Chapter 1. SME structure and business dynamism: Trends and performance in productivity and wages

This chapter provides an overview of trends in the small and medium-sized enterprises (SME) sector and in business dynamism, and offers insights, where data allow, on cross-country and cross-sectoral differences. While the general structure of the SME population in OECD countries has remained stable in the past decade, this chapter shows that dynamic changes are occurring in activities highly exposed to digitalisation, or able to capitalise on it. The chapter highlights that in recent years the majority of new enterprise entries and the resulting job creation occurred in sectors with below average productivity levels. It presents evidence that more jobs in lower-productivity activities translated into more lower-paid jobs, weighing down on material well-being. The chapter also highlights that, apart for exceptions in the services sector, productivity gaps are observed between SMEs and large firms that translate in lower pay in SMEs. The findings reveal that gaps in productivity and wages are smaller for SMEs that export, and that global value chains provide scope for technology and knowledge spill-overs but also increase competition. The chapter demonstrates that current official statistics are able to provide important insights, in particular with respect to structural heterogeneity, but also illustrates the importance of continuing to expand the statistical boundary, not least to tackle emerging issues.

    

The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

Highlights
  • Although the general structure of the SME population in OECD countries has remained stable in recent years, dynamic changes are occurring in activities that are highly exposed to, or able to capitalise on, digital transformation.

  • In the services sector, SMEs, and in particular medium-sized firms, outperform large enterprises in terms of productivity levels in many countries. This can be observed especially in wholesale and retail trade across almost all OECD countries but also in more knowledge-intensive services sectors, like professional, scientific and technical activities, with the notable examples of Austria, Belgium, Germany or Poland. However, in manufacturing activities characterised by capital-intensive production processes, large firms show consistently higher levels of productivity than SMEs.

  • Digitalisation provides scope for SME growth but the pace of adoption varies across countries and sectors. However, the take up of digital tools is to a large extent still confined to basic services, while in many countries there are large adoption gaps compared with large firms in cloud computing services, which would provide scope for cost savings compared to the fixed costs of ICT investment.

  • Between 2010 and 2016, in many economies, most new entries and job creation took place in sectors with below average productivity levels: for instance, accommodation and food services in Greece, Ireland and the United Kingdom; the construction sector in Italy and Norway; and wholesale and retail trade in most countries.

  • More lower-productivity jobs has resulted in more lower-paid jobs, weighing down on material well-being. Between 2010 and 2016, close to 90% of all new jobs in France were created in activities with below average wages, close to two-thirds in Germany and the United Kingdom, and over three-quarters in the United States.

  • Lower productivity in SMEs also translates into lower pay. SMEs typically pay employees around 20% less than in large firms, even in large SMEs. But gaps are smaller for exporting SMEs.

  • Global value chains provide new avenues to tap into global markets both directly through exports of intermediates to larger downstream firms but also indirectly as upstream suppliers to larger direct exporters, providing scope for technology and knowledge spill-overs.

  • But global value chains also increase competition, in particular in labour intensive activities in OECD countries, from low-wage economies. In sectors such as textiles for example, increased competition has had a profound impact on SME producers, forcing many to close but also acting as a stimulus to upgrade and move up the value chain.

  • Understanding the direction and potential impact of these mega-trends is essential in developing sound policies, and stresses the importance of continuing to develop stronger statistical capacities to measure, identify, and highlight SME heterogeneity.

Introduction

As the predominant form of business and a major employer in the business economy, small and medium-sized enterprises (SMEs) are key actors for economic resilience, productivity and inclusiveness. The OECD work on the productivity-inclusiveness nexus documents that SMEs are central to the collective goal of increasing productive potential, reducing inequality and ensuring that the benefits from increased globalisation and technological progress are enhanced and shared (OECD, 2017[1]; Blanchenay, Criscuolo and Calvino, 2016[2]). The OECD Declaration on Strengthening SMEs and Entrepreneurship for Productivity and Inclusive Growth1 recognises that strengthening SMEs and entrepreneurship policies is key to achieving more inclusive societies and growth.

Major transformations in the economy and society, such as increased globalisation, digitalisation, the next production revolution, the changing nature of work, demographic changes, the circular economy and the transition to a low-carbon economy, have far-reaching consequences for productivity and income distribution, including through their impact on SMEs and entrepreneurship. Against this backdrop governments are seeking innovative solutions to harness opportunities and address risks that are emerging, and, SMEs and entrepreneurs have an important role to play.

However, in looking at the impact of these mega-trends on businesses and on the types of policies needed to navigate them, it is important to recognise that a one-cap fits all approach may be too blunt an instrument. The SME population is composed of very diverse businesses, in terms of age, size, ownership, business models, and entrepreneurs’ profiles, motivations and aspirations. It is increasingly becoming clear that firm heterogeneity matters in thinking about innovation, productivity, job creation and income, but it matters equally when thinking about the response and adaptation of economies to mega-trends.

This chapter aims to shed light on these issues using currently available data while also flagging up areas where improved statistics are needed (see Box 1.1). It provides detailed empirical evidence on the composition and economic contributions of SMEs, and on the changes that have occurred in SME structure and performance over the last decade. Where data allow, it also offers insights on cross-country and cross-sectoral differences in SME performance and business dynamics, and insights into how digitalisation and globalisation are impacting on SME performance. In exploring these issues, the chapter reinforces the need to focus on heterogeneity in analysing SMEs and in developing policies around them, while also highlighting the importance for improved data to better capture firm heterogeneity (Annex 1.A).

Given the breadth of issues influencing the structure and performance of SMEs, following guidance by the Working Party on SMEs and Entrepreneurship (WPSMEE), this first edition of the SMEE Outlook focuses in particular on trends in SME productivity and wage performance. The chapter sets the analysis against the backdrop of major engines of structural change, namely, digitalisation and globalisation.2 The intention is for future editions of the Outlook to continue to focus on topical and priority issues, in consultation with the WPSMEE.

Box 1.1. Definition of SMEs

One particular challenge concerning the comparison of SMEs across countries is the definition of SMEs. Definitions in a national context can vary significantly across countries, and also within countries across policy domains, with smaller countries typically setting lower thresholds than larger countries. This creates challenges for international statistics on SMEs. In this publication, unless otherwise stated, the approach is to consider SMEs all enterprises with less than 250 persons employed. This provides a robust framework to compare statistics across countries but caveats are needed in interpretation. For example care is needed in drawing conclusions with analyses of statistics on firms with less than 250 persons employed and the impact of policies targeting a policy driven definition of SMEs, which may differ. Moreover, although the definition of an enterprise as the main unit in the country where decisions are made, risks are taken and reward accrue, for foreign owned enterprises (foreign affiliates) in particular, this may not be the case. Efforts to separately identify independent and dependent SMEs are being made in some countries but greater up-take is needed in many others.

The structure of the SME population

SMEs account for a significant proportion of employment and GDP…

Across most OECD economies, SMEs account for nearly all firms (over 99%), provide over half of all business sector employment, employing on average five workers, and generate over half of all business sector GDP, creating on average around 270 thousand USD of value-added per firm (Figure 1.1). SME shares tend to be smaller the larger the economy – reflecting economies of scale, and the greater scope for operating larger firms in larger economies.

Figure 1.1. SMEs and their contribution to the business economy
Percentage 2016, or latest available year
Figure 1.1. SMEs and their contribution to the business economy

Source: (OECD, 2018[3]), Highlights of Entrepreneurship at a Glance 2018, OECD, www.oecd.org/sdd/business-stats/EAG-2018-Highlights.pdf.OECD.

 StatLink https://doi.org/10.1787/888933923583

However, despite these “framework” differences, there is broad comparability in the structure of SMEs across countries, i.e. in their sectoral distribution and economic weight in broad activities.

SME employment is concentrated in specific services sectors, notably wholesale and retail trade, and construction, where SMEs account for high shares of total jobs, reflecting relatively low requirements to operate, in terms of skills, capital and financing. In 2016, the wholesale and retail trade and construction sectors accounted, respectively, for between 19% and 30%3 and between 7% and 18% of all employment in SMEs across OECD economies (Figure 1.2); equivalent to between half to 100% of all jobs in those sectors (see Annex Figure 1.B.2).

Conversely, across OECD economies, SMEs operating in manufacturing sectors -which are typically more capital-intensive than services - account for a lower share of employment and, in particular, value added. In 2016, manufacturing sectors accounted for between 12% and 29% of all employment in SMEs.

…but with cross-country differences in specific activities

Despite the broad similarities across countries, significant differences in SME participation emerge in some sectors, including at the broad industry level. In the United Kingdom for example, the share of SMEs operating in professional and scientific activities in 2016 was three times that of Korea, while in Poland very few SMEs operated in the hotels and accommodation sectors.

As noted above, this can reflect a number of “framework” differences that drive specialisation, as well the size of the economy, and in relation, outsourcing strategies driven by lead (typically large) firms in value chains, which may be outward looking (including through investment abroad, generating or sustaining SMEs in other countries that displace domestic SMEs) or through the development of domestic clusters.

In Germany for example, the vehicle and transport equipment sector has created strong upstream supply chains through outward investment in neighbouring, typically lower-wage economies, such as Poland and the Czech Republic, and SMEs’ share of employment in the sector in those countries is around twice – 20% - that in Germany. In Korea however, where there have been targeted actions to promote linkages between domestic SMEs and larger Chaebols, the SME share of employment in the vehicle and transport sector is significantly higher (57.8%), despite the fact that the foreign content of exports in the sector in both Germany and Korea is around one-third (source: OECD Trade in Value-Added database).

Structures have remained broadly stable over time

While the crisis had a significant impact on employment, affecting some sectors more than others, there have not been significant structural changes in the contributions of SME across broad industry groupings.

Between 2010 and 2016 for example, across OECD countries, the top five sectors, accounting for around 60% of all SME employment, were unchanged (Figure 1.2): i) wholesale and retail trade, accounting for one in four of all persons employed in SMEs; ii) manufacturing; iii) construction; iv) accommodation and food services; and v) professional, scientific and technical activities.

In addition, over the same period, the top five sectors where SMEs accounted for more than 80% of employment in the sector also remained the same: i) advertising, market research, other professional, scientific and technical activities; ii) real estate activities; iii) construction; iv) accommodation and food services; and v) legal, accounting and management services (see Annex Figure 1.B.2).

Figure 1.2. Broad specialisation of SMEs has changed little in recent years
Percentage, SME total employment=100
Figure 1.2. Broad specialisation of SMEs has changed little in recent years

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923792

But changes are occurring at the sub-sector level

While the broad structure of the SME population has changed relatively little in recent years, more granular data at the sub-sector level point to important dynamic changes, particularly in those sub-sectors, such as the ICT sector, that are highly exposed or able to capitalise on the ongoing digital transformation, including through access to cheaper ICT goods and the ability to scale without mass.

Across OECD countries, the average employment share of SMEs in the ICT sector grew from 3.8% to 4.7% over the period spanning 2010-16, driven in large part by new entries, with SME shares of value-added increasing in nearly all OECD countries (Figure 1.3).

Figure 1.3. Share of SMEs in value added
Percentage of total sector value added
Figure 1.3. Share of SMEs in value added

Note: Information and communication services cover two-digit sectors 58 to 63 of ISIC Rev. 4.

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923944

A more detailed breakdown of the ICT sector reveals additional insights. In many countries, for example, SMEs saw substantial increases in their share of overall value-added in publishing activities (58), and also in telecommunications (61) (Figure 1.4).

In the publishing sector, which includes activities such as newspapers, increases generally took place against a backdrop of an overall contraction in the industry. The overall contraction in the publishing sector, which is partly explained by the appearance of new forms of media content and media providers in other sectors, appears to have had a mixed impact on SMEs across countries, possibly depending on the ability of many SMEs to occupy niche spaces in the sector; albeit, perhaps, with limited growth potential. In Italy for example, although value-added in publishing activities declined by over 10 percent between 2010 and 2016, SMEs, which make up 57% of total value added, increased their share by over 10 percentage points.

In telecommunications on the other hand, despite the strong association with only a handful of key (and large) firms in the sector, SMEs have increased their share of value added in many countries, in an expanding industry, which appears to reflect upstream specialised services to larger firms but also other entry routes, such as purchasing access and network capacity from larger operators (i.e. entries of new smaller scale niche providers).

On the other hand, SMEs appeared to underperform relative to larger firms in the growing information service activities (63), with the exception of France, Denmark and the United Kingdom, where SMEs increased their share of value added of the sector.

Figure 1.4. Digitalisation provides scope for SME growth but the pace of adoption varies across countries and sectors
Share of SMEs value added and total sector value added growth, percentage
Figure 1.4. Digitalisation provides scope for SME growth but the pace of adoption varies across countries and sectors

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923963

Also, in spite of declining costs of ICT capital, many countries have experienced a declining economic weight of SMEs in computer programming activities (62), a sector which is expanding overall. The decline was particularly evident in Hungary, Poland, and the Slovak Republic.

Business dynamics and SMEs

New entries of SMEs have been an important driver of employment growth in the services sector in most countries but size still matters in manufacturing

Although there has been relatively limited change in the structure of SMEs across most economies, SMEs were a significant driver of overall employment growth in the market services sector between 2010 and 2016, mainly reflecting new entries (Figure 1.5 and Annex Table 1.C.1) and possibly a rebound effect as the impact of the crisis began to wane.

This, in part, reflects SMEs’ relative share of overall employment in the services sector, with typically smaller contributions the larger the economy, but even in larger economies, such as France, Germany and the United Kingdom, SMEs were important drivers of job growth.

In the United States, on the other hand, nearly all employment growth in services was driven by an increase in the number of larger enterprises (including by SMEs that ‘graduated’ to large), with the average size of SMEs also increasing, whereas the average size of SMEs decreased in France and the United Kingdom. In Italy, on the other hand, the smaller average size of SMEs and the decline in the number of SMEs weighed negatively on job creation in the sector.

SME performance has been more muted in manufacturing, with larger firms driving employment growth in nearly all countries where manufacturing employment grew between 2010 and 2016. And in those countries where manufacturing employment contracted, such as in Spain, Italy and Greece, this was almost entirely through SME closures.

Figure 1.5. In most countries SMEs have been the primary drivers of employment growth in the service sector
Employment growth by sector and size class, contributions and percentage change between 2010 and 2016
Figure 1.5. In most countries SMEs have been the primary drivers of employment growth in the service sector

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923982

Figure 1.6. Enterprise creations are back to pre-crisis levels
New enterprise creations, annual levels, index 2012=100
Figure 1.6. Enterprise creations are back to pre-crisis levels

Note: For all countries data refer to creations of corporations or of other legal forms of employer enterprises.

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933924001

Despite the general positive trends in the number of firm entries, some care is needed in interpretation as birth rates (i.e. the ratio of enterprise creations over the total stock of enterprises) remain below pre-crisis rates in many countries, even if recent trends in birth rates are pointing upwards (such as in the UK) or showing signs of stabilisation (such as in the US) (Figure 1.7) (see also Annex 1.C).

Of particular relevance to conclusions that can be drawn on underlying business dynamism is the fact that in many of the countries where birth rates improved, this was associated with a smaller average size of entries (OECD, 2017[4]), meaning that changes in the share of jobs created by enterprise births, as a percentage of total employment, has generally lagged behind changes in birth rates.

Smaller average sizes may reflect productivity gains and the ability to capitalise on new labour saving (and digital) technologies, but it may also reflect other factors not necessarily conducive to longer-term productivity growth, for example, if higher firm creations are driven by push factors such as fiscal tightening and lower levels of social security payments, rather than by pull factors e.g. business opportunities. Of note in this respect are the substantial differences across countries. For example, although job creation by enterprise births represented on average 4% of total employment, job creation rates ranged from as high as 7-10% in Turkey to as low as 1% in Ireland in 2014 and 2015 (Figure 1.7).

Figure 1.7. Enterprise birth rates have risen in many countries but job creation rate remains low in some
Birth of and job creation by employer enterprises, percentage rates, business economy
Figure 1.7. Enterprise birth rates have risen in many countries but job creation rate remains low in some

Note: Birth rate is measured as the ratio of employer enterprise births over the population of active employer enterprises. Job creation rate is measured as the ratio of jobs created by employer enterprise births over the number of jobs in all active employer enterprises.

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933924020

SMEs and productivity

Many new SME jobs are in low productivity activities

Although there are signs emerging that business entries have begun to pick up in recent years, new employment is not universally created in high-productivity high-growth sectors. A higher pace of entries in lower-productivity activities can therefore work to weigh down on productivity levels and indeed on productivity growth observed across developed economies in recent decades (OECD, 2018[5]).

Across major economies, on average, between 2010 and 2016, increases in employment in activities with below average labour productivity were about two to four times higher than in those with above average labour productivity. In the United States, for example, 9.7 million more jobs in activities with below average labour productivity levels existed in 2016 compared to 2010, which is over four times the additional 2.4 million jobs in above average labour productivity activities. Comparable figures in other major economies were: 0.5 million and 0.2 million in Canada, 0.4 million and 0.2 million in France, 1.5 million and 0.6 million in Germany; minus 0.02 million and minus 0.2 million in Italy; and 1.9 and 0.6 million in the United Kingdom.

In nearly all major OECD economies, the top three sectors generating the largest net employment gains over the period 2010 to 2016 had below average labour productivity, with restaurants, health and residential care activities featuring highly in most economies (Table 1.1, Panel A). Only France saw a sector, namely legal, accountancy, management consultancy, with above average labour productivity in the top three sectors. On the other hand, in sectors that had lost most (net) jobs over the same period, most major economies had at least one above average labour productivity sector in the top three; all three in the case of the United States (Table 1.1, Panel B).

Table 1.1. Net employment creation and destruction between 2010 and 2016 (or latest available year)

 

ISIC Rev.4 code

Activity label

Net employment creation

Labour productivity level of the sector

Panel A. Three sectors with largest net employment creation, G7 countries, thousands of persons

CAN

G47

Retail trade, except of motor vehicles and motorcycles

141

Below average labour productivity

I56

Food and beverage service activities

64

Below average labour productivity

P85

Education

50

Below average labour productivity

FRA

Q_87_88

Residential care activities; social work activities without accommodation

128

Below average labour productivity

Q86

Human health activities

114

Below average labour productivity

M_69_70

Legal and account activities; activities of head offices; management consultancy activities

94

Above average labour productivity

DEU

Q86

Human health activities

357

Below average labour productivity

Q_87_88

Residential care activities; social work activities without accommodation

306

Below average labour productivity

N_80_82

Security and investigation activities; services to buildings and landscape activities; office administrative, office support and other business support activities

189

Below average labour productivity

ITA

I_55_56

Accommodation and food service activities

214

Below average labour productivity

T_97_98

Activities of households as employers; undifferentiated goods- and services- producing activities of private households for own use

135

Below average labour productivity

Q_87_88

Residential care activities; social work activities without accommodation

86

Below average labour productivity

GBR

I_55_56

Accommodation and food services activities

334

Below average labour productivity

N_80_82

Security and investigation activities; services to buildings and landscape activities; office administrative, office support and other business support activities

292

Below average labour productivity

M_69_70

Legal and account activities; activities of head offices; management consultancy activities

249

Below average labour productivity

USA

Q86

Human health activities

1 457

Below average labour productivity

F_41_42_43

Construction

1 251

Below average labour productivity

I56

Food and beverage service activities

1 214

Below average labour productivity

Panel B. Three sectors with largest net employment destruction, G7 countries, thousands of persons

CAN

N80

Security and investigation activities

-8

Below average labour productivity

O84

Public administration and defence; compulsory social security

-12

Above average labour productivity

N82

Office administrative, office support and other business support activities

-15

Below average labour productivity

FRA

G45

Wholesale and retail trade and repair of motor vehicles and motorcycles

-40

Below average labour productivity

T_97_98

Activities of households as employers; undifferentiated goods- and services- producing activities of private households for own use

-42

Below average labour productivity

F_41_42_43

Construction

-76

Below average labour productivity

DEU

J58

Publishing activities

-43

Below average labour productivity

S96

Other personal service activities

-44

Below average labour productivity

O84

Public administration and defence; compulsory social security

-180

Below average labour productivity

ITA

A01

Crop and animal production, hunting and related service activities

-66

Below average labour productivity

O84

Public administration and defence; compulsory social security

-120

Above average labour productivity

F

Construction

-403

Below average labour productivity

GBR

C18

Printing and reproduction of recorded media

-27

Below average labour productivity

K64

Financial service activities, except insurance and pension funding

-46

Above average labour productivity

O84

Public administration and defence; compulsory social security

-260

Below average labour productivity

USA

G46

Wholesale trade, except of motor vehicles and motorcycles

-164

Above average labour productivity

J60

Programming and broadcasting activities

-173

Above average labour productivity

O84

Public administration and defence; compulsory social security

-296

Above average labour productivity

Note: Average labour productivity is measured as gross value added per person employed in the total economy.

Source: (OECD, 2018[5]), OECD Compendium of Productivity Indicators 2018, OECD Publishing, Paris, https://doi.org/10.1787/pdtvy-2018-en.

 StatLink http://dx.doi.org/10.1787/888933924229

As noted above, a significant part of overall job creation has been through new entries, and in many economies these have been in sectors with below average productivity levels (see Figure 1.8 and Annex Figure 1.C.3): for instance: accommodation and food services in Greece, Ireland and the United Kingdom; construction in Italy and Norway; and wholesale and retail trade in most countries.4

Figure 1.8. Newly created jobs are mainly in lower productivity sectors
Percentage of jobs created by births in sectors with above-median productivity, as a share of all employment created by employer enterprise births
Figure 1.8. Newly created jobs are mainly in lower productivity sectors

Note: Median productivity (value added per person employed) was calculated at the sectoral level (in ISIC REV.4: section-level and a selection of division-level manufacturing sectors) separately for each country and year.

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933924039

…which weighs down on wages

Labour compensation levels correlate highly with labour productivity levels; hence, having more jobs in lower labour productivity activities has resulted in more jobs with below average wages in most economies, working to weigh down on average salaries in the economy as a whole. Between 2010 and 2016, for example, close to 90% of all new jobs in France were created in activities with below average wages; close to two-thirds in Germany and the United Kingdom and over three-quarters in the United States (Table 1.2).

Table 1.2. Change in employment over the period 2010-2016, or latest available year
Thousands of persons

Country

Below average labour compensation and below average labour productivity in 2010

Below average labour compensation and above average labour productivity in 2010

Above average labour compensation and below average labour productivity in 2010

Above average labour compensation and above average labour productivity in 2010

CAN

228

-7

245

197

FRA

479

-2

-101

169

DEU

1 247

21

157

624

ITA

-99

1

5

-175

GBR

1 498

72

414

515

USA

8 752

626

1 039

1 785

Note: Data for Canada refer to 2010-2013. Data for France, Germany and Italy refer to 2010-2015. At the time of writing the OECD Compendium of Productivity Indicators, data for the period 2010-2016 in Italy were available only at the whole economy level and show net employment creation equal to 56 thousand of persons.

Source: (OECD, 2018[5]), OECD Compendium of Productivity Indicators 2018, OECD Publishing, Paris, https://doi.org/10.1787/pdtvy-2018-en.

 StatLink http://dx.doi.org/10.1787/888933924248

Growth in real wages, adjusted for inflation using the consumer price index, has also been lagging behind labour productivity growth in many countries (Figure 1.9) (OECD, 2018[5]; Schwellnus, Kappeler and Pionnier, 2017[6]). Indeed, real labour compensation per hour worked, adjusted for the CPI (which provides for a better measure of real purchasing power from a household perspective, compared to the GDP deflator), declined between 2010 and 2016 in Portugal, Spain and the United Kingdom. However, in some countries, such as Germany and the United States, real labour compensation has begun to rise in line with (albeit slow) labour productivity growth in recent years, helping to reverse pre-crisis decoupling.

Figure 1.9. Real wages have lagged productivity growth in many countries
GVA per hour worked and average hourly employee labour compensation, percentage change at annual rate (2010-2016), total economy
Figure 1.9. Real wages have lagged productivity growth in many countries

Source: (OECD, 2018[5]), OECD Compendium of Productivity Indicators 2018, OECD Publishing, Paris, https://doi.org/10.1787/pdtvy-2018-en.

 StatLink http://dx.doi.org/10.1787/888933924058

Decoupling of wages and productivity can exacerbate inequalities

Even in countries where there is only limited decoupling of wages from labour productivity growth at the whole economy level, this can mask significant divergences within sectors (OECD, 2018[5]). In France, for example, where there has been limited decoupling at the whole economy level, out of 63 sectors, (according to ISIC Rev. 4 classification), 41 saw a decoupling in the post-crisis period; with the largest decoupling occurring in the water transport services and telecommunication services sectors. Similarly in the United Kingdom and Italy, over half of all sectors saw real average labour compensation grow at a slower pace than labour productivity, with the largest gaps in the fishing and aquaculture and education sectors in the United Kingdom and basic metals and non-metallic mineral products in Italy. In the United States and Germany however, mirroring the improvement seen at the whole economy level, most sectors saw real wage growth outpacing productivity growth (40 out of 58 in the United States and 37 out of 63 in Germany). Across all major economies, sectors that saw the highest net increase in employment gains also saw wages outpacing, or keeping pace with, labour productivity growth.

Many of the sectors where wage growth has lagged productivity have relatively high shares of SMEs. With average wages in SMEs typically, considerably lower than average wages in larger firms (Figure 1.10), and in some countries such as Mexico, significantly so, decoupling could exacerbate existing income inequalities between employees in SMEs and larger firms.

Figure 1.11 provides a more granular analysis of the gap between SMEs and large firms in terms of productivity and compensation per employee, highlighting, in turn, that, in some countries and sectors SMEs can, in fact, have higher labour productivity than larger firms. However, even in sectors where SMEs have on average higher labour productivity than larger firms they almost always pay on average lower salaries. In the Austrian chemicals sector for example, SMEs have over one third higher labour productivity than larger firms, but 20% lower salaries.

Of note is that within sectors, gaps between SMEs and larger firms vary considerably across countries, revealing real potential for improved productivity performance in countries with larger gaps.

Figure 1.10. Labour productivity and compensation per employee by enterprise size, manufacturing
Value added per person employed (VAPE) and compensation per employee (COE), index 250+ = 100, 2016, or latest available year
Figure 1.10. Labour productivity and compensation per employee by enterprise size, manufacturing

Source: OECD Structural and Demographic Business Statistics Database, 2018 (http://dx.doi.org/10.1787/sdbs-data-en).

 StatLink http://dx.doi.org/10.1787/888933923602

Figure 1.11. SME labour productivity and compensation per employee, by industry sector
Value added per person employed and compensation per employee, index 250+ = 100, 2016, or latest available year
Figure 1.11. SME labour productivity and compensation per employee, by industry sectorFigure 1.11. SME labour productivity and compensation per employee, by industry sector

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923621

…and productivity gaps between large and small firms are growing…

Since 2008, many countries have seen an increase in productivity gaps between SMEs and large companies (Figure 1.12), which could reflect also increased market concentration. The growth in gaps was particularly notable in Switzerland, the Netherlands, Italy and Turkey.

In manufacturing activities characterised by capital-intensive production processes, large firms show consistently higher levels of productivity than SMEs. In the United Kingdom and Germany gaps have increased slightly in the last decade but in other large European economies such as Spain and Poland they have decreased, in particular, between medium-sized firms and large enterprises (Figure 1.12, Panel A). Interestingly in the United States, between 2007 and 2012 (the period for which data are available), the relative labour productivity of manufacturing SMEs showed a marginal increase, suggesting that if productivity spill-overs have stalled, the impact on the manufacturing sector (where concerns about concentration have been less prevalent) may have been at most limited.

…but specialised high-skilled SMEs can outperform larger firms

In professional, scientific and technical activities, however, where SME entries have been relatively high in many economies (which include the activities of advertising agencies and consulting companies, including legal services, architectural services, etc.,) SMEs can be as productive as, or indeed more than, large firms, with micro firms in France, Sweden, and the United Kingdom (Figure 1.12, Panel B) performing as well as large firms.

Although most countries, whatever the sector, saw their micro SMEs perform significantly behind larger companies, the experience of France, Sweden and the United Kingdom in professional activities suggests that significant productivity gains could be made.

Figure 1.12. Labour productivity of SMEs
Index, productivity of large enterprises =100
Figure 1.12. Labour productivity of SMEs

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923640

SME participation in global markets

SME activity in international trade can help reduce the wage gap with larger firms

In recent years, there have been growing concerns that the benefits from globalisation have not been spread evenly within economies, possibly exacerbating long-standing wage gaps between large and smaller firms (OECD, 2017[7]). Average compensation per worker across OECD economies is considerably smaller, the smaller the firm size, with remuneration levels, even in large SMEs, around 20% lower than in large firms. This reflects, in large part, correspondingly lower productivity, but the level of direct exports by SMEs also appears to play a role. In countries where SMEs have a relatively high share of exports for example, differences in average salaries between SMEs and larger firms are smaller (Figure 1.13).

Figure 1.13. Manufacturing wage gaps and SME trade, 2014
Log trend
Figure 1.13. Manufacturing wage gaps and SME trade, 2014

Notes: Current prices, US Dollars - converted using period average exchange rate.

SME wage gaps are measured as the difference between average salaries per employee in large firms and SMEs as a ratio of average salaries in SMEs. Export propensity of SMEs is measured as the share of exports by SMEs divided by the share of output by SMEs.

Sources: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en and Trade by Enterprise Characteristics Database, 2018, https://doi.org/10.1787/eeefdd40-en.

 StatLink http://dx.doi.org/10.1787/888933923659

Relative to their share of overall activity and employment, SMEs account for only a small proportion of exports. As noted above, in most OECD economies SMEs account for 99% of all firms, around two-thirds of total employment and over half of business sector value-added. The SME contribution to overall exports is for most countries similar to their contribution to value added (Figure 1.14).

Figure 1.14. SME export activity, value added and employment shares
As a percentage of exports (value added, business employment), total business economy, 2015 or latest available year
Figure 1.14. SME export activity, value added and employment shares

Sources: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en and Trade by Enterprise Characteristics Database, 2018, https://doi.org/10.1787/eeefdd40-en.

 StatLink http://dx.doi.org/10.1787/888933923678

The relatively low contribution of SMEs to overall exports reflects their lower contribution in particular to mining and manufacturing (industry), where economies of scale play a role. Indeed, the share of industrial SMEs engaged in exports is notably lower than the corresponding share for large firms. In most economies, for example, the vast majority (when not the totality) of large industrial firms export, whereas only between 5%-40% of SMEs do so (Figure 1.15).

Figure 1.15. Industrial firms engaged in exports, 2016 or earliest available year
As a percentage of total firms by size class
Figure 1.15. Industrial firms engaged in exports, 2016 or earliest available year

Sources: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en and Trade by Enterprise Characteristics Database, 2018, https://doi.org/10.1787/eeefdd40-en.

 StatLink http://dx.doi.org/10.1787/888933923697

…and SME participation in global value chains offers access to foreign markets and new sources of growth

Evidence suggests that, in OECD countries, looking only at direct exports by SMEs under-represents the actual engagement of small firms in a country’s gross exports. When the role of SMEs as suppliers of inputs to larger direct exporters is taken into account, the importance of SMEs as exporters increases considerably. In the Slovak Republic, for example, SMEs account for 34% of gross exports, but for 56% of the total value added in the country’s exports (Figure 1.16).

Figure 1.16. Direct and indirect exporting activity of SMEs in OECD countries
As a percentage of gross exports, 2014
Figure 1.16. Direct and indirect exporting activity of SMEs in OECD countries

Source: OECD (2018), “Accounting for firm heterogeneity in global value chains: The role of Small and Medium sized Enterprises”, OECD Working Party on International Trade in Goods and Trade in Services Statistics.

 StatLink http://dx.doi.org/10.1787/888933923716

The significance of indirect channels is especially important for independent SMEs (i.e. those not owned by a larger domestic firm or foreign firm). For example, while only 4% of total value added generated by independent micro SMEs in Norway is exported directly, an additional 23% of their value added is indirectly embodied in exports by other firms (Figure 1.17).

Indirect exports by SMEs are particularly large in sectors where GVCs are important and where scale matters. In the transport equipment sector, for example, SMEs accounted for over 40% of total US value-added exported, with nearly all of that contribution reflecting upstream component and services suppliers to the transport equipment industry (OECD, 2017[7]). This indirect mode of internationalisation provides SMEs access to foreign markets and new sources of growth, but without incurring trade related costs.

SMEs may also benefit from GVCs on the input side (Lopez Gonzalez, 2016[8]) and (López González and Jouanjean, 2017[9]). Recent studies have found evidence that firms which use more imported goods and services are more productive and better able to face the costs of exporting (Bas and Strauss-Kahn, 2015[10]) and (Bas and Strauss-Kahn, 2014[11]). SMEs, including non-exporters, can increase their productivity by drawing on cheaper and more sophisticated imports; by exploiting new technologies embodied in new and cheaper capital products; as well as through improved access to new technologies from engagement with internationally-oriented firms, including through linkages arising from foreign investment. All of these channels can also help to target specialisation in parts of the value-chain, where SMEs have comparative advantages and can in turn foster upgrading.

Benefits from GVC participation, including in terms of productivity growth, depend on the position of the firm within global production networks and the nature of inter-firm linkages. Firms and industries positioned at the centre of complex production networks have access to a greater variety of foreign inputs, and potentially a broader range of technologies, compared to those at the periphery. Smaller firms display faster productivity growth in those sectors that have become more central to global production, from those on the periphery, and also in sectors with stronger linkages to more productive foreign buyers/ suppliers (Criscuolo and Timmis, 2018[12]).

Figure 1.17. Direct and indirect exporting activity of SMEs in Nordic countries
As a percentage of total value-added, 2013
Figure 1.17. Direct and indirect exporting activity of SMEs in Nordic countries

Source: OECD (2017), Nordic Countries in Global Value Chains, report by the OECD Statistics Directorate and the National Statistical Offices in the Nordic countries, https://www.dst.dk/Site/Dst/Udgivelser/GetPubFile.aspx?id=28140&sid=nordglobchains.

 StatLink http://dx.doi.org/10.1787/888933923735

Closer global integration also has implications for firms that operate in local markets, through increased competition, which can have disruptive effects on local economies and requires enhanced market knowledge and competitiveness by small businesses.

GVCs amplify the importance of goods and services trade policies. Trade and investment openness, trade facilitation, intellectual property protection, and infrastructure and institutional quality, are all key to SME engagement in global markets. However, while some trade costs have fallen significantly in recent years, due also to the expansion of digital platforms, others remain. Reform of slow or cumbersome border procedures can cut costs of trading by 12%-18%, depending on a country’s level of development (Blanchenay, Criscuolo and Calvino, 2016[2]). OECD analysis shows that opening up services markets would primarily benefit SMEs. For instance, for cross-border exports of services, an average level of services trade-restrictiveness represents the equivalent of an additional 14% tariff for SMEs relative to large firms (OECD, 2017[1]).

Statistics for analysis on this topic are currently scarce, but where they have been developed, the insights generated are of high relevance. For example, data are available for Nordic economies, showing that SMEs consistently source a lower share of foreign goods and services to produce exports than larger firms (Figure 1.18). The evidence also highlights that dependent SMEs have higher integration from an import perspective than independent SMEs, which indicates that they leverage on those links to overcome barriers to import trade.

Figure 1.18. Foreign value added share of exports in Nordic economies
As a percentage
Figure 1.18. Foreign value added share of exports in Nordic economies

Source: OECD (2017), Nordic Countries in Global Value Chains, report by the OECD Statistics Directorate and the National Statistical Offices in the Nordic countries, https://www.dst.dk/Site/Dst/Udgivelser/GetPubFile.aspx?id=28140&sid=nordglobchains.

 StatLink http://dx.doi.org/10.1787/888933923754

…and increase competitive pressures in local markets

While global value chains provide opportunities to access new markets (see also Chapter 3) and in turn growth, whether directly or indirectly for SMEs, they also create avenues for increased foreign competition in their home markets, in particular in lower productivity (i.e. labour intensive) sectors and activities, as larger firms and especially multinationals capitalise on international sourcing of intermediate parts from countries with lower wages costs - and often lower regulations. The evidence from databases such as the OECD-WTO Trade in Value-Added database confirms this, with increasing foreign content, typically from lower skilled activities (such as assembly) from lower income countries with an abundance of cheap labour, in the sourcing patterns of high income countries.

This foreign competition may have a disproportionate impact on SMEs in the upstream part of the value chain (e.g. producing parts for larger firms), especially if the competition emerges from larger foreign firms that are able to capitalise on economies of scale. Substantial differences in average salaries, even within OECD economies, suggest that the scope for larger firms to capitalise on GVCs is indeed not insignificant (Figure 1.19). The result is that, for example, employees in manufacturing micro firms in France earned on average almost twice the compensation of employees in large firms in Portugal in 2015. In the case of computers and electronics the differences between countries are even larger, with Finnish micro firms paying on average more than twice the salaries of the United States.

Figure 1.19. International competitiveness in wages
Compensation per employee, thousands of USD per employee, 2016
Figure 1.19. International competitiveness in wages

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923773

To some extent, this exaggerates the potential scale of the challenge for high wage SMEs, as outsourcing decisions are not solely based on relative differences in salaries, there are many other factors that determine a firm’s sourcing pattern (Just-in-Time delivery, trade related costs – behind the border, tariffs and transportation – regulatory costs, reliability, etc), but one particularly important factor is relative productivity. In general, in countries with lower relative labour costs, labour productivity is also relatively lower (Figure 1.20).

Figure 1.20. International competitiveness through productivity
Value added in 000 current USD per person employed, 2016
Figure 1.20. International competitiveness through productivity

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923811

However, often relative differences in labour productivity are smaller than relative differences in labour costs. For example, large Portuguese manufacturers have about the same labour productivty as French micro firms. Looking at unit labour cost levels (taking the share of compensation per employee over productivity) Austria, Estonia and Germany present similar levels regardless of the size of the enterprise Figure 1.21).

Figure 1.21. Labour income shares in manufacturing
Total compensation as a share of gross value added, total manufacturing, percentage, 2016 or latest available year
Figure 1.21. Labour income shares in manufacturing

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923830

Textiles is one sector that has been particularly transformed by GVCs in the last twenty years, in large part reflecting the fact that the assembly of clothing at the end of the value chain remains a labour intensive task, and so in many countries low-skilled activities have been outsourced to low wage countries. At least for now, assembly has not been greatly affected by automation, with firms in the sector in developed economies specialising in design, often fabric and high-quality assembly; and firms in lower income countries specialising in the basic assembly. This is a factor that has seen the number of SMEs in the sector decline considerably in many OECD countries over the last 16 years, especially in Italy, Spain, Denmark and Chile (Figure 1.22) with contractions in employment even more severe.

However, precipitous though the declines have been, SMEs increased their domestic market shares across all countries, as a result, in many, of a focus on niche activities of higher value. France for example, where SMEs have increased their share of production in the sector from 75% in 2000 to 89% in 2016, has specialised in higher value technical textiles (accounting for one-quarter of European production), and so although the number of individuals in the sector fell signifcantly, from just under 92 000 in 2000 to just under 35 000 in 2016, the number of firms increased from 5.5. to 6.6 thousand, turnover per employee was up over 40%, exports per employee nearly doubled, with wages (amongst the highest in the OECD) increasing from 10% below the whole economy average to 10% over.

Figure 1.22. SMEs in the manufacturing textiles sector
Percentage, 2016 or latest available year
Figure 1.22. SMEs in the manufacturing textiles sector

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923849

Other OECD countries have adopted different adaptive strategies to the competition from cheaper imports. In Spain for example, which also saw precipitous falls in numbers, and employment in SMEs, exports rose over 50% over the period, as firms capitalised on cheaper intermediate imports of textile products (with the foreign content of Spain’s exports rising from 27.5% in 2000 to over 35% in 2014) but, arguably only through considerable reductions in costs, as relative salaries fell from 97% of the economy average in 2000 to 85% in 2016.

Digitalisation presents opportunities for SMEs to strengthen their performance in terms of growth, innovation and internationalisation

Digital technologies are evolving rapidly and combining in often unforeseeable ways, with large scale effects on market structures and competitive conditions for SMEs (see also chapter 3 on market conditions). Shifts in client demands and supply-chain processes are exerting pressure to reshape business models to become more compatible with the digital era of continuous connection and instantaneous global reach. The impact of advanced digital technologies has also transformed and disrupted many sectors traditionally dominated by SMEs, notably transportation (e.g. Uber), restaurants (Deliveroo), real estate (via a whole range of on-line platforms), or travel and accommodation (Expedia, Booking.com, AirBnb) (see also chapter 7 on access to innovation assets).

Digitalisation is therefore playing a major role in shaping market conditions and SME performance, whether through cheaper digital tools (ICT equipment) that provide scope for new innovative firms to enter the market, the provision of digital services (which reduce the space between consumers and producers), or access to new (including international) market places via digital intermediation platforms, such as Amazon and Task Rabbit, and other dedicated company websites.

Indeed, digitalisation represents an important vehicle for SMEs to be “born global” and opens new opportunities to enhance competitiveness, through product or service innovation and improved production processes. Furthermore, Big Data and data analytics can enable a better understanding of the processes within the firm, the needs of their clients and partners, and the overall business environment.

Digitalisation has also transformed the possibilities of scaling up, and different forms of business growth are emerging, with some companies able to achieve significant scale, market share and high-productivity without affording large investment in tangible assets. “Lean start-ups” are emerging that leverage the Internet to lower fixed costs and outsource many aspects of the business to stay agile and responsive to the effect market (OECD, 2017[13]).

The use of digital technologies can also ease SMEs’ access to skills and talent, through better job recruitment sites, outsourcing and online task hiring, as well as connection with knowledge partners (OECD, 2017[13]). It can facilitate access to a range of financing instruments. Mobile banking and online payments have had an important impact on traditional SME financing, and digitalisation has allowed new financial services to emerge, with innovative solutions to address information asymmetries and collateral shortages (see also chapters 5-6 on access to finance and skills).

Indeed SME value-added and employment growth in high-digital intensity5 activities have outpaced those in low digital intensity activities (Figure 1.23).

Figure 1.23. High-digital intensity SMEs have higher growth
Ratio of high over low digital intensity sectors growth rate of employment and value added
Figure 1.23. High-digital intensity SMEs have higher growth

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933923868

But many SMEs struggle to seize the emerging opportunities

To date, a large number of SMEs have still not capitalised on these possibilities. SMEs often lack the vision and resources to seize the opportunities opened up by the digital transformation.

In most countries, gaps with larger firms are narrow with respect to simple connectivity and web presence, but large in e-commerce and, especially, as concerns the uptake of more sophisticated applications. For instance, across OECD countries, enterprise resource planning (ERP) software applications to manage business information flows are popular among large firms (78% adoption rate in 2016) but considerably less so among SMEs (less than 28%). Computer programming activities are being more frequently outsourced by SMEs, due to a facilitated access to ready-to-go software, processing power and storage capabilities offered by large firms in their cloud computing services (see Chapter 7 on Access to Innovation). Still, in many countries there are also large adoption gaps in cloud computing services, which provide scope for cost savings compared to the fixed costs of ICT investment (Figure 1.24).

The adoption lag of SMEs is also related to a lack of investment in complementary knowledge-based assets, such as R&D, human resources, organisational changes and process innovation, and this lag has implications for their capacity to turn technological change into innovation and productivity growth. Furthermore, SMEs face specific challenges in managing digital security and privacy risks, mainly due to lack of awareness, resources and expertise to assess and manage risk effectively (see Chapter 6 on access to skills).

Figure 1.24. SMEs lag in the adoption of more sophisticated digital technologies
Enterprises using cloud computing services, by firm size (2016), as a percentage of enterprises in each employment size class
Figure 1.24. SMEs lag in the adoption of more sophisticated digital technologies

Source: OECD ICT Access and Usage by Businesses (database), http://oe.cd/bus (accessed June 2017).

 StatLink http://dx.doi.org/10.1787/888933923887

For those SMEs and start-ups that are responding to the challenges of digitalisation, the evidence, where available, points to a positive impact on business confidence and performance. For example, data from the Future of Business Survey reveal that among SMEs with a web presence (i.e. a Facebook page) those using a range of digital on-line tools to advertise and sell their products are more likely:

  1. 1. to have a positive assessment of future employment growth (Figure 1.25); and

  2. 2. to be involved in international trade (Figure 1.26).

Figure 1.25. Firms planning to increase employment in the next six months
Percentage of all firms, July 2017 – April 2018
Figure 1.25. Firms planning to increase employment in the next six months

Source: Facebook-OECD-World Bank, Future of Business Survey.

 StatLink http://dx.doi.org/10.1787/888933923906

Figure 1.26. Firms trading internationally, by the use of online tools
Percentage of all firms, July 2017 – April 2018
Figure 1.26. Firms trading internationally, by the use of online tools

Source: Facebook-OECD-World Bank, Future of Business Survey.

 StatLink http://dx.doi.org/10.1787/888933923925

Box 1.2. The Future of Business Survey

The Future of Business Survey is a survey launched in February 2016 by Facebook, the OECD and the World Bank, which jointly work at its design and continuous development. The survey covers the population of enterprises whose digital presence includes a Facebook Page and was run, as of September 2018, in 42 developed and emerging economies.

The Future of Business Survey provides timely information on how firms assess the current state and future outlook of their business, the job creation perspectives, and the main challenges. Regular survey questions allow profiling of businesses according to their size, age, gender of business ownership/management, participation in trade and use of online tools. The survey also features ad hoc modules, where specific topics are investigated such as motivations for starting-up and sources of business financing.

The survey constitutes an innovative experiment of public-private partnership in data development and collection that contributes to deliver new insights on SMEs in the digital economy.

Conclusions and looking ahead

SMEs are key players in the economy and the wider eco-system of firms. They provide the main source of employment, and often value added, across countries. However, their contributions vary widely across and within countries and sectors. In this regard, improving knowledge about heterogeneity is essential for analysis and evidence-based policy making, not least during a period of rapid digital transformation, set against a backdrop of two decades of increasing globalisation, which may now be slowing as trade tensions rise and as digitalisation and automation begin to impact on global value chains – i.e. global fragmentation of production – that have been at the vanguard of globalisation.

The evidence shows that a broad view of the SME sector can mask the impact of these mega-trends, as they affect more some SMEs and sectors than others. The structure of the SME population as a whole has been broadly stable in recent years, with the bulk of SMEs in most countries engaging in activities with relatively low entry costs such as in the distribution and construction activities. Important dynamic changes are however occurring in sub-sectors highly exposed or able to capitalise on the digital transformation, such as in the information and communication sector.

The same is true of exposure to globalisation. While global value chains (GVCs) have provided significant opportunities for SMEs to participate and specialise in tasks within value chains that offer scope for access to foreign markets, either as direct or indirect exporters, they have also increased competition within domestic markets, from lower-cost producers in other parts of the world, particularly in emerging economies. In sectors such as textiles for example, this increased competition has had a profound impact on SME producers in developed economies with higher-wage costs, forcing many to close but also acting as a stimulus to upgrade and move up the value chain through the provision of higher value, and higher productivity, tasks. Similar impacts have almost certainly occurred in other sectors exposed to low-wage competition. However, GVCs have also presented opportunities for high-skilled, knowledge intensive SMEs in developed economies, and exploring the evidence on them will form an important part of the broader work programme as the data are developed.

Both of these mega-trends have a clear impact on wages and productivity, but they are not the only drivers. The productivity slowdown in the post-crisis period has concerned both large and smaller firms. Slower productivity growth reduces the potential for wage growth, and it is no coincidence that wage growth has been weak in the post-crisis period, with real wages barely above their pre-crisis levels in many countries.

There is an on-going debate and considerable analyses on the potential causes of the productivity slowdown, including through a slowing in technology diffusion, winner takes all dynamics, lagged effects of new digital innovations, and slower business dynamism. The evidence presented in this chapter suggests that an additional factor may also be at play. In most OECD economies the level of firm entries is rising again but are these entries increasingly in high-growth high-productivity activities? The answer appears to be no. In all OECD economies most new entries are in activities with below average labour productivity, and below average wages; and this may be working to weigh down on overall labour productivity and wages.

This of course is not new and reflects the traditional gravitational lure of lower-entry cost sectors to budding entrepreneurs, whether they are pushed or pulled into their orbit. Digitalisation, and in particular gig or sharing economy factors, may have added a new dimension, so that many of these new entries have very limited growth potential and little desire from the ‘entrepreneurs’ behind them to scale-up.

With SMEs paying on average less than 20% lower compensation per worker than in larger firms (and significantly more in some countries), this matters, especially in considering inequalities. Policies predicated on delivering higher-growth and higher productivity start-ups necessarily require a focus on those sectors that can capitalise on the digitalisation and globalisation mega-trends; especially as they can drive higher wages and reduce inequalities. Exporting SMEs typically have higher productivity and pay higher wages than non-exporting SMEs, and although the causal relationship between higher-productivity and exports is complex, it is clear that both are important targets for reducing inequalities.

At the same time, strong impacts on economic growth and inclusion can be achieved if small established enterprises in traditional sectors of the economy, which represent the vast majority of SMEs, are given the means and opportunities to upgrade their productivity levels, including through the adoption of digital technologies. Indeed, SMEs across sectors are likely to benefit from greater mass customisation and the reduction of distance and time to market enabled by digital technologies. For many SMEs, however, the lag in the adoption of digital technologies and investment in complementary knowledge based assets, including skills and practices to adequately manage digital risks, could jeopardise their transition towards the next production revolution and participation in global markets, as well as limit the benefits they can accrue from the rise of open innovation.

As illustrated in this Outlook, part of the policy toolkit to enhance productivity in traditional SMEs and sustain business dynamism in high growth activities already exists, for example with regard to integrating SME-related considerations in regulatory policy making and reducing administrative burden, using Big Data and digital tools to deliver higher quality and more customised public services to businesses, consolidating and expanding ICT infrastructures to ensure better SME connectivity, and improving access by SMEs to strategic resources, such as diversified sources of finance, innovation assets, a broader pool of talent and upskilling opportunities for workers and managers. However, the pace of pro-growth structural reforms has slowed in recent years and large differences remain across the SME population in their capacity to seize the benefits of the digital transition, including in public services. In addition, a changing policy landscape, particularly with regards to trade, and the development of new technologies, such as blockchain and automation, requires flexibility and adaptability from firms.

Many of these developments are pulling in the direction of reshoring, which could provide scope for SME growth in sectors that have seen slow growth or retrenchment in recent years, but they may also provide threats if they increase the costs of entry, either through direct acquisitions of capital or through the need for higher skilled employees.

More broadly, the effects of the digital transformation may only be beginning to be felt and sectors with high SME participation today, such as retail and real estate may be significantly affected in the future. Understanding the direction and potential impact of these changes is essential in developing sound policies, especially as these threats also present opportunities for SMEs that are able to embrace or capitalise on digital technologies – and the evidence shows they are currently lagging.

These new developments require new statistics to address policy relevant questions. What for example is the motivation behind new entries – push or pull? Have firms exposed to international competition upgraded, through higher innovation, or developed survival strategies through wage cuts? How are SMEs capitalising on digitalisation - how many SMEs use digital technologies to reach new markets and/or generate efficiencies? Are successful SMEs really SMEs or are they affiliates of larger firms, or in ‘control’ relationships? Can SMEs improve participation in GVCs through links to larger domestic exporters? Are SMEs able to penetrate emerging, and higher growth markets?

This chapter demonstrates that current official statistics are able to provide important insights, in particular with respect to structural heterogeneity, but also illustrates the importance of continuing to expand the statistical boundary, not least to tackle emerging issues.

Data sources

Chapter 1 primarily builds on i) the OECD databases of official business statistics, in particular the Structural and Demographic Business Statistics (SDBS) and Trade by Enterprise Characteristics (TEC), databases that are built by the OECD in cooperation with national statistical offices and Eurostat; ii) indicators developed from the macro-linking of data from these two databases with other official statistics; and iii) the OECD Productivity Statistics database containing productivity measures computed using the OECD National Accounts and Employment and Labour Market Statistics databases. The chapter also relies on the OECD Timely Indicators of Entrepreneurship database, covering official statistics and administrative data, and on enterprise statistics from new data sources, such as Facebook-OECD-World Bank Future of Business Survey, as relevant for the analysis.

References

[14] Ahmad, N. and J. Ribarsky (2018), Towards a Framework for Measuring the Digital Economy, Working Paper for the 16th conference of the International Association of Official Statisticians (IAOS), http://oecd.org/iaos2018/programme/IAOS-OECD2018_Ahmad-Ribarsky.pdf.

[10] Bas, M. and V. Strauss-Kahn (2015), Input-trade Liberalisation, export prices and quality upgrading, Journal of International Economics, https://doi.org/10.1016/j.jinteco.2014.12.005.

[11] Bas, M. and V. Strauss-Kahn (2014), Does importing more inputs raise exports? Firm-level evidence from France, https://doi.org/10.1007/s10290-013-0175-0.

[2] Blanchenay, P., C. Criscuolo and F. Calvino (2016), “Cross-country evidence on business dynamics over the last decade: from boom to gloom?”, OECD Science, Technology and Industry Working Papers, OECD Publishing, Paris.

[17] Calvino, F. et al. (2018), “A taxonomy of digital intensive sectors”, OECD Science, Technology and Industry Working Papers, No. 2018/14, OECD Publishing, Paris, https://dx.doi.org/10.1787/f404736a-en.

[12] Criscuolo, C. and J. Timmis (2018), “GVC centrality and productivity: Are hubs key to firm performance?”, OECD Productivity Working Papers, No. 14, OECD Publishing, Paris, https://dx.doi.org/10.1787/56453da1-en.

[15] Decker, R. et al. (2016), Declining Business Dynamism: Implications for Productivity?, Brookings Institution, Washington, DC.

[16] Haltiwanger, J. (2016), Top ten signs of declining business dynamism and entrepreneurship in the United States, Kauffman Foundation, http://www.kauffman.org/neg/section-3.

[8] Lopez Gonzalez, J. (2016), “Using Foreign Factors to Enhance Domestic Export Performance: A Focus on Southeast Asia”, OECD Trade Policy Papers, No. 191, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jlpq82v1jxw-en.

[9] López González, J. and M. Jouanjean (2017), “Digital Trade: Developing a Framework for Analysis”, OECD Trade Policy Papers, No. 205, OECD Publishing, Paris, https://dx.doi.org/10.1787/524c8c83-en.

[3] OECD (2018), Highlights of entrepreneurship at a Glance 2018, OECD, http://www.oecd.org/sdd/business-stats/EAG-2018-Highlights.pdf.

[5] OECD (2018), OECD Compendium of Productivity Indicators 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/pdtvy-2018-en.

[19] OECD (2018), Opportunities for All: A Framework for Policy Action on Inclusive Growth, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264301665-en.

[18] OECD (2018), The Productivity-Inclusiveness Nexus, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264292932-en.

[4] OECD (2017), Entrepreneurship at a Glance 2017, OECD Publishing, Paris, https://dx.doi.org/10.1787/entrepreneur_aag-2017-en.

[7] OECD (2017), “Making trade work for all”, OECD Trade Policy Papers, No. 202, OECD Publishing, Paris, https://dx.doi.org/10.1787/6e27effd-en.

[13] OECD (2017), OECD Digital Economy Outlook 2017, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264276284-en.

[1] OECD (2017), Services Trade Policies and the Global Economy, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264275232-en.

[6] Schwellnus, C., A. Kappeler and P. Pionnier (2017), “Decoupling of wages from productivity: Macro-level facts”, OECD Economics Department Working Papers, No. 1373, OECD Publishing, Paris, https://dx.doi.org/10.1787/d4764493-en.

Annex 1.A. Developing new data

Although the definition of an SME varies by country, partly reflecting the size of the economy, statistics on the size of firms provide a basis for some comparability. For the purpose of Chapter 1, SMEs are considered to be firms with less than 250 persons employed.

Within the universe of SMEs lays considerable heterogeneity, not just in terms of their core industrial activity, i.e. what they produce, which is well covered in the statistical information system, but also in terms of why and how they produce, where the coverage of statistics is currently patchy. Heterogeneity in this sense concerns a number of firm characteristics and business models such as: hobby activities; subsistence activities; social enterprises; informal firms; high-growth firms; unicorns; gazelles; firms that are part of a multinational; firms with a multitude of customers, firms locked-in to supplying to one customer; firms that export; firms that sell only to local markets; and so on.

To understand the performance of SMEs, help them grow and enable them to improve well-being, better knowledge is needed of specific groups of SMEs. This chapter demonstrates what can be done with current data, using detailed information on the composition of the business population and its characteristics, in particular as provided by OECD databases of official business statistics compiled in cooperation with National Statistics Offices of member and partner countries.

But the chapter is also designed to illustrate the importance of continuing to expand the statistical boundary, not least to tackle emerging issues. Two of the most pressing current challenges concern digitalisation and globalisation. The chapter tries to provide insights on the potential impact these may be having on the SME population but while the current range of statistics provide useful insights, there remain significant data gaps, in part reflecting the way, and the reasons, that national statistics on structural business statistics and business demography were originally developed.

Typically the focus has been on what the eventual output of a firm was and what types of inputs were used in production, primarily as inputs to developing the national accounts; with firms classified to sectors largely on the basis of what their final output was. Globalisation, and now digitalisation, are beginning to question this approach, and whether new data need to be collected within information systems. Both phenomena have resulted in considerable heterogeneity on what firms actually do, even when classified to the same industrial activity. For example a vertically integrated firm producing computers will find itself classified to the same sector as a firm only assembling computer parts but the underlying use of capital, labour, inputs, and human and organisational capital will be significantly different, not to mention exposure to international trade and take-up of digital tools.

There have of course been, and continue to be, important innovations and changes in national statistics information systems over the years to address these challenges, including for example the development of Trade by Enterprise Characteristics databases, and indeed one should not forget that even the development of Business Demography statistics is in itself a relatively recent innovation designed to meet growing demands for better data and insights on entrepreneurship and business dynamism. More recently there have been significant advances to better understand the role of SMEs in GVCs, being managed by the OECD Expert Group on Extended Supply-Use Tables and in developing a stronger understanding on the impact of digitalisation (Ahmad and Ribarsky, 2018[14]), which will provide the basis for improved insights on digital intermediaries and also on the gig economy. Yet, more can be done, especially in the area of SMEs and business dynamics.

To advance in new areas and to sustain momentum in on-going areas, this Annex makes the following recommendations:

Consider the development of modules to existing surveys, such as the Labour Force Survey, to estimate whether individuals were pushed or pulled into self-employment, and/or whether that employment (secondary activity or otherwise) was part of the gig economy.

Develop statistical business registers that differentiate between independent and dependent (i.e. affiliates that are owned and managed by a larger firm) SMEs.

Explore the ability of statistical business registers to identify (and group) firms on the basis of ages (alongside conventional and complementary – industry and size – characteristics) and tabulate regularly statistics based also on this dimension.

Develop and disseminate composite indicators from existing statistics that can be disclosed without breaching statistical confidentiality – for example productivity distributions of SMEs within sectors (or indeed other aggregations, including the whole economy), concentration indices.

Ensure that structural business statistics cover all economic activities (including financial services) and all variables (for example wages in the services sector).

Explore the possibility of further developing structural and demographic business statistics broken down by region (i.e. regular production, extended coverage of variables, improved international comparability).

Explore the scope to link various datasets and registers, in particular employment and labour related registers with firm level registers

Explore, or improve, access to types of data that remain underexploited, notably open data from public and private sources, and continue to explore innovative approaches to data development, such as public-private partnerships, webscraping or crowdsourcing methods.

Annex 1.B. Sectoral specialisation
Annex Figure 1.B.1. SME employment by economic activity, selected countries, 2015
Percentage
Annex Figure 1.B.1. SME employment by economic activity, selected countries, 2015Annex Figure 1.B.1. SME employment by economic activity, selected countries, 2015

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933924077

Annex Figure 1.B.2. SME share of employment within sectors
Percentage, total employment in sector x =100
Annex Figure 1.B.2. SME share of employment within sectors

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933924096

Annex 1.C. Enterprise creation rates

In recent years there has been a considerable debate, set against a backdrop of declining trends in productivity, on a possible ‘secular decline’ in enterprise creation rates. The debate has focused primarily on US data, where relatively long time series going back to the 1980s are available (Decker et al., 2016[15]; Haltiwanger, 2016[16]), but similar studies, although with shorter time series (Blanchenay, Criscuolo and Calvino, 2016[2]) have drawn similar conclusions for other countries.

It is instructive to highlight the statistical nature of the construction of enterprise creation rates and how this may need to be interpreted. In many analyses, statistics on creation rates of businesses are viewed analogously to birth rates in the general human population. However, it is often forgotten that, unlike with general population measures, existing firms do not typically give birth to new entries, and where they do engage in the creation of new firms, these are often recorded as ‘growth’ in the existing firm and not new creations.

The use of the number of existing firms as the denominator in measures of enterprise creation rates is a convenient choice (and generates consistent measures of rates of enterprise death where also the denominator is the population of existing firms), but does come with statistical caveats that can impact on comparability over time and across countries.

For example, two countries with exactly the same numbers of new creations in a given year can have very different creation rates if the population of firms differ. For this reason, it is helpful to consider also levels of creations and not just the rates.

Panel A of Annex Figure 1.C.1shows the number of employer establishments in the United States over the last quarter of a century, revealing a strong upward trend, notwithstanding the crisis dip, which presents the secular decline story in a slightly more nuanced context (US Small Business Administration data). The strong growth in the population of large enterprises (with more than 500 employees) registered in the United States increased market concentration and may have crowded out potential new entrants (Annex Figure 1.C.2). However, the number of non-employer establishments increased by around 60% in the last fifteen years (Annex Figure 1.C.1Panel B).

Annex Figure 1.C.1. Number of establishments in the United States
Millions
Annex Figure 1.C.1. Number of establishments in the United States

Source: US Small Business Administration.

 StatLink http://dx.doi.org/10.1787/888933924115

Annex Figure 1.C.2. Number of employer establishments with 500+ employees in the United States
Number
Annex Figure 1.C.2. Number of employer establishments with 500+ employees in the United States

Source: US Small Business Administration.

 StatLink http://dx.doi.org/10.1787/888933924134

Annex Figure 1.C.3 provides an additional illustration of the statistical nature of rates by comparing enterprise birth rates in France, Sweden and the United Kingdom, showing that flat or declining trends in birth rates (i.e. graphs on the right side) can go hand-in-hand with rising numbers of births (i.e. graphs on the left side).

Annex Figure 1.C.3. Number of enterprise births, number of enterprises and birth rate, selected countries
Thousands for births and stock of enterprises; percentage for birth rates
Annex Figure 1.C.3. Number of enterprise births, number of enterprises and birth rate, selected countries

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933924153

Finally, increases in the population of firms can also be associated with decreases in the number of creations and decreases also in failures, which would be a sign of lower levels of creative destruction and by extension entrepreneurialism. The focus on the number of active firms, which has increased significantly in many countries despite lower levels of start-ups, helps to inform the analysis; specifically, it may suggest that the state of entrepreneurialism in its broadest sense has been less bleak than that suggested by creation rates alone (Annex Table 1.C.1).

Annex Table 1.C.1. Number of enterprises, employment and enterprise births
Business economy

 

Number of enterprises

Persons employed

Enterprise births

 

2005

2010

2015

2005

2010

2015

2005

2010

2015

AUS

1 614 586

1 649 734

1 678 411

7 480 000

7 934 000

7 912 000

242 861

231 024

248 898

AUT

313 885

426 815

413 929

2 581 345

2 871 123

3 004 647

24 568

34 198

28 311

BEL

483 809

597 850

642 130

2 551 169

2 771 320

2 824 001

32 218

41 162

41 102

BRA*

 

1 936 862

2 244 939

 

31 431 860

35 797 020

 

330 419

297 494

CAN*

789 290

738 880

808 330

9 502 575

9 817 862

10 571 770

92 560

48 810

64 720

CHE

293 746

384 559

395 608

2 647 914

3 416 639

3 451 971

10 684

11 071

27 677

CZE

889 726

969 801

1 026 355

4 081 346

3 986 570

3 701 741

77 672

110 880

85 645

DEU

2 810 118

2 958 720

2 795 899

22 650 420

23 334 510

28 071 530

283 105

258 076

198 135

DNK

205 145

212 593

217 960

1 413 589

1 331 449

1 754 365

26 939

23 266

24 283

ESP

3 047 021

3 102 016

2 970 947

13 780 000

12 508 930

11 711 550

317 273

242 228

274 172

EST

62 149

70 302

82 769

432 706

408 069

455 287

6 440

7 794

8 512

FIN

239 381

286 432

291 722

1 374 143

1 418 660

1 593 226

21 253

28 424

19 623

FRA

2 220 897

2 947 623

3 492 052

14 840 630

16 999 170

16 056 040

224 819

376 631

328 884

GBR

1 966 355

2 013 225

2 326 020

18 583 780

18 731 650

20 466 460

265 545

210 950

343 550

GRC

799 040

847 055

777 268

2 575 832

2 768 305

2 552 875

69 716

72 186

39 896

HUN

580 885

563 368

531 121

2 672 856

2 533 662

2 695 977

52 646

56 370

56 799

IRL

203 083

195 431

248 843

1 483 966

1 237 385

1 402 981

11 954

11 237

18 100

ISL

23 107

23 774

26 039

109 067

95 934

112 913

 

3 289

3 221

ISR*

162 793

188 695

217 737

1 615 088

1 922 201

2 256 815

14 800

16 664

28 253

ITA

3 966 758

3 985 434

3 819 956

15 637 520

16 010 810

14 806 370

308 307

265 060

279 132

JPN*

2 001 152

2 033 692

2 139 380

 

 

 

87 966

91 300

109 202

KOR

 

4 717 796

4 946 304

11 490 520

13 348 190

11 888 300

 

701 123

714 902

LTU

113 201

120 830

185 954

932 629

904 828

1 007 609

30 807

25 463

34 490

LUX

23 194

27 611

31 906

250 896

290 425

315 848

2 225

2 629

2 989

LVA

63 529

82 650

110 310

644 569

565 660

651 593

7 278

13 803

19 003

NLD

638 118

970 457

1 112 691

4 972 465

5 570 939

5 683 695

62 040

101 002

107 946

NOR

243 776

268 949

295 204

1 320 552

1 502 994

1 678 695

24 811

20 758

26 753

NZL*

102 984

101 733

107 586

1 268 693

1 222 571

1 366 556

12 765

9 390

13 467

POL

1 667 934

1 957 113

2 059 967

8 156 535

9 532 762

9 644 727

195 970

270 271

249 815

PRT

889 084

875 083

818 120

3 356 756

3 344 375

3 093 225

116 920

103 859

130 156

SVK

323 836

374 114

446 471

1 640 182

1 318 844

1 562 409

43 278

49 354

53 899

SVN

95 554

123 467

141 118

610 824

633 904

613 502

8 579

12 757

15 154

SWE

581 622

667 421

740 182

2 712 453

2 944 008

3 113 746

41 212

50 214

53 185

TUR

 

3 088 887

3 210 972

 

12 652 270

15 972 460

 

717 573

387 385

USA

6 523 644

6 460 877

6 596 243

114 186 342

109 805 388

118 016 100

785 419

596 872

654 444

Note: Countries marked with asterisk (*) refer to employer business demography data, which excludes non-employing businesses and births of enterprises with no employees. On the other hand a transition from non-employer enterprise to employer enterprise counts as birth in employer business demography.

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933924267

Annex Figure 1.C.4. Birth rates, job creation and productivity
Percentage (employer enterprise births rates), labour productivity (value added in thousands of national currency per person employed), and number of persons employed in births, 2016
Annex Figure 1.C.4. Birth rates, job creation and productivityAnnex Figure 1.C.4. Birth rates, job creation and productivity

Note: The red line corresponds to the median productivity calculated for 2-digit ISIC Rev. 4 sectors.

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933924172

Annex 1.D. Labour productivity of SMEs by sector
Annex Figure 1.D.1. Labour productivity of SMEs
Index, Large enterprises =100
Annex Figure 1.D.1. Labour productivity of SMEsAnnex Figure 1.D.1. Labour productivity of SMEsAnnex Figure 1.D.1. Labour productivity of SMEsAnnex Figure 1.D.1. Labour productivity of SMEs

Source: OECD Structural and Demographic Business Statistics Database, 2018, http://dx.doi.org/10.1787/sdbs-data-en.

 StatLink http://dx.doi.org/10.1787/888933924191

Annex Figure 1.D.2. Labour productivity in 2010 and employment growth over 2010-2016 (or latest available year)
Selected countries, detailed sectors (ISIC Rev 4), national currency, current prices and percentage change at annual rate
Annex Figure 1.D.2. Labour productivity in 2010 and employment growth over 2010-2016 (or latest available year)

Note: Bubbles in dark blue denote employment creation; bubbles in grey denote employment destruction. The size of the bubbles increases with the absolute change in employment on a person’s count basis in the different sectors. Quadrants are relative to the average labour productivity levels in 2010 and average employment growth between 2010 and 2016. Measurement of value added, and hence labour productivity, in the primary sector, real estate services, public administration, health, education and other non-market activities is subject to a number of conceptual and measurement issues.

Source: (OECD, 2018[5]), OECD Compendium of Productivity Indicators 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/pdtvy-2018-en.

 StatLink http://dx.doi.org/10.1787/888933924210

Notes

← 1. www.oecd.org/cfe/smes/ministerial/SME-Ministerial-Declaration-ENG.pdf.

← 2. (OECD, 2018[19]; OECD, 2018[18]; OECD, 2017[13]).

← 3. 30% refers to Greece.

← 4. For top sectors with the highest job creation rate, see (OECD, 2018[3]).

← 5. (Calvino et al., 2018[17]) propose a taxonomy of digital sectors, where high digital intensity sectors include not only ICT but also: transport equipment, legal and accounting activities, scientific research and developments, advertising and market research; other business services, administrative and support service activities, and other service activities (94-96 of ISIC REV. 4). Also finance and insurance is identified as a high digital intensity sector, but data for this sector are not covered in Figure 1.23.

End of the section – Back to iLibrary publication page