Structural transformation has been an important driving force of economic development over the last decades. According to classic economic theory, (such as Kuznets (1957), Chenery (1960) and Fourastié (1963)) development is driven by a shift from the extraction of raw materials and primary sector activities to increasingly complex technical transformation processes, commonly referred to as manufacturing. The sources of that transition include, on the supply side, the development of know-how, increase in high-skilled labour and technological advancement, enabling application of new production methods. On the demand side, the rising standard of living induces a shift from the consumption of food and other primary commodities towards consumer goods, that are usually manufactured. This transformation leads to higher value added and greater economic welfare. In line with this thinking, SDG target 9.2 promotes inclusive and sustainable industrialization and aims to significantly raise industry’s share of employment and GDP by 2030.
In recent years, we have seen a sectoral shift from manufacturing to services. Once a certain standard of living is reached, the demand for services increases relative to the demand for physically produced goods. According to Haraguchi and Rezonja (2010), this level is reached when GDP per capita amounts to around US$13,000 (in 2005 prices). At that stage, manufacturing usually accounts for around one fifth of value added. Based on these estimates, UNIDO (2017) considers countries industrialized when their manufacturing value added, adjusted to purchasing power parities, exceeds US$2,500 per capita.
Rapid industrialization in developing economies of Asia and Oceania
In 2017, manufacturing value added per capita amounted to US$6,167 (in 2010 prices) in developed economies (see figure 1). It was around twelve times as high as in developing Asia and Oceania (US$502) and almost six times the level of the developing economies of America (US$1,098) and transition economies (US$1,036). It exceeded the value in Africa (US$202) by 30 times.
Over the last 20 years, the region of developing Asia and Oceania has been steadily catching up – its manufacturing value added more than doubled between 1997 and 2017 – while in Africa and the developing economies of America it remained almost constant. In developed economies manufacturing value added per capita has not significantly increased either.
Notes: Logarithmic scale.
Changing structure of value added and employment
In addition to manufacturing value added per capita, the 2030 Agenda measures progress in industrialization by the share of manufacturing in total value added and employment. As these indicators show, over the last 20 years, structural change is evident in employment, but less so in value added. Most developing regions, as well as transition and developed economies, witnessed a decreasing proportion of manufacturing value added (see figure 2).
Africa and Asia and Oceania are the only regions, over the past 20 years, where developing economies have experienced higher growth in manufacturing than in total employment. As outlined above (figure 1), Asia and Oceania has recorded a substantial increase in the level of manufacturing value added per capita. This highlights a growing disparity in average productivity growth between it and Africa. Apparently, productivity in Asian and Oceanian developing economies has boosted not only within the manufacturing sector, but for the economy as a whole. In developed economies, the manufacturing value added per capita has been increasing slightly, accompanied by stagnating shares of manufacturing in value added and employment.1
Does structural transformation advance in LDCs?
Figure 1 shows that LDCs registered particularly low manufacturing value added per capita. In 2017, their manufacturing sector produced on average US$109 per head in constant prices, about half of the average produced in Africa. The manufacturing value added per capita has steadily increased in LDCs over the last 20 years, at a pace almost as fast as in developing Asia and Oceania. Nevertheless, LDCs’ value added remains below the levels, in per capita terms, produced by other groups of economies.
Figure 2 suggests that the increase in manufacturing value added per capita was strongly employment driven. The share of manufacturing in employment increased from 4.3 to 8.3 per cent between 1997 and 2017. By contrast, the manufacturing share in value added, which is in the focus of target 9.2 for LDCs, did not rise much. From 1997 to 2010, the share remained constant at slightly less than 11 per cent. Since then, it has been rising almost continuously, reaching 12. 8 per cent in 2017 (see figure 3).
Extrapolating the growth after 2005 into the future, the pace appears to be too slow to achieve the SDG target of doubling the manufacturing share in value added by 2030. Since 2005 onwards, an increase of 0.43 percentage points would have been required on average each year to reach the target. Increases of that amount were indeed achieved over the last three years. However, during the period from 2005 to 2017, the actual average increase was 0.17 percentage points per year. The findings above suggest that new innovations and policies towards industrialization are needed in LDCs to accelerate structural transformation.
Technology gap in manufacturing widening
In the process of economic development, structural transformation happens not only across broad economic sectors, such as primary production, manufacturing and services, but also at the more detailed industry level and within industries. Within manufacturing, we can observe diversification and a shift from low-productivity to high-productivity activities, raising the average value added per worker. Research and innovation play a crucial role in this transformation by providing the grounds for the use of new and more efficient technologies. The 2030 Agenda promotes technological development through research and innovation, especially in developing economies. Progress towards the achievement of that target is measured by the proportion of medium and high-tech industry value added over total manufacturing value added.
This indicator shows that, in the developed world, medium and high-tech industry accounts for higher shares of manufacturing value added than for developing and transition economies (see figure 4). When looking at weighted regional averages, around half of developed economies’ manufacturing output is obtained in medium and high-tech industries. Among developing countries, the weighted rate varies considerably across regions. In developing Asia and Oceania it is almost as high as in developed economies (50 per cent), while the rate reaches 32 per cent in developing America, but only 21 per cent in Africa. For transition economies the level is only slightly above that of Africa.
Over the last 15 years, the gap between developing and developed economies has widened slightly. While developed economies managed to increase the proportion of medium and high-tech manufacturing, from 48 per cent in 2001 to 50 per cent in 2016, the rate fell slightly for developing America (from 35 to 32 per cent) and in Africa (from 23 to 21 per cent). Only in the developing economies of Asia and Oceania has it remained constant, at around 43 or 44 per cent. Transition economies have experienced a relatively strong reduction: from 30 to 26 per cent. These figures suggest that developed countries have cemented their lead somewhat in the development and application of new technologies, while in developing and transition countries, except for Asia and Oceania, manufacturing is shifting further towards lower-technology sectors.
Figure 4 highlights the considerable variation across individual economies, especially in Asia. This region encompasses, on one hand, the two economies with the world’s most innovative manufacturing sectors, namely Singapore (78 per cent) and Taiwan, Province of China (69 per cent); on the other hand, it includes several countries, primarily LDCs and, in which the share of medium and high-tech industries in value added has persistently remained below three percent.
Notes: A violin plot illustrates the distribution of individual countries’ medium and high-tech industry shares in manufacturing value added within each country group and year. The coloured areas depict the distribution of countries’ rates smoothed by kernel density estimates. Kernel density estimation is a non-parametric way to estimate the probability density function of a random variable. It can be useful for visualizing the “shape” of data, as it estimates the probability of seeing an observation in each point. The wider the shape, the higher the possibility to find an observation in that location. The dots within the violin shapes represent the regional weighted average of countries’ medium and high-tech industry shares in manufacturing value added. Their location shows that typically larger economies have higher medium and high-tech industry shares.
Considerable spread in the medium and high-tech industry share of manufacturing value added is also found within the group of developed economies. Some developed countries reach less than one third of the rates recorded by countries such as Switzerland and Germany that belong to the top medium and high-tech developed countries.
Many LDCs and SIDS are characterized by low shares of medium and high-tech manufacturing. However, this is changing. Noteworthy development, among SIDS, include Trinidad and Tobago, where the medium and high-tech share in manufacturing value added increased from 29 per cent in 2001 to 40 per cent in 2016. In addition, in Barbados the rate has remained high, at 38 per cent, over the last 15 years (see UNIDO (2019)).
Convergence in medium and high-tech manufactured exports
Contrary to the changes observed in domestic productive activities, the share of medium and high-tech products exported by developing countries has been increasing in recent years, while it remained constant in the developed world (see figure 5). Developing economies in America and in Asia and Oceania reached a share of almost 60 per cent in 2016, three to five percentage points more than in 2005. Africa recorded an increase from 29 to 38 per cent over the same period of time. As a result, Africa has been catching up in the structural transformation of manufactured exports, and the gap between the developing and developed world has narrowed. However, transition economies are lagging behind. In 2001, they exported the same proportion of high-tech manufactured goods as Africa; but by 2016 their relative share had fallen behind.
Modest growth in R&D intensity across the world
In the 2030 Agenda, governments pledged to substantially increase public and private spending on research and development (R&D). This is an essential determinant of structural transformation and a shift to high-tech manufacturing, as described above. Since the turn of the millennium, the global gross expenditure on R&D has increased from US$730 billion to an estimated US$1.9 trillion in 2016, adjusted to(see table 1). Over 80 per cent of world R&D spending is taking place in the ten leading economies. In PPP-adjusted value terms, the leading countries are the United States (US$511 billion), China (US$451 billion), Japan (US$169 billion), and Germany (US$118 billion).
From 2010 to 2016, world R&D investment rose by an estimated 5.8 per cent annually, on average. China’s R&D expenditure has grown more rapidly than that of the other leaders over recent years: between 2010 and 2016 the expenditures increased in China by 12.8 per cent on average, annually. Since 2010, among the economies with high spending on R&D, particularly strong growth was observed in Turkey, Poland, Egypt and the United Arab Emirates.
|Investors||PPP US$ billions|| Annual average growth percentage|
|Percentage of GDP||Percentage of world total|
|Republic of Korea||79||6.1||4.2||4.1|
Global R&D investments have grown in absolute terms over the recent years. However, R&D intensity – one of the SDG indicators – saw only modest increases. In 2016, global gross expenditure on R&D stood at 1.7 per cent of GDP, marking a slight increase compared to 1.5 per cent observed in 2000 (see figure 6). Among countries, Israel (4.3 per cent) and the Republic of Korea (4.2 per cent) are the most prominent R&D investors relative to GDP, followed by Switzerland (3.4 per cent) and Sweden (3.3 per cent). The United States invested 2.7 per cent of its GDP in R&D. Only two developing economies, the Republic of Korea and China, reported R&D intensity above the world average (see table 2).
|Investors||GERD as a percentage of GDP|
|Republic of Korea||4.2|
|United Arab Emirates||1.0|
|China, Hong Kong SAR||0.8|
Notes: World total is an estimate.
Northern America is investing more of its GDP in R&D than any other region, but it is Eastern and South-Eastern Asia where the increase in R&D spending relative to GDP has been largest, growing from 1.5 per cent in 2000 to 2.1 per cent in 2016. Europe, on the other hand, has consistently been investing 1.8 per cent of its GDP in R&D since 2010, remaining well below the 3 per cent goal set by the European Commission, 2010; UNECA, 2018).. Among the EU member states, only Sweden and Austria are surpassing the three-per-cent target; Germany and Denmark are close, at 2.9 per cent. Regional blocks, like the and the EU, frequently promote R&D initiatives by endorsing assessable targets for member states (see
However, it remains difficult for countries to follow up on their R&D aspirations, owing to budgetary, socio-economic, and infrastructure constraints. The AU set the R&D intensity target for its member states at one per cent (UNECA, 2018). Available data indicate that three Sub-Saharan economies are close to that objective: Kenya, Senegal, and South Africa, with about 0.8 per cent of GDP each.
For Sub-Saharan Africa as a whole, R&D received an estimated 0.4 per cent of GDP, while Northern Africa and Western Asia invested around 0.8 per cent of GDP in R&D. The developing economies of America advanced from 0.5 per cent in 2000 to 0.7 per cent of GDP in 2016. The figure of 1.3 per cent, reported by Brazil in 2016, represents an R&D intensity twice as high as any other country in the region. According to UNESCO Institute for Statistics (2019), Oceania’s R&D stood at 1.8 per cent of GDP in 2016, dropping from a peak of 2.2 per cent observed between 2008 and 2011. In LDCs, an estimated 0.2 per cent of GDP were allocated to R&D.
Looking at the number of persons directly employed in R&D inper million inhabitants, which is measured by SDG indicator 9.5.2, the top rankings were dominated by European countries, in particular Denmark, Switzerland, Iceland, Sweden, and by Israel and the Republic of Korea. These economies reported over 9,000 persons per million employed on R&D. It should be noted that these figures included not only researchers, but also technical and supporting staff. According to data available for 90 economies, on average, 37 per cent of the persons employed in R&D were female, but with significant variation among economies.
With the spread of multinational enterprises and globalized production chains, R&D services are increasingly traded across countries. World R&D services exports expanded by an estimated 7.1 per cent annually between 2010 and 2016. Hence, when comparing to data from UNCTAD (2019), R&D exports growth clearly outpaced the total services trade, the latter scoring only a 2.6 per cent expansion in the same period. In 2016, world exports of R&D services amounted to US$150 billion. The top-ten R&D exporters accounted for 75 per cent of R&D exports, led by the United States and followed by Germany and France (see table 3). Seven out of ten leading R&D services exporters also belonged to the top-ten R&D services importers, as well as to world leading recipients (exporters) of charges for the use of intellectual property.
|Annual average growth of exports, percentage, |
|Ranking in GERD,
R&D is financed by public and private (mainly corporate) funds. According to OECD (2018)., public spending on R&D has decreased since 2010 in the OECD member states, not only as percentage of GDP, but also as a share of total government expenditure. With increased private R&D funding and wider use of competitive funding instruments, R&D gets steered towards more narrowly defined purposes, where advances are directly measurable, but tend to be incremental. Public and non-competitive funds are needed to support riskier, potentially transformative R&D projects and long-term undertakings where the expected results would benefit societies at large, especially in the social and environmental fields
- The share of services in value added increased in developed economies from 70.3 per cent in 1997 to 76.1 per cent in 2017 (UNCTAD, 2019).
- Chenery HB (1960). Patterns of industrial growth. American Economic Review. 50(4):624–654.
- European Commission (2010). Europe 2020: A Strategy for Smart, Sustainable and Inclusive Growth. European Commission, No. COM(2010) 2020.
- Fourastié J (1963). Le Grand Espoir Du XXe Siècle. Gallimard. Paris.
- Haraguchi N and Rezonja G (2010). Search of general patterns of manufacturing development. Development Policy and Strategic Research Branch Working Paper No. 02/2010. UNIDO.
- ILO (2019). ILOStat database. Available at https://www.ilo.org/ilostat (accessed 15 April 2019).
- Kuznets S (1957). Quantitative aspects of the economic growth of nations: II. Industrial distribution of national product and labor force. Economic Development and Cultural Change. 5(4):1–111.
- OECD (2015). Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development. The Measurement of Scientific, Technological and Innovation Activities. OECD Publishing. Paris.
- OECD (2018). OECD Science, Technology and Innovation Outlook 2018: Adapting to Technological and Societal Disruption.
- UNCTAD (2019). UNCTADStat. See https://unctadstat.unctad.org/ (accessed 10 April 2019).
- UNECA (2018). Towards achieving the African Union’s recommendation of expenditure of 1% of GDP on research and development. ECA Policy Brief. ECA/18/004. Addis Ababa.
- UNESCO Institute for Statistics (2019). UNESCO Institute for Statistics database. Available at http://uis.unesco.org/ (accessed 15 April 2019).
- UNIDO (2017). Industrial Development Report 2018, Demand for Manufacturing: Driving Inclusive and Sustainable Industrial Development. United Nations publication. Sales No. E.18.II.B.48. Vienna.
- UNIDO (2019). UNIDO databases. Available at http://stat.unido.org (accessed 4 May 2019).
- United Nations (2008). International Standard Industrial Classification of All Economic Activities (ISIC). Statistical papers / Department of International Economic and Social Affairs, Statistical Office of the United Nations Series M, No. 4, Rev. 4. United Nations publication. Sales No. E.08.XVII.25. New York, NY.
- United Nations et al. (2012). Manual on Statistics of International Trade in Services 2010. United Nations publication. Sales No. E.10.XVII.14. Geneva.
- United Nations (2019). SDG indicators: Metadata repository. Available at https://unstats.un.org/ (accessed 14 May 2019). /metadata/
- United Nations, European Commission, , OECD and World Bank (2009). System of National Accounts 2008. Studies in methods, No. 2. United Nations publication. Sales No. E.08.XVII.29. New York.
Click on a star to rate this page!
Average rating / 5. Vote count
We are sorry that this post was not useful for you!
Let us improve this post!
Thanks for your feedback!