In 2024, small and medium-sized enterprises (SMEs) employing less than 250 persons in the business sector displayed significantly lower productivity levels than large firms with 250 or more persons employed. Across OECD and accession candidate countries, unweighted average SME labour productivity was only 65% of that of large firms. The size-productivity gap, defined as the difference in labour productivity between SMEs and large firms, varied substantially across countries, ranging from nearly 60 percentage points (p.p.) in Ireland to just 2 p.p. in Switzerland. Since 2013, this gap has widened in most countries (20 out of 32) and across most industries.
Among SMEs, labour productivity increases with firm size, with micro firms (fewer than 10 persons employed) exhibiting the lowest productivity. In manufacturing, productivity remains strongly scale dependent. By contrast, in business services, medium-sized enterprises (50 to 249 persons) often match or even outperform large firms.
Large productivity gaps also exist among firms of similar size. Evidence from the OECD MultiProd project shows that productivity dispersion within size classes explains around 95% of total productivity variance within a given industry.
Firm-level data from Italy and Spain reveal large productivity differences between firms in different regions of the same country, even within a given industry and firm size class, underscoring the role of local factors beyond firm size and industrial composition.
OECD Compendium of Productivity Indicators 2026
A large body of evidence points to a significant productivity gap between large firms and small and medium‑sized enterprises (SMEs), particularly in manufacturing (Cunningham et al., 2022[1]). Key explanations for productivity differences across firms include economies of scale, access to capital, and skills of workers and managers (Ferrando and Ruggieri, 2018[2]; Criscuolo, Gal and Freund, 2024[3]; Criscuolo et al., 2021[4]). Ownership and management structures are also relevant, with professional management often associated with higher productivity levels (Aguilera et al., 2024[5]). In addition, firm internationalisation contributes to variation in productivity performance, as large firms, particularly multinationals, are more likely to engage in exporting, importing, and foreign production, potentially reinforcing both scale and productivity advantages (Bernard et al., 2018[6]).
This chapter examines differences in labour productivity beyond industry-level patterns, comparing SMEs (firms with 1 to 249 employees) with large firms (250 or more employees). To provide a comprehensive picture of productivity differences between firms, it uses several data sources that are more granular than those underpinning standard industry and total‑economy productivity statistics. Due to data gaps and limited harmonisation of key variables across countries (see Box 5.1), productivity measures in this chapter are not directly comparable to those in other chapters of the Compendium.
The first part of this chapter relies on the OECD Structural and Demographic Business Statistics database, with labour productivity measured as turnover per person employed in nominal terms. It provides an economy‑wide overview of productivity patterns by firm size and their evolution between 2013 and 2024. The discussion then examines how firm-size productivity gaps differ across industries, highlighting a weaker relationship between size and productivity in services than in manufacturing. Complementary firm-level evidence relies on data from the OECD MultiProd project, where labour productivity is measured as value added in constant prices per worker (see Box 5.2.). The final part of the chapter explores differences in productivity performance across firms in regions of Italy and Spain. Drawing on firm-level data from Bureau van Dijk’s Orbis database, this last part uses both labour productivity and multifactor productivity (MFP) measures.
In 2024, SMEs in the business sector operated at significantly lower labour productivity levels than large firms. Productivity of SMEs, on average across 33 OECD and accession candidate countries, was only 65% of that of large firms, measured by turnover per person employed, ranging from 30% in Ireland to 98% in Switzerland (Figure 5.1). Large firms typically benefit from size-related cost advantages, easier access to finance, stronger managerial and HR capabilities, and deeper integration into foreign markets, contributing to their higher productivity (Marchese et al., 2019[7]).
SMEs in the business sector consistently exhibit lower productivity than large firms. The size-productivity gap generally increased between 2013 and 2024 in 20 of 32 countries with available data. In ten countries, the productivity differential between large and small firms has narrowed slightly over the past decade, suggesting either slower productivity growth among large firms or catch-up among SMEs driven by a small group of rapidly growing firms or ‘scalers’ (OECD, 2025[8]). For example, in Korea, where SME catch-up was most pronounced, large firms experienced average annual productivity growth of 0.8% during this period. SMEs’ productivity grew almost five times faster, at 3.7% on average, in nominal terms.
In contrast, countries with a strong presence of multinational enterprises (MNEs) have seen a substantial increase in the productivity gap between large firms and SMEs between 2013 and 2024. The gap almost doubled in Ireland, where the productivity of large firms grew almost three times faster over this period than that of SMEs, at approximately 9.6% per year. In Luxembourg, the size‑productivity gap widened sharply between 2013 and 2024, shifting from a situation in which SMEs were 17% more productive than large firms to one where SMEs reached only 62% of large‑firm productivity. This shift in Luxembourg was driven by annual productivity growth of about 5.3% in large firms, compared with near‑zero growth among SMEs.
Per cent of large firm productivity, as measured by turnover over total employment
Note: The Business sector corresponds to ISIC, Revision 4, Sections B to J and M to N. SMEs are defined as firms with 1 to 249 persons employed, except in the US, where they correspond to 1 to 299 persons employed. Data for Costa Rica, Korea and the United States correspond to 2022, and for the United States, the diamond corresponds to 2012.
Source: OECD SDBS database, Business population estimates of the UK Department of Business and Trade, and U.S. Census Bureau Statistics of US Businesses.
Firm‑size‑specific productivity measurement is based primarily on business surveys and economic censuses rather than national accounts. Compared with national‑accounts‑based productivity measures, this implies a less integrated and less harmonised statistical framework, which introduces a number of measurement challenges:
First, size class definitions are not fully harmonised. Some countries apply national firm‑size classifications tailored to domestic enterprise structures, improving internal relevance but reducing international comparability.
Second, reporting thresholds vary across countries as well as by industry within countries. Employment‑based thresholds, such as including only firms with at least ten employees, are common but may lead to systematic coverage differences across industries. Large firms are consistently included, while micro‑firms are often excluded or only partially sampled, despite representing a substantial share of employment in certain industries. Alongside employment-based cut-offs, thresholds may also be based on turnover, adding to the complexity of cross‑country comparisons.
Third, data on hours worked, especially in services sectors, are often unavailable. Moreover, some countries provide firm-size data only for economic census years, resulting in gaps in the time series. Similarly, some countries do not report value added by firm-size class.
Finally, all data are reported in nominal prices, as firm-size-specific deflators are not available.
Due to these limitations, the labour productivity metrics in this chapter are not entirely comparable to those in other chapters. To maximise country coverage, the main analysis based on the OECD Structural and Demographic Business Statistics database measures labour productivity as turnover in current prices per person employed. This is complemented by two analyses based on micro data – Box 5.2 (firm‑level) and a subsection focused on firms in different regions – where productivity is instead measured as value added in constant 2005 USD PPPs per employee or person employed.
Productivity gaps between SMEs and large firms vary markedly across industries. Most research has historically focussed on the manufacturing sector, highlighting a noticeable productivity gap between SMEs and large enterprises (Bartelsman, Haltiwanger and Scarpetta, 2013[9]; Cunningham et al., 2022[1]). This is partly driven by measurement issues in productivity estimates for services (also see Box 4.1 in Chapter 4). In non-financial business services, the link between firm size and productivity is less pronounced (Berlingieri, Calligaris and Criscuolo, 2018[10]; Mourougane and Kim, 2020[11]).
Microdata evidence offers an additional perspective on differences across sectors (Box 5.2). In services, productivity variability is overwhelmingly driven by differences in productivity among firms within the same sector and size class, which account for around 95% of total variance. In manufacturing, productivity differences within firm size classes explain a slightly smaller share of variance (around 92%), pointing to stronger scale effects.
Over the past decade, the unweighted average size-productivity gap between SMEs and large firms across 33 OECD and accession candidate countries with available data has widened by nearly 5.5 percentage points in the business sector (Figure 5.2). In 2024, SMEs, on average, reached about 65% of the productivity level of large firms, down from 70.5% in 2013. In 2024, this productivity gap was more pronounced in the industrial sector – comprising Mining and Quarrying, Manufacturing and Utilities – than in the business services sector. On average, SMEs in the industrial sector did not even reach 50% of the productivity of large firms.
The disparity was most noticeable in manufacturing, where even the most productive SMEs did not attain half the productivity of large enterprises. Additionally, the gap expanded by approximately 1.5 p.p. in both the industrial and manufacturing sectors over the past decade. This trend could reflect that large, capital-intensive companies are increasingly outperforming smaller producers, a trend that has been attributed to large firms’ more favourable access to finance and intangible capital (Chen and Lee, 2023[12]).
The link between firm size and productivity is more tenuous in business sector services, with the average size-productivity gap across 33 OECD and accession candidate countries around 13 p.p. in 2024, meaning SMEs reached about 87% of the productivity of large firms. Overall, this gap has grown in business sector services by over 11 p.p., with SMEs’ productivity level having fallen from 98% of the level of large firms in 2013 to 87% in 2024. A notable productivity gap exists in the information and communication industry, where SMEs’ productivity was less than 64% of large firms in 2024. The gap has, however, narrowed since 2013, when SMEs achieved just 53% of large‑firm productivity.
The firm-size productivity gap is, on average, closer to parity in industries with lower value added, where SMEs in several countries outperform large firms on average. In 2013, SME productivity in transport services was, on average across countries, 46% higher than that of large firms. Administrative and support services is the only industry that consistently shows higher productivity among SMEs than among large firms across countries and over time, except in Germany. In 2024, SMEs in this industry were more than twice as productive as large firms on average across countries. However, the relative advantage of SMEs has diminished over time, reflecting faster productivity growth among large firms.
Unweighted average of turnover over total employment across 33 OECD and accession candidate countries
Note: The figure shows the unweighted average for 2024 as bars, the interquartile range (between the 25th and 75th percentiles), and the unweighted average for 2013 as a diamond, by economic activity. Distribution statistics (i.e. the median and percentiles) are computed across 33 OECD and accession candidate countries. The Business sector corresponds to ISIC revision 4 Sections B to J and M to N, the Industry (except construction) to B to E, Manufacturing C, Construction F, Business sector services G to J and M to N, Wholesale and retail G, Transport and storage H, Accommodation and food I, Information and communication to J, Professional, scientific and technical activities to M and Administrative and support activities to N. SMEs are firms with 1 to 249 persons employed, except in the United States, where they are firms with 1 to 299 persons employed.
Source: OECD SDBS database, Business population estimates of the UK Department of Business and Trade, and U.S. Census Bureau Statistics of US Businesses.
Furthermore, productivity in the business sector generally increases with firm size. Micro firms in this sector have less than half the productivity of large firms on average across countries, while medium-sized firms are nearly as productive, reaching up to 96% at the 75th percentile.
However, significant differences exist between SMEs in manufacturing and business services (Figure 5.3). In 2024, the most productive manufacturing SMEs at the 75th percentile achieved only about 61% of the productivity of large firms, with micro firms averaging only a quarter of that. In the business services sector, a similar pattern occurred, though with some differences. Medium-sized firms, with 50 to 249 employees, were, on average, 23% more productive than large firms across countries. Small firms were close to parity with large firms, with only a three-percentage-point difference.
This suggests that, while size matters for average productivity levels of firms, it explains only a limited portion of the overall variability in productivity across firms – a point explored further in Box 5.2.
Unweighted average of turnover over total employment across 33 OECD and accession candidate countries by size category
Note: The figure shows the unweighted average for 2024 as bars, the interquartile range (between the 25th and 75th percentiles), and the unweighted average for 2013 as a diamond, by economic activity and size. Distribution statistics (i.e. the median and percentiles) are computed across 33 OECD and accession candidate countries. The business sector corresponds to ISIC revision 4 Sections B to J and M to N, Manufacturing C, Business sector services G to J and M to N. SMEs are firms with 1 to 249 persons employed, except in the United States, where they are firms with 1 to 299 persons employed.
Source: OECD SDBS database, Business population estimates of the UK Department of Business and Trade, and U.S. Census Bureau Statistics of US Businesses.
Using value-added-based labour productivity (as opposed to turnover-based measures used above), firm-level data reveal large productivity performance gaps among firms operating in the same industry. The OECD MultiProd project, a collaboration of the OECD Directorate for Science, Technology and Innovation with experts in participating countries, exploits firm-level data across countries to provide granular insights into productivity differences and complements conventional productivity statistics by industry and firm size. The project complements conventional productivity statistics by using data that either cover the full population of firms or by reweighting representative samples to reflect the demographic structure of the business population. The evidence from MultiProd indicates that the 10% most productive firms are, on average across 15 OECD countries, around seven times more productive than the 10% least productive ones within the same country and industry. However, size explains only a small share of these productivity gaps (OECD, 2026, forthcoming[13]).
Productivity gaps are large even among firms within similar size groups in the same industry. Further to differences between size groups documented in this chapter, the interquartile ranges, for the period between 2010 and 2023, illustrate that firms at the 75th percentile within each size group are more than twice as productive as those at the 25th percentile within the same group (Figure 5.4). While larger firms tend to be more productive on average, highly productive micro firms (2-4 workers) can outperform the median small firm (20-49 workers) and approach the productivity of the median firm with 250 or more workers. Conversely, firms in the bottom quartile of the largest size class (250+) are less productive than the median small firm.
Productivity differences within size groups account for the vast majority of productivity dispersion within industries. A variance decomposition shows that on average, the within-size group component (capturing variation among firms of comparable size) explains around 95% of total variance of productivity within a given industry, while differences in average productivity between size groups play a relatively minor role. This share varies somewhat across industries and is lower where productivity is more strongly related to firm size, such as in the case of manufacturing and information and telecommunications (where the within-group component accounts for approximately 91.5% and 92.5% of total variance, respectively).
These gaps within size groups point to risks of resource misallocation and barriers to firm growth and productivity catch-up. The coexistence of low-productivity large firms and high-productivity small firms undermines the contribution of allocative efficiency to aggregate productivity. These patterns might, to some extent, point to constraints – potentially related to capabilities or incentives – that prevent more efficient firms from scaling up, or less productive ones from downsizing or exiting. At the same time, these gaps highlight the potential for stronger diffusion of technologies, knowledge and best practices among firms of the same size operating within the same industries. This implies that while policies should account for the specific characteristics of micro, small, medium, and large firms, targeting policies based on size alone is not sufficient to effectively address productivity gaps.
Average, median and interquartile range of labour productivity by size groups, relative to the average productivity of firms with 20-49 workers; 2010-2023 (or latest year)
Note: The figure shows the average, median, and interquartile range (between the 25th and 75th percentile) of labour productivity, by firm size. Distribution statistics (i.e. the average and percentiles) are computed across firms within country-industry (SNA A38)-year and firm size or age class. The statistics are normalised by the average productivity in the reference category (firms with 20-49 workers) and are then, for each size class, aggregated within the country-year using industries’ share of total employment as weights. The figure reports an unweighted average over years (for the period from 2010 to the latest available year, which ranges from 2021 to 2023, and 2019 for Croatia and Italy) and across countries. Labour productivity refers to value-added per worker. Depending on the country, employment is measured as either headcounts or full-time equivalents, of either persons engaged or employees – in case multiple measures exist, preference is given to headcounts of persons engaged for computing LP as this is more widely available across countries (more information in OECD 2026 referenced below). Data cover manufacturing (section C of ISIC Rev. 4, excluding division 19 – “Coke and refined petroleum”) and non-financial market services (sections G-N, excluding section K - financial services - and section L - real estate) for 15 countries: Belgium, Canada, Croatia, Estonia, Finland, France, Hungary, Italy, Lithuania, Latvia, Norway, Portugal, Slovenia, Spain, United Kingdom.
Source: Based on OECD (2026, forthcoming[13]).
Across the OECD, productivity gaps between less and more developed regions within countries are as wide as productivity gaps among countries (OECD, 2023[14]). Moving from labour productivity to MFP using Orbis data, this part of the chapter focusses on regions at level two of the OECD territorial classification, i.e. typically the highest administrative level under the national level. Across several large OECD countries, such as Germany, the United Kingdom and the United States, productivity in the most productive region is about twice as high as in the least productive region. This within-country pattern holds also for Italy and Spain, where less developed regions, defined as those with GDP per capita below 75% of the EU average, remain about 30% less productive than more developed regions with GDP per capita above 100% of EU average, with no significant convergence observed since 1995.
An analysis of the financial statements of incorporated firms helps to determine whether the regional productivity gap reflects differences in the composition of firms between places, e.g. by sector, size or age. This information relies on Orbis database, which has been widely used to analyse firm-level productivity dynamics across countries (Gal and Hijzen, 2016[15]; OECD, 2021[16]; OECD, 2024[17]). A regional average of firm-level labour productivity (value added over employment) from the Orbis database is strongly correlated with labour productivity statistics from the OECD regional economic statistics database, and closely aligned in monetary levels (Menon and Vermeulen, 2026[18]). This reflects the good coverage of firm-level data in Orbis for Italy and Spain (Bajgar et al., 2020[19]). With the rich information on firms’ balance sheets in Orbis, however, the analysis can move from labour productivity to multifactor productivity (MFP) (Gal, 2013[20]). For manufacturing and tradeable sectors, firm-level MFP estimates are used to investigate if regional productivity gaps exist within firm-size groups (Menon and Vermeulen, 2026[18]). Firms in manufacturing and tradeable services sectors provide the most reliable estimates on MFP as these sectors are dominated by private sector firms for which observed revenues are based on market prices.
There is a 20-30% multifactor productivity gap between firms in more and less developed regions in Italy and Spain, where estimates vary within that range across firm size. In Italy, small firms with less than 50 employees and medium sized firms with 50 to 99 employees are 20% less productive in less developed regions than in developed regions (Figure 5.5). Even micro-firms, those with less than 10 employees, show a similar productivity gap relative to micro-firms in developed regions. The largest firms in Italy, those with 100 or more employees, have a larger productivity gap, amounting to over 25%. In Spain, a similar picture emerges, but the smallest firms have a larger productivity gap than the other size classes.
Percentage point gap of firms in less developed regions in multifactor productivity by size class, 2019
Note: Point estimates and standard errors from an OLS regression on firm level log MFP controlling for, log of age, age categories and two-digit industries. The coefficients and standard errors are transformed to percentage terms using the delta method.
Source: Menon and Vermeulen (2026[18]).
References
[5] Aguilera, R. et al. (2024), “Ownership, Control, and Productivity: Family Firms in Comparative Perspective”, Journal of Management, Vol. 51/8, pp. 3321-3351, https://doi.org/10.1177/01492063241259964.
[19] Bajgar, M. et al. (2020), “Coverage and representativeness of Orbis data”, OECD Science, Technology and Industry Working Papers, No. 2020/06, OECD Publishing, Paris, https://doi.org/10.1787/c7bdaa03-en.
[9] Bartelsman, E., J. Haltiwanger and S. Scarpetta (2013), “Cross-Country Differences in Productivity: The Role of Allocation and Selection”, American Economic Review, Vol. 103/1, pp. 305-334, https://doi.org/10.1257/aer.103.1.305.
[10] Berlingieri, G., S. Calligaris and C. Criscuolo (2018), “The Productivity-Wage Premium: Does Size Still Matter in a Service Economy?”, AEA Papers and Proceedings, Vol. 108, pp. 328-333, https://doi.org/10.1257/pandp.20181068.
[6] Bernard, A. et al. (2018), “Global Firms”, Journal of Economic Literature, Vol. 56/2, pp. 565-619, https://doi.org/10.1257/jel.20160792.
[12] Chen, S. and D. Lee (2023), “Small and vulnerable: SME productivity in the great productivity slowdown”, Journal of Financial Economics, Vol. 147/1, pp. 49-74, https://doi.org/10.1016/j.jfineco.2022.09.007.
[3] Criscuolo, C., P. Gal and L. Freund (2024), “Human capital at work: Five facts about the role of skills for firm productivity, growth, and wage inequality”, OECD Productivity Working Papers, No. 36, OECD Publishing, Paris, https://doi.org/10.1787/486c0289-en.
[4] Criscuolo, C. et al. (2021), “The human side of productivity: Uncovering the role of skills and diversity for firm productivity”, OECD Productivity Working Papers, No. 29, OECD Publishing, Paris, https://doi.org/10.1787/5f391ba9-en.
[1] Cunningham, C. et al. (2022), “Dispersion in Dispersion: Measuring Establishment‐Level Differences in Productivity”, Review of Income and Wealth, Vol. 69/4, pp. 999-1032, https://doi.org/10.1111/roiw.12616.
[2] Ferrando, A. and A. Ruggieri (2018), “Financial constraints and productivity: Evidence from euro area companies”, International Journal of Finance & Economics, Vol. 23/3, pp. 257-282, https://doi.org/10.1002/ijfe.1615.
[20] Gal, P. (2013), “Measuring total factor productivity at the firm level using OECD-ORBIS”, OECD Economics Department Working Papers, No. 1049, OECD Publishing, Paris, https://doi.org/10.1787/5k46dsb25ls6-en.
[15] Gal, P. and A. Hijzen (2016), “The short-term impact of product market reforms: A cross-country firm-level analysis”, OECD Economics Department Working Papers, No. 1311, OECD Publishing, Paris, https://doi.org/10.1787/5jlv2jm07djl-en.
[7] Marchese, M. et al. (2019), “Enhancing SME productivity: Policy highlights on the role of managerial skills, workforce skills and business linkages”, OECD SME and Entrepreneurship Papers, No. 16, OECD Publishing, Paris, https://doi.org/10.1787/825bd8a8-en.
[18] Menon, C. and W. Vermeulen (2026), “What drives regional productivity differentials? Evidence from firm-level data in Italy and Spain”, OECD Local Economic and Employment Development (LEED) Papers, No. 2026/03, OECD Publishing, Paris, https://doi.org/10.1787/8fba4b12-en.
[11] Mourougane, A. and E. Kim (2020), “Boosting productivity in the United Kingdom’s service sectors”, OECD Economics Department Working Papers, No. 1629, OECD Publishing, Paris, https://doi.org/10.1787/78f4022e-en.
[8] OECD (2025), Unleashing SME Potential to Scale Up: Helping SMEs Scale Up, OECD Studies on SMEs and Entrepreneurship, OECD Publishing, Paris, https://doi.org/10.1787/ea948a58-en.
[17] OECD (2024), “Bringing Trentino’s productivity growth back on track: A comparison with OECD “peer” regions”, OECD Local Economic and Employment Development (LEED) Papers, No. 2024/03, OECD Publishing, Paris, https://doi.org/10.1787/0e74a691-en.
[14] OECD (2023), OECD Regional Outlook 2023: The Longstanding Geography of Inequalities, OECD Publishing, Paris, https://doi.org/10.1787/92cd40a0-en.
[16] OECD (2021), “The spatial dimension of productivity in Italian co-operatives”, OECD Local Economic and Employment Development (LEED) Papers, No. 2021/02, OECD Publishing, Paris, https://doi.org/10.1787/4f9b9d77-en.
[13] OECD (2026, forthcoming), The productivity dynamics of different laggard firms: New evidence from the OECD MultiProd project.