiveness. The export quality, MQual, istherefore estimated on the basis of the share ofMHT manufacturing exports in total manufactur-ing exports, (MHXsh), and the share of manufac-turing exports in total exports (MXsh).

1.2.4 World impact


Finally, the CIP Index considers a country’s posi-tion in the global market in terms of its manufac-tured goods. This indicator consists of the coun-try’ shares in world MVA (ImWMVA) and in theworld trade of manufactured goods (ImWMT ).The higher the values of these shares, the higherthe country’s ability to benefit from agglomera-tion, scope and scale effects. These benefits mayinclude sharing investments in infrastructure and

greater negotiating power in trade agreements.All indicators of the CIP Index are used to

compute the final CIP values. Appendix A pro-vides definitions, data sources and further infor-mation for each of the components, as well asa detailed description of the methodology usedto deal with missing values and to calculate theCIP Index.

1.3 Sustainable Development Goal 91.3.1 Measuring SDG progress

The 2030 Agenda for Sustainable Developmentemphasizes the importance of inclusive and sus-tainable industrial development. The SustainableDevelopment Goal 9 (SDG 9) aims to “Build re-silient infrastructure, promote inclusive and sus-tainable industrialization and foster innovation”.

All SDGs have specific targets and indica-tors that allow countries to measure the progressin achieving global development. Targets spec-ify the specific goals and indicators represent

the metrics by which the world aims to trackwhether these targets are being achieved.

In particular, SDG 9 has 8 targets and 12indicators that cover the economic, social andenvironmental dimensions of ISID. From these12 indicators, 6 indicators are directly linked toUNIDO’s mandate, among which three are di-rectly included in the CIP. These 6 indicatorswill be mention in the sections below.

1.3.2 SDG 9.2.1 Manufacturing value added as a proportion of GDP and per capita

The level of industrialization is central to SDG9. It is often measured as the share of MVA inGDP and MVA per capita because it reflects acountry’s manufacturing production capabilities.SDG Target 9.2 aims to “significantly increase”the level of industrialization in developing coun-tries. In LDCs, the target is even more ambitious,namely doubling the share of manufacturing in

GDP; Box 1.2 discusses the prospects of achiev-ing this target.

Manufacturing value added as a share ofGDP and MVA per capita are included in thefirst and second dimensions of the CIP Index asthe capacity to produce manufactured goods isalso a key aspect of competitiveness.

1.3.3 SDG 9.2.2 Manufacturing employment as a proportion of total employment

Although not covered in the CIP Index, the shareof manufacturing employment as a share of to-tal employment provides an indication of struc-tural change within the economy as countriesmove from labour- to capital-intensive modesof production. It also reflects the share of thepopulation that directly benefits from the coun-try’s industrial sector. Figure 1.6 shows that theshare of the workforce employed in manufac-turing in industrialized economies fell by morethan one-third from 1990 to 2016, from 21 per

cent to 13 per cent. Similarly, there was a slightreduction in the share of manufacturing employ-ment in emerging industrial economies and otherdeveloping countries. Although an increase inmanufacturing employment share was registeredin LDCs between 1990 and 2016, it remainedvery low: the share of the labour force engagedin the manufacturing sector increased from 6.0per cent to 7.2 per cent in the world’s poorestcountries.

Openness to trade and participation in global

1.3 Sustainable Development Goal 9 19

Figure 1.6: Manufacturing employment as a share of total employment by country group, 1990-2015Source: ILO, 2018

value chains leads to specialization of productionin specific goods that countries have a compara-tive advantage in. As discussed above, this stim-ulates competitiveness and increases net welfare.However, the effects are not homogeneous acrossthe labour force. Foreign competition forcesworkers in uncompetitive industries to reallocateto high-productivity sectors. Their ability to doso depends on whether they are able to acquirethe specific skills necessary to be competitive ina different sector.

As a result, there might be “losers” from

trade within an economy. They are likely to befirms and workers who are unable to compete inglobal markets and have difficulties transition-ing to industries with high levels of productivity.Economic policies are central in fostering thetransmission of new skills and technologies inorder for workers to be able to enter into moreproductive sectors or industries. This facilitatesthe successful reallocation of the labour force toother jobs. However, this may be a long-termtransition and require several years.

1.3.4 SDG 9.3.1 Proportion of small-scale industries in total industry value added

Small-scale industries can act as importantdrivers of competitiveness, in particular in de-veloping and emerging economies. Micro, smalland medium enterprises are important sources ofvalue added and jobs. They are therefore key topoverty alleviation and the provision of incomes.Moreover, small-scale industries can easily par-ticipate in local markets, do not require huge

investments or advanced technology and buildon local know-how to respond flexibly to chang-ing market conditions. However, small firms areoften characterized by lower levels of productiv-ity and wages due to internal inefficiencies andan unsupportive business environment. Effectivepolicies are therefore necessary to support small-scale industries in accessing finance, interacting

https://www.ilo.org/ilostat/


with suppliers and customers and benefiting froman effective regulatory environment. This will

allow them to be drivers of innovation in specificniches that directly benefit the population.

1.3.5 SDG 9.3.2 Proportion of small-scale industries with a loan or line of credit

Despite the comparatively low capital necessaryto set up small-scale industries, they often lackaccess to financing to realize their productivepotential. This may be attributable to lack of fi-nancial infrastructure in the given country. More-over, banks are less likely to provide credit forindividuals who lack collateral, financial literacyand even bank accounts – these factors add upand make small-scale industries more likely todefault on their debt. As a result, these firmsoften only have access to informal credit sys-tems, which can be considerably more expen-

sive than those offered in the formal bankingsector. Policies can support the provision ofcredit with schemes by backing communitiesin the credit market, ensuring that women canalso access credit and providing the necessaryinfrastructure and skills to promote credit marketactivity. This encourages entrepreneurship andinnovative activity to exploit market opportuni-ties. Consequently, access to loans and creditcan increase the competitiveness of small-scalefirms, thereby enabling their integration in localand global value chains.

1.3.6 SDG 9.4.1 CO2 emissions per unit of value added

The manufacturing sector is responsible foraround one-third of global CO2 emissions. In-dustrialization must therefore be adapted to becompatible with the realization of environmen-tal targets. This includes limiting greenhousegas emissions by upgrading infrastructure andadopting more efficient technologies. SDG 9highlights the need to reduce the emission inten-sity of production and the emissions per unit ofMVA.

The CO2-adjusted CIP Index, presented inChapter 3, takes into consideration the environ-mental damage from industrial production. Thisindex provides an alternative perspective of theCIP Index, whereby the adoption of the mostefficient technologies, the production of lessemissions-intensive goods and investments inpollution abatement positively affect a country’slevel of competiveness.

1.3.7 SDG 9.B.1 Proportion of MHT industry value added in total value added

SDG 9.B.1 assesses the technological deepen-ing of a country’s industrial sector based on theshare of medium and high-tech (MHT) industryvalue added in GDP. This reflects a country’scapability to innovate and absorb new technolo-gies, and hence its level of competitiveness inniches in global value chains. The ability of theeconomy to generate a large share of value addedin medium- and high-tech industries is pivotal tolong-term competitiveness and welfare.

SDG 9.B.1 is captured in the second dimen-sion of the CIP Index. As discussed above, agreater share of MHT industry in total produc-tion generates higher levels of productivity inthe use of machinery, as well as in productionprocesses and structures. Moreover, the potential

for knowledge spillovers across industries andsectors increases.

The diffusion of technological advances de-pends on both firm-level characteristics and theexternal economic environment. Firms must in-vest in specific technologies that alter the pro-ductivity of machinery, the way knowledge ismanaged and the organization of the firm. At thesame time, interaction with other firms, researchinstitutions, workers and further external bod-ies in domestic and foreign markets allows newtechnologies to spread throughout the economy.Whether the firm’s environment is conducive togrowth depends on a number of factors, such asopenness to trade, institutions and policies. Ap-plying more advanced technologies leads to radi-

1.4 How to use the CIP Index 21

Figure 1.7: Countries’ progress in achieving SDG 9, 2015Source: UNIDO, 2018d

cal changes in the manufacturing sector’s modesof production. Currently, a trend towards ad-vanced modes of production with systems basedon cloud-computing, artificial intelligence and

cyber physical systems in the framework of thefourth industrial revolution is emerging (UNIDO,2017b).

1.3.8 Countries’ progress in achieving SDG 9

Applying the same methodology as that usedin the CIP Index, UNIDO calculates a compos-ite index to measure a country’s performance inreaching the SDG 9 targets. The composite indexcovers four indicators (included in the SDG 9targets) that are related to industrial developmentand for which there is reasonable country cover-age: share of MVA in GDP and MVA per capita;manufacturing employment as a share of totalemployment; CO2 emissions per unit of valueadded; and the share of medium- and high-techindustry value added in total value added. The

data used by UNIDO as the custodian agencyof SDG 9 to monitor the indicators were usedto calculate the resulting SDG 9 performanceindex.

Figure 1.7 presents the results of the com-posite analysis for 2015, ranking 122 countriesbased on their performance in achieving inclu-sive and sustainable industrialization. The fivecountries that performed best on the compositeSDG 9 progress index were: Germany, Republicof Korea, Switzerland, Czechia and Japan.

1.4 How to use the CIP Index1.4.1 The CIP index as analytical tool

The 2018 CIP Index assesses and benchmarksthe industrial competitiveness of 150 countries.Each country’s outcome is a reflection of its per-formance across the three dimensions of the CIPIndex: (1) The capacity to produce and export

manufactured goods; (2) Technological deepen-ing and upgrading, and (3) World impact. TheCIP Index enables cross-country comparisons ofindustrial competitiveness. Box 1.3 presents thefunctions of the CIP Index as an analytical tool



in three steps.First, the index allows countries to identify

comparator countries. To provide a differenti-ated representation of competitiveness, the CIPreport presents outcomes by different categoriesdepending on the stage of industrialization, geo-graphical region and indicator. Comparator coun-tries can include neighbours, immediate competi-tors, potential competitors or role models. Theymay be comparable due to similarities in geogra-phy, availability of production factors, or typesof goods produced.

Second, countries that perform best acrossthe three dimensions of the CIP Index can serveas a benchmark for their comparators, given theirspecific circumstances. By highlighting areas inwhich other countries achieve higher CIP scores,

the Index can support and guide policies for fu-ture industrial development. For example, themanufacturing sectors of countries that performpoorly in the CIP Index are characterized byinefficiencies in the allocation of factors of pro-duction, such as labour and capital.

Third, the CIP Index serves as a guideline,with an intuitive starting point to more detailedanalyses for identifying and tackling these in-efficiencies, thereby contributing to widespreadproductivity growth and structural change by us-ing feasible targets that depend on the countries’circumstances. As structural change is a long-term process, changes in the CIP Index are likelyto be reflected several years after policies aimedat increasing competitiveness have been imple-mented.

1.4 How to use the CIP Index 23

Box 1.2: LDCs’ prospects of achieving SDG 9.2

Progress on SDG 9.2 on inclusive and sustain-able industrial development consists of two ma-jor targets to be reached by 2030 in LDCs: first,doubling the share of manufacturing value added(MVA) both as a percentage of GDP and in percapita terms and second, doubling manufactur-ing employment as a percentage of total em-ployment (UNIDO, 2017). Determining whetherLDCs’ industrial sectors are on the growth tra-jectories necessary to achieve these targets iscrucial for developing industrial policies that cancontribute to structural change and development.Forecasts based on previous trends indicate thatdoubling the 2015 share of MVA in GDP by2030 is highly unlikely in LDCs as a whole. Onaverage, reaching the required annual growthrate of 4.7 per cent seems improbable based onthe projected value of only 0.8 per cent – evenwhen considering the high level of uncertaintyassociated with forecasting.The results also point to considerable regionalheterogeneity. Even if this is insufficient forreaching the ambitious target, LDCs in South-East Asia, such as Cambodia, Myanmar and LaoPeople’s Democratic Republic, will likely regis-ter high growth rates of their share of MVA inGDP by 2030. Figure below presents the forecastfor LDCs in the Asia-Pacific region. The modelprojects a share of MVA in GDP of around 24per cent by 2030, still some way off the MVA inGDP target share of 32 per cent. A high level of

uncertainty surrounds this estimate, leading toa large bandwidth of the 95 per cent confidenceinterval. The target growth trajectory of 4.7 percent per year is far greater than the forecasted an-nual growth rate of 2.6 per cent, but falls withinthis confidence interval.By contrast, some sub-Saharan Africaneconomies have actually experienced deindustri-alization. This implies that their share of MVAin GDP has fallen since 1990. Thus, the indus-trial sectors of countries such as Burundi, Chador Malawi are moving in the opposite directionof that intended by SDG Target 9.2. Consideringthe evidence of the positive effects of industrial-ization in eradicating poverty, this developmentis of particular concern. Understanding the fun-damental causes behind these divergent trendsand developing effective industrial policies thatpromote sustainable industrialization is thereforeimperative.Moreover, while positive interlinkages betweenindustrialization and fighting poverty are clear,it is also important to understand how the futuregrowth of industrial sectors in the global Southrelates to other SDG targets, such as those con-cerning the environment or natural resources. Atthe same time, potential synergies may emergethrough improved institutions, better infrastruc-ture and greater welfare. This can provide addi-tional momentum for the growth of manufactur-ing in LDCs.

Figure: Forecast for reaching SDG 9.2 target in LDCs in the Asia-Pacific region by 2030Source: Nice, 2018 . Note: The dotted red line depicts the targeted MVA share in GDP; the dashed green line denotes

the required growth trajectory; the solid green line represents the growth model forecast and the shaded area is the 95 per

cent confidence interval.


Box 1.3: The CIP Index as an analytical tool1. Identify comparatorsIdentify neighbours Which comparators can provide usefulIdentify immediate competitors information?Identify potential competitors For which activities are the comparators useful?Identify role models What is a manageable number of comparators?2. Benchmark performanceCompare overall industrial How has the country performedperformance over time in global or regional rankings?Compare basic indicators of industrial Is the industrial structure suited to growth andperformance make the best use of local resources andTrace competitive strengths and capabilities?weaknesses How far from or close to the selectedwith respect to different sets of benchmarks is the country?comparators In which aspect of performance does the country

lead or lag?Does the performance of comparators suggestcause for concern about any aspect ofperformance?Is there a need for more detailed technicalbenchmarking of particular industries,clusters or technologies?

3. Benchmark driversCompare individual elements of drivers What are the relative strengths and weaknesses

in the capabilities of the selected country?Trace competitive strengths and weaknesses Do the general indicators capture the underlyingwith respect to different sets of comparators drivers at work? If not, how can they be refined?Assess which drivers are most important Which drivers constitute the most criticalfor improved performance constraints to industrial growth andAdd new data and analysis as necessary competitiveness?

Is there enough information to evaluatenon-quantifiable variables such as linkages,institutions and governance?If not, how can more information be obtained?

Source: UNIDO, 2002 and UNIDO, 2017a, page 15

II

2 Highlights of the CIP report 2018 . . . . . . 272.1 Basic facts about the CIP 2018 edition . . . . . . 272.2 By development stage . . . . . . . . . . . . . . . . . . 342.3 By geographical region . . . . . . . . . . . . . . . . . 372.4 By indicator . . . . . . . . . . . . . . . . . . . . . . . . . . . 61


2. Highlights of the CIP report 2018

2.1 Basic facts about the CIP 2018 edition2.1.1 Country coverage

The 2018 edition of the CIP Index assesses 150economies, six more than in the previous edi-tion. In total, US$ 12.3 trillion of value addedwas generated by the manufacturing sectors ofthese economies in 2016, corresponding to 15.6per cent of global GDP. Chapter 2 provides anoverview of the CIP rankings and presents moredetailed results by development group, geograph-ical region and CIP dimension.

Economies are thus able to identify compara-tors and benchmark their performance within

their specific group. This approach furthermoreexposes the extent of global inequalities in man-ufacturing performance. Long-term trends thatcontribute to competitiveness within a countryor region are discussed, and selected cases ofeconomies with dynamic manufacturing sectorsand noteworthy results are examined in detail.This chapter focuses on highlights of the CIPIndex and paints a picture of the global manufac-turing sector.

2.1.2 Data sources and compilation

All the data used in this report are available inthe UNIDO Statistics Data Portal. The websiteprovides online access to different sets of datacompiled by UNIDO Statistics, including thedata set for the CIP.

The database is updated annually and is pri-marily based on the UN Statistical Division’sNational Accounts Main Aggregates database,the World Bank’s World Development Indica-tor database, OECD STAN Structural Analysisdatabase and the UN Comtrade database. Other

supplementary sources include databases main-tained by regional agencies such as the Asian De-velopment Bank, the African Development Bank,the Economic Commission for Latin Americaand the Caribbean and databases of national sta-tistical offices. Occasionally, non-official datasources are used to cross-check the consistencyof the data. Population data are provided by theUN Population Division.

Appendix A.1 provides a brief summary ofmore extensive methodological notes included in

28 Chapter 2. Highlights of the CIP report 2018

the previous editions of the CIP reports. For fur-ther details, including the treatment of missingvalues and outliers and the normalization proce-

dure, please refer to UNIDO, 2013 and UNIDO,2017a.

2.1.3 The CIP ranking

Table 2.1 presents the complete results of the2018 CIP Index, with each economy ranked ac-cording to its composite score. Economies aregrouped into quintiles of the CIP Index—top,upper middle, lower middle and bottom—whichare highlighted in the table. Appendix C.1 pro-vides detailed results of the CIP Index for eacheconomy in each of the three dimensions.

The colour of the rank depicts the econ-omy’s stage of development, differentiating be-tween least developed countries, other develop-ing economies, emerging industrial economiesand industrialized economies. There is a correla-

tion between stage of development and competi-tiveness. The top quintile of the CIP Index con-sists almost entirely of industrialized economies,while the majority of LDCs are concentratedin the bottom quintile. There are some excep-tions, however. For example, the Philippinesand Viet Nam are both classified as other de-veloping economies; yet in the CIP Index, theyperform better than a number of industrializedeconomies and emerging industrial economiesand are ranked in the upper middle quintile ofthe CIP.

Industrialized economiesEmerging industrialized economiesOther developing economiesLeast developed countries

Quintile Rank 2016 Country Score 2016 Rank 2015

Top quintile

1 Germany 0.5234 12 Japan 0.3998 23 China 0.3764 44 United States of America 0.3726 35 Republic of Korea 0.3667 56 Switzerland 0.3207 67 Ireland 0.3172 78 Belgium 0.2807 89 Italy 0.2733 910 Netherlands 0.2707 1011 France 0.2679 1112 Singapore 0.2573 1213 China, Taiwan Province 0.2547 1314 Austria 0.2389 1415 Sweden 0.2254 1616 United Kingdom 0.2191 1517 Czechia 0.2148 1818 Canada 0.2074 1719 Spain 0.2044 1920 Mexico 0.1786 2021 Denmark 0.1715 2122 Malaysia 0.1662 22

2.1 Basic facts about the CIP 2018 edition 29

Table 2.1 continued from previous pageQuintile Rank 2016 Country Score 2016 Rank 2015

23 Poland 0.1651 2324 Slovakia 0.1604 2425 Thailand 0.1536 2526 Hungary 0.1493 2627 Finland 0.1457 2728 Israel 0.1318 2829 Turkey 0.1242 2930 Australia 0.1199 30

Upper middle quintile

31 Slovenia 0.1109 3432 Russian Federation 0.1047 3133 Norway 0.1042 3234 Portugal 0.1026 3735 Brazil 0.1019 3336 Saudi Arabia 0.1018 3537 Romania 0.1015 3638 Indonesia 0.0907 3839 India 0.0830 4040 Lithuania 0.0818 3941 United Arab Emirates 0.0735 4142 Luxembourg 0.0728 4243 Philippines 0.0725 4344 Viet Nam 0.0724 4645 South Africa 0.0694 4446 New Zealand 0.0659 4547 Belarus 0.0657 4948 Estonia 0.0647 4849 Argentina 0.0633 5050 Qatar 0.0631 4751 Chile 0.0606 5152 Greece 0.0591 52

Middle quintile

53 Croatia 0.0552 5454 Bulgaria 0.0524 5755 Bahrain 0.0515 5556 Trinidad and Tobago 0.0499 5657 Kuwait 0.0491 5958 Iran (Islamic Republic of) 0.0482 5359 Latvia 0.0474 5860 Peru 0.0426 6261 Tunisia 0.0418 6362 Serbia 0.0416 6563 Morocco 0.0415 6664 Ukraine 0.0407 6465 Malta 0.0398 7066 Oman 0.0392 6067 Costa Rica 0.0389 6868 Venezuela (Bolivarian Republic of) 0.0382 6169 Kazakhstan 0.0372 67



70 Colombia 0.0369 6971 Iceland 0.0345 7172 Bangladesh 0.0340 7273 Egypt 0.0331 7474 Guatemala 0.0309 7375 Panama 0.0308 7576 El Salvador 0.0303 7777 Sri Lanka 0.0298 7978 The f. Yugosl. Rep of Macedonia 0.0291 7879 Uruguay 0.0281 7680 Jordan 0.0267 8081 Bosnia and Herzegovina 0.0257 8682 Pakistan 0.0245 8283 Brunei Darussalam 0.0245 8384 Swaziland 0.0243 8585 Botswana 0.0238 8886 Mauritius 0.0222 8487 China, Hong Kong SAR 0.0220 8788 Cambodia 0.0212 9189 Ecuador 0.0196 9090 Lebanon 0.0188 89

Lower middle

91 Myanmar 0.0186 9792 Honduras 0.0159 9293 Cyprus 0.0159 9494 Algeria 0.0149 9595 Namibia 0.0147 9396 Paraguay 0.0136 9697 Georgia 0.0135 9898 Bolivia (Plurinational State of) 0.0134 9999 Armenia 0.0128 103100 Jamaica 0.0121 101101 Lao People’s Dem Rep 0.0115 104102 Mongolia 0.0109 107103 Kenya 0.0108 105104 Barbados 0.0107 102105 Côte d’Ivoire 0.0106 108106 Albania 0.0105 100107 Azerbaijan 0.0101 115108 Senegal 0.0101 109109 Gabon 0.0097 112110 Republic of Moldova 0.0097 111111 State of Palestine 0.0096 110112 Syrian Arab Republic 0.0093 81113 Congo 0.0093 113114 Suriname 0.0092 106115 Nigeria 0.0092 114116 Fiji 0.0091 116



117 Cameroon 0.0087 117118 Bahamas 0.0080 119119 Zambia 0.0080 118120 Papua New Guinea 0.0066 121

Bottom quintile

121 Kyrgyzstan 0.0066 124122 Montenegro 0.0066 123123 Ghana 0.0064 125124 Zimbabwe 0.0061 122125 Mozambique 0.0055 129126 Madagascar 0.0054 120127 United Republic of Tanzania 0.0053 126128 Belize 0.0049 127129 Uganda 0.0045 128130 Angola 0.0039 134131 Nepal 0.0037 133132 Central African Republic 0.0035 131133 Tajikistan 0.0035 132134 Malawi 0.0034 135135 Saint Lucia 0.0034 137136 Cabo Verde 0.0030 136137 Haiti 0.0030 138138 Bermuda 0.0027 140139 Yemen 0.0026 139140 Niger 0.0022 143141 Rwanda 0.0021 141142 Maldives 0.0018 144143 Ethiopia 0.0015 148144 Afghanistan 0.0013 142145 Gambia 0.0011 149146 Iraq 0.0009 130147 China, Macao SAR 0.0008 145148 Burundi 0.0000 146149 Eritrea 0.0000 147150 Tonga 0.0000 150

Table 2.1: 2018 CIP Index resultsSource: UNIDO, 2018a

Figure 2.1 presents the scores and ranks ofthe top performing countries in the 2018 CIPIndex. Germany achieved the highest compositescore and thus tops the CIP rank – as it has for allbut one year since 1990. It is followed by Japanin 2nd place and China in 3rd place. China’scompetitiveness has continued to surge, risingfrom rank 5 in 2014 and 22 in 2000. China’s

climb in the CIP ranks ejected the United Statesof America from the top 3 to rank 4 and theRepublic of Korea down to 5th place. Figure2.1 also provides an overview of those coun-tries setting the competitiveness benchmark inseven geographic regions and in four develop-ment groups.



Figure 2.1: Scores and ranks of the top performing countries in the 2018 CIP IndexSource: UNIDO, 2018a. Note: If a country is already listed in the top three, the runner-up is highlighted in the group of

regional leaders. Similarly, if a country is included in the group of regional leader, the runner-up will come in first

among the development group leaders. See Appendix B for country classifications.



Figure 2.2: Relationship between CIP scores and ranksSource: UNIDO, 2018a.

Overall, the CIP Index can range between 0and 1. Yet the highest score (achieved by Ger-many) is only 0.52. This reflects the fact that nocountry leads in all eight CIP indicators. At thesame time, the CIP scores are distributed very un-equally. Few countries achieve high scores andthus do not substantially outrival others in termsof industrial competitiveness; low CIP scoresbelow 0.1 are far more frequent.

Figure 2.2 depicts the non-linear relationshipbetween countries’ scores and their ranks in the2018 CIP Index. For example, the gap betweenthe CIP scores of Germany in 1st place and thecountry in 12th place, Singapore, is around 0.26and thereby just as large as the distance betweenthe CIP scores of Singapore and the countriesranked 150th, Burundi, Eritrea and Tonga. Aslower CIP scores are more congested, minor in-creases in competitiveness result in large shiftsin those countries’ ranking. There is thus morecompetition among countries that rank lower interms of competitiveness. For example, Alba-nia’s CIP rank improved by nine positions year-

over-year, with an absolute increase in its CIPscore of less than 0.001. Such countries arelikely to replicate technologies in a bid to “catch-up” with innovative countries at the frontier, asthey lack the capabilities to act as pioneers them-selves. This also implies that a positive changein a country’s CIP score is not the same as achange in its CIP rank – both values are impor-tant to obtain an accurate picture of a country’soverall performance.

One example of a country that has witnesseda considerable improvement in its rank and scorein the top quintile of the CIP Index is China– this is discussed in detail below. Yet whileChina attained the highest absolute increase interms of CIP score in recent years, other coun-tries achieved even larger relative increases intheir CIP ranking between 2000 and 2016. Coun-tries that were able to increase their CIP scoremost relative to their direct competitors includeMyanmar, which moved up 39 places in the CIPrank, Kazakhstan, which moved up 37 placesand Viet Nam, which moved up 35 places.



2.2 By development stage2.2.1 Main findings by development country group

There are major differences in competitivenessbetween countries at different stages of develop-ment. The CIP Index only evaluates outcomesin a given year; a more dynamic perspective,however, is necessary to assess long-term pro-cesses of structural change. Box 2.1 presentstrends in the dimensions of competitiveness ofcountries at different development stages overthe past quarter century.

Figure 2.3 presents the median CIP ranks ofthe four development groups used in this analy-sis for the period 1990 and 2016. There was littlechange during this 26-year period. One slightexception is the small increase in the medianranks of industrialized economies and emerg-ing industrial economies. This slight increasein the median of the CIP rankings reflects thefact that few countries were able to displace oth-ers at the top of the competitiveness rankings.Although Box 2.1 indicates that industrializa-tion occurred in emerging industrial economiesduring this period, it did not suffice for a sys-tematic increase in the competitiveness of theentire group. The positive upward shift of a fewcountries with exceptionally high growth rates,as China, India, Oman and Poland, could not sus-tain the median as it was counterbalanced withthe loss in competitiveness in more countries,among them: Suriname, Cyprus, Ukraine andVenezuela. Thus, the overall mean of the groupexperienced a decrease in competitiveness.1

Countries are assigned to developmentgroups ex-post (see Appendix B.1 for countryclassifications), and transitions between develop-ment stages are therefore not considered. Thiscould have an effect on the median CIP rank ifincreases in competitiveness—and thus of indus-trialization—led to transitions of countries be-tween groups. Let us take a country that jumpedfrom rank 100 to 50, in the CIP ranking, for ex-ample. If that country was grouped in the “OtherDeveloping Economies” in 1990 and continuesto be in the same category in 2016, the median of

the group “Other Developing Economies” willmost likely improve.

However, when as a result of the increasein its industrial competitiveness it also movesfrom its previous category to the group “Emerg-ing Industrial Economies”, then its shift to thenew category will imply that the improvement inthe “Other Developing Economies” group willbe removed, as the country belongs to anothergroup.

Figure 2.4 differentiates the rankings basedon the three dimensions of the CIP Index. Thishighlights areas in which a country group is par-ticularly (un)competitive. It indicates, for ex-ample, that industrialized economies are morecompetitive in the first dimension of the CIP In-dex, which measures countries’ capacity to pro-duce and export manufactured goods, than theyare in others. By contrast, least developed coun-tries’ performance in the first dimension, whichis most closely linked to profound processes ofstructural change, was very poor. The world’spoorest country group is also uncompetitive interms of its world impact.

Inequality in competitiveness is least pro-nounced (albeit still strong) in the dimension oftechnological deepening and upgrading. Thisis an indication that although the diffusion ofadvanced technologies can support an increasein competitiveness, it does not suffice to drivehigher competitiveness across all relevant dimen-sions. For technological diffusion to truly con-tribute to greater competitiveness in manufactur-ing, the country’s institutions, infrastructure, hu-man capital, business environment and other fac-tors must also be conducive to structural change(Rodrik, 2018).

Figure 2.4 also shows that emerging indus-trialized economies have shown a systemic im-provement in technological deepening, shorten-ing considerably the distance with industrializedeconomies.

1Interestingly enough, the average rank of the emerging industrial economies has sligtly improved during this period,from 57 in 1990 to 56 in 2016. Thus, while most of the countries in this group have suffered some loss in competitiveness,in average, the high performers have gained more competitiveness than what the low performers have lost.

2.2 By development stage 35

Figure 2.3: Median CIP ranks by country group, 1990 and 2016Source: UNIDO, 2018a

Figure 2.4: Median CIP rank in CIP dimensions by country group, 1990 and 2016Source: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening

and upgrading; Dimension 3: World impact.




Box 2.1: Trends in global manufacturing production and exports, by development stage

Globally, industrial output increased rapidlyand—with the exception of the financial cri-sis—steadily from US$ 5.8 trillion to US$ 12.6trillion (at 2010 prices) from 1990 until 2016.This MVA was, however, generated very un-equally between country groups. Figure belowillustrates that in 2016, 56 per cent of globalMVA was produced in industrialized economies,which are home to just 15 per cent of the world’spopulation. The share of global MVA producedin industrialized economies fell by over 22 per-centage points between 1990 and 2016. Overthe same period, the share of manufacturing netoutput from emerging industrialized economiesincreased substantially. In 2016, these countriesproduced 40 per cent of global MVA, more thandouble the share in 1990. This industrial growthwas largely driven by India, China, Brazil, In-donesia and Mexico (UNIDO, 2014). In thesecountries, the share of MVA in GDP has in-creased considerably in recent years to reacharound 24 per cent.The share of global MVA of LDCs was a mere0.8 per cent in 2016, although these countriesare home to over one-tenth of the world’s popu-lation. This is a slight increase from the 0.5 percent share of global MVA of LDCs in 1990, butmanufacturing growth in these countries laggedfar behind that in emerging industrial economies.The share of LDCs’ MVA in GDP was around

12 per cent in 2016. In comparison, the agricul-tural sector and extractive industries contributedaround one-third of LDCs’ GDP, while in emerg-ing industrial economies, structural change ledto a reversal of these shares (UNIDO, 2018b).These results are an indication of an inverted U-shaped relationship between the share of MVA inGDP and GDP per capita, as empirically demon-strated by Haraguchi, Cheng, and Smeets, 2017.Very broadly speaking, this denotes that the shareof developing countries’ MVA in GDP is low;transition economies have undergone a processof structural change, with factors of productionshifting to manufacturing; and industrializedeconomies have deindustrialized and shifted toservices-based economies. However, there isonly limited evidence supporting the theory ofunconditional convergence (McMillan and Ro-drik, 2011; Rodrik, 2012 and Rodrik, 2011), ac-cording to which countries with the smallest rel-ative manufacturing sector should exhibit thehighest growth rates in the sector – instead, leastdeveloped countries have had relatively limitedmanufacturing growth in recent decades. Thisgeneralization does not, however, apply to allleast developed countries. These countries rep-resent a heterogeneous category with varyingdegrees of structural change, as discussed in Box1.2.

Figure: Share of global MVA by country group, 1990-2016.Source: UNIDO, 2018a


2.3 By geographical region 37

2.3 By geographical region2.3.1 Main findings by geographical region

Assessing competitiveness by geographical re-gion facilitates comparisons with neighbour-ing countries which may have several commonsocio-economic and geographical characteristics.Countries within the same region are likely tohave similar features, such as institutional lega-cies, endowments with natural capital and cul-tures. Commonalities may extend to free trade ar-eas, such as under NAFTA, in the EU or ASEANand even to common currencies, such as the euroor the franc-CFA in West and Central Africa.

Beyond these shared features, the competi-tiveness of a neighbouring country is likely toaffect domestic competitiveness through spatialspillovers. Proximity to competitive countriesincreases the likelihood of high levels of domes-tic competitiveness, as it is easier to trade withcountries that are in close proximity. As dis-cussed previously in this report, greater tradeis closely linked to higher levels of productiv-ity: greater demand means that the country cansell more goods and exploit economies of scale;if the country faces greater competition fromabroad, its firms are forced to innovate and to be-come more competitive to survive on the market;if there is trade between neighbouring countries,it is also likely that there is a greater flow of cap-ital and workers, thus supporting the diffusion

of knowledge. This corresponds to the grav-ity model of international trade that is used toexplain flows of goods, investments and people.The gravity model deems that trade flows dependon both the size of the economies (and thus onone of the dimensions of their competitiveness)and the proximity between the two countries.

Figure 2.5 shows the geographic distribu-tion of countries in different CIP Index quintilesacross the world. The figure provides evidencethat countries with high levels of competitive-ness are likely to be grouped within the sameregions. Clusters of highly competitive countriesare concentrated in North America, Western Eu-rope and East Asia. By contrast, the majority ofcountries with the lowest levels of competitive-ness are located in sub-Saharan Africa.

The remainder of this sub-section assessesthe CIP Index results for seven regions: EastAsia, Europe, Latin America, the Middle Eastand North Africa, South and South East Asia,sub-Saharan Africa and Other Asia and Pacific.2

The following tables present the regional andglobal rankings of the countries within each ge-ographical region, as well as the changes theyunderwent over time. In addition to each table, afigure illustrates the regional score distributionwithin the region.

2.3.2 East Asia

Overall, East Asian economies rank high in theCIP Index. Three countries from the region arein the top 5 of the global ranking: Japan, Chinaand the Republic of Korea. Table 2.2 shows that5 of the top-6 countries in the ranking improvedtheir positions in the CIP ranking relative to 1990.China’s jump of 29 places in the CIP ranking isparticularly remarkable; Box 2.2 assesses thedrivers of this increase in competitiveness in de-tail.

Both Australia and New Zealand, despite be-ing high income countries, have comparatively

low levels of manufacturing competitiveness.This follows decades of contraction in the manu-facturing sectors of both countries, accompaniedby a drop in the CIP ranks as they deindustrial-ize.

Figure 2.6 shows the distribution of scoresin each of the three dimensions of the CIP Index,with the median depicted by the colour changein each bar. There is a high level of heterogene-ity in the scores of East Asian countries. Chinadominates the World Impact dimension, withthe highest share of both world MVA and world

2See Appendix B.1 for a detailed classification of countries based on geographical regions.


Figure 2.5: Map of CIP performance (quintiles), 2016Source: UNIDO, 2018a.

trade in manufactured goods. At the same time,the Republic of Korea and Singapore have veryhigh scores in the second dimension, Technolog-ical Deepening and Upgrading.

These countries can act as role modelsfor achieving high rates of structural changeover the past 60 years, and for having success-

fully transitioned to high-technology, high value-added forms of production. With the exceptionof Singapore, the performance of East Asianeconomies is comparatively weaker in the firstdimension of the CIP Index, which measures thecapacity to produce and export.



Absolute change in rank

Regional rank Country Global rank 2010-2016 1990-20161 Japan 2 0 02 China 3 -3 -293 Republic of Korea 5 1 -124 Singapore 12 5 -15 China, Taiwan Province 13 1 -16 Malaysia 22 0 -67 Australia 30 2 88 New Zealand 46 0 89 China, Hong Kong SAR 87 14 6610 China, Macao SAR 147 13 102

Table 2.2: Regional and global 2018 CIP rank, East AsiaSource: UNIDO, 2018a.

Figure 2.6: Score distribution, East AsiaSource: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening





Box 2.2: China’s increasing competitiveness

China has entered the top three in the CIP rank-ing. This follows a considerable increase in thecountry’s position from 32nd in 1990 to 22nd in2000. Between 1990 and 2016, China’s CIP In-dex score increased from 0.09 to 0.37 – by farthe greatest absolute increase in the CIP scoreof any country. Figure A presents the change inrank and score over this period.Figure B presents the normalized results ofthe six indicators used to calculate the CIPscores for China in 1990, 2000, 2010 and2016. The largest contributor to China’shigh CIP score—particularly between 2000 and2010—was the increase in the world impact ofboth MVA and manufacturing exports. In 2016,China accounted for 24 per cent of global MVA(up from 3 per cent in 1990) and 17 per cent ofworld trade in manufactured goods (up from 3per cent in 1990).Multiple factors have played a role in China’srise to becoming the global leader in both theproduction and export of manufactured goods.One of these factors includes China’s low labourcost, providing it with a comparative advantagein low-technology industries such as textiles; ithas furthermore made large-scale investmentsin transport infrastructure for shipping goodsabroad and introduced effective industrial poli-cies to boost industries with high potentials and

the adoption of efficient technologies.Compared to other large, resource-rich countries,where exports are typically less important dueto the size of the domestic market, exports playquite a remarkable role for China’s manufactur-ing sector. For example, from 1990 to 2016, man-ufacturing exports growth in the United Stateswas 4.3 per cent, whereas it was 13.8 per centin China. Exports of tradable goods have beenthe main contributor to China’s recent economicgrowth (Guo and N’Diaye, 2009).In China, MVA contributes around 32 per cent oftotal value added and may therefore have alreadypeaked. This means that in future, structuralchange will likely lead to a further deindustrial-ization of the economy (Haraguchi, Cheng, andSmeets, 2017). The quality of Chinese man-ufacturing exports also increased considerablybetween 1990 and 2016. Export quality in theCIP Index is measured as the mean of the shareof MHT manufacturing exports in total exportsand the share of manufacturing exports in totalexports. One potential problem of China’s ex-port orientation is the resulting vulnerability todemand shocks abroad. If there is a recessionin other countries, as was the case during thefinancial crisis of 2007/08, it can have consid-erable knock-on effects on domestic productionand consequently competitiveness.

Figure A: China’s CIP rank and score, 1990-2016Source:UNIDO, 2018ae



Moreover, as wages in the manufacturing sec-tor increase, China’s comparative advantage inlabour-intensive production has fallen. To re-main competitive and achieve higher levels ofindustrialization, the manufacturing sector hasto adapt (Labaye et al., 2013). This means con-tinuing to increase productivity levels. Furtherinvestments in the quality of exports is crucialto continue the transition to industries furtherup the value chain and to increase value added.Such a development is essential for China toavoid the “middle-income trap” – this situationarises when productivity levels are insufficientfor industries to compete in high-tech goods mar-kets with highly competitive economies, whilelabour-intensive production processes are relo-cated to other emerging industrial economieswith lower wage levels (Eichengreen, Park, andShin, 2013).

Productivity-enhancing policies, such as re-search and development, education and high-tech infrastructure are therefore key if countriesare to avoid the middle-income trap. The compar-atively high share of MHT technology productsin exports is an indication that China’s industrialsector is adapting and may have the potential toovercome the middle-income trap.Figure B highlights areas in which future in-creases in competitiveness could be achieved.Compared to leading industrialized economies,China performs poorly in the first CIP dimensionmeasured as the country’s capacity to produceand export manufactured goods per capita. Thus,despite China’s tremendous impact on worldmanufacturing, there is still considerable roomfor improvements in inclusive and sustainableindustrialization and to provide the workforcewith high incomes.

Figure B: China’s indicator scores, 1990-2016Source:UNIDO, 2018a

2.3.3 Europe

European countries have maintained their strongposition in the CIP rankings, occupying sixplaces in the global top 10. Most notably, Ire-land’s rank improved considerably relative tothe 2014 CIP Index. It moved up 7 positions,mainly due to a large increase in the net-outputof its manufacturing sector (see table 2.3). Thereasons for this are discussed in detail in Box2.3.

Yet while the group of European economieshosts some of the world’s most competitive

economies, competitiveness has not spread toall countries. This is supported by Figure 2.7:there is a large dispersion in European countryscores within each of the CIP Index’s three di-mensions. While Germany, Switzerland, Irelandand other countries in the top quintile of the CIPIndex performed well, particularly in Dimen-sions 1 and 2, this is not the case for a number ofEastern European countries. In countries such asAlbania, Moldova and Montenegro, integrationwith Western Europe since their transition from



centralized to market-based economies has onlypartially resulted in structural change. Despiteminor increases in their CIP rank in recent years,

these countries’ performance in production ca-pacity and manufacturing exports is fairly poor.

Figure 2.7: Score distribution, EuropeSource: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening





Regional rank Country Global rank 2010-2016 1990-20161 Germany 1 0 02 Switzerland 6 1 13 Ireland 7 -7 -124 Belgium 8 -2 -25 Italy 9 1 56 Netherlands 10 -1 17 France 11 2 58 Austria 14 -2 29 Sweden 15 2 410 United Kingdom 16 1 911 Czechia 17 -2 -1212 Spain 19 1 413 Denmark 21 0 514 Poland 23 -3 -3015 Slovakia 24 -5 -1516 Hungary 26 -1 -917 Finland 27 7 918 Slovenia 31 -3 119 Russian Federation 32 -1 820 Norway 33 3 1321 Portugal 34 -1 922 Romania 37 1 023 Lithuania 40 -3 -2224 Luxembourg 42 -2 1525 Belarus 47 5 -126 Estonia 48 -3 -1127 Greece 52 5 1628 Croatia 53 -2 1929 Bulgaria 54 -5 030 Latvia 59 -3 431 Serbia 62 -10 632 Ukraine 64 7 2333 Malta 65 2 1634 Iceland 71 0 1035 The f. Yugosl. Rep of Macedonia 78 -11 536 Bosnia and Herzegovina 81 -3 -2637 Cyprus 93 1 2238 Georgia 97 -3 -439 Albania 106 -1 -940 Republic of Moldova 110 -9 3041 Montenegro 122 -1 12

Table 2.3: Regional and global 2018 CIP rank, EuropeSource: UNIDO, 2018a.



Box 2.3: Ireland’s jump in MVA

Between 2014 and 2015, Ireland’s share of MVAin GDP increased from 17.7 per cent to 29.9per cent as MVA per capita more than doubled.This large jump is mostly the result of the con-try’s economic situation and the taxation policies,which have been considered very favorable tobusinesses and corporations (European Commis-sion Eurostat, 2016; OECD, 2016). Followingthese incentives, many multinational firms withvery high revenues -especially those engagedin R&D and innovation – were registered here.With this movement, many highly valuable in-tangible assets owned by these companies wereregistered in Ireland’s accounts. Examples ofthese assests are intellectual property rights inthe pharmaceutical industry, which are worthlarge sums of money. The relocation of thesefirms thus had a large effect on the comparativelysmall Irish manufacturing sector.The jump in Irish MVA highlights the impor-tance of considering countries’ relative positionsin global value chains. In globalized productionnetworks, multinational companies often movetheir headquarters to countries in which they canreduce their tax obligations. This can lead to verylarge shifts in countries’ MVA levels because,from the perspective of national accounting, thelocation of the headquarters is often where thegreatest value is added to the production process,even if physical manufacturing does not takeplace in that country. Equally, comparatively few

individuals may benefit from this value added inthe form of higher wages or consumption. There-fore, MVA per capita is just one dimension ofinclusive and sustainable industrialization: andunfortunately, large increases in value do notnecessarily translate into greater well-being ofthe population if it does not lead to additionaljobs or higher incomes.The figure below shows normalized MVA fora group of European countries, after Ireland’sjump in MVA (that is, from 2015 to 2018): Ire-land, the United Kingdom, the three countrieswith the greatest MVA in the EU – Germany,France and Italy, all other EU countries and theEU overall. The figure shows that even after thevery large increase in MVA mentioned earlier,MVA in Ireland continued to grow at a consider-ably faster rate than in other European countries– increasing by 18 per cent over the three ob-served years compared to 8 per cent in the restof the EU. At the same time, MVA in the UnitedKingdom contracted slightly. This may indicatethat firms chose to not expand manufacturingproduction in the United Kingdom in the faceof economic and political uncertainty linked tothe United Kingdom’s plan to leave the EU in2019. Indeed, according to the Bank of England,it is expected that the reduction in trade opennessbetween UK and EU will reduce the UK’s pro-ducetive capacity and economic growth (Bankof England, 2018).

Figure: Index numbers of MVA in selected European countries, 2015-2018Source:UNIDO, 2018a



2.3.4 Latin America

The manufacturing sectors in Mexico, Braziland Argentina continue to be the most compet-itive in Latin America, while the majority ofCaribbean countries are positioned in the lowestquintiles of the CIP Index. Table 2.4 presents theCIP ranks for Latin America.

There have been diverging trends evenwithin the group of most competitive LatinAmerican countries. While Mexico continuedits upward trajectory, Brazil and Argentina havenot been able to prevent their slide down in theglobal CIP ranks. Figure 2.8 depicts the threecountries’ indicator scores compared to 1990.Each of the three countries experienced minorand even negative growth rates across the CIPIndex’s dimensions towards the end of the 1990s– particularly during Argentina’s Great Depres-sion and a recession in Mexico. Yet on the whole,Mexico performed far better in each of the threedimensions of the CIP Index, particularly in itscapacity to produce and export manufacturedgoods and its world impact. Additionally, Mex-ico has the third highest share of MHT exportsworldwide. This is a result of the establishmentof NAFTA in 1994, facilitating manufacturing

exports to the large markets in the United Statesand Canada as well as a flow of FDI into Mexico.

Brazil’s CIP scores dropped in all threedimensions and Argentina’s CIP scores alsodropped in its capacity to produce and exportmanufactured goods and its world impact, whilebarely improving in its technological deepeningand upgrading. These two countries have wit-nessed a significant slump since the global finan-cial crisis of 2007/08. Furthermore, the presenteconomic conditions in these countries are notyet robust enough to recover and for their manu-facturing sectors to enter into a solid growth path(Padilla, 2018).

Figure 2.9 shows that Latin American coun-tries performed particularly poorly in the firstand third dimension of the CIP Index, largely dueto the lack of integration in global value chains.Similarly, Latin American countries performedconsiderably worse in the second dimension thanother emerging industrial economies, most no-tably from Asia. Mexico is the only country inthe region that has closed the gap to the techno-logical frontier, attributable largely to very highlevels of FDI.

Box 2.4: Panama’s rise in competitiveness

Panama achieved the greatest improvements incompetitiveness of any Latin American countryin recent years. This development has not beena continuous upward trend. Between 1990 and2009, Panama’s CIP rank dropped from 104th to120th. Panama witnessed a considerable increasein its rank between 2010 and 2016, moving upto rank 74. This is partly a reflection of the clus-tering of a large number of countries with lowlevels of competitiveness at low CIP values. Allthe same, important policy implications can bederived from Panama’s increase in competitive-ness.Panama’s strong manufacturing growth is mainlyattributed to spillovers from a boom in large-scale investment projects. This occurred in indus-tries related to construction, particularly thoseinvolved in the expansion of the Panama Canal,which was completed in 2016. A number ofother infrastructure projects were carried out

over the same period, such as motorway andunderground expansions. Beyond large govern-ment and private investment, the business envi-ronment also supported growth in manufactur-ing productivity. The overall quality of trans-port, energy and communications infrastructureis very high for the region, which further sup-ports investment and economic growth (Beatonand Hadzi-Vaskov, 2017).Improving the quality of education and the effi-ciency of institutions is central to guaranteeingthat future increases in competitiveness can beachieved, as Panama still lags far behind the re-gional and global frontiers. At the same time,effective policies will be necessary to guaranteethat high growth rates are sustainable once largeinfrastructure projects are completed and thatworkers can be effectively transferred from con-struction to other manufacturing sectors (Haus-mann, Espinoza, and Santos, 2016).


Figure 2.8: Performance in the three CIP dimensions, Brazil, Argentina and Mexico, 1990-2016(1990=100)Source: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening





Regional rank Country Global rank 2010-2016 1990-20161 Mexico 20 -3 -112 Brazil 35 4 93 Argentina 49 10 34 Chile 51 2 -75 Trinidad and Tobago 56 6 -126 Peru 60 -1 -157 Costa Rica 67 1 -78 Venezuela (Bolivarian Republic of) 68 16 189 Colombia 70 5 1310 Guatemala 74 -1 -311 Panama 75 -43 -2912 El Salvador 76 -1 -1113 Uruguay 79 3 1414 Ecuador 89 4 -1315 Honduras 92 -2 -2416 Paraguay 96 -9 -1317 Bolivia (Plurinational State of) 98 1 -1018 Jamaica 100 4 3119 Barbados 104 2 1520 Suriname 114 16 3821 Bahamas 118 10 -1522 Belize 128 2 823 Saint Lucia 135 -1 324 Haiti 137 0 23

Table 2.4: Regional and global 2018 CIP rank, Latin AmericaSource: UNIDO, 2018a.

Figure 2.9: Score distribution, Latin AmericaSource: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening





2.3.5 Middle East and North Africa

Israel, Turkey and Saudi Arabia continue to leadthe CIP rankings in the MENA region. The mostnoticeable change in manufacturing in the regionover the past decades has been the large increasein the industrial competitiveness of oil-exportingcountries, such as Saudi Arabia, the United ArabEmirates and Oman. At the same time, politicaland social unrest in a number of countries in theMENA region have contributed to reduced com-petitiveness – for example, in the Syrian ArabRepublic and Yemen (see Table 2.5).

Figure 2.10 presents the average scores ofthe MENA region in the three CIP dimensions incomparison with two regions at similar levels ofcompetitiveness, Latin America and South andSouth East Asia. The composition of the CIPscore in the three regions is similar: all achievestrong scores for technological deepening andupgrading and the weakest ones in world impact.Although all MENA countries have low levelsin the world impact, the MENA region is morecompetitive than the other two regions in this di-mension, but has a slighly weaker technologicaldeepening than South and South East Asia and

smaller production and export of manufacturedgoods than Latin America.

Figure 2.11 shows that—as is the case inother regions—there are major differences in theperformance of countries from the MENA re-gion in the dimensions of the CIP Index. Themanufacturing sector plays the most importantrole for the economy in oil-importing countriessuch as Turkey and Egypt. Yet these countriesproduce comparatively low-technology, labour-intensive goods with low margins. As a result,they are unable to contribute with a large shareof the value added in global value chains.

By contrast, Israel’s competitiveness islargely based on its adoption of advanced tech-nologies in manufacturing, and it is therefore theregional leader in Dimension 2 of the CIP Index.This is closely linked to Israel’s high levels ofinvestment in education, research and develop-ment and targeted support of information andcommunication technologies (The Economist In-telligence Unit, 2018a; The Economist Intelli-gence Unit, 2018b).


Figure 2.10: Mean CIP dimension scores in MENA, Latin America and South and South East Asia,2016Source: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening


Figure 2.11: Score distribution, MENASource: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening






Regional rank Country Global rank 2010-2016 1990-20161 Israel 28 3 52 Turkey 29 -3 -113 Saudi Arabia 36 -1 -64 United Arab Emirates 41 -13 -435 Qatar 50 -6 66 Bahrain 55 10 37 Kuwait 57 9 -158 Iran (Islamic Republic of) 58 0 -279 Tunisia 61 1 -210 Morocco 63 -6 -411 Oman 66 -2 -3412 Egypt 73 3 -513 Jordan 80 6 -1114 Lebanon 90 9 -815 Algeria 94 3 1316 State of Palestine 111 -4 017 Syrian Arab Republic 112 17 1518 Yemen 139 11 019 Iraq 146 -3 24

Table 2.5: Regional and global 2018 CIP rank, MENASource: UNIDO, 2018a.



2.3.6 North America

Table 2.6 shows that the leading industrial-ized economies in North America, i.e. the UnitedStates and Canada, have continued to experiencereductions in their levels of industrial compet-itiveness. Both dropped further in the globalrankings, overtaken by China and Czechia, re-spectively. Compared with other economies inthe top quintile of the CIP Index, the UnitedStates and Canada have not been able to keep upwith other countries’ increases in manufacturingcompetitiveness in the top quintile of the CIPIndex.

Figure 2.12 shows the performance ofBermuda, Canada and United States. While theresult of the comparison between these threecountries is quite obvious - as Bermuda has con-sistently been at the bottom quintile of the CIPIndex; its manufacturing sector is very small,with most of the country’s high GDP per capitabeing generated in the financial and insurancesectors by offshore firms - interesting insightscan be drawn from the analysis of the countries’trends over time.

In the first dimension, Canada performsslightly better than United States and much betterthan Bermuda. Yet, all three countries suffered adrop in their capacity to produce and export man-fuactured goods. The poor performance of theUnited States is directly due to a considerablelack of growth in MVA per capita and manufac-turing exports per capita. The United States’ lowexport activity is mainly attributable to the sizeof its domestic market and the importance of ser-

vices for the economy, which are less tradable in-ternationally. The low growth in manufacturingnet output is part of a general trend towards dein-dustrialization in the United States and Canada,which is also reflected in a falling share of MVAin GDP in both countries.

This also has an impact on Dimension 2 ofthe CIP Index, which measures technologicaldeepening and upgrading. Although manufactur-ing firms in the United States and Canada havehistorically been close to the technological fron-tier, they are facing increasing competition fromabroad, in particular from East Asia, and havebeen unable to keep up with increases in com-petitiveness. It is also remarkable to witness theimprovement shown in the Bermuda’s CIP scorein this dimension.

Finally, although the United States still hasa very large impact on world manufacturing dueto the size of its economy, its score in the thirddimension of the CIP Index has also fallen dra-matically due to the growth of China’s manu-facturing sector. Moreover, it is clear that theheterogeneity of these countries is much moreprofound here than in the other dimensions ofthe CIP.

Figure 2.13 shows that this region still per-forms best in technological upgrading, and thatthis is an area in which there is potential to over-come the current trends in the contraction of theirmanufacturing sectors. This could, in particular,occur through spillovers from high-tech firmsinto the manufacturing sector.


Regional rank Country Global rank 2010-2016 1990-20161 United States of America 4 1 12 Canada 18 1 103 Bermuda 138 3 -9

Table 2.6: Regional and global 2018 CIP rank, North AmericaSource: UNIDO, 2018a.



Figure 2.12: Scores in the three CIP dimensions, Bermuda, Canada and USASource: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening




Figure 2.13: Score distribution, North AmericaSource: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening




2.3.7 South and South East Asia

Table 2.7 shows that countries in South andSouth East Asia represent multiple groups. Thefirst, consisting of the top five countries in theregional rank, are found in the upper quintiles ofthe CIP Index. Led by Thailand, the region hasexperienced the strongest growth in manufactur-ing exports per capita of any region, as shownin Figure 2.14. From 2010 to 2016, the value ofmanufacturing exports in South and South EastAsia nearly tripled.

Figure 2.14 also shows that while globaltrade was strongly affected by the financial crisis

of 2007-2009, South and South East Asia re-covered quickly in comparison to other regions,achieving average annual growth rates in manu-facturing exports of 5.9 per cent between 2010and 2016. As a result, most countries in Southand South East Asia have been able to move upthe CIP ranks since 1990. The major challengeremains the integration of those economies thathave not experienced increased competitivenessin global value chains, to follow the paths ofother South and South East Asian countries suchas Thailand and Indonesia.


Regional rank Country Global rank 2010-2016 1990-20161 Thailand 25 1 -82 Indonesia 38 0 -133 India 39 -2 -214 Philippines 43 -10 -45 Viet Nam 44 -23 -506 Bangladesh 72 -8 -347 Sri Lanka 77 -1 -138 Pakistan 82 3 39 Brunei Darussalam 83 0 -1310 Cambodia 88 -13 -3811 Myanmar 91 -8 -4712 Lao People’s Dem Rep 101 -19 -2713 Nepal 131 1 214 Maldives 142 -1 215 Afghanistan 144 4 20

Table 2.7: Regional and global 2018 CIP rank, South and South East AsiaSource: UNIDO, 2018a.

The most competitive countries in South andSouth East Asia improved their performance inDimension 3 of the CIP Index considerably be-tween 2010 and 2016. That said, the region stillhas a relatively small impact on global produc-tion and trade in manufactured goods.

Despite the size of its economy and recentgrowth, India’s case is somewhat unusual in thatthe country’s structural change moved the econ-omy away from agriculture but it has not nec-essarily been accompanied by high productivity

growth in manufacturing. Instead, its servicessector has registered strong increases in pro-ductivity (Goel and Restrepo-Echavarria, 2015).The share of MVA in GDP has increased com-paratively little in recent decades and for its size,India has little impact on global production andmanufacturing exports. However, India’s poli-cies are now increasingly focusing on the promo-tion of the manufacturing sector (The Economist,2014).

Further improvements in competitiveness in



Figure 2.14: Manufacturing exports per capita by geographic region, 2000-2016Source: UNIDO, 2018a.

the region depend on whether manufacturingin labour-intensive industries continues to shiftfrom China to countries such as Viet Nam andBangladesh, particularly in the apparel indus-try (Lopez-Acevedo and Robertson, 2016). Al-though this is at comparatively low stages ofglobal value chains, it could be a first step to-wards increasing both MVA per capita and man-ufacturing exports per capita. Figure 2.15 shows

that the region’s performance is particularly poorin this regard (see dimension1). The majority ofcountries in South and South East Asia haveinvested heavily in transport infrastructure andeducation in recent years (World Economic Fo-rum, 2016). This increases the likelihood thatthe growth of manufacturing production and ex-ports will continue with greater capabilities toabsorb new firms and investments.



Figure 2.15: Score distribution, South and South East AsiaSource: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening




2.3.8 Sub-Saharan Africa

The manufacturing sectors of some countriesin sub-Saharan Africa (SSA) have witnessedstrong growth rates in recent years. The coun-tries with the greatest expansion of their man-ufacturing sectors and the highest increases incompetitiveness between 1990 and 2016 includeBotswana, Ethiopia, Kenya, Mozambique andRwanda. Their development is illustrated in Fig-ure 2.16. Maintaining consistently high MVAgrowth rates poses a considerable challenge,even for the most successful countries in sub-Saharan Africa.

A large gap in industrial competitiveness be-tween sub-Saharan Africa and other regions isstill evident: with the exception of the top per-

formers South Africa, Swaziland, Botswana andMauritius, countries in sub-Saharan Africa arefound in the bottom quintiles of the CIP Index(see Table 2.8). What is more, the majority ofLDCs (14 out of 22 in the CIP Index) are insub-Saharan Africa and all, with the exceptionof Senegal, are positioned in the lowest quin-tile of the CIP Index. Many of these countrieshave undergone a process of deindustrializationsince 1990, as described in Box 1.2. This high-lights the strong need for increases in compet-itiveness to drive structural change in Africa,thereby generating economic growth and provid-ing jobs (Fox, Thomas, and Haines, 2017).

Figure 2.16: Average annual growth rates of MVA per capita, selected countries in SSASource: UNIDO, 2018a.

Figure 2.17 shows that the performance ofsub-Saharan African countries in CIP Dimen-sions 1 and 3 is particularly poor. Yet somecountries have made some advances in techno-

logical deepening and upgrading. This resultis driven by comparatively sophisticated manu-facturing production systems in Swaziland andSouth Africa.




Regional rank Country Global rank 2010-2016 1990-20161 South Africa 45 5 22 Swaziland 84 2 143 Botswana 85 -5 34 Mauritius 86 0 205 Namibia 95 8 96 Kenya 103 -1 07 Côte d’Ivoire 105 12 68 Senegal 108 2 -49 Gabon 109 -2 -1210 Congo 113 1 -2111 Nigeria 115 27 -812 Cameroon 117 8 1213 Zambia 119 2 -614 Ghana 123 -4 415 Zimbabwe 124 8 3116 Mozambique 125 -14 -1717 Madagascar 126 1 -118 United Republic of Tanzania 127 6 -1019 Uganda 129 -2 -1520 Angola 130 -2 -1121 Central African Republic 132 -10 222 Malawi 134 1 -223 Cabo Verde 136 -2 124 Niger 140 -1 925 Rwanda 141 -3 -826 Ethiopia 143 -4 -527 Gambia 145 -1 028 Burundi 148 3 529 Eritrea 149 1 3

Table 2.8: Regional and global 2018 CIP rank, sub-Saharan AfricaSource: UNIDO, 2018a.



Figure 2.17: Score distribution, sub-Saharan AfricaSource: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening




2.3.9 Other Asia and Pacific

Table 2.9 shows that the region of other Asiaand Pacific is composed by a heterogeneousgroup of countries that still shares similar weak-nesses in industrial competitiveness. With theexception of Kazakhstan, all countries of this re-gion rank in the bottom or lower middle quintileof the ranking. Yet, Armenia and Mongolia haveshown some major improvements in their indus-trial competitiveness during the current decade,2010-2016.

A closer look into the different dimensions ofcompetitiveness shows that this group of coun-tries faces major difficulties at producing andexporting their manufactured goods and theirchallenge in even bigger in international marketas they have almost no impact. Moreover, de-spite having a limited technological deepening,these countries seem to perform much better inthis dimension than in the other two (see Figure2.18).


Regional rank Country Global rank 2010-2016 1990-20161 Kazakhstan 69 5 52 Armenia 99 -11 113 Mongolia 102 -11 -164 Azerbaijan 107 4 245 Fiji 116 2 216 Papua New Guinea 120 -2 37 Kyrgyzstan 121 -3 298 Tajikistan 133 4 209 Tonga 150 0 0

Table 2.9: Regional and global 2018 CIP rank, North AmericaSource: UNIDO, 2018a.

Figure 2.18: Score distribution, Other Asia and PacificSource: UNIDO, 2018a. Note: Dimension 1: Capacity to produce and export; Dimension 2: Technological deepening




2.4 By indicator 61

2.4 By indicator2.4.1 Main findings by indicator

The following subsection assesses aggregate re-sults in each of the three dimensions of the CIPIndex as well as the indicators they cover. It alsopresents longer-term trends to determine howcompetitiveness is changing at the global level,as well as the drivers of those changes and theirpolicy implications.

Table 2.10 presents a summary of the mean

ranks within each CIP indicator for a combina-tion of the development groups and geographicregions considered in the previous sections ofthe CIP report. The vertical lines separate thethree dimensions of the CIP Index. The follow-ing subsection discusses each dimension and thecorresponding indicators in detail, with referenceto Table 2.10.

2.4.2 Dimension 1: Capacity to produce and export manufactured goods

The lower the mean CIP rank within a specificsubgroup in Table 2.10, the greater the averagelevel of its countries’ competitiveness. The tablereveals a number of important findings from theCIP Index. First, the average gap in competitive-ness between industrialized economies and therest of the world is particularly large for both in-dicators measuring the capacity to produce andexport manufactured goods. The distance be-tween a few countries in the top quintile and therest is visible in Figure 2.19. The figure plotsthe relationship between CIP Indicators 1 and2, MVApc and MXpx, and demonstrates that fewcountries dominate both indicators. The frontierin Dimension 1 consists of Ireland, Switzerland,Singapore and Belgium.

Only LDCs were able to partially reduce theaverage gap in the rankings of at least one of

the two indicators of the first CIP dimension be-tween 2010 and 2016: manufacturing exports percapita. This was mainly due to the remarkableexport performance of Bangladesh and Myan-mar. Emerging industrial economies, by contrast,fell behind in the rankings of the first indicatordue to the drop in export performance whichwas particularly high across some countries fromLatin America and from the Middle East. Thisdynamic was not the same across regions, how-ever. For example, the positions of Europeanindustrialized economies, emerging industrialeconomies and other developing economies inthe first two indicators improved. At the sametime, emerging industrial economies from EastAsia (mainly driven by China) made substantialgains in the ranking.

Figure 2.19: Indicator 1 and 2 performance, 2016Source: UNIDO, 2018a.



Cou

ntry

Man

ufac

turi

ngM

anuf

actu

ring

Indu

stri

aliz

atio

nM

ediu

man

dSh

are

ofE

xpor

tM

ediu

man

dSh

are

ofIm

pact

onIm

pact

ongr

oup

valu

ead

ded

expo

rtpe

rin

tens

ityhi

gh-

MVA

inqu

ality

high

-tec

hm

anuf

actu

ring

wor

ldw

orld

perc

apita

capi

ta(M

Xpc

)(I

ND

int)

tech

valu

eG

DP

(MX

Qua

l)ex

port

expo

rts

inm

anuf

actu

ring

man

ufac

turi

ng(M

VApc

)ad

ded

shar

e(M

VA

sh)

shar

ein

tota

lto

tale

xpor

ts(I

mW

MV

A)

expo

rts

into

talM

VAex

port

s(M

X sh)

(Im

WM

T)

(MH

VA

sh)

(MH

X sh)

Indu

stri

aliz

edE

cono

mie

s27

2842

3461

4745

5646

40E

astA

sia

3841

5039

6557

5263

4342

Eur

ope

2219

3430

5034

3444

4134

Lat

inA

mer

ica

4032

2733

3368

7479

8678

ME

NA

2933

5845

9110

184

9958

57N

orth

Am

eric

a36

6071

3910

743

4351

5555

Em

ergi

ngIn

dust

rial

Eco

nom

ies

6066

5863

5767

6374

5356

Eas

tAsi

a41

575

272

1328

61

1E

urop

e61

5161

5767

4351

4970

59L

atin

Am

eric

a58

7564

7162

9280

9851

60M

EN

A62

7251

5358

8064

8842

50So

uth

and

Sout

hE

astA

sia

5977

3961

2353

4575

3446

Sub-

Saha

ran

Afr

ica

6263

7889

5865

9147

7071

Oth

erA

sia

and

Paci

fic62

8010

198

9010

259

124

5367

Oth

erD

evel

opin

gE

cono

mie

s10

299

102

101

9494

9986

102

101

Eur

ope

9990

106

101

101

7895

7312

010

9L

atin

Am

eric

a90

8893

8991

8494

7010

610

4M

EN

A10

710

510

497

101

9095

8986

91So

uth

and

Sout

hE

astA

sia

9996

6877

5670

9061

6766

Sub-

Saha

ran

Afr

ica

108

105

105

111

9911

211

010

399

106

Oth

erA

sia

and

Paci

fic11

210

812

612

510

710

798

105

126

117

Lea

stD

evel

oped

Cou

ntri

es13

613

111

011

990

103

103

9311

111

8L

atin

Am

eric

a13

314

412

213

189

8814

053

125

136

ME

NA

141

145

128

145

104

104

7311

610

813

8So

uth

and

Sout

hE

astA

sia

124

116

9211

958

9412

269

8893

Sub-

Saha

ran

Afr

ica

140

136

115

116

102

108

9510

512

012

5

Tabl

e2.

10:C

hang

esin

mea

nin

dica

torr

anks

betw

een

2010

and

2016

,by

deve

lopm

ents

tage

and

geog

raph

icre

gion

Sour

ce:U

NID

O,2

018a

.Not

e:R

edde

note

sa

dow

nwar

dsh

ift(

dete

rior

atio

n)in

the

mea

nra

nkgr

eate

rtha

non

epl

ace,

and

gree

nan

upw

ard

shif

t(im

prov

emen

t)gr

eate

rtha

non

epl

ace.

Fora

deta

iled

desc

ript

ion

onth

ehe

ader

sof

each

colu

mn,

plea

sese

eA

ppen

dix

A.1


2.4 By indicator 63

Figure 2.20: Indicator 3 and 4 performance, 2016Source: UNIDO, 2018a.

2.4.3 Dimension 2: Technological deepening and upgrading

East Asian economies, with their successful in-dustrial policies of export-orientation and techno-logical adoption—discussed in previous sectionsof the report—are at the frontier of Dimension2, which measures technological deepening andupgrading. However, East Asia was unable toincrease its average competitiveness in this di-mension between 2010 and 2016 and they evendropped some positions in the export qualityranking. Figure 2.20 shows that the countriesat the frontier of the two indicators of Dimen-sion 2 are all from the East Asia region, with theRepublic of Korea, Taiwan ROC and Singaporeleading the ranks in technological deepening andupgrading.

Table 2.10 indicates that despite the cleardominance of emerging industrial East Asiancountries, there is considerably less inequalitybetween groups at different development stagesin the second dimension of the CIP Index than

in the first one. This is particuarly valid whenlooking at the export quality indicator (and itssubindicator of share of manufacturing exportsin total export), as it was also characterized bygreater convergence between 2010 and 2016.The average rank of industrialized economiesin this indicator worsened. The opposite tookplace in LDCs, and therefore, the distance in ex-port quality between the most developed and lessdeveloped countries was considerably reduced.

The industrialization intensity indicatorshows mixed results. Indeed, while industrial-ized economies improved their industrializationintensity between 2010 and 2016, other develop-ing economies worsened due to the fall in per-formance of Latin American and MENA coun-tries. Despite the improvement of Haiti and otherLDCs in the South and South East Asian region,the group of LDCs was not able to move up inits ranks in this indicator.



2.4.4 Dimension 3: World impact

Finally, rankings in Dimension 3 of the CIP In-dex, World Impact, are highly correlated with acountry’s level of industrialization as well asthe size of its economy. Figure 2.21 showsthat the five most competitive countries accord-ing to the 2018 CIP rankings—Germany, Japan,China, the United States and the Republic of Ko-rea—dominate share of world MVA and shareof world manufacturing exports.

Those five countries are responsible for 58per cent of global MVA and 44 per cent of globaltrade in manufactured goods (China alone ac-counted for 24 per cent and 17 per cent, respec-tively). The top quintile dominates both indica-tors in general, with 81 per cent of world MVAand 87 per cent of world manufacturing trade.There is a continuing concentration of globalMVA and trade in manufactured goods amongEast Asian countries. While the Republic of Ko-rea kept its share constant and China increased

its share considerably between 2010 and 2016,the other three countries witnessed reductions inboth indicators. Few other countries were ableto increase their shares and thus “keep up” withincreases in competitiveness in China. Theseinclude India, Ireland (see Box 1.3), Indonesiaand Poland – countries with particularly strongmanufacturing growth between 2010 and 2016.

Countries in the bottom quintile of the CIPIndex are responsible for less than 1 per cent ofglobal MVA and trade in manufactured goods.Moreover, Table 2.10 shows that the compet-itiveness of LDCs in the indicators measuringworld impact shows an small improvement. This,however, is the result of the improvement of fewcountries (mainly from South and South EastAsia) rather than a systemic movement of theentire LDCs. As a result, manufacturing produc-tion and trade continue to be very much concen-trated in industrialized countries.

2.4 By indicator 65

Figure 2.21: Shares of world MVA and world manufacturing trade, 2016Source: UNIDO, 2018a.


III

3 CO2-adjusted CIP Index . . . . . . . . . . . . . . 693.1 Introduction to the CO2-adjusted CIP Index . . 693.2 Emission efficiency and carbon footprint . . . . 723.3 Calculating the CO2-adjusted CIP Index . . . . . 823.4 CIP adjustment . . . . . . . . . . . . . . . . . . . . . . . . 84


3. CO2-adjusted CIP Index

3.1 Introduction to the CO2-adjusted CIP Index3.1.1 General considerations

Industrialization is a key driver of structuralchange, increased incomes and greater welfare.Yet manufacturing production also contributesto the degradation of local and global ecosys-tems through resource use and pollution (UnitedNations Environment Programme, 2011). Theprevious chapter of the CIP report approximatedthe relative success of a country’s manufactur-ing sector on the basis of its manufacturing netoutput, value of exports and other economic mea-sures. These are important indicators of a coun-try’s success in increasing economic prosperity;but the economic value of production does notfully reflect the contribution of a sector to over-all prosperity. Moreover, MVA does not accountfor the external, uncompensated environmentaleffects of manufacturing. These effects can dam-age ecosystems, put their essential services atrisk and thus decrease human welfare.

Industrial production can have numerousnegative environmental effects, with impactsranging from the local to the global scale. Man-ufacturing may contribute to the overstepping of

planetary boundaries (Rockström et al., 2009).For example, when natural habitats vanish tobe replaced by infrastructure for production ornatural resource extraction, biodiversity is jeop-ardized. Similarly, an over-extraction of naturalresources such as water or timber for productionputs pressure on natural systems. The manufac-turing material footprint, which measures thesector’s total material requirements, is thus animportant indicator for assessing the natural re-sources used in production across value chains.Beyond the inflow of natural resources for pro-duction, the outflow of waste products from pro-duction can pollute soil and water sources withchemicals or increase the concentration of green-house gases in the atmosphere.

Degradation of “natural capital” has highsocial and economic costs (Nordhaus, 2017).These are difficult to measure and vary depend-ing on the sector of the economy, the locationand the time scale (Tol, 2018). For example,global warming can negatively impact indus-trial competitiveness1 and production capabili-

1Although some industries may benefit from climate change, the adjusted CIP Index assumes that the long-termmacroeconomic effect on industrial competitiveness is negative (Tol, 2018).

70 Chapter 3. CO2-adjusted CIP Index

ties through frequent climate catastrophes, whichlead to the destruction of infrastructure andother physical capital. Equally, the social, eco-nomic and political uncertainty resulting fromclimate change may divert investment away fromproductivity-enhancing innovations. It is impor-tant to note, however, that manufacturing is lessvulnerable to climate change than agriculture.

The present chapter focuses on the rela-tionship between industrial competitiveness andCO2 emissions, the main cause of anthropogenicglobal warming. The adjustment extends the CIPIndex as a measure to also incorporate the nega-tive environmental effects of manufacturing. Tothis end, it incorporates CO2 emissions data forthe manufacturing sector collected by the Inter-national Energy Agency (IEA). The adjustmentaccounts for the damage caused to ecosystems bylarge, emissions-intensive industries by reducingthe CIP scores accordingly.2

The adjustment of the CIP Index on the basisof CO2 emissions rearranges the CIP rankings tothe benefit of countries that effectively protectnatural capital by realizing measures for pollu-tion abatement in manufacturing. Conversely,countries in which manufacturing is particularlyharmful to the environment receive lower scoresin the adjusted CIP Index and move down therankings. There is a lot of potential—and anenvironmental need—to reduce emissions frommanufacturing industries in these countries. Theadjusted CIP Index serves as an additional instru-ment in cross-country comparisons of compet-itiveness by taking environmental externalitiesinto account.

This reflects the notion that there might be atrade-off between MVA growth and emissions re-ductions, for example, when countries introducestringent environmental policies or carbon taxes(Cifci and Oliver, 2018). These countries may,

as a result, have smaller manufacturing sectors,and may appear less competitive in the unad-justed CIP Index compared to countries withhigh levels of pollution. Yet by acknowledgingand compensating for the environmental costsof production, countries that invest in pollutionabatement mitigate pressure on ecosystems, thevalue of which is not otherwise accounted for.

The premise of the adjusted CIP Index isthat greater competitiveness can be achieved byimproving emission efficiency3, i.e. reducingemissions per unit of MVA. Efficiency gains canbe achieved with technological innovations, newforms of cooperation or improvements in insti-tutions and management. Greater efficiency, inturn, can reduce the quantity of carbon-basedresources as inputs in production – the primarypath to lower CO2 emissions from manufactur-ing (United Nations Environment Programme,2011).4 At the same time, promoting a circulareconomy that retains value within productionprocesses by reusing, repairing and refurbishingexisting manufactured goods can reduce depen-dency on the production of emissions-intensivematerials without negatively impacting welfare(Nasr et al., 2018).

In addition to production’s emission inten-sity, the adjustment of the CIP Index is alsobased on countries’ manufacturing carbon foot-prints. This reflects the notion that the respon-sibility for reducing emissions is not the sameglobally. Some countries contribute a far greatershare to overall global natural capital depletionand thus play a key role in the global effort to re-duce manufacturing emissions. Economic mea-sures are less suited to adequately evaluate thecontribution of manufacturing to welfare in coun-tries that cause greater environmental harm.

The section 3.2 of this chapter presents anoverview of trends in emission efficiency and

2Limiting the analysis to CO2 emissions only exposes one dimension of the relationship between competitiveness andthe negative environmental externalities of manufacturing. This is problematic because the results may differ for othertypes of emissions, such as Sulphur oxides, nitrogen oxides or particulate matter (Naqvi and Zwickl, 2017; OECD, 2002).Brinkley and Less, 2010, for example, finds that CO2 emissions reduction in the manufacturing sectors of Denmark andthe Netherlands was the result of substituting coal and oil for natural gas. Although natural gas emits less CO2, it alsoreleases large amounts of methane, another greenhouse gas linked to global warming. However, due to a lack of data onother environmental externalities from the manufacturing sector, our analysis is limited to CO2 emissions.

3Emission efficiency is defined as the inverse of emission intensity – countries with low CO2 emissions per unit ofMVA have low emission intensity and high emission efficiency.

4UNIDO’s 2011 Industrial Development Report focuses on the importance of energy efficiency for sustainableindustrial development in detail UNIDO, 2011.

3.1 Introduction to the CO2-adjusted CIP Index 71

total emissions from manufacturing across quin-tiles of the CIP Index. It thereby extends theanalysis of previous sections of the CIP reportby focusing on the two dimensions that deter-mine country-specific impacts of manufacturing,

i.e. CO2 emissions on the environment. The sec-tion 3.3 describes the method and reasoning usedto adjust the CIP Index on the basis of CO2 emis-sions. And the section 3.4 discusses the resultsof the adjustment and their implications.

Box 3.1: The CO2-adjusted CIP Index in the context of the SDGs

Reconciling industrial development and emis-sions reductions is key to achieving the 17 SDGs.As discussed in Chapter 1, SDG Target 9.2 aimsto “significantly raise industry’s share of employ-ment and national product” by 2030. In leastdeveloped countries, the goal is to double theshare of MVA in GDP and the share of manufac-turing employment in total employment. At thesame time, environmental targets are embeddedin the SDGs, for example, in SDGs 12 (Respon-sible Consumption and Production), 13 (ClimateAction) and 15 (Life on Land).The need to limit global warming by reducingCO2 emissions is reiterated in numerous inter-national treaties. In the 2015 Paris Agreement,for example, 175 countries have committed tolimiting global warming to below 2oC above pre-industrial levels. A recent IntergovernmentalPanel on Climate Change (IPCC) report goeseven further: it argues that a temperature riseof 1.5oC above pre-industrial levels in the com-ing decades already poses considerable climaterisks for humans and ecosystems. This is ahighly likely scenario unless emissions reduc-tions can be achieved. By 2017, human activityhad already caused average global temperaturesto increase by approximately 1oC relative to pre-industrial levels (Masson-Delmotte et al., 2018).There are complex feedback processes betweenhuman and natural systems. Global natural cap-ital depletion through emissions threatens thesocial foundation industrialization seeks to im-prove. For example, if climate change reduces

crop yields, leads to the destruction of physicalcapital and threatens livelihoods, the purpose ofstructural change and increased manufacturingcompetitiveness is undermined. In short, sus-tainable development implies that environmentalgoals cannot be separated from greater humanwelfare (through industrialization) (Robert, Par-ris, and Leiserowitz, 2005). This is of particularsignificance because climate risks are not dis-tributed equally: the most vulnerable populationgroups are also those in LDCs who have mostto gain from poverty reduction, from increasedgender equality and other benefits of industrial-ization (Tol, 2018).The CO2-adjusted CIP Index facilitates the inte-gration of environmental considerations in in-dustrial policy design. It serves to comparecountries’ relative success in achieving emission-efficient industrialization and synergies in theSDGs. Yet achieving these goals simultaneouslyis extremely challenging. Reducing CO2 emis-sions can be thought of as contributing to theglobal public good of a “clean” atmosphere witha limited amount of greenhouse gases (Samuel-son, 1954). This means that individual coun-tries do not reap the full benefits of their owninvestment in pollution abatement and there isan incentive for countries to not reduce theiremissions. It is therefore important to highlightthose countries that set the benchmark in achiev-ing industrialization with minimal environmentaldamage.


3.2 Emission efficiency and carbon footprint3.2.1 The global distribution of production and emissions

A disproportionate share of global CO2 emis-sions comes from industrial production: in 2015,manufacturing contributed around 16 per centof global total value added, but 36 per cent ofCO2 emissions (IEA, 2018). CO2 is releasedthrough two channels in industrial production.Most manufacturing emissions are the result ofenergy generation – either directly from com-bustion of fossil fuels, such as oil, coal and gasor indirectly by purchasing electricity. Chemi-cal processes in various forms of production, forexample, in the cement, chemical and metal in-dustries, also cause considerable CO2 emissions(Andrew, 2018).

The production of manufactured goods is

central to industrialization and structural change.Yet the type of goods being produced, as wellas their production processes, can vary consid-erably between countries and across time. As aresult, there are large variations in the extent towhich manufacturing negatively impacts ecosys-tems. The following section presents a descrip-tive analysis of the two dimensions used to ad-just the CIP Index: emission intensity, measur-ing both the efficiency of the technologies usedto generate value added and the energy require-ments of goods; and the carbon footprint, mea-suring each country’s per capita impact on globalenvironmental degradation.

3.2.2 Emission intensity

Emission intensity measures the quantity of CO2emitted per unit of MVA. Figure 3.1 depicts therelationship between countries’ emission inten-sity and MVA per capita. There is a negative cor-relation: on average, countries with the lowestlevels of industrialization and the least compet-itive industries are those with the highest emis-sion intensities. For example, countries in thebottom quintile of the CIP Index emit, on aver-age, around one metric tonne of CO2 per unitof MVA. The average emission intensity is lessthan one-quarter of that in countries that are inthe top quintile of the CIP Index. The emissionintensity of production decreases when movingfrom less competitive countries with low levelsof industrialization to more competitive, highlyindustrialized countries.

This relationship is determined by changesin sector specialization and technology adoptionthat accompany structural change: a country’semission intensity thus depends on the featuresof its production (Borghesi, Cainelli, and Maz-zanti, 2015; Balibey, 2015). Both of these dimen-sions determine the relationship between indus-trialization and emission efficiency. Manufactur-ing in low-income countries is largely concen-trated in the food and beverages, textiles and ap-parel industries. These industries are at the bot-

tom of global value chains, with little value beingadded to production (UNIDO, 2018b). These in-dustries do not have particularly large levels ofoverall emissions (as discussed in the follow-ing sub-section), yet the particularly low valueadded means that production is comparativelyemissions-intensive.

In countries in the middle and upper middlequintiles of the CIP, MVA per capita is consider-ably higher. Manufacturing production shifts tomedium-technology goods, for example, metals,cement and chemicals. These industries havehigh levels of overall emissions, but also con-tribute with comparatively high levels of valueadded to the economy. As a result, the emis-sion efficiency remains fairly constant in the CIPIndex’s bottom four quintiles.

As countries reach high levels of competi-tiveness, their products, processes and organiza-tional structures become more efficient throughecological innovations. This leads to reductionsin emission intensity. Moreover, an additional“natural” structural change in manufacturing pro-duction evolves as countries industrialize. Indus-trialized countries with greater (human) capitalavailability specialize in more high-technology,capital-intensive goods, such as electronics. Theemission level of high-technology industries is

3.2 Emission efficiency and carbon footprint 73

Figure 3.1: MVA per capita and CO2 emissions from manufacturing per unit of MVA, 2000-2015Source: UNIDO, 2018a and UNIDO, 2018d. Note: Larger circles correspond to quintile means.

low while their MVA is high. Hence, the emis-sion intensity of manufacturing is low. The threecountries with the greatest emission efficiency

in 2015 were European countries in the top CIPquintile: Ireland, Switzerland and Denmark.

3.2.3 The manufacturing carbon footprint

Total annual CO2 emissions from manufacturingcorrespond to the manufacturing sector’s carbonfootprint. Both the scale and efficiency of pro-duction determine the size of the manufacturingcarbon footprint: the net output of manufactur-ing goods requires a specific amount of energy,most often through combustion of CO2 emittingresources, while emission intensity determinesthe quantity of CO2 emitted per unit of MVA(Nordhaus, 2017).

In 2015, 5.8 billion tonnes of CO2 were emit-ted in global manufacturing production to gen-erate US$ 11.8 trillion of MVA. The manufac-turing sector thus contributed over one-third to

total global CO2 emissions, when accounting forindirect emissions from electricity and heating(IEA, 2018). Figure 3.2 shows that total CO2emissions from manufacturing increased by 59per cent between 2000 and 2015.

Carbon footprints and emission intensity aredistributed highly unequally between countrygroups (Piketty and Chancel, 2015). Figure 3.2shows the level of variation in total manufactur-ing emissions between the CIP Index’s quintiles.Countries in the top quintile were responsiblefor 70 per cent of global CO2 emissions frommanufacturing in 2015. In absolute terms, theCO2 emissions of the top CIP quintile increased



Figure 3.2: Total CO2 emissions from manufacturing by CIP Index quintile, 2000-20155Source: UNIDO, 2018a and UNIDO, 2018d.

from 2.5 billion tonnes to 4.0 billion tonnes be-tween 2000 and 2015. Yet, as already mentioned,these countries also had the greatest emission ef-ficiency, producing 83 per cent of global MVAin 2015. At the same time, Figure 3.2 showsthat the average growth rate of CO2 emissions incountries found in the CIP Index’s upper quin-tile has not been constant over time. Whileemissions rose considerably between 2000 and2010—driven mainly by manufacturing growthin China—the annual emissions in the top CIPquintile essentially stagnated between 2010 and2015.

The manufacturing sector of the countriesin the CIP Index’s upper middle quintile wereresponsible for over one-third of global man-ufacturing CO2 emissions and one-quarter ofglobal MVA. From 2000 to 2015, their shareof the world’s manufacturing CO2 emissions in-creased by five percentage points. By contrast,the countries in the bottom CIP quintile, con-sisting mostly of countries with the least com-petitive industries (the majority of which are insub-Saharan Africa) were responsible for just

0.3 per cent of global CO2 emissions and 0.2 percent of MVA in 2015. There was little change ineither share between 2000 and 2015.

In addition to an increase of one-third in to-tal manufacturing output, other factors are alsolinked to the rise in the carbon footprints ofemerging industrial economies, including India,Mexico, Indonesia and China (which is in theupper quintile of the CIP Index). Rapid urbaniza-tion has taken place in developing and transitioneconomies, particularly China. This has led to adoubling of the highly emissions-intensive pro-duction of cement, as demand for constructionmaterials has increased (IPCC, 2014; Andrew,2018). The cement industry alone is respon-sible for 5 per cent of global CO2 emissions(IEA, 2018). Thus, while structural change hascontributed to large-scale poverty reduction inemerging industrial economies over the past 50years, it has also been accompanied by an in-crease in both the scale and emission intensity ofindustrial production in those countries (IPCC,2014; UNIDO, 2018b).



Box 3.2: Comparing annual and historical emissions in the adjusted CIP Index

The adjusted CIP Index only accounts for CO2emissions within a given year. As such, it doesnot consider the historical responsibility of coun-tries to emissions abatement – for example, themajority of industrialized economies have emit-ted CO2 for up to 150 years, while emergingindustrial economies have only contributed sev-eral decades of manufacturing emissions. Be-fore achieving high levels of emission efficiency,most countries underwent a process of structuralchange that involved the production of emissions-intensive goods. As the stock of CO2 particles inthe atmosphere has caused the rise in the level of

anthropogenic warming, these countries can beconsidered to have contributed proportionatelymore to climate change than less developed coun-tries. The main purpose of the CO2-adjusted CIPIndex, however, is to supplement the ranking ofcountries’ competitive industrial performancein a given year. Countries that have achievedhigh levels of emission efficiency should serve asbenchmarks to others. It is therefore important toconsider countries’ performance in a given yearto also be able to track improvements throughenvironmental policies.

Figure 3.3 illustrates that CO2 emissions percapita and MVA per capita follow a shallowinverted-U shape. This implies that the processof industrialization is initially accompanied byan increase in the manufacturing sector’s carbonfootprint. However, the curve flattens at a percapita MVA of around US$ 2,000 in (constant2010 prices). This corresponds to the averageMVA per capita of countries in the upper middlequintile of the CIP Index. Beyond this level of in-dustrialization, the curve measuring the averagemanufacturing carbon footprint flattens out, andeven falls in countries with a per capita MVA ofabove US$ 10,000.

The manufacturing sectors of most countrieswith emission-efficient industries in the CIP In-dex’s top quintile still have higher emissions percapita than those of the least competitive, low-technology countries due to a greater scale ofindustrial production. Therefore, the initial in-crease and subsequent drop in emission intensitywith industrial development does not mean thatsimply encouraging the diffusion of more effi-cient technologies leads to a reduction in totalemissions. Instead, it is important to considerthat emissions rise if the scale effect of increasedproduction through greater incomes dominatesefficiency gains, particularly if there is no changein the type of goods being produced.

The shallow inverted-U shape depicting therelationship between MVA per capita and CO2emissions corresponds roughly to that of theenvironmental Kuznets curve (Dasgupta et al.,2002). Although production in low-income coun-

tries is comparatively inefficient, as shown inFigure 3.3, the very low level of output domi-nates, meaning that the least industrialized coun-tries have the lowest carbon footprints. The en-vironmental Kuznets curve predicts that as low-income countries with small manufacturing sec-tors industrialize, per capita emissions increase.

Structural change leads to greater outputin the manufacturing sector as productivity in-creases. In addition to increased production, atransition in the type of goods produced takesplace towards products with higher emission in-tensity, as discussed above. At the same time,manufacturing firms in less competitive coun-tries are likely to have low levels of productivity;therefore, they may not have the means to in-vest in pollution abatement and more efficienttechnologies. Equally, governments may nothave the institutional capabilities or incentivesto introduce and enforce environmental policies,especially if accompanied by reductions in eco-nomic growth. As countries industrialize, thislack of regulation may also contribute to increas-ing emissions.

According to the environmental Kuznetscurve, the marginal increase in CO2 emissionsfrom manufacturing falls as countries move tohigh levels of industrial output. At higher MVAlevels, the composition of the economy changesand a country’s manufacturing sector is likelyto become highly productive and specialized inless emissions-intensive goods. Finally, the eco-nomic significance of manufacturing overall de-clines as structural change leads to an expansion


Figure 3.3: MVA per capita and CO2 emissions from manufacturing per capita, 2000-2015Source: UNIDO, 2018a and UNIDO, 2018d. Note: Larger circles correspond to quintile means.

of the services sector and to deindustrialization.Consequently, governments may be more will-ing to support emissions reductions by imposingenvironmental regulations on the heaviest pol-

luters. Overall, both efficiency and scale effectscontribute to the decrease in the carbon footprintof industrialized economies.

3.2.4 Linking emission efficiency and the carbon footprint

The previous sections have shown that a coun-try’s per capita manufacturing carbon footprintis defined as the product of emissions per unitof MVA and MVA per capita. Both of these di-mensions are combined in Figure 3.4 to assessthe different groups described in the second partof this report. The horizontal axis in the figuresdenotes MVA per capita and the vertical axis sig-nifies the emission intensity of production. Theper capita carbon footprint from manufacturing(in kilogrammes of CO2) is thus the product ofthe two, represented as the area of the circles.The figures provide a number of important find-

ings and highlight the significance of consideringboth emission intensity and the carbon footprintwhen evaluating a country’s environmental im-pact (Ritchie, 2018).

When differentiating countries by stageof development, we find that industrializedeconomies have by far the highest per capitamanufacturing carbon footprint. The compara-tively high level of emission efficiency does notsuffice to compensate for the volume of produc-tion. Industrialized economies therefore beara large responsibility for further investments inpollution abatement to reduce the environmental



impact of production. Yet at the same time, afurther expansion of manufacturing industriesin industrialized economies would have a com-paratively small effect on global CO2 emissionsdue to their high level of emission efficiency.By contrast, emission intensity is highest inother developing countries and emerging indus-trial economies: greater manufacturing outputin those groups of the same goods and using thesame technologies to produce them would havethe largest negative environmental impact. It istherefore particularly important for the projectedgrowth in the manufacturing industries of thoseeconomies to occur with less emissions-intensivegoods and production processes.

Determining countries’ carbon footprint isalso important in the context of global valuechains. As goods production at the lower end ofthe value chain has shifted from industrializedto transition economies, production emissionshave increased. This is because the “outsourc-ing” of production entails the use of less efficienttechnologies and energy is more likely to comefrom more emissions-intensive fuels, such ascoal, than is the case in economies that use moreadvanced technologies (Liu et al., 2016). Furtherdeindustrialization in industrialized economiesand industrialization elsewhere must thereforebe accompanied by the diffusion of cleaner tech-

nologies and higher environmental standards;otherwise, the global manufacturing carbon foot-print will expand.

The graph at the bottom of Figure 3.4 distin-guishes seven geographic regions. Countries inthe Middle East and North Africa have the high-est emission intensity and the second highest car-bon footprint of any region – mostly attributableto oil exporting countries such as the UnitedArab Emirates, Saudi Arabia and Oman. Furthergrowth in countries’ manufacturing sectors willbe particularly problematic with regard to green-house gas emissions, unless gains in emissionefficiency can be achieved.

The manufacturing industries of South andSouth East Asia have the second highest emis-sion intensity, but a comparatively small carbonfootprint due to the low level of value added.However, strong manufacturing growth in theregion may change this, depending on whethera shift to more high-technology production pro-cesses and goods takes place as the region in-dustrializes, thereby reducing emission intensity.The lowest emission intensity was registered inNorth America, even though the region has thelargest carbon footprint. This again indicates thesignificance of distinguishing between the scaleand the emission efficiency of production whenmeasuring environmental impacts.

3.2.5 Changes in emission efficiency and carbon footprint over time

To determine whether countries’ manufacturingsectors are moving towards greater levels of en-vironmental sustainability, changes in net manu-facturing output and CO2 emissions from man-ufacturing over time must be assessed. Glob-ally, the amount of required energy per unit ofmanufacturing value added fell by 20 per centbetween 1990 and 2015, particularly in industri-alized economies (IEA, 2018). Yet the IPCC es-timates that the current energy intensity of man-ufacturing could be reduced by a further 25 percent if the most efficient innovations were univer-sally implemented (IPCC, 2014). Moreover, ad-vances in technology and renewable substitutes,new forms of sharing infrastructure between in-dustries and changing demand for “green” prod-

ucts could also contribute to greater emissionefficiency in industrial production.

Figure 3.5 presents the trends of MVA andCO2 emissions from manufacturing within dif-ferent CIP quintiles between 2000 and 2015. Incountries belonging to the top quintile of theCIP Index, excluding China, MVA increased bynearly one quarter. Over the same period, CO2emissions from manufacturing shrank by a simi-lar share. Yet in all other groups, manufacturingemissions increased over the observed time pe-riod. This is particularly the case in China, whereMVA was over four times greater in 2015 than in2000, and the level of emissions was over threetimes higher. Figure 3.5 also shows that China’semission efficiency climbed considerably from


Figure 3.4: MVA per capita and CO2 emissions from manufacturing per capita by developmentstage (top) and region (bottom), 2015Source: UNIDO, 2018a and UNIDO, 2018d. Note: The area of the circles and the number within the circles represent

the carbon footprint, CO2 emissions from manufacturing per capita in kg.



2010: in the preceding decade, the rise in emis-sions was higher than MVA, but then sloweddown while net manufacturing output continuedto increase.

In the upper middle quintile, MVA and CO2emissions from manufacturing both increasedby approximately a factor of 1.8 between 2000and 2015, implying that there was no change inthe emission intensity of production over thatperiod. In lower quintiles, CO2 emissions frommanufacturing increased by around one quarter,which, however, was considerably lower than theincrease in MVA.

If MVA growth is higher than the increase ofmanufacturing emissions over a given period,manufacturing’s emission intensity decreasesand a decoupling takes place in this sector. De-coupling is defined by the OECD, 2002, page 4as “breaking the link between ‘environmentalbads’ and ‘economic goods’”. In this context,decoupling occurs when the manufacturing sec-tor’s net output grows at a greater rate than itsCO2 emissions.

Two different cases of decoupling can be dis-tinguished. If manufacturing emissions indicatenegative or no growth while MVA shows a pos-itive growth—as in the case of the CIP Index’stop quintile (excluding China)— absolute decou-pling takes place. This means that an increase inindustrial production is not linked to increases inenvironmental pressures through CO2 emissions.

If relative decoupling occurs, emissions increase,albeit at a slower rate than MVA – as in the caseof all other quintiles of the CIP Index and China.This reflects increased emission efficiency, butmay still be considered problematic for the en-vironment if the growth in production cancelsout the efficiency gains. In this case, the carbonfootprint would continue to rise.

Whether greater emission efficiency actuallyleads to a reduced carbon footprint depends thuson the size of the seemingly paradoxical reboundeffect. The rebound effect applies when greaterresource efficiency leads to economic responsesthat effectively increase net manufacturing out-put, thereby leading to an increase in resourcedemand. This is the case when more efficientproduction means that goods become cheaper,incomes increase, machines can be used moreintensively or technology becomes more widelyavailable and production increases (??). Hence,if the scale effect is larger than the realized effi-ciency gains, there is greater resource demandin manufacturing and the carbon footprint in-creases. This implies that a reduction in emissionintensity is dominated by rising MVA per capita,thereby increasing the per capita carbon footprint(i.e. the area of the circles in Figure 3.4). Over-all, increased industrial production can only beconsidered to not contribute to greater CO2 emis-sions in the presence of absolute decoupling.


Figure 3.5: MVA and CO2 emissions from manufacturing by country group, 2000-2015Source: UNIDO, 2018a and UNIDO, 2018d. Note: MVA in 2010 USD and CO2 emissions from manufacturing relative

to the respective values in 2000, (index numbers, year 2000 =100).



Box 3.3: Measuring countries’ rates of decoupling

The Figure below presents the rate of decouplingin each country for which data is available from2010 to 2015. The quadrants correspond to dif-ferent states of decoupling over the five-year pe-riod. In the bottom-right quadrant, MVA growthis positive and CO2 growth is non-positive – thisdenotes cases of absolute decoupling. In otherwords, increased manufacturing output did notlead to greater CO2 emissions over the observedperiod. The majority of countries in which ab-solute decoupling took place are found in thetop quintiles of the CIP Index. There was com-paratively positive MVA growth and negativeemissions growth in Spain and Denmark, forexample.Both MVA and CO2 emissions increased be-tween 2000 and 2015 in countries in the top-right quadrant of the Figure below. Below thedotted identity line, relative decoupling occurred,as the MVA growth rate is higher than the CO2growth rate. This means that emission intensityfell, as the MVA growth rate was greater thanthat of emissions. The majority of countries thatachieved a relative decoupling of CO2 emissionsfrom MVA were those in the upper quintiles ofthe CIP Index. However, some countries clas-sified as having low levels of competitivenessalso achieved high MVA growth rates and thusrelative decoupling. For example, in Angola,total CO2 emissions from manufacturing roseconsiderably between 2010 and 2015, and yet ef-ficiency gains grew even more. This reflects thesignificance of distinguishing between relativeand absolute decoupling.Countries in which coupling took place are found

above the dotted line. In those countries, theemission intensity of manufacturing increasedbetween 2010 and 2015. This was the casein the majority countries in the middle, lowermiddle and bottom quintiles of the CIP Index,for example, in Bangladesh, Ethiopia and Zam-bia. Yet the greatest increase in emissions withrelatively low MVA growth was registered inBahrain. The emission intensity of manufactur-ing also increased in other industrialized, oil ex-porting countries, such as Kuwait and the UnitedArab Emirates. It is thus important to considerheterogeneity when determining the decouplingrates of countries at different levels of industri-alization. Countries in which negative couplingoccurred are located in the bottom-left quadrant,i.e. both emissions and MVA decreased – this,however, is only the case in very few countriesin exceptional circumstances, such as the SyrianArab Republic.The countries that achieved the highest rates ofdecoupling were Central and Eastern Europeancountries: Czechia, Poland and Slovakia. Thisfollows the methodology used by the OECD’s(2002) indicator to measure decoupling. Theunadjusted CIP Index is an “outcome indicator”whereas decoupling measures “progress”; hencethe decoupling indicator is not suitable to adjustthe CIP Index. It does, however, give an indica-tion of how the adjusted CIP Index is likely todevelop: countries with high rates of decouplingare expected to increase their adjusted competi-tiveness as the emission intensity of productionfalls.


Figure: MVA and CO2 emissions growth rates, 2010-2015Source: UNIDO, 2018a and UNIDO, 2018d. Note: Based on Naqvi and Zwickl, 2017.

3.3 Calculating the CO2-adjusted CIP Index3.3.1 A non-linear adjustment

The impact of industrialization on ecosys-tems—along with the essential services they pro-vide—depends on multiple factors (United Na-tions Environment Programme, 2011). The pre-vious section discussed the relationship betweenthe emission intensity of manufacturing, emis-sions per capita and the level of manufacturingnet output. The main adjustment of country-

specific values of the CIP Index is based onemission intensity. A new CIP Index value isthereby calculated for each country for whichemissions data is available.5

Industrial development that minimizes envi-ronmental externalities depends on the ability ofa country to achieve low levels of emission in-tensity. The first indicator for adjusting the CIP

5See Appendix A.2 for a detailed discussion on the methodology used to calculate the adjusted CIP Index.


3.3 Calculating the CO2-adjusted CIP Index 83

Index is therefore emissions per unit of MVA,normalized so all countries can be ranked rela-tive to the most efficient country, which takes avalue of 1. Countries that experience a substan-tial change from the unadjusted to the adjustedCIP Index can thus be considered to be furtheraway from the frontier of emission efficiency inindustrial production. As discussed above, thereare two main reasons for variations in emissionintensity: the use of different technologies andthe production of different goods.

Moreover, for a given level of emission in-tensity, the adjustment of the CIP Index variesdepending on the country’s carbon footprint. Thecarbon footprint serves as a scaling factor in theadjustment of the CIP Index. The countries withthe lowest per capita manufacturing emissionsare Niger, Cambodia and Cameroon – countrieswith very low levels of production, and hence thescale effect dominates the efficiency effect in thecalculation of the carbon footprint. The countrieswith the greatest scaling factor due to their highlevels of per capita CO2 emissions are Qatar,Oman and the United Arab Emirates. A highscaling factor therefore reflects the need to moveto more emission-efficient modes of productionin manufacturing industries that are responsiblefor the greatest per capita environmental exter-nalities. Efficiency gains in these countries arekey to reducing global manufacturing emissions.

The carbon footprints of countries presentedin the adjusted CIP Index reveal which of thesecountries’ manufacturing sectors have the high-est negative per capita impact on the environ-ment. The Index also reflects the fact that theenvironmental impact of a country’s given emis-sion intensity depends on the ecosystem’s abilityto absorb emissions in carbon sinks and therebylimit the negative externalities of manufactur-ing on the natural environment. The carryingcapacity of carbon sinks refers to the amountof CO2 emissions that forests, the ocean, soilsand other carbon “reservoirs” can capture. Thissequestration of greenhouse gases slows atmo-spheric warming. Accounting for the carryingcapacity of carbon sinks is therefore an importantdimension for estimating the negative impact ofmanufacturing emissions (Nordhaus, 2017).

The interrelationship of carbon sinks and

CO2 emissions is highly complex: the adjust-ment factor is merely a stylized representationof this relationship. The final adjustment fac-tor is non-linear, with an initially increasingmarginal penalty for greater emissions. Due topositive feedback loops, evidence suggests thatclimate change does not have a constant effecton the ability of carbon sinks to capture emis-sions (Rockström et al., 2009). For example, asthe ocean becomes more acidic through carbonabsorption, its ability to sequestrate additionalCO2 is reduced. Similarly, with more frequentwildfires on account of global warming, the stockof carbon stored as biomass in plants and treesenters the atmosphere. This increases the CO2particles in the atmosphere and leads to a reduc-tion in the availability of carbon sinks (Ciais etal., 2013).

The non-linearity of the adjustment factormeans that initially, an increase in emission in-tensity only leads to minor increases in the ad-justment of MVA per capita. If production ishighly efficient, carbon sinks may be able tocompensate for a reasonable share of manufac-turing emissions. The marginal penalty from anincrease in emission intensity expands to reflectthe increasing negative environmental impact oflagging further behind the efficiency frontier.

At the same time, the country’s carbon foot-print determines an emissions threshold. Themarginal negative adjustment is greatest at thisthreshold. This reflects the fact that at a given“tipping point”, which is highly dependent onthe amount of emitted CO2, i.e. the carbon foot-print, the marginal cost of an increase in emis-sions per unit of MVA is highest. For countriesabove this specific emission intensity threshold,the marginal penalty decreases again, i.e. eachincrease in the percentage of emission intensityhas less effect on the adjustment of the CIP valuethan the previous increase in intensity percent-age. This reflects the results of Figure 3.1. Forleast competitive countries—whose manufactur-ing sectors are characterized by high emissionintensity—efficiency improvements through in-vestments in pollution abatement or by shiftingto other product types are assumed to be cheap.Thus, large emission reductions and an improve-ment in efficiency are comparatively easy to


achieve for least efficient manufacturers. How-ever, the efficiency gains from each unit of in-vestment in pollution abatement are decreasing,which implies that as emission intensity declines,greater marginal investments are necessary tofurther reduce the manufacturing sector’s emis-sion intensity. This is visible in countries inthe CIP Index’s middle quintile, where the man-ufacturing sector’s emission intensity grew atthe same rate as manufacturing output between2000 and 2015. Each marginal reduction in emis-sions above the emission intensity tipping pointresults in a lower marginal negative impact oncompetitiveness. It is assumed that this can beeasily changed. For countries with a medium-high emission intensity, however, large invest-ments are necessary to further reduce emissionintensity.

The initially increasing and then decreasingmarginal adjustment factor beyond the tippingpoint also has methodological reasons. It im-poses a lower bound of 0 and an upper boundof 1 on the adjustment factor as emission inten-sity becomes infinitely large or converges to 0.As the unadjusted CIP score is multiplied withthe adjustment factor to compute the adjustedCIP score, imposing those bounds means thatthe adjusted CIP value cannot be larger than theunadjusted CIP value, but can also not be smallerthan 0.

It is necessary to adjust the final CIP indexes,rather than, for example, specific indicators priorto normalization. Adjustment prior to normaliza-tion would change the index value for all coun-tries if the maximum or minimum value werechanged. For example, if there is a large down-ward adjustment for a country with the highestper capita MVA, the index value would increasefor other countries as their relative distance tothe country with the highest MVA per capitafalls, even if they also have a high emission in-tensity. This is not an intuitive result, and hencethe final indexes are multiplied by the adjust-ment factor. Figure 3.6 presents the ratio of theadjusted and unadjusted CIP Index values ac-cording to a ranking of the adjustment factors(i.e. the country with the smallest downward ad-justment in its CIP score ranks first, while coun-tries with least emission-efficient industries havethe highest downward adjustment). The figurepresents the described non-linear relationship: atthe efficiency frontier, there is little downwardadjustment in CIP scores. The marginal adjust-ment increases up to around the 25th countryin the ranking. Beyond this tipping point, themarginal adjustment decreases. This highlightsthe fact that very few countries are at the effi-ciency frontier and have a comparatively smallcarbon footprint.

Box 3.4: The sustainable modernization indicator

The UNIDO Industrial Development Report2016 developed an alternative measure of indus-trialization’s environmental compatibility: thesustainable modernization indicator. This entails

two factors – the emission efficiency gap rela-tive to a “benchmark” country and the extentof structural change, measured as the share ofemployees in the non-agricultural sector.

3.4 CIP adjustment3.4.1 Adjustment in ranks and scores

Table 3.1 presents the CO2-adjusted CIP ranking.The percentage change from the unadjusted tothe adjusted CIP Index is shown in the table’sfourth column. The countries with the largestrelative adjustment to MVA per capita on thebasis of emission intensity are Oman, the UnitedArab Emirates and Ukraine. These countrieshave both high emission intensities and large

carbon footprints. Efficiency increases couldthus have a considerable effect on reductions inenvironmental harm caused by those countries’manufacturing sectors. China’s CIP value alsodecreased significantly, resulting in a drop inits CIP rank of 16 places, from 4th to 20th, thelargest shift in the CIP Index’s top quintile.

Figure 3.7 illustrates the ranks in the adjusted

3.4 CIP adjustment 85

Figure 3.6: Ratio of adjusted and unadjusted CIP Index at different CO2 emission levelsSource: UNIDO, 2018a and UNIDO, 2018d. Note: Larger circles correspond to quintile means.

and unadjusted CIP Indexes by stage of develop-ment. The majority of countries that experienceda downward shift in the CIP rank belonged to thegroup of other developing economies and LDCs.Consequently, emerging industrial economies isthe only group to clearly experience a worsen-ing of their mean rank, moving from an averageposition of 56th to 61st .

The countries with the smallest adjustmentof their CIP scores are all in the top quintile ofthe CIP Index. This indicates that they have alow emission intensity and comparatively lowper capita CO2 emissions from manufacturing.The most competitive country according to theadjusted CIP ranking remains unchanged: Ger-many. Its CIP value decreases by around 11 percent, nonetheless. Due to the German manufac-turing sector’s large lead in CIP scores, how-ever, there is no drop in its ranking. Over-all, those countries with the highest energy effi-ciency—and thus the smallest change in unad-justed to adjusted MVA per capita—are some ofthe most competitive countries according to theCIP Index. This reflects evidence of the distribu-tion of emission intensities (Figure 3.3).

The top 3 of the adjusted CIP Index are Ger-many, Switzerland and Ireland. Both Switzer-land and Ireland are closest to the emissionefficiency frontier and their CIP scores weretherefore not adjusted downward. The coun-tries above them in the unadjusted CIP ranking,namely Japan, the United States, China and theRepublic of Korea, all experienced a consider-able downward adjustment of at least one-thirdand therefore moved down the ranks. This al-lowed Switzerland and Ireland to enter the top3.

Figure 3.7 reveals that the mean CIP rank ofindustrialized economies barely changes whenadjusting for CO2 emissions. This is partly be-cause industrialized economies occupy the ma-jority of the top positions in the ranking, henceminor shifts among countries within the samegroup do not lead to changes in the average rank.At the same time, it is also a reflection of het-erogeneity within the group of industrializedeconomies as regards their emission efficiencyand carbon footprint: while there was very littleadjustment in some countries, other large pol-luters such as the United Arab Emirates, Kuwait



and Qatar experienced major downward shifts intheir CIP scores.

The largest average shift in the rankings isobserved in LDCs, with their mean CIP posi-tion increasing from 119th to 96th. The groupclosed the gap to other developing countries, forwhich the mean rank increased from 94th to 88th.The move of the poorest—and least competi-

tive—countries up the CIP ranks following ad-justment is in part due to the clustering of smallCIP scores described in Chapter 1 of the CIPreport. Minor changes in the scores lead to largevolatility in the CIP ranks. Overall, the CO2 ad-justment reduces the inequality in CIP scoresacross country groups.

Rank Country Adjusted Unadjusted Change Changeadj. CIP CIP CIP (%) in rank

1 Germany 0.4628 0.5196 -10.93 02 Switzerland 0.3181 0.3181 0 -43 Ireland 0.3049 0.3049 0 -44 Japan 0.2776 0.3939 -29.53 25 Italy 0.2433 0.2720 -10.56 -46 United States of America 0.2373 0.3815 -37.81 37 France 0.2272 0.2685 -15.38 -48 Republic of Korea 0.2205 0.3725 -40.8 39 Sweden 0.2201 0.2217 -0.68 -710 Austria 0.1929 0.2275 -15.22 -411 Denmark 0.1701 0.1704 -0.13 -1012 United Kingdom 0.1678 0.2242 -25.16 -313 Singapore 0.1606 0.2654 -39.5 114 Spain 0.1600 0.2016 -20.63 -515 Netherlands 0.1578 0.2709 -41.76 516 Belgium 0.1453 0.2797 -48.05 817 Israel 0.1339 0.1345 -0.5 -1118 Czechia 0.1215 0.2070 -41.31 019 Finland 0.1047 0.1459 -28.25 -720 China 0.1010 0.3803 -73.45 1621 Canada 0.1007 0.2105 -52.15 422 Norway 0.0917 0.1132 -18.96 -923 Hungary 0.0879 0.1454 -39.55 -424 Mexico 0.0807 0.1790 -54.92 425 Poland 0.0791 0.1618 -51.13 226 Lithuania 0.0748 0.0808 -7.49 -1327 Slovenia 0.0741 0.1023 -27.54 -828 Slovakia 0.0708 0.1545 -54.15 429 Portugal 0.0692 0.1024 -32.44 -530 Malaysia 0.0657 0.1666 -60.57 831 Estonia 0.0629 0.0640 -1.75 -1832 Turkey 0.0583 0.1243 -53.10 333 Thailand 0.0525 0.1524 -65.56 834 Australia 0.0519 0.1212 -57.19 435 Brazil 0.0476 0.1042 -54.38 236 Philippines 0.0442 0.0722 -38.81 -737 Romania 0.0427 0.0986 -56.71 1


Table 3.1 continued from previous pageRank Country Adjusted Unadjusted Change Changeadj. CIP CIP CIP (%) in rank

38 Malta 0.0345 0.0379 -8.92 -3139 Indonesia 0.0335 0.0883 -62.07 140 Belarus 0.0330 0.0691 -52.32 -441 Latvia 0.0323 0.0475 -32.08 -1742 Luxembourg 0.0322 0.0727 -55.68 043 Russian Federation 0.0319 0.1127 -71.66 1144 Chile 0.0299 0.0616 -51.5 -745 Costa Rica 0.0298 0.0387 -22.92 -2246 Croatia 0.0285 0.0538 -47.02 -947 Greece 0.0275 0.0605 -54.45 -548 Uruguay 0.0273 0.0302 -9.66 -2849 New Zealand 0.0269 0.0650 -58.54 150 Saudi Arabia 0.0234 0.0967 -75.84 1351 Iceland 0.0233 0.0346 -32.59 -2052 Argentina 0.0233 0.0657 -64.57 553 Botswana 0.0217 0.0224 -3.14 -3454 El Salvador 0.0214 0.0302 -28.99 -2355 India 0.0199 0.0808 -75.39 1556 Bahrain 0.0195 0.0543 -64.02 257 Cambodia 0.0194 0.0194 -0.13 -3458 Guatemala 0.0186 0.0311 -40.09 -1759 Nigeria 0.0186 0.0244 -23.73 -2560 South Africa 0.0183 0.0688 -73.41 1561 Bulgaria 0.0175 0.0509 -65.64 462 Qatar 0.0172 0.0634 -72.82 1263 Peru 0.0171 0.0409 -58.09 164 United Arab Emirates 0.0163 0.0728 -77.61 2365 Morocco 0.0159 0.0404 -60.57 066 Viet Nam 0.0156 0.0677 -76.98 2067 Mauritius 0.0154 0.0229 -32.57 -1968 Colombia 0.0153 0.0373 -58.99 -269 Brunei Darussalam 0.0141 0.0223 -36.94 -1970 Paraguay 0.0137 0.0137 -0.2 -2771 Trinidad and Tobago 0.0131 0.0526 -75.13 1572 Kuwait 0.0127 0.0558 -77.23 1973 Venezuela (Bolivarian Rep. of) 0.0125 0.0410 -69.42 1274 Serbia 0.0125 0.0400 -68.72 875 Tunisia 0.0121 0.0412 -70.58 1576 Jordan 0.0113 0.0276 -59.14 -477 Bangladesh 0.0112 0.0330 -66.04 478 Iran (Islamic Republic of) 0.0109 0.0457 -76.28 1979 Egypt 0.0101 0.0339 -70.15 780 Lebanon 0.0093 0.0205 -54.49 -981 Ukraine 0.0090 0.0404 -77.65 1782 Oman 0.0088 0.0405 -78.38 19


Table 3.1 continued from previous pageRank Country Adjusted Unadjusted Change Changeadj. CIP CIP CIP (%) in rank

83 Cameroon 0.0087 0.0089 -1.36 -3484 Kazakhstan 0.0087 0.0384 -77.28 1685 The f. Yugosl. Rep of Macedonia 0.0086 0.0282 -69.52 686 Panama 0.0085 0.0322 -73.56 1287 Congo 0.0080 0.0096 -16.7 -2588 Ecuador 0.0078 0.0201 -61.27 -289 Honduras 0.0077 0.0160 -51.45 -490 Myanmar 0.0071 0.0138 -48.73 -691 Bosnia and Herzegovina 0.0067 0.0250 -73.01 1092 Pakistan 0.0062 0.0246 -74.99 1093 Armenia 0.0062 0.0114 -46.19 -1094 China, Hong Kong SAR 0.0056 0.0239 -76.68 995 Cyprus 0.0052 0.0156 -66.76 096 Jamaica 0.0051 0.0125 -58.84 -497 Zambia 0.0051 0.0080 -35.8 -2298 Algeria 0.0044 0.0157 -71.73 499 Côte d’Ivoire 0.0044 0.0106 -58.22 -8100 Bolivia (Plurinational State of) 0.0038 0.0125 -69.35 1101 Azerbaijan 0.0038 0.0120 -68.09 -1102 Georgia 0.0037 0.0135 -72.47 4103 Kenya 0.0036 0.0108 -66.31 -2104 Senegal 0.0034 0.0102 -66.49 -5105 United Republic of Tanzania 0.0033 0.0071 -54.1 -15106 Mongolia 0.0030 0.0120 -74.96 5107 Suriname 0.0029 0.0107 -73.21 1108 Republic of Moldova 0.0027 0.0098 -72.3 -3109 Albania 0.0026 0.0094 -72.37 -6110 Mozambique 0.0024 0.0041 -41.48 -19111 Gabon 0.0023 0.0096 -76.02 -2112 Montenegro 0.0023 0.0065 -64.64 -11113 Angola 0.0022 0.0037 -39.16 -20114 Syrian Arab Republic 0.0021 0.0094 -77.49 0115 Zimbabwe 0.0021 0.0065 -67.58 -9116 Ghana 0.0021 0.0064 -68.13 -9117 Niger 0.0020 0.0020 0 -26118 Kyrgyzstan 0.0016 0.0066 -76.17 -3119 Haiti 0.0010 0.0030 -67.08 -18120 Nepal 0.0010 0.0038 -74.38 -10121 Yemen 0.0010 0.0027 -64.47 -18122 Iraq 0.0009 0.0037 -76.86 -9123 Ethiopia 0.0000 0.0000 0 -25

Table 3.1: Adjusted and unadjusted CIP Index ranks and scoresSource: UNIDO, 2018a and UNIDO, 2018d



Figure 3.7: Ranks in the adjusted and unadjusted CIP Indexes by development stageSource: UNIDO, 2018a and UNIDO, 2018d

Figure 3.8 depicts the ratio of the adjustedand unadjusted CIP Index scores at different lev-els of MVA per capita, as well as the CIP quin-tile means. If the ratio is closer to 1, then thedownward adjustment is comparatively small;the country is close to the emission efficiencyfrontier and has a small carbon footprint. Mov-ing from countries in the bottom to those in theCIP Index’s middle quintile, the average ratioof the adjusted and unadjusted CIP scores de-creases. Although there is little change in emis-sion efficiency between these groups, the manu-facturing sectors of the countries in the bottomquintile have a lower emissions impact due totheir level of net output.

The fitted trend line, which depicts the ra-tio of the adjusted and unadjusted CIP values,increases beyond the MVA per capita thresh-old of the countries in the middle quintile. Atgreater levels of MVA per capita, it approachesa value of 1—meaning the countries with thehighest levels of industrialization, i.e. thosethat have specialized in the production of high-technology goods at the highest node of globalvalue chains—also have the highest emissionefficiency levels.

The ratio of the adjusted and unadjustedMVA per capita values corresponds to the inverse

of the stylized inverted-U shape of the environ-mental Kuznets curve. Countries with the great-est environmental impact, i.e. those close to themaximum point of the environmental Kuznetscurve, experience the highest downward shift inCIP values.

As discussed above, the adjustment of theCIP Index is solely based on annual emissions– it does not consider the historical costs ofemissions in different countries. The 1992United Nations Framework Convention on Cli-mate Change states that countries have a “com-mon but differentiated responsibility” to com-bat climate change. While the indicator mea-suring the carbon footprint captures differencesin countries’ contributions to natural capital de-pletion through manufacturing emissions, it isquestionable whether the same adjustment factorshould be applied to both LDCs and industrial-ized economies which have emitted CO2 relatedto manufacturing production over the past 150years. The adjusted CIP Index therefore merelyserves as a benchmark and performance mea-sure. It should not be interpreted as a reasonfor countries with the most emission-efficient in-dustries to reduce their commitment to decreasethe environmental impact of their manufacturingsectors.

3.4.2 By regions



Figure 3.8: Ranks in the adjusted and unadjusted CIP Indexes by development stageSource: UNIDO, 2018a and UNIDO, 2018d. Note: The further the distance from the dashed line, the greater the relative

change from the unadjusted to the adjusted CIP Index.

Figure 3.9 shows that countries with similar ad-justment factors are clustered together. Globally,the countries that experienced the greatest adjust-ment in their CIP values were oil-exporting coun-tries in the Middle East, such as Saudi Arabia,the United Arab Emirates, Oman and Kuwait.These countries have both high emission inten-sities and large carbon footprints. This is partlydue to the industries these countries specialize in.Oil refinery is particularly emissions intensivedue to the energy requirements and chemical pro-cesses during the transformation of crude oil intopetroleum exchanged on international markets(Concawe, 2008). At the same time, if fossil fu-els are very cheap, countries may be less likely toinvest in pollution abatement and improvementsin emission efficiency as there is less monetaryincentive to reduce dependence on fuels that emitlarge amounts of CO2.

In Asia, Viet Nam, India, China and manyothers registered a large downward adjustmentof their CIP Index scores. These are all transi-tion economies that have experienced major in-creases in manufacturing net output which havenot been compensated through gains in emissionefficiency. The adjustment of CIP scores on thebasis of emission intensity and the carbon foot-print highlights the potential of these economiesto catch up with the technological frontier.

In Europe, Ukraine, Bosnia and Herzegov-ina, Russia and other countries in the East wit-

nessed the greatest adjustment to their CIP val-ues. Yet there are also tremendous differencesamong highly-industrialized economies in West-ern Europe. For example, Luxembourg, Belgiumand the Netherlands face considerably higherpenalties than the more efficient industries of, forexample, the Scandinavian countries. Similarly,the United States and Canada have significantlylower levels of emission efficiency relative tocountries at comparable levels of industrializa-tion.

With the exception of oil exporting coun-tries such as Gabon, there is comparatively littledownward adjustment of the CIP Index’s val-ues in sub-Saharan Africa. This, in part, is dueto low total manufacturing production and theresulting small carbon footprint in the major-ity of countries. The adjusted CIP Index showsthat sub-Saharan countries at a low level of in-dustrialization, producing low-technology, low-emissions goods, should aim to find alternativemodes of industrialization that do not necessarilyfollow the path taken by the countries of Eastand South Asia. One possibility for countrieswith low levels of industrialization and low emis-sions is leapfrogging, i.e. adopting advanced,low-emissions technologies from highly com-petitive countries in a process of radical change.For example, decentralized energy generationcan make small-scale firms independent of apotentially unreliable, emissions-intensive, cen-



Figure 3.9: Ranks in the adjusted and unadjusted CIP Indexes by development stageSource: UNIDO, 2018a and UNIDO, 2018d.

tralized energy generating infrastructure (Bank,2017). Such innovations may allow developingcountries to industrialize without necessitatinglarge investments in heavily polluting, fossil fuel-based energy systems. However, if existing de-velopment problems such as the lack of access toeducation, poor quality of infrastructure, ineffec-tive institutions, etc. continue, then it is unclearto what extent leapfrogging can be consideredto contribute to large-scale industrialization andstructural change with limited negative environ-mental effects.

In Latin America, Trinidad and Tobago ex-perienced the largest downward shift in its CIP

score when adjusting for environmental external-ities. It is considered a regional leader in man-ufacturing competitiveness (see Chapter 2.3 ofthis report), yet its industries are built on cheapnatural resources – as such, there is consider-able potential for investments in pollution abate-ment. Other countries that performed poorlyin the adjusted CIP Index include Panama andVenezuela: Box 2.4 explained that Panama’s in-dustrial growth is closely linked to the emissions-intensive construction industry, while Venezuelais an oil exporting economy. By contrast,Paraguay and Uruguay witnessed the smallestadjustment of their CIP scores in the region.


IV

Concluding remarks . . . . . . . . . . . . . . . . . 95

Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

References . . . . . . . . . . . . . . . . . . . . . . . . . 111

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

Concluding remarks

Concluding remarks

Inclusive and sustainable industrial develop-ment (ISID) encompasses the economic, socialand environmental dimensions on which overallprosperity depends. The CIP Index measures acentral component of ISID, namely the economicefficiency of processes related to industrial de-velopment. It is constructed in such a way as toreflect the multiple dimensions of competitive-ness, incorporating the capacity of economies toproduce and sell their goods domestically andinternationally, to adopt advanced technologiesand to generate a greater share of value added.

In general, economies with high levels ofcompetitiveness tend to perform well in each ofthe three key dimensions of competitiveness. Yetthe gap between the most competitive economiesand those trying to catch up is not the sameacross those dimensions. There are major in-equalities in economies’ capacities to produceand export manufactured goods and in theirworld impact. Both dimensions indicate a largegap between a few highly competitive economiesthat dominate globally integrated markets formanufactured goods and a majority of less com-petitive countries.

Performance in technological deepening andupgrading is less uneven. This implies that whilethere is diffusion of technologies to less compet-

itive economies, it is not necessarily accompa-nied by an improvement in other aspects of thoseeconomies’ competitiveness. Building the capac-ities required to convert technological deepen-ing and upgrading into greater overall industrialcompetitiveness - for example, through improve-ments in infrastructure, greater human capital ormore effective institutions - is a key challengeto achieving structural change and ISID, partic-ularly in LCDs. Industrialization in LDCs isfundamental to SDG Target 9.2, namely the dou-bling of the shares of both MVA in GDP and ofmanufacturing employment in total employmentby 2030. This objective can be fostered throughinvestments and capacity building from abroad;yet more needs to be done to increase LDCs’potential to achieve the target.

It is thus possible to broadly distinguishbetween two groups of economies based ontheir performance in developing competitive in-dustries. First, a group of global leaders thatcompetes to innovate and thereby generatesthe largest share of value added at the top ofglobal value chains. Second, economies withinthe group of technological adopters specializein lower-technology intermediate manufacturedgoods, generating smaller amounts of valueadded, and thus compete to position themselves

96

at higher stages of global value chains. The ma-jority of economies fall within the second group,yet there is little dispersion in their levels of com-petitiveness. Minor improvements may thereforesuffice for countries to leapfrog others in theyear-over-year competitiveness rankings. On thewhole, however, very few countries have beenable to transition from the group of technologicaladopters to innovators.

Long-term trends in CIP scores serve to iden-tify those countries that have been particularlysuccessful in improving the competitiveness oftheir manufacturing sectors. It is thus importantto go beyond a static analysis within a given yearto identify successful comparators and role mod-els. Over the past 25 years, China has achievedthe highest absolute increase in both CIP scoreand rank of any country. This increase is at-tributable to the progress made in the dimensionsmeasuring world impact and technological deep-ening and upgrading, both of which are linkedto its strong export orientation and integration inglobal value chains.

Yet China’s improvement in industrial com-petitiveness is only a reflection of the economicdimension of industrialization. A more com-plete evaluation of ISID requires a perspectivethat takes into account environmental sustain-ability as well as socio-economic considerations.Therefore, the 2018 edition of the CIP reportextends the Index to adjust for environmentaldamages resulting from manufacturing, approx-imated by CO2 emissions. The CO2-adjustedCIP Index can support industrial policies thatpromote forms of production that are more com-patible with the preservation of natural capital.

For example, as China industrializes, theCO2 emissions and environmental impact of itsmanufacturing sector have increased consider-ably – much like in other industrialized and tran-sition economies as their manufacturing output

increases. There is an urgent global need topromote industrial development while reducinggreenhouse gas emissions. This can be achievedthrough the use of more efficient technologiesor alternative energy sources or the change ofthe specialization pattern towards the produc-tion of more environmentally-friendly goods, allof which can reduce the damage manufacturingcauses to the world’s natural capital. Otherwise,long-term productivity and welfare are at risk.

Accordingly, the CO2-adjusted CIP Indexdefines industrial competitiveness not just on thebasis of economic measures, but also with regardto the environmental sustainability of manufac-turing production. The adjusted Index identi-fies countries near the frontier of emission ef-ficiency that can serve as a benchmark for oth-ers. Through technological progress and furtherstructural change, this frontier can shift furthertowards green forms of production that minimizeindustrialization’s environmental externalities,while contributing to greater welfare.

Preservation of natural systems, which is aglobal public good on which the welfare of ev-ery country depends, is imperative. As in manyother cases with public goods, the free marketallocation will be socially inefficient and there-fore, developed countries should have incentivesto help developing and transition economies, par-ticularly LDCs, to follow a more environmen-tally friendly path of industrialization than theythemselves did. Thus, global cooperation andintegration are central to achieving sustainableindustrialization. Global policies in support ofenvironmentally sustainable manufacturing canlead to a form of structural change that doesnot put excessive stress on natural systems, andthereby does not undermine the benefits of com-petitive industries for economic growth and so-cial inclusion.

Appendix

Appendix A

A.1 Methodology of the CIP IndexThe CIP index is constructed as a weighted geometric average of six indicators, using the followingformula:

CIPjt = (6

∏i=1

Ii jt)(1/6) (3.1)

where Ii jt is the value of the indicator i for country j and year t. The CIPjt values range from 0 to 1.The six indicators are included in the dimensions of the CIP Index, mentioned in Chapter 1, Section2. They are the following:

Dimension 1: Capacity to produce and export manufactured goods

Indicator 1: Manufacturing value added per capita (MVApc) is the relative value of total netmanufacturing output to a country’s population (POP):

MVApc = MVA/POP (3.2)

Indicator 2: Manufacturing exports per capita (MXpc) is the value of manufacturing exportsrelative to the population:

MXpc = MX/POP (3.3)

Dimension 2: Technological deepening and upgrading

Indicator 3: Industrialization intensity (INDint) is a composite indicator calculated as thearithmetic average of two other indicators, namely manufacturing value added share in total GDP

98

(MVAsh = MVA/GDP) and the medium- and high-tech manufacturing value added share in totalmanufacturing value added (MHVAsh = MHVA/MVAtotal):

INDint = (MVAsh +MHVAsh)/2 (3.4)

Indicator 4: Export quality (MXQual) is also a composite indicator. It is calculated as thearithmetic average of two further indicators measuring the share of manufacturing exports in totalexports (MXsh= MX/Xtotal) and the share of medium- and high-tech manufacturing exports in totalmanufacturing exports (MHXsh=MHX/MX):

MXQual = (MXsh +MHXsh)/2 (3.5)

Dimension 3: World impact

Indicator 5: Impact of an economy on world manufacturing value added (ImWMVA) is theshare of an economy in global manufacturing value added:

ImWMVA = MVA/MVAworld (3.6)

Indicator 6: Impact of a country on world trade (ImWMT) is the share of an economy in globalmanufacturing trade:

ImWMT = MX/MXworld (3.7)

A.2 Methodology of the CO2-adjusted CIP IndexAs presented in the text, two dimensions are used to adjust the CIP Index on the basis of emissions.The first dimension covers the emission intensity of manufacturing, measured as CO2 emissionsper unit of MVA per capita in country i. It is calculated as:

EIi = Ei/MVAi (3.8)

with E denoting CO2 emissions from manufacturing in a given year. Countries with a small indexvalue in the first dimension of the adjustment function can be considered to use more efficienttechnologies and produce “greener” goods. The second dimension of the adjustment function iscalculated as CO2 emissions per capita and thus represents the per capita manufacturing carbonfootprint of country i:

CFi = Ei/Popi (3.9)

The second dimension thereby measures the negative impact of a given country’s manufacturingsector on the global stock of natural capital. This reflects the notion that countries with a larger percapita manufacturing carbon footprint, for example, because the manufacturing sector producesgoods that are particularly emissions intensive but important for the economy, assume a greaterresponsibility to invest in pollution abatement. Each country’s adjustment factor is calculated asfollows. Both emission indicators are normalized according to the distance-to-reference method(Joint Research Centre-European Commission, 2008). The country with the lowest emissionintensity at the emissions frontier serves as the reference for other countries:

IIEi = 1− (EImin/EIi) (3.10)

Similarly, for the carbon footprint:

ICFi = 1− (CFmin/CFi) (3.11)

99

Both emission indicators are thus considered to measure the “bads” of manufacturing ratherthan the “goods” as in the other CIP indicators. The final adjustment factor is calculated accordingto the following formula that contains the product of the two indexes:

Ad ji = 1− IEIi

1+αICFi(IEIi)β

(3.12)

α and β are values selected to determine the scale of the adjustment. In the analysis above,both parameters are arbitrarily set to a value of 2. Starting from this value, a marginal increase(reduction) of β , would change the relative weight of IIEi respect to ICFi , increasing (reducing)thus the penalty for emissions intensity; while a marginal increase (reduction) of α would increase(reduce) the adjustment factor for each polluting country.

The precise shape of the adjustment factor curve is mainly determined by a country’s emissionintensity. If the emission intensity is small, then there is a larger bandwidth of “acceptable” carbonfootprint at which the adjustment factor remains small. If, in contrast, the country has a largeemission intensity, the marginal penalty increases quickly. Increasing the carbon footprint therebyleads to a very small adjusted CIP score sooner. In other words, being further away from theemission efficiency frontier is penalized more severely. This reflects the aim of the adjusted CIP topoint out the potential for emissions reductions relative to the technologies available in the leadingcountries.

The CO2 adjusted CIP Index is calculated as:

CIPAd ji = Ad ji ·CIPi (3.13)

with the unadjusted CIP Index calculated as in Appendix A.1. If a country has both a lowemission intensity and carbon footprint value, its adjustment factor is close to 1 and there is thereforelittle change between the adjusted and the unadjusted CIP scores. If, however, a country is furtherfrom the efficiency frontier and has a large carbon footprint, the adjustment factor approaches 0and the adjusted CIP score is therefore a smaller proportion of the unadjusted score. The adjustedCIP Index is thus also within the range [0; 1] and is necessarily no larger than the unadjusted CIPIndex. The relative downward shift in the adjusted CIP Index value is a reflection of the state ofenvironmental damages and efficiency potential within a country.

100

Appendix BB.1 Country classificationsCountries by development stage

Industrialized Economies

Australia Austria BahrainBelarus Belgium BermudaCanada China, Hong Kong SAR China, Macao SARChina, Taiwan Province Czechia DenmarkEstonia Finland FranceGermany Hungary IcelandIreland Israel ItalyJapan Kuwait LithuaniaLuxembourg Malaysia MaltaNetherlands New Zealand NorwayPortugal Qatar Republic of KoreaRussian Federation Singapore SlovakiaSlovenia Spain SwedenSwitzerland Trinidad and Tobago United Arab EmiratesUnited Kingdom United States of America

Emerging Industrial Economies

Argentina Brazil Brunei DarussalamBulgaria Chile ChinaColombia Costa Rica CroatiaCyprus Egypt GreeceIndia Indonesia Iran (Islamic Republic of)Kazakhstan Latvia MauritiusMexico Oman PeruPoland Romania Saudi ArabiaSerbia South Africa SurinameThailand FYR Macedonia TunisiaTurkey Ukraine UruguayVenezuela

Other Developing Economies

Albania Algeria AngolaArmenia Azerbaijan BahamasBarbados Belize BoliviaBosnia and Herzegovina Botswana Cabo VerdeCameroon Congo Côte d’IvoireEcuador El Salvador FijiGabon Georgia GhanaGuatemala Honduras IraqJamaica Jordan KenyaKyrgyzstan Lebanon MaldivesMongolia Montenegro MoroccoNamibia Nigeria PakistanPanama Papua New Guinea Paraguay

101

Philippines Republic of Moldova Saint LuciaSri Lanka State of Palestine SwazilandSyrian Arab Republic Tajikistan TongaViet Nam Zimbabwe

Least Developed Countries

Afghanistan Bangladesh BurundiCambodia Central African Republic EritreaEthiopia Gambia HaitiLao People’s Dem. Rep. Madagascar MalawiMozambique Myanmar NepalNiger Rwanda SenegalUganda United Republic of Tanzania YemenZambiaSource: UNIDO, 2017a, UNIDO, 2018c, UNIDO, 2019.

Countries by geographic region

Europe

Albania Austria BelarusBelgium Bosnia and Herzegovina BulgariaCroatia Cyprus CzechiaDenmark Estonia FinlandFrance Georgia GermanyGreece Hungary IcelandIreland Italy LatviaLithuania Luxembourg MaltaMontenegro Netherlands NorwayPoland Portugal Republic of MoldovaRomania Russian Federation SerbiaSlovakia Slovenia SpainSweden Switzerland FYR MacedoniaUkraine United Kingdom

Latin America

Argentina Bahamas BarbadosBelize Bolivia (Plurinational State of) BrazilChile Colombia Costa RicaEcuador El Salvador GuatemalaHaiti Honduras JamaicaMexico Panama ParaguayPeru Saint Lucia SurinameTrinidad and Tobago Uruguay Venezuela (Bolivarian Rep. of)

MENA

Algeria Bahrain EgyptIran (Islamic Republic of) Iraq IsraelJordan Kuwait LebanonMorocco Oman QatarSaudi Arabia State of Palestine Syrian Arab Republic

102

Tunisia Turkey United Arab EmiratesYemen

North America

Bermuda Canada United States of America

South and South East Asia

Afghanistan Bangladesh Brunei DarussalamCambodia India IndonesiaLao People’s Dem Rep Maldives MyanmarNepal Pakistan PhilippinesSri Lanka Thailand Viet Nam

Sub-Saharan Africa

Angola Botswana BurundiCabo Verde Cameroon Central African RepublicCongo Côte d’Ivoire EritreaEthiopia Gabon GambiaGhana Kenya MadagascarMalawi Mauritius MozambiqueNamibia Niger NigeriaRwanda Senegal South AfricaSwaziland Uganda United Republic of TanzaniaZambia Zimbabwe

Other Asia and Pacific

Armenia Azerbaijan FijiKazakhstan Kyrgyzstan MongoliaPapua New Guinea Tajikistan TongaSource: UNIDO, 2017a, UNIDO, 2019.

103

B.1 Country classificationsTechnology classification of exports

Type of export SITC Rev. 3

Resource-based 016, 017, 023, 024, 035, 037, 046, 047, 048, 056, 058, 059, 061, 062, 073,098, 111, 112, 122, 232, 247, 248, 251, 264, 265, 281, 282, 283, 284, 285,286, 287, 288, 289, 322, 334, 335, 342, 344, 345, 411, 421, 422, 431, 511,514, 515, 516, 522, 523, 524, 531, 532, 551, 592, 621, 625, 629, 633, 634,635, 641, 661, 662, 663, 664, 667, 689

Low technology 611, 612, 613, 642, 651, 652, 654, 655, 656, 657, 658, 659, 665, 666, 673,674, 675, 676, 677, 679, 691, 392, 693, 694, 695, 696, 697, 699, 821, 831,841, 842, 843, 844, 845, 846, 848, 851, 893, 894, 895, 897, 898, 899

Medium technology 266, 267, 512, 513, 533, 553, 554, 562, 571, 572, 573, 574, 575, 579, 581,582, 583, 591, 593, 597, 598, 653, 671, 672, 678, 711, 712, 713, 714, 721,722, 723, 724, 725, 726, 727, 728, 731, 733, 735, 737, 741, 742, 743, 744,745, 746, 747, 748, 749, 761, 762, 763, 772, 773, 775, 778, 781, 782, 783,784, 785, 786, 791, 793, 811, 812, 813, 872, 873, 882, 884, 885

High technology 525, 541 542, 716, 718, 751, 752, 759, 764, 771, 774, 776, 792, 871, 874,881, 891

Source: UNIDO, 2017a, page 90.

Medium high- and high-technology (MHT) manufacturing categoriesDescription ISIC Rev. 3

Manufacture of chemicals and chemical products 24Manufacture of machinery and equipment 29Manufacture of office, accounting and computing machinery 30Manufacture of electrical machinery and apparatus 31Manufacture of radio, television, and communication equipment and apparatus 32Manufacture of medical, precision and optical instruments, matches and clocks 33Manufacture of motor vehicles, trailers and semi-trailers 34Manufacture of other transport equipment, excluding: 35–ISIC revision 3:Building and repairing of ships and boats = 351–ISIC revision 4:Building of ships and floating structures = 3011Building of pleasure and sporting boats = 3012Repair of transport equipment, except motor vehicles = 3315Source: OECD, 2003, UNIDO, 2010 and UNIDO, 2017a, page 90.

104

Appendix CC.1 CIP Index 2018, detailed tablesRegional scores and ranks

Table 3.5: East Asia

Country Group Global Index Dimension Dimension Dimensionrank rank score 1 2 3

Japan 1 2 0.3998 17 10 4China 2 3 0.3764 48 9 1Republic of Korea 3 5 0.3667 13 1 5Singapore 4 12 0.2573 4 3 28China, Taiwan Province 5 13 0.2547 16 2 15Malaysia 6 22 0.1662 35 14 23Australia 7 30 0.1199 33 98 30New Zealand 8 46 0.0659 35 99 58China, Hong Kong SAR 9 87 0.0220 82 111 84China, Macao SAR 10 147 0.0008 124 150 147Source: UNIDO, 2018a.

105

Table 3.6: Europe


Germany 1 1 0.5234 7 5 3Switzerland 2 6 0.3207 2 13 17Ireland 3 7 0.3172 1 4 26Belgium 4 8 0.2807 3 21 12Italy 5 9 0.2733 19 23 7Netherlands 6 10 0.2707 5 30 11France 7 11 0.2679 23 20 6Austria 8 14 0.2389 6 16 24Sweden 9 15 0.2254 9 17 27United Kingdom 10 16 0.2191 31 27 9Czechia 11 17 0.2148 11 7 25Spain 12 19 0.2044 30 33 13Denmark 13 21 0.1715 10 29 33Poland 14 23 0.1651 37 24 21Slovakia 15 24 0.1604 14 6 35Hungary 16 26 0.1493 22 8 32Finland 17 27 0.1457 15 34 39Slovenia 18 31 0.1109 12 19 52Russian Federation 19 32 0.1047 61 76 18Norway 20 33 0.1042 18 69 45Portugal 21 34 0.1026 36 46 40Romania 22 37 0.1015 44 18 38Lithuania 23 40 0.0818 29 37 59Luxembourg 24 42 0.0728 8 58 72Belarus 25 47 0.0657 51 22 57Estonia 26 48 0.0647 25 38 70Greece 27 52 0.0591 49 70 54Croatia 28 53 0.0552 45 43 65Bulgaria 29 54 0.0524 53 50 60Latvia 30 59 0.0474 41 54 75Serbia 31 62 0.0416 64 44 66Ukraine 32 64 0.0407 92 57 51Malta 33 65 0.0398 28 39 93Iceland 34 71 0.0345 20 94 104The f. Yugosl. Rep of Macedonia 35 78 0.0291 59 36 90Bosnia and Herzegovina 36 81 0.0257 69 66 89Cyprus 37 93 0.0159 74 63 125Georgia 38 97 0.0135 96 81 112Albania 39 106 0.0105 101 110 115Republic of Moldova 40 110 0.0097 113 83 126Montenegro 41 122 0.0066 102 100 138Source: UNIDO, 2018a.

106

Table 3.7: Latin America


Mexico 1 20 0.1786 47 15 10Brazil 2 35 0.1019 65 53 16Argentina 3 49 0.0633 63 65 41Chile 4 51 0.0606 54 107 46Trinidad and Tobago 5 56 0.0499 39 47 80Peru 6 60 0.0426 76 102 50Costa Rica 7 67 0.0389 58 56 77Venezuela (Bolivarian Republic of) 8 68 0.0382 66 120 48Colombia 9 70 0.0369 80 85 49Guatemala 10 74 0.0309 84 59 71Panama 11 75 0.0308 56 64 74El Salvador 12 76 0.0303 75 51 81Uruguay 13 79 0.0281 57 106 85Ecuador 14 89 0.0196 91 134 78Honduras 15 92 0.0159 105 79 95Paraguay 16 96 0.0136 99 122 98Bolivia (Plurinational State of) 17 98 0.0134 107 133 92Jamaica 18 100 0.0121 97 88 121Barbados 19 104 0.0107 71 55 137Suriname 20 114 0.0092 73 131 135Bahamas 21 118 0.0080 83 60 139Belize 22 128 0.0049 104 121 141Saint Lucia 23 135 0.0034 98 80 143Haiti 24 137 0.0030 137 95 132Source: UNIDO, 2018a.

107

Table 3.8: Middle East and North Africa


Israel 1 28 0.1318 24 25 37Turkey 2 29 0.1242 50 41 22Saudi Arabia 3 36 0.1018 43 72 31United Arab Emirates 4 41 0.0735 38 124 44Qatar 5 50 0.0631 26 86 63Bahrain 6 55 0.0515 32 77 79Kuwait 7 57 0.0491 42 118 62Iran (Islamic Republic of) 8 58 0.0482 78 82 42Tunisia 9 61 0.0418 70 32 64Morocco 10 63 0.0415 88 42 56Oman 11 66 0.0392 52 97 73Egypt 12 73 0.0331 103 67 47Jordan 13 80 0.0267 86 49 82Lebanon 14 90 0.0188 89 71 94Algeria 15 94 0.0149 112 143 68State of Palestine 16 111 0.0096 110 91 123Syrian Arab Republic 17 112 0.0093 125 116 87Yemen 18 139 0.0026 143 123 128Iraq 19 146 0.0009 133 149 102Source: UNIDO, 2018a.

Table 3.9: North America


United States of America 1 4 0.3726 27 28 2Canada 2 18 0.2074 21 48 14Bermuda 3 138 0.0027 87 62 149Source: UNIDO, 2018a.

108

Table 3.10: South and South East Asia


Thailand 1 25 0.1536 46 12 19Indonesia 2 38 0.0907 77 45 20India 3 39 0.0830 108 40 8Philippines 4 43 0.0725 81 11 34Viet Nam 5 44 0.0724 72 31 29Bangladesh 6 72 0.0340 114 61 43Sri Lanka 7 77 0.0298 85 84 67Pakistan 8 82 0.0245 126 68 53Brunei Darussalam 9 83 0.0245 40 78 117Cambodia 10 88 0.0212 100 73 76Myanmar 11 91 0.0186 115 90 69Lao People’s Dem Rep 12 101 0.0115 111 89 105Nepal 13 131 0.0037 144 96 127Maldives 14 142 0.0018 117 147 146Afghanistan 15 144 0.0013 140 145 124Source: UNIDO, 2018a

109

Table 3.11: Sub-Saharan Africa


South Africa 1 45 0.0694 67 52 36Swaziland 2 84 0.0243 62 35 111Botswana 3 85 0.0238 55 93 83Mauritius 4 86 0.0222 60 75 107Namibia 5 95 0.0147 79 126 109Kenya 6 103 0.0108 130 101 86Côte d’Ivoire 7 105 0.0106 122 127 91Senegal 8 108 0.0101 127 87 101Gabon 9 109 0.0097 90 146 119Congo 10 113 0.0093 106 74 106Nigeria 11 115 0.0092 121 135 55Cameroon 12 117 0.0087 123 119 96Zambia 13 119 0.0080 129 125 108Ghana 14 123 0.0064 131 144 97Zimbabwe 15 124 0.0061 132 113 122Mozambique 16 125 0.0055 139 92 116Madagascar 17 126 0.0054 136 128 118United Republic of Tanzania 18 127 0.0053 138 142 99Uganda 19 129 0.0045 141 130 113Angola 20 130 0.0039 116 148 88Central African Republic 21 132 0.0035 135 26 140Malawi 22 134 0.0034 142 115 131Cabo Verde 23 136 0.0030 120 109 145Niger 24 140 0.0022 145 117 130Rwanda 25 141 0.0021 146 139 136Ethiopia 26 143 0.0015 149 141 120Gambia 27 145 0.0011 147 114 148Burundi 28 148 0.0000 150 132 142Eritrea 29 149 0.0000 148 140 144Source: UNIDO, 2018a.

Table 3.12: Other Asia and Pacific


Kazakhstan 1 69 0.037172835 68 104 61Armenia 2 99 0.012790185 95 103 114Mongolia 3 102 0.010934582 94 138 103Azerbaijan 4 107 0.010140587 109 137 100Fiji 5 116 0.00907017 93 105 134Papua New Guinea 6 120 0.006566948 118 136 110Kyrgyzstan 7 121 0.00656189 128 108 129Tajikistan 8 133 0.003536148 134 112 133Tonga 9 150 0 119 129 150Source: UNIDO, 2018a.

References

Andrew, Robbie M (2018). “Global CO 2 emissions from cement production”. In: Earth SystemScience Data 10.1, page 195 (cited on pages 72, 74).

Balibey, Mesut (2015). “Relationships among CO2 emissions, economic growth and foreign directinvestment and the EKC hypothesis in Turkey”. In: International Journal of Energy Economicsand Policy 5.4, pages 1042–1049 (cited on page 72).

Bank of England (2018). EU withdrawal scenarios and monetary and financial stability: A responseto the House of Commons Treasury Committee. Available at:https://www.bankofengland.co.uk/ (cited on page 44).

Beaton, Ms Kimberly and Metodij Hadzi-Vaskov (2017). Panama’s Growth Prospects: Determi-nants and Sectoral Perspectives. International Monetary Fund (cited on page 45).

Borghesi, Simone, Giulio Cainelli, and Massimiliano Mazzanti (2015). “Linking emission tradingto environmental innovation: evidence from the Italian manufacturing industry”. In: ResearchPolicy 44.3, pages 669–683 (cited on page 72).

Brinkley, Andrew and Simon Less (2010). “Carbon omissions”. In: Consumption-based accountingfor international carbon emission. Policy exchange research note. London: Policy Exchange(cited on page 70).

Ciais, Philippe et al. (2013). “Climate change 2013: the physical science basis. Contribution ofWorking Group I to the Fifth Assessment Report of the Intergovernmental Panel on ClimateChange”. In: K., Tignor, M., Allen, SK, Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, PM,Eds (cited on page 83).

Cifci, Eren and Matthew Oliver (2018). “Reassessing the Links between GHG Emissions, EconomicGrowth, and the UNFCCC: A Difference-in-Differences Approach”. In: Sustainability 10.2,page 334 (cited on page 70).

Concawe. Environmental Science for European Refining (2008). CO2 emissions from EU refineries.Available at:https://www.concawe.eu/ (cited on page 90).

Dasgupta, Susmita et al. (2002). “Confronting the environmental Kuznets curve”. In: Journal ofeconomic perspectives 16.1, pages 147–168 (cited on page 75).

https://www.bankofengland.co.uk/report/2018/eu-withdrawal-scenarios-and-monetary-and-financial-stability

https://www.concawe.eu/publication/co2-emissions-from-eu-refineries/

112

Dosi, Giovanni, Keith Pavitt, Luc Soete, et al. (1990). The economics of technical change andinternational trade. Harvester Wheatsheaf Press/New York University Press, London/New York(cited on page 15).

Eichengreen, Barry, Donghyun Park, and Kwanho Shin (2013). Growth slowdowns redux: Newevidence on the middle-income trap (cited on page 41).

European Commission Eurostat (2016). Irish GDP revision. Available at:https://ec.europa.eu/ (cited on page 44).

Fox, Ms Louise, Mr Alun H Thomas, and Cleary Haines (2017). Structural Transformation inEmployment and Productivity: What Can Africa Hope For? International Monetary Fund (citedon page 57).

Goel, Manisha and Paulina Restrepo-Echavarria (2015). “India’s Atypical Structural Transforma-tion”. In: Economic Synopses 23 (cited on page 54).

Guo, Kai and Papa N’Diaye (2009). “Is China’s Export-Oriented Growth Sustainable?” In: IsChina’s Export-Oriented Growth Sustainable? 9.172, pages 1–21 (cited on page 40).

Haraguchi, Nobuya, Charles Fang Chin Cheng, and Eveline Smeets (2017). “The importanceof manufacturing in economic development: Has this changed?” In: World Development 93,pages 293–315 (cited on pages 36, 40).

Haraguchi, Nobuya and Gorazd Rezonja (2011). “Emerging patterns of manufacturing structuralchange”. In: UNIDO working papers 04/2010 (cited on page 14).

Hausmann, Ricardo, Luis Espinoza, and Miguel Santos (2016). “Shifting gears: A growth Diagnos-tic of Panama”. In: Research Working Paper Series 16-045 (cited on page 45).

Intergovernmental Panel on Climate Change (2014). Mitigation of climate change. Contribution ofWorking Group III to the Fifth Assessment Report of the Intergovernmental Panel on ClimateChange (cited on pages 74, 77).

International Energy Agency (2018). CO2 Emissions from Fuel Combustion 2018. Paris: IEA (citedon pages 72–74, 77).

International Labour Organization (2018). ILOSTAT 2018 Database. Available at:https://www.ilo.org/ilostat/. Geneva (cited on page 19).

Kaldor, Nicholas (1966). Causes of the slow rate of economic growth of the United Kingdom: aninaugural lecture. Cambridge University Press (cited on page 14).

Krugman, Paul (1994). “Competitiveness: a dangerous obsession”. In: Foreign Affairs 73, page 28(cited on page 10).

Labaye, Eric et al. (2013). “A new dawn: Reigniting growth in Central and Eastern Europe”. In:McKinsey Global Institute (cited on page 41).

Lewis, W Arthur (1954). “Economic development with unlimited supplies of labour”. In: Themanchester school 22.2, pages 139–191 (cited on page 14).

Liu, Zhu et al. (2016). “Targeted opportunities to address the climate–trade dilemma in China”. In:Nature Climate Change 6.2, page 201 (cited on page 77).

Lopez-Acevedo, Gladys and Raymond Robertson (2016). Stitches to Riches?: Apparel Employment,Trade, and Economic Development in South Asia. The World Bank (cited on page 55).

Malik, Adeel and Jonathan RW Temple (2009). “The geography of output volatility”. In: Journalof Development Economics 90.2, pages 163–178 (cited on page 10).

Masson-Delmotte, V et al. (2018). Global warming of 1.5 C An IPCC Special Report on the impactsof global warming of 1.5 C above pre-industrial levels and related global greenhouse gasemission pathways, in the context of strengthening the global response to the threat of climatechange, sustainable development, and efforts to eradicate poverty. Summary for PolicymakersEdited by Science Officer Science Assistant . . . (cited on page 71).

McMillan, Margaret S and Dani Rodrik (2011). Globalization, structural change and productivitygrowth. National Bureau of Economic Research (cited on pages 12, 14, 36).

https://ec.europa.eu/eurostat/documents/24987/6390465/Irish_GDP_communication.pdf

https://www.ilo.org/ilostat/

113

Naqvi, Asjad and Klara Zwickl (2017). “Fifty shades of green: Revisiting decoupling by economicsectors and air pollutants”. In: Ecological Economics 133, pages 111–126 (cited on pages 70,82).

Nasr, Nabil et al. (2018). “Re-defining Value–The Manufacturing Revolution: Remanufacturing,Refurbishment, Repair and Direct Reuse in the Circular Economy; a Report of the InternationalResources Panel”. In: Nairobi, International Resource Panel (cited on page 70).

Nice, Thomas (2018). “Forecasting Industrialization in Developing Countries”. In: UNIDO workingpapers forthcoming (cited on page 23).

Nordhaus, William D (2017). “Revisiting the social cost of carbon”. In: Proceedings of the NationalAcademy of Science. Volume 114, pages 1518–1523 (cited on pages 69, 73, 83).

OECD (2002). Indicators to measure decoupling of environmental pressure from economic growth.The OECD enviromental programme. Available at:https://www.oecd.org/ (cited on pages 70, 79).

— (2003). OECD Science, Technology and Industry Scoreboard 2003. Paris (cited on page 103).— (2008). Staying competitive in the global economy: compendium of studies on global value

chains. Organisation for Economic Co-operation and Development (cited on page 12).— (2016). Irish GDP up by 26.3% in 2015? Available at:

https://www.oecd.org/ (cited on page 44).Padilla, Emilio (2018). “Factors leading to sustainable consumption (and barriers to it)”. In: UNIDO

working papers 15/2018 (cited on page 45).Piketty, Thomas and Lucas Chancel (2015). “Carbon and inequality: from Kyoto to Paris”. In:

Trends in the Global Inequality of Carbon Emissions (1998-2013) and Prospects for AnEquitable Adaptation Fund. Paris: Paris School of Economics (cited on page 73).

Ricardo, David (1817). On the principles of political economy and taxation. London: John Murray(cited on page 9).

Ritchie, Hannah (2018). Global inequalities in CO2 emissions. Available at:https://ourworldindata.org/co2-by-income-region (cited on page 76).

Robert, Kates W, Thomas M Parris, and Anthony A Leiserowitz (2005). “What is sustainabledevelopment? Goals, indicators, values, and practice”. In: Environment: science and policy forsustainable development 47.3, pages 8–21 (cited on page 71).

Rockström, Johan et al. (2009). “Planetary boundaries: exploring the safe operating space forhumanity”. In: Ecology and society (cited on page 69).

Rodrik, Dani (2011). The future of economic convergence. Technical report. National Bureau ofEconomic Research (cited on page 36).

— (2012). “Unconditional convergence in manufacturing”. In: The Quarterly Journal of Economics128.1, pages 165–204 (cited on page 36).

— (2018). New Technologies, Global Value Chains, and Developing Economies (cited on pages 15,34).

Samuelson, Paul A (1954). “The pure theory of public expenditure”. In: The review of economicsand statistics, pages 387–389 (cited on page 71).

The Economist (2014). Time to make in India? Available at:https://www.economist.com/ (cited on page 54).

The Economist Intelligence Unit (2018a). Irish GDP up by 26.3% in 2015? Available at:https://www.eiu.com/ (cited on page 48).

— (2018b). Israel’s technology sector faces challenges. Available at:https://www.eiu.com/ (cited on page 48).

Tol, Richard SJ (2018). “The economic impacts of climate change”. In: Review of EnvironmentalEconomics and Policy 12.1, pages 4–25 (cited on pages 69, 71).

http://www.oecd.org/environment/indicators-modelling-outlooks/1933638.pdf

http://www.oecd.org/sdd/na/Irish-GDP-up-in-2015-OECD.pdf

https://ourworldindata.org/co2-by-income-region

https://www.economist.com/banyan/2014/09/25/time-to-make-in-india

http://country.eiu.com/article.aspx?articleid=466755630&Country=Israel&topic=Economy&subtopic=Long-term+outlook&subsubtopic=Long-term+trends&aid=1&oid=1486755532

http://www.eiu.com/industry/article/416900425/israels-technology-sector-faces-challenges/2018-07-03

114

United Nations (2015). “Transforming our world: The 2030 agenda for sustainable development”.In: Resolution adopted by the General Assembly (cited on page 10).

United Nations Conference on Trade and Development (2018). Handbook of Statistics 2017.Geneva: United Nations (cited on page 10).

United Nations Environment Programme (2011). Decoupling natural resource use and environmen-tal impacts from economic growth. UNEP/Earthprint (cited on pages 69, 70, 82).

United Nations Industrial Development Organization (2002). Industrial Development Report2002/2003: Competing through innovation and learning. Vienna: UNIDO (cited on page 24).

— (2010). Industrial Statistics: Guidelines and Methodology. Vienna: UNIDO (cited on page 103).— (2011). Industrial Development Report 2011: Industrial energy efficiency for sustainable wealth

creation. Capturing environmental, economic and social dividends. Vienna: UNIDO (cited onpage 70).

— (2013). The industrial competitiveness of nations: Looking back, forging ahead. Vienna: UNIDO(cited on pages 10, 16, 17, 28).

— (2014). “How industrial development matters to the well-being of the population – Somestatistical evidence”. In: Working paper series 04 (cited on pages 11, 36).

— (2017a). Competitive Industrial Performance Report 2016. Volume I. Vienna: UNIDO (cited onpages 24, 28, 101–103).

— (2017b). “Industry 4.0 – Opportunities Behind the Challenge”. In: Partnering for impactachieving SDGs, Background paper for the UNIDO General Conference 17 (cited on page 21).

— (2018a). CIP 2018 Database. Available at:https://stat.unido.org/. Vienna (cited on pages 13, 15, 16, 31–33, 35, 36, 38–44, 46, 47, 49–63,65, 73, 74, 76, 78, 80, 82, 85, 88–91, 104–109).

— (2018b). Industrial Development Report 2016: The Role of Technology and Innovation inInclusive and Sustainable Industrial Development. Vienna: UNIDO (cited on pages 12, 36, 72,74).

— (2018c). International Yearbook of Industrial Statistics 2018. Cheltenham: Edward ElgarPublishing (cited on page 101).

— (2018d). SDG 2018 Database. Available at:https://stat.unido.org/. Vienna (cited on pages 21, 73, 74, 76, 78, 80, 82, 85, 88–91).

— (2019). International Yearbook of Industrial Statistics 2019I. Vienna: UNIDO (cited onpages 101, 102).

World Bank Group; China Development Bank (2017). Leapfrogging : The Key to Africa’s Develop-ment? Washington, DC: World Bank (cited on page 91).

World Economic Forum (2016). What makes South Asia the fastest growing region in the world?Available at:https://www.weforum.com/ (cited on page 55).

— (2017). The Global Competitiveness Report 2017-2018. Genevra: WEF (cited on page 10).



https://www.weforum.org/agenda/2016/09/what-makes-south-asia-the-fastest-growing-region-in-the-world

Index

A non-linear adjustment, 82Adjustment in ranks and scores, 84Appendix A, 97Appendix B, 100Appendix C, 104

By regions, 89

C.1 CIP Index 2018, detailed tables , 104Capacity to produce and export, 16Changes in emission efficiency and carbon foot-

print over time, 77Competitiveness and industrial development, 9Composition of the CIP Index, 15Concluding remarks, 95Countries’ progress in achieving SDG 9, 21Country classifications, 100, 103Country coverage, 27

Data sources and compilation, 27Dimension 1: Capacity to produce and export

manufactured goods, 61Dimension 2: Technological deepening and up-

grading, 63Dimension 3: World impact, 64

East Asia, 37Emission intensity , 72Europe, 41

General considerations, 69

International embeddedness, 12

Latin America, 45Linking emission efficiency and the carbon

footprint, 76

Main findings by development country group,34

Main findings by geographical region, 37Main findings by indicator, 61Measuring SDG progress, 18Methodology of the CO2-adjusted CIP Index,

98Methodology of the CIP Index, 97Middle East and North Africa, 48

North America, 51

Other Asia and Pacific, 60

SDG 9.2.1 Manufacturing value added as a pro-portion of GDP and per capita, 18

SDG 9.2.2 Manufacturing employment as aproportion of total employment, 18

SDG 9.3.1 Proportion of small-scale industriesin total industry value added, 19

SDG 9.3.2 Proportion of small-scale industrieswith a loan or line of credit, 20

SDG 9.4.1 CO2 emissions per unit of valueadded, 20

116 INDEX

SDG 9.B.1 Proportion of MHT industry valueadded in total value added, 20

South and South East Asia, 54Sub-Saharan Africa, 57

Technological absorption, 14Technological deepening and upgrading, 16The CIP index as analytical tool, 21The CIP ranking, 28The global distribution of production and emis-

sions, 72The manufacturing carbon footprint, 73

World impact, 17

competitive industrial performance report 2018

Documents