This chapter presents indicators and methods that provide insights into the market introduction, adoption and diffusion of knowledge for sustainable growth. It brings together measures of environmental innovation within firms from both official and experimental sources with measures of market diffusion and the use of different types of intellectual property. Indicators of start-up and venture capital activity provide a view of environmental entrepreneurship and financing. This chapter concludes with measures of the changing economic footprint of environmental technology goods production and international trade that illustrate the emergence of the “green” economy and its offering of new goods and services in support of sustainable growth.
Measuring Science and Innovation for Sustainable Growth
Abstract
The scaling-up and broad commercial uptake of new and improved solutions is key to addressing climate change, biodiversity loss and pollution. Therefore, policymakers need to measure how ideas are developed and how they can transform organisations, local markets, countries, the global economy and the very fabric of society in the face of resource and environmental challenges.
Key findings in this chapter suggest that the adoption and diffusion of technologies and practices related to environmental sustainability continue to accelerate but an uneven pace:
In the median OECD country, over 40% of innovative companies report they have introduced at least one innovation with environmental benefits in the previous three years. Canada and Korea are leaders in this area.
The incidence of environmental innovation in OECD countries varies depending on firm size and research and development (R&D) activity. It rises to 54% for large companies and to 45% for innovative R&D active companies, compared to 31% for innovative non-R&D active firms.
About one-third of small and medium-sized enterprises (SMEs) are environmentally engaged, albeit with considerable variations across countries. Medium-sized firms are nearly twice as likely to be greening as small firms and three or more times as likely as micro firms. Solar energy installations, recycling, renewables and circularity are among the most popular topics mentioned by such “greening” firms.
Environmental and energy International Organization for Standardization (ISO) standards are adopted mainly by companies in European countries, the EU27 and the United Kingdom, altogether accounting in 2023 for 45% of the total of valid ISO 14001 certificates and 80% of valid ISO 50001 certificates. Countries in North America rely more on specific standards for energy management systems.
Many climate change mitigation and adaptation patents (17.4%) cross at least one border, which is slightly above the average for all technologies (16.2%). During the surveyed period 2015-2021, technology transfer in climate change mitigation technologies, which occurs when the country where patent protection for the invention is sought is different from the inventor’s country or inventors’ countries, was heavily concentrated among high-income countries and China.
The number of patent filings within a jurisdiction reflects the degree of optimism about a given market’s future importance. The United States and the People’s Republic of China (hereafter “China”) started from a similar base in 2010, both in terms of the absolute count of patent filings in environment-related technologies and their share of total patent filings. The number of environment-related patent filings in China has since increased rapidly, coupled with an increasing share of these filings in the total. Nearly 25% of patent filings in China are related to the pursuit of improved environmental outcomes, compared to only 10% in the United States.
Trademark data from the European Union Intellectual Property Office (EUIPO), the United States Patent and Trademark Office (USPTO) and the Japan Patent Office (JPO) reveal that the share of trademarks covering climate-related goods and services has grown at a relatively steady pace since 2000. The proportion has more than tripled in the United States (from 1% to 4%) and Europe (from 2.6% to 9.7%) and has more than doubled in Japan (from 2.7% to 6%).
Of 43 surveyed countries, 33 have seen an increase in the share of green start-ups among their start-up populations.
A substantial increase in venture capital has been channelled into start-ups developing environment-related products, peaking at approximately EUR 74 billion (euros) in 2021.
The absolute volume of sustainability-relevant venture capital has grown in OECD countries, the European Union, the United States and China; however, the share going to environmental sustainability-related start-ups has only increased in China and the European Union, with growth being particularly steep in China, from just over 5% in 2010 to over 30% in 2022.
The low-carbon mobility activity area has been particularly dynamic, with significant growth of venture capital investment, particularly in China and the United States, and comparable levels of investments in both countries in terms of volume. In China, this activity area grew as a share of total venture capital from 10% in 2018 to 12% in 2022. In contrast, in the United States and the European Union, it declined in terms of share of total venture capital, from 7% to 3% and from 11% to 7%, respectively.
The results confirm the rapid emergence and consolidation of China as an important low-carbon technology market vis-à-vis OECD countries.
Measurement rationale
Addressing current and emerging economic, social and environmental challenges requires not only new ideas but also putting those ideas into practice and transforming them into products, including both goods and services, that are offered in the economy, as well as processes that can be deployed in production. Indicators of inventions based on patents presented in the previous chapter provided a view of the capacity for generating knowledge suitable to solving energy and environmental problems. However, those do not necessarily translate into measures of actual change. Several additional considerations – e.g. the need for additional investments and demand availability – constrain the practical application of those ideas. Furthermore, there are multiple pathways for companies and organisations to provide themselves with the knowledge required to transform their productive capabilities and offerings that patents may not reflect.
This calls for direct measures of capability and change and underpins the motivation for specific measures of innovation, adoption and diffusion in general (Box 3.1), and among those, innovations that have environmental benefits. Such measures can help policymakers understand economic and social changes, assess the contribution (positive or negative) of innovation to social and economic goals, and monitor and evaluate the effectiveness and efficiency of their policies. These measures also need to evolve and adapt to capture the environmental impacts of innovation.
The definition and measurement of business innovation
The concept of innovation gravitates around the notion, both the activity and the outcome, of putting new ideas into practical use. It encompasses the central role of knowledge, the requirements of novelty, utility, and value creation or preservation as goals. The implementation requirement differentiates innovation from concepts such as invention, as the latter may never be used or made available for others to use. This leads to the Oslo Manual definition of business innovation as a “new or improved product or business process (or combination) that differs significantly from the firm’s previous products or business processes and that has been introduced on the market or brought into use by the firm.” Intended for use in statistical surveys, it characterises firms by the incidence of an event – an innovation – that results from their “innovation activities”. These “include all developmental, financial and commercial activities undertaken by a firm that are intended to result in an innovation for the firm”. Rather that collecting records of individual “innovations”, statistical surveys collect information on the incidence of various types of innovations and activities. Databases record innovation features of firms (the units of analysis) and can be connected with business performance indicators for analysis.
The adoption of existing technology is a basic form of process innovation in firms
Innovation activity is not restricted to R&D. The minimum requirement for an innovation to be considered as such is that the new or improved product or business process must-have features that are significantly different and better than those in the products or processes previously offered by or used by the firm. This applies both to innovations that a firm develops by itself and those first developed by other firms, organisations or individuals, with little or no additional modification. Therefore, the definition of innovation also includes the adoption of technologies and practices that imply diffusion processes. To distinguish simple adopters from other innovative firms, firms can be statistically profiled by combining information on these different elements.
Source: Authors, based on OECD/Eurostat (2018[1]), Oslo Manual 2018: Guidelines for Collecting, Reporting and Using Data on Innovation, 4th Edition, https://doi.org/10.1787/9789264304604-en.
Measures of business innovation with environmental benefits
Innovation has often been regarded as “too fuzzy” a concept to be measured and accounted for in a systematic fashion. However, statistical methods exist (OECD/Eurostat, 2018[2]) that make this challenge manageable at the level of economic systems and enable better interpretation of indicators such as R&D or inventive activity, which do not directly show whether and how scientific and technological knowledge is put into practice. Innovation surveys provide a method of collecting and reporting data on business innovations and can be customised to identify properties of innovations that are relevant to environmental goals (Box 3.2).
While it is reasonable to presume that businesses only introduce innovations if they perceive they are to their benefit and add value, the core definition of innovation does not assume that it has a positive value for the firm, let alone for society as a whole. Such neutrality helps prevent speculative value judgements in measurement and reduces the scope for assessing directionality and impacts if not complemented by additional measures. While there is no international consensus on the precise recurring measurement and reporting of innovation according to its societal impacts, several countries have collected data that help identify companies that have introduced innovation with environmental benefits across a spectrum of potential impacts.
Innovation with environmental benefits (IWEB)
Paraphrasing the general definition of innovation, an “IWEB” has been defined by Eurostat and several countries as referring “to a new or enhanced product or business process of the enterprise that produces favourable environmental outcomes or reduces adverse impacts when contrasted with the firms’ earlier products or business processes, and that is accessible to potential users or has been put into practice.” National surveys measure IWEB through items covering different types of environmental benefits, from the perspective of where such benefits are realised, their nature and significance:
Environmental benefits within the enterprise (i.e. within business processes and supply chain):
reduced material, resources (other than energy) or water use per unit of output
reduced energy use or CO2 “footprint” (i.e. reduced total CO2 emissions)
reduced soil, noise, water or air pollution OR replaced a share of materials with less polluting or hazardous substitutes OR recycled waste, water, or materials
replaced a share of fossil energy with renewable energy sources.
Environmental benefits realised during the consumption or use of a good or service (end user):
reduced energy use or CO2 “footprint”
reduced air, water, soil or noise pollution
facilitated recycling of the product after use OR extended product life and durability.
Source: Authors, based on OECD (2023[3]), OECD Business Innovation Indicators: Statistical Highlights, https://www.oecd-ilibrary.org/content/dam/oecd/en/data/datasets/business-innovation-statistics-and-indicators/business-innovation-indicators-2023-highlights.pdf.
The 2023 OECD Business Innovation Statistics reported for the first time on the incidence of IWEBs (or “environmental innovation” for short). For the median country responding to the OECD data call, slightly over one-third of innovative firms report having introduced at least one environmental innovation in 2018 20 (Figure 3.1). The incidence of environmental innovation varies depending on firm size and R&D activity. It rises to 54% for large companies and 45% for innovative R&D active companies, compared to 31% for innovative non-R&D active firms. Environmental innovation appears to be slightly more oriented towards innovations that yield environmental benefits within the enterprise and its supply chain (31%) than innovations whose environmental benefits are realised downstream by consumers or end users (25%). This finding is consistent with the pattern that general business process innovations are more common than product innovations. Judging by co-occurrence patterns, both types of environmental innovations appear to be complementary.
By type of environmental innovation (IWEB), as a percentage of innovative firms
Note: Innovative firms report one or more innovations in the reference period (2018-20). Innovations with environmental benefits can be obtained either within the enterprise's business processes and supply chain (Internal/Upstream) or generated during the use of the enterprise’s goods and services by consumers or end users (User/Downstream), or both. See Box 3.2 for details.
Source: Authors, based on OECD (2023[3]), OECD Business Innovation Indicators: Statistical Highlights, https://www.oecd-ilibrary.org/content/dam/oecd/en/data/datasets/business-innovation-statistics-and-indicators/business-innovation-indicators-2023-highlights.pdf.
The environmental innovation profile of five countries that reported more detailed information on the various types of environmental innovation provides further insights on observed international differences, helping to make the case for further internationally co-ordinated work in this area (Figure 3.2). For example, Canada reports high performance across all types of environmental benefits, with the exception of internal use renewables, where Germany and Portugal attain higher innovation rates. Across all countries depicted, innovations resulting in pollution reduction are more common for internal and supply chain processes than in the downstream.
This publication extends the indicators previously published by the OECD with additional analysis of environmental innovation patterns. Analysing data published by Eurostat at the level of European countries, industries and firm size, it is possible to show that there is no one-size-fits-all when it comes to adopting innovations with environmental benefits while accounting for other systematic differences between countries, including industrial composition.
Firms in different industries tend to specialise in different types of innovations with environmental benefits. Firms in the energy sector are the most likely to introduce the most types of environmental innovations, except for innovations such as recycled waste, water, materials and products that are mostly introduced in the water supply, sewerage, waste management and remediation activities sector and innovations extending product life, which are mainly introduced in the manufacturing sector (Figure 3.3).
By detailed type of environmental innovation, as a percentage of innovative firms within each country
Note: Innovative firms report one or more innovations in the reference period (2018-20). See notes to Figure 3.1.
Source: Authors, based on OECD (2023[3]), OECD Business Innovation Indicators: Statistical Highlights, https://www.oecd-ilibrary.org/content/dam/oecd/en/data/datasets/business-innovation-statistics-and-indicators/business-innovation-indicators-2023-highlights.pdf and Eurostat (2023[4]), Community Innovation Survey 2020 (CIS2020), https://ec.europa.eu/eurostat/cache/metadata/en/inn_cis12_esms.htm.
Relative likelihood of introducing a given type of environmental innovation, by broad industry group
Note: Values displayed depict standardised (min-max standardisation) industry fixed effects of the probability of an environmental innovation among innovation-active companies in each country*industry group*firm-size group, also controlling for country and firm-size effect.
Source: OECD analysis of semi-aggregated innovation data at country, industry and firm-size levels, based on Eurostat (2023[4]), Community innovation survey 2020 (CIS2020), https://ec.europa.eu/eurostat/cache/metadata/en/inn_cis12_esms.htm.
There are notable intra-sectoral differences within the services sector. Firms operating in transportation and in wholesale and retail trade subsectors have a higher probability of introducing innovations with environmental benefits compared to the average services firm. Conversely, firms in financial and insurance activities, as well as information and communication subsectors, show a lower likelihood of adopting such innovations relative to the sector average. In contrast, manufacturing subsectors exhibit more homogeneity in their probabilities to introduce innovations with environmental benefits. However, certain hard-to-abate subsectors (e.g. manufacture of basic metals and fabricated metal products) demonstrate a lower probability than the average manufacturing firm (Figure 3.4).
Relative average likelihood of introducing any type of environmental innovation, by detailed industry group
Note: Panel A represents the average probability of an innovation with environmental benefits among innovation-active companies in services subsectors, compared to the total average probability for firms in services sector. Panel B represents the average probability of an innovation with environmental benefits among innovation-active companies in manufacturing subsectors, compared to the total average probability for firms in the manufacturing sector. Country and firm-size fixed effects are accounted for.
Source: OECD analysis of semi-aggregated innovation data at country, industry and firm-size levels, based on Eurostat (2023[4]), Community Innovation Survey 2020 (CIS2020), https://ec.europa.eu/eurostat/cache/metadata/en/inn_cis12_esms.htm.
Innovation survey data lends itself to empirical analysis of factors underpinning environmental innovation. The empirical literature on the determinants of environmental innovations accentuates the role of cost savings, customer benefits and the role of policies like regulation. For example, in their study of German firms, Horbach, Rammer and Rennings (2011[5]) found that customer requirements are another important source for environmental innovations, particularly with regard to products with improved environmental performance and process innovations that increase material efficiency, reduce energy consumption and waste, and the use of dangerous substances. A more recent study of Irish companies by Siedschlag et al. (2019[6]) of the propensity of firms to introduce environmental innovations suggests that environmental regulations, in-house R&D and acquisition of capital assets are major drivers of environmental innovation and confirms that larger firms are more likely to introduce “green” innovations. Chapter 4 provides further examples of how innovation survey data can be used to assess the impact of governmental innovation support policies. This report also assesses some of the main limitations of innovation survey data (Box 3.3) and makes recommendations under its measurement agenda in Chapter 6.
There are several challenges when implementing questions on innovations with environmental benefits (IWEBs) and interpreting results. First, the array of potential environmental benefits must be comprehensively laid out, with input from environmental and energy policy experts. Given the breadth of possible impacts, missing out on specific items or introducing duplications is easy. For example, several countries have left out of the set of possible choices offered to respondents for reporting innovations, resulting in reduced material resources and water use by end users. Questions can sometimes refer to environmental benefits on an absolute basis or on a relative efficiency basis. Firms may also face some ambiguity when interpreting benefits that are generated within a firm’s own processes, for instance, if they work with suppliers to ensure a reduced environmental footprint.
Firms’ responses are not always consistent with responses to general innovation questions. This may reflect the role of providing prompts on concrete and understandable impacts of innovation activity. Biases can arise if firms apply different novelty thresholds in the presence of significant environmental benefits or a perceived duty to appear “green”. Cognitive testing can help fine-tune the formulation of questions and develop a good appreciation of what questions companies of different types are prepared to answer as intended.
There are additional interpretation considerations once survey results are available. It is important to recall that innovation indicators only capture the launch of new products and the adoption of new business processes within a reference period but do not measure the extent to which a firm’s capability allows it to operate in relation to its full potential to deliver environmental benefits. Such capability may have been implemented before the reference period, so the indicators may capture technological catch-up among laggards.
Furthermore, as noted in Chapter 1 and Kemp et al. (2019[7]), the achievement of natural resource or environmental efficiency and efficacy improvements at the level of an individual firm does not necessarily translate into aggregated outcomes at the level of an industry, region or country. For example, energy or resource efficiency improvements can drive an increase in demand that, in turn, drives increases in overall resource consumption.
Source: Authors’ own elaboration.
Measures of technology adoption and capability
There is growing policy attention to understanding the ability of firms to use or develop emerging and enabling technologies, particularly those with applications across multiple industries and those with high transformational potential. At present, the only technology domain for which there is systematic and co‑ordinated, regular survey-based measurement of adoption across most OECD countries is that of information and communication technologies (ICTs). However, several countries have, at some point, put in place measurement initiatives focused on other technologies, and some among them are also conducting them with some regularity.
Policy interest in adopting and using advanced green technologies has prompted Canada to include questions on adopting and using “clean technologies” in successive waves of its survey of Advanced Technology conducted in 2017, 2019 and 20221. The Canadian survey defines clean technologies as “processes, devices or applications designed to mitigate the effects of human activity on the environment or promote the sustainability of ecosystems” (Box 3.4).
Canada’s Survey of Advanced Technology provides statistical information on the use of advanced technologies by Canadian firms in all industries except for personal and non-market services. The information compiled from this survey is used by the Canadian and provincial governments to better understand innovation activities linked to the development and adoption of technology, as well as to inform how policies can help businesses improve their productivity and competitiveness by using the technology.
Main questions
The survey includes a multi-item matrix-based question regarding the actual, past or planned use of an extensive list of clean technologies, comprising: 1) air and environmental protection or remediation; 2) waste management, reduction or recycling; 3) water or wastewater treatment; 4) alternative fuels; 5) non-emitting energy supply, e.g. solar, wind, hydro, nuclear; 6) bio-products; 7) smart grid; 8) energy storage; 9) energy management and efficiency improvements; 10) water management or recycling; 11) agriculture, aquaculture, forestry or biodiversity improvements; 12) sustainable mining; 13) energy-efficient transportation; 14) energy-efficient equipment or appliances; and 15) advanced or lightweight materials.
Firms are also asked about whether the use of clean technologies has resulted in the achievement of a list of possible objectives, comprising: 1) develop new or improved products or extended product range; 2) develop new or improved processes or operations; 3) increase market share or improve or develop new marketing strategies; 4) develop new or improved organisational practices or organisational flexibility; 5) improve competitiveness; and 6) compliance with regulatory standards.
Finally, the survey section on clean technologies includes a question on the main obstacles to adoption and measures adopted to address those.
Source: Authors, based on Statistics Canada (2022[8]), Survey of Advanced Technology, 2022, https://www.statcan.gc.ca/en/statistical-programs/instrument/4223_Q1_V1.
The 2022 survey found that advanced design and information control technologies (35.0%) and clean technologies (33.4%) were the most commonly adopted technologies. The utilities sector has the highest rate of respondents who have used clean technology, averaged across all of the technologies (18.8%), followed by agriculture, forestry, fishing (11.2%) and mining, quarrying, and oil and gas extraction (11.1%). Averaged across all sectors, the three most used clean technology types are technologies related to waste management, reduction or recycling (26.0%), air and environmental protection or remediation (11.5%) and energy-efficient equipment and appliances (11.0). As shown in Figure 3.6, a significant fraction of firms not using relevant green technologies plan to adopt them in the coming two years, particularly those aiming to increase resource efficiency.
Notes: Percentage of enterprises that have “used the technologies”. Non-applicable responses treated as not used.
Source: Authors, based on Statistics Canada (2023[9]), Adoption of clean technologies, by industry and enterprise size, Table: 27-10-0362-01, https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=2710036201.
Companies planning to adopt green technologies in the coming two years, as percentage of companies in surveyed industries not using relevant applicable technologies in 2022
Source: Authors, based on Statistics Canada (2023[9]), Adoption of clean technologies, by industry and enterprise size, Table: 27-10-0362-01, https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=2710036201.
Experimental web-based indicators of “greening” practices in firms
The Internet is a rich source of information on the activities of firms. Firms’ websites are a particularly interesting “big data” source for innovation research, as they publicly convey information about potentially innovative products, services and co-operation with other firms (Kinne and Axenbeck, 2020[10]; Rammer and Es-Sadki, 2023[11]). Extracting innovation-related information from these websites' content can provide researchers and policymakers with a cost-effective way to survey millions of businesses and gain insights into their innovation activity, co-operation, and applied technologies. Although public web disclosure may be biased in different directions (with the possibility of both ‘greenwashing’ and ‘greenhushing’), business websites convey information about firms in ways that statistical surveys, which must guarantee confidentiality, cannot. There is emerging innovation measurement literature that works to develop and test innovation and technology adoption indicators, drawing on business websites and the use of text mining to explore business activities.
Although not focused on innovation or technology adoption, a relevant recent OECD study (Wildnerova et al., 2024[12]) has developed an experimental indicator to identify environmental engagement, also referred to as “greening”, using machine learning to analyse the content of over 1 million websites of firms from 15 OECD countries. Greening is identified based on firms’ self-declared information about products or processes on their websites (Box 3.5).
Using information available on 1 million corporate websites, the indicator developed by Wildnerova et al. (2024[12]) provides information on firms’ (self-declared) environmental strategies and sustainable initiatives. The approach takes segments of text from a sample of firms’ websites, manually classified to “greening” actions, which form the basis of an automated machine learning model applied to nearly 1 million firms. The “action” keyword is identified as some concrete solution for improving the environmental footprint of the firm, as opposed to “generic” keywords such as environment, ISO 14001, or sustainable. The keywords and their variations are considered in all languages. Based on this approach, 36% firms in the sample are identified as environmentally engaged, meaning that they mention their environmental engagement on the website at least once. The indicator is available for nearly one-third of small firms and half of the medium-sized firms in the private business sector of 15 OECD countries in 2022. In some countries, the coverage of the indicator is substantially below that of others. If the selection of the firms favours more environmentally engaged firms, the indicator may overstate the performance of less well-represented countries.
Source: OECD, based on Wildnerova et al. (2024[12]), “Which SMEs are greening?: Cross-country evidence from one million websites”, https://doi.org/10.1787/ddd00999-en.
About one-third of SMEs are environmentally engaged, albeit with considerable variations across countries. Medium-sized firms are nearly twice as likely to be greening as small firms and three or more times as likely as micro firms. There is a large gap in the share of greening firms between micro or small SMEs and larger firms across all countries in the sample. On average, across countries, the share of small firms with a working website that declares some environmental engagement (38%) is about 14 percentage points lower than the share of medium-sized firms that are greening (52%) (Figure 3.7).
As a percentage of all firms with functioning websites in 2022
Note: Based on a total of 904 817 observations.
Source: OECD calculations, based on OECD-Orbis database and ISTARI.ai web-scraped data. Extracted from Wildnerova et al. (2024[12]), “Which SMEs are greening?: Cross-country evidence from one million websites”, https://doi.org/10.1787/ddd00999-en.
Greening SMEs are more productive, pay higher wages, and their sales grow faster than non-greening SMEs. Solar energy installations, recycling, renewables, and circularity are among the most popular topics greening firms mention (Figure 3.8). The analysis of the keywords mentioned on the greening SMEs’ websites illustrates the diffusion of the different greening strategies.
As a percentage of all environmentally engaged firms
Note: See Box 3.5.
Source: OECD calculations, based on OECD-Orbis database and ISTARI.ai web-scraped data. Extracted from Wildnerova et al. (2024[12]), “Which SMEs are greening?: Cross-country evidence from one million websites”, https://doi.org/10.1787/ddd00999-en.
Measurement rationale
Environmental standards play a crucial role in driving the adoption of cleaner and more sustainable technologies. By setting clear, common guidelines for reducing emissions, waste and resource consumption in a demonstrable fashion, they can encourage business investment in innovations, such as energy-efficient systems, renewable energy sources and waste reduction technologies (Box 3.6). Environmental standards create market demand for sustainable technologies when they set industry-wide benchmarks that facilitate co-ordinated action and interoperability without locking industries into outdated technology. Companies that adhere to these standards can gain a competitive advantage by demonstrating their commitment to green practices, attracting environmentally conscious consumers and investors. Additionally, governments and regulatory bodies may offer incentives, subsidies or grants for adopting compliant technologies, further accelerating innovation. Standard setting is particularly important for certain environmental and energy technologies with network externalities (Vollebergh and van der Werf, 2014[13]), e.g. green hydrogen or electrical vehicle charging infrastructure. As a result, standards act as catalysts for technological transformation, encouraging industries to develop and implement solutions that reduce environmental impact while improving long-term business sustainability.
Definition and main types
A standard is a document that specifies characteristics of technical design or rules of behaviour. A wide range of standards exists, all of which can contribute to the innovation and diffusion of eco-innovation. Examples include:
Standards for measurement and reference: Benchmarks allow more efficient information processing by reducing transaction costs and mitigating information asymmetries. Common measurement criteria help businesses and consumers assess products. For example, as technical standards, energy labels help consumers identify key product features and enable producers to communicate compliance (see Chapter 5). Standards such as the Global Reporting Initiative, the Task Force on Climate-related Financial Disclosures, and the Sustainability Accounting Standards Board ensure transparency and accountability in corporate sustainability efforts.
Standards for (minimum) quality and safety: This includes regulatory standards like environmental policies, which can be technical or behavioural in nature. Their design – whether targeting emissions, inputs or outputs – affects not only a company’s operations and profitability but also broader social welfare. Quality and safety standards for technologies and products may require specific industrial equipment or set product specifications. Behavioural standards define acceptable input, output or emission levels, allowing businesses flexibility on how to comply.
Compatibility and interface standards: Also referred to as interoperability standards, these are crucial in industries reliant on interconnected networks, such as electricity or natural gas infrastructure, where producers and consumers rely on compatible technical systems for product delivery. Promoting the uptake of green hydrogen requires standardisation of guarantees of origin, hydrogen purity, design of liquefaction/conversion and regasification/reconversion facilities, equipment specifications and integration into the gas grid. Plugs for electric cars across vehicles and charging stations also fall under this category.
Source: Authors, based on Vollenbergh and van der Werf (2014[13]), “The role of standards in eco-innovation: Lessons for policymakers”, https://doi.org/10.1093/reep/reu004, and OECD (2025[14]), “Fostering convergence in SME sustainability reporting”, https://doi.org/10.1787/ffbf16fb-en.
Indicators of the adoption of ISO environmental standards
Environmental and energy-related International Standardization Organisation (ISO) standards (Box 3.7) are adopted mainly by companies in Europe, in China, in Japan and in Korea, altogether accounting in 2023 for 87% of the total of valid ISO 14001 certificates and valid ISO 50001 certificates. Countries in North America rely more on specific standards for energy management systems (e.g. Energy Star certificates in the United States), which explains the limited number of valid ISO certificates reported (Figure 3.9).
Number of valid certificates and covered sites
Note: This figure shows the number of valid ISO 14001 (environmental management systems) and ISO 50001 (energy management systems) certificates as of 31 December 2023 for the top 20 countries (representing 96% of total valid certificates for ISO 14001 and 93% for ISO 50001). The number of sites covered (i.e. permanent locations where an organisation carries out work or provides a service) is also depicted. The ISO estimates the number of certifications and covered sites using a voluntary survey of certification bodies, whose participation can fluctuate over time, so these numbers should be regarded as conservative. Moreover, high-energy demand countries such as the United States have their own specific standards for the management of energy use by companies, such as the Energy Star Program, to improve energy efficiency. This helps explain the low levels of ISO certification compared to European countries (Sousa Lira, Salgado and Beijo, 2019[15]).
Source: OECD calculations based on ISO (n.d.[16]), ISO Survey 2023, https://www.iso.org/the-iso-survey.html.
The pattern of adoption of environmental and energy ISO standards across sectors mirrors sectoral specificities in terms of energy consumption and waste generation. Companies’ sites operating in the transport, storage and communication sector and in energy-intensive manufacturing subsectors (e.g. concrete, cement, lime, plaster, etc.; basic metal and fabricated metal products; and food products, beverage and tobacco) are prevalent among sites covered by an energy management ISO standard (ISO 50001) and represent 45% of the total. For environmental management ISO standard (ISO 14001), the construction sector ranks first at 12% (Figure3.10).
Industry percentage distribution of sites covered by valid ISO 14001 and ISO 50001 certificates
Note: This figure shows the distribution of sites (i.e. permanent locations where an organisation carries out work or provides a service) covered by valid ISO 14001 (environmental management systems) and ISO 50001 (energy management systems) certificates as of 31 December 2023. The distribution is displayed for the subset of sites that can be allocated to a specific sector of the European Accreditation classification (65% for ISO 14001 and 71% for ISO 50001). Based on ISO estimates using a voluntary survey of certification bodies.
Source: OECD calculations based on ISO (n.d.[16]), ISO Survey 2023, https://www.iso.org/the-iso-survey.html.
The International Organization for Standardization (ISO) is an independent, non-governmental international organisation that develops and publishes standards to ensure quality, safety, efficiency, and interoperability across various industries. ISO energy and environmental standards, such as ISO 14001 (environmental management systems) and ISO 50001 (energy management systems), provide structured approaches for businesses to identify, manage and reduce their environmental footprints. ISO publishes an annual Survey of Certifications reporting the number of valid certificates to environmental and energy ISO management system standards worldwide. Data are provided by certification bodies accredited by the International Accreditation Forum (IAF) Multilateral Recognition Agreement (MLA) Members on a voluntary basis. Participation can fluctuate over time. For instance, China’s accreditation body did not contribute to the ISO Survey 2023, which can affect country-level results, despite reporting by other bodies for that country. Furthermore, the use of alternative standards may be significant in major economies.
Source: Authors, partly based on ISO (n.d.[16]), ISO Survey 2023, https://www.iso.org/the-iso-survey.html.
Measurement rationale
As discussed in Chapter 1, science and innovation possess unique features as a domain for policy analysis, owing to the nature of knowledge and how it is created and diffused, leaving few reliable traces. In this context, intellectual property (IP) rights serve a unique role as a marker that allows, albeit imperfectly, for invention and the adoption and diffusion of technologies to be traced across industries and markets (Box 3.8). IP rights data, particularly patents and trademarks, offer a promising alternative to survey data by providing an objective, transparent, efficient and cost-effective method for assessing firms’ innovation activities (Block et al., 2025[17]). Box 3.8. Beyond invention: Using patents to measure technology adoption and diffusion
In Chapter 2, patent data were used for indicators of knowledge creation, and therefore, patents were allocated to the inventor’s country based on fractional counts. Patent legal event data, which relate to information on the events during the lifetime of a patent application, can, however, also serve as a basis for indicators of technology adoption and diffusion, which in turn can then be deployed to analyse trends in commercialised innovation, which is more likely to deliver tangible positive environmental impact (OECD, 2010[18]). Three main types of legal events can be used to examine different facets of adoption and diffusion:
Patent filings outside the inventor’s/owner’s country: Since patenting is costly, patents tend to be filed in all patent offices where the inventor wants to protect the technology. Therefore, patenting outside the inventor’s/owner’s country provides information on all countries where the technology is expected to be used (Probst et al., 2021[19]).
Patent assignment: A patent assignment is a transfer, by a seller to a buyer, of the rights, title and interest in one or more granted patents or patent applications. A focus on the number of patent families in climate change-related technologies that undergo one or several assignments gauges the level of commercial interest in an innovation (Dussaux, Agnelli and Es-Sadki, 2023[20]).
Patent licensing: Another method to examine relevant commercialised innovation is to measure license agreements involving patents with underlying climate change-related technologies (or patents relevant to environmental innovations). Licensing deals strongly signal economic interest in a particular technology (Dussaux, Agnelli and Es-Sadki, 2023[21]). Data on licensing are, however, limited.
Patent-based measures of technology diffusion in markets
Diffusion based on patent filings across jurisdictions
Diffusion of energy and environment-related technologies can be proxied by the count of inventions that have sought patent protection in a jurisdiction other than the inventors’ (as evidenced by registered patent applications, not necessarily by granted patents) based on international patent families. While rich academic literature exists using patent family counts as a measure of invention in energy and environment-related technologies, in which patent families tend to be allocated on the basis of the inventor’s or owner’s location, the potential use of patent data to measure technology diffusion in technologies relevant to sustainable growth has received limited attention thus far. Various data issues complicate this type of analysis, namely the fact that regional offices are a frequently chosen route for patenting, which in some instances precludes measurement of the number of patents filed within individual countries covered by that regional office. The uneven quality of patents and a greater propensity to initially file in one’s home market might also limit the scope for international comparison without appropriate analytical adjustments.
Indicators in Probst et al. (2021[19]) and Dechezleprêtre et al. (2020[22]) provide valuable insights with respect to the diffusion of climate change mitigation and adaptation technologies. The former contribution shows the geographical distribution of cross-country invention transfers by country income groups. Updating this indicator for the period 2015-21 shows that technology transfer in climate change mitigation technologies continues to be heavily concentrated among high-income countries (Figure 3.11) and China, confirming the findings of Probst et al. (2021[19]).
Patents filed in a jurisdiction other than the inventor’s
Note: A transfer has occurred when the jurisdiction (right) where patent protection for the invention is sought is different from the inventor’s country or inventors’ countries (left). Patents filed at the European Patent Organisation (EPO) by one of European Patent Convention (EPC) members are not considered as a transfer. Only flows in excess of 5000 are shown.
Source: OECD, STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats, June 2025.
The United States and China lead on the recipient side in the period 2015-21, with 150 097 and 123 716 patents filed, respectively, by inventors from other countries. The two largest flows of climate change mitigation and adaptation technologies occurred between Japan and the US and Japan and China (40 729 and 37443 patent filings, respectively).
Dechezleprêtre et al. (2020[22]) compare technology transfer shares for three technology groups: climate change adaptation, climate change mitigation, and all technologies during the period 2010–15. They find that only 17% of adaptation inventions cross at least one border, which is significantly below the average for all technologies (24%) and about half that of mitigation technologies (31%) Updating this indicator for the period 2015-2021 (Figure 3.12) shows that climate change mitigation are still more likely to cross at least one border than the average, while climate change adaptation technologies substantially less. The differences are generally narrower, however (17.4% for climate change mitigation technologies, 16.2% average across all technologies and 9.7% for climate change adaptation technologies). The lower rates of technology transfer in climate change adaptation technologies possibly reflect the local nature of adaptation challenges and consequently, also the technologies addressing them. Given the rapid development of markets related to environmental and energy technologies, notably the increasing importance of China as a jurisdiction where relevant technologies are developed, there is a need to update the analysis regularly to examine whether there has been any change in the trend with respect to diffusion.
International patents as a share of total
Note: International patents are defined as those that are developed by inventors residing in a given country that were protected in a different country.
Source: OECD, STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats, June 2025.
Records of simple patent filings allow for only a rudimentary overview of the relative importance of various markets rather than technology diffusion since they also include filings by domestic inventors; nevertheless, indicators based on filing data offers an indication of the degree of optimism with respect to the commercialisation of environment-related technologies in different geographies. As shown in (Figure 3.13), the United States and China started from a similar base in 2010, both in terms of the absolute count of patent filings in environment-related technologies and in terms of their share of total patent filings. Since then, however, the two have diverged significantly, with a rapidly increasing number of environment-related patent filings in China, which is also coupled with an increasing share of these filings in the total. Meanwhile, there has been a stagnation in the United States with respect to both measures.
Note: The indicator is based on simple counts of filings of ENV-TECH patents in the respective patent offices.
Source: OECD (2022[23]), Patents on environment technologies, https://www.oecd.org/en/data/indicators/patents-on-environment-technologies.html
Patent assignments
A complementary method for assessing the commercialisation of climate change-related innovation is to analyse patent assignments, which involve transferring ownership rights from one party to another for granted patents or pending applications. Patent assignments may happen for several reasons, but they do help identify high-value innovations that are more likely to be commercialised and utilised compared to the average patented invention. Research by Serrano (2010[24]) indicates that patents with a greater number of forward and backward citations have a significantly higher likelihood of being traded. Similarly, De Marco et al. (2017[25]) found that patents related to emerging technologies – characterised by greater technological uncertainty and proximity to basic research – are also more frequently traded.
A recent OECD working paper (Dussaux, Agnelli and Es-Sadki, 2023[21]) offers several indicators based on the patent assignment of climate change mitigation and adaptation patent families. Due to data limitations described in Box 3.9 further below, these indicators only focus on the United States. Between 2014 and 2018, 7.8% of US reassigned inventions were related to climate change, compared to only 4.5% between 2000 and 2004. This increase is observed in all sectors but is particularly significant in the energy sector, where the share of US commercialised inventions related to climate change increased from 12% in 2000‑04 to 22% in 2014-18. The second and third largest increases occurred in the transport and construction sectors, where commercialised climate change-related innovation rose by 9 and 6 percentage points, respectively (Figure 3.14).
As % of total assigned inventions, by sector
Note: The year for non-assigned patented inventions corresponds to the application year of the first patent in the family. Only the patent family’s first assignment is included. Climate change-related patented inventions are defined as patent families having at least one Y02 code. Assignments are defined as a change in ownership, notably “assignment” (ownership transferred to another entity) or “merger” (ownership change due to merger). To avoid double-counting, only the first assignment of each patent family is counted. The total “all sectors” is not equal to the sum across all individual sectors because some inventions appear in different sectors. Patented inventions are classified into six economic sectors: agriculture, construction, energy, ICT, manufacturing and transport. The classification is based on an ad hoc methodology using Cooperative Patent Classification (CPC) codes to assign an invention to the respective economic sector.
Source: Dussaux, Angnelli and Es-Sadki (2023[21]), “Exploring new metrics to measure environmental innovation”, https://doi.org/10.1787/e57a8a13-en based on USPTO (n.d.[26]), Patent Assignment Dataset, https://www.uspto.gov/ip-policy/economic-research/research-datasets/patent-assignment-dataset and OECD (n.d.[27]), STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats (accessed May 2022)
The degree of commercialisation of climate change-related technologies, measured by the share of inventions assigned at least once, is slightly lower than observed for all patents. In recent years, 7.6% of all US-patented inventions related to climate change have been assigned at least once, below 8.1% for all patented inventions (Figure 3.15). The degree of commercialisation of climate change-related innovation varies across sectors. Compared to all technologies, US climate change-related inventions in 2000-18 were relatively more commercialised in agriculture (10.1% against 8.7%) and construction (9.3% against 7.3%). In transport and energy, climate change-related innovation tends to be slightly less commercialised than the average technology (5.3% against 5.8%). In manufacturing and ICT, the degree of commercialisation does not differ significantly.
Assigned inventions as a percentage of climate and all patented inventions, within each industry
Note: See note to Figure 3.14.
Source: Dussaux, Angnelli and Es-Sadki (2023[21]), “Exploring new metrics to measure environmental innovation”, https://doi.org/10.1787/e57a8a13-en. based on USPTO (n.d.[26]), Patent Assignment Dataset, https://www.uspto.gov/ip-policy/economic-research/research-datasets/patent-assignment-dataset and OECD (n.d.[27]), STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats (accessed May 2022)
Patent owners are not required to disclose patent transactions to patent offices but may have incentives to do so. For example, patent owners facing litigation in court are only protected against subsequent assignments if they have recorded the transfers of patents at the US Patent Office (USPTO). These recordings are registered by the USPTO and made available through the USPTO Patent Assignment Dataset (UPAD). UPAD contains detailed information on 8.6 million patent assignments and other transactions recorded at the USPTO since 1970, involving roughly 14.9 million patents and patent applications. While other patent offices also keep records of patent assignments, the USPTO makes this data publicly available and user-friendly by applying a methodology categorising patent assignments according to the “nature of conveyance” (Marco et al., 2017[25])
There are several patent‐asset conveyances recorded, including assignments or reassignments, mergers and acquisitions, licenses to government and several other types (see Graham, Marco and Myers (2018[28]) for an overview). Many USPTO assignment records reflect assignments conducted in the ordinary course of business, between inventor employees and their firm employers (employer assignments) or to different units of a multinational. However, a substantial number of patent rights have been transferred to other firms. Graham, Marco and Myers (2018[28]) identified about 700 000 such patent rights transfers to other organisations from 1970 to 2014. By linking the Patent Assignment and USPTO Patent Datasets, it is possible to study assignments by technology fields and areas such as climate mitigation.
The main limitation of using patent assignments as an indicator of technology deployment, besides the coverage issues mentioned above, is how to interpret them. Transfers occur for several strategic, commercial or reporting reasons rather than for actual technology applications. Patents may be automatically reassigned when companies planning to implement their technology fail and then remain unused or as a result of “killer acquisitions” to halt a competitor’s innovation activity (Cunningham, Ederer and Ma, 2021[29]). As opposed to licensing, patent assignment transfers IP enforcement rights, which might be used to extract punitive charges or block developments by competitors who need to draw on the protected technology. These practices can lead to overestimating genuine interest in a technology based on assignment numbers.
Source: Dussaux, Angnelli and Es-Sadki (2023[21]), “Exploring new metrics to measure environmental innovation”, https://doi.org/10.1787/e57a8a13-en.
Licensing
Analysis of licensing deals involving climate change mitigation and adaptation patents (or any patent relevant for environmental innovations) is yet another approach to measuring environmental technology commercialisation. Licensing deals are closer to commercialisation than patents and signal economic interest in a particular technology. The analysis of technology commercialisation using licensing data is hindered by a lack of comprehensive data, as disclosure is generally not compulsory and agreements are often subject to non-disclosure agreements among the parties. A recent OECD working paper (Dussaux, Agnelli and Es-Sadki, 2023[20]) presents several relevant indicators based on licensing deal data from private provider ktMINE. While the data are not comprehensive (Box 3.10), they offer useful insights regarding the degree of confirmed commercial interest in environment-related innovations. Given that ktMINE data contains a fraction of total licensing deals, it is more appropriate to look at the relative share of climate change-related deals rather than the absolute number. The share of climate change-related deals has fluctuated between 7% and 21% between 1995 and 2017. The most consistent increase took place during the global financial crisis between 2007 and 2012. Despite the overall economic downturn, the share of licensing deals in climate change related technologies has increased from 12% before 2007 to 17% in 2008.
Source: Dussaux, Angnelli and Es-Sadki (2023[21]), “Exploring new metrics to measure environmental innovation”, https://doi.org/10.1787/e57a8a13-en based on ktMINE data.
The analysis in Dusseaux, Agnelli and Es-Sadki (2023[21]) presented here relies on the private database ktMINE, which gathers intangibles-related data from publicly available sources, such as the Securities and Exchange Commission, the USPTO, news sources, as well as Freedom of Information Act requests. The database links sources where firms report IP transactions. It allows users to search for particular technologies and agreements, controlling for several variables such as industry, Cooperative Patent Classification (CPC) codes, etc. The database consists of more than 120 000 agreements, of which about 20 000 contain information on royalty rates. It covers the period from about 1980 to 2025. There are different types of license agreements for which ktMINE has data, including asset purchases, distribution, joint development, cross-license, franchise, manufacturing or process. Of note, a license agreement can be of more than one type, having, for instance, a manufacturing and distribution element. Each record includes a filing date and an entry into effect date, allowing for trend analysis of licensing deals related to environmental innovation technologies.
Two ktMINE databases were used in the Dusseaux, Agnelli and Es-Sadki (2023[21]) analysis. The first includes detailed information on 913 agreements involving climate change-related technologies. The second includes information on 7 028 license agreements pertaining to all technologies (not only climate change-related) and provides the associated CPC codes for each. Detailed CPC information is available up until the sub-class level (e.g. soil working in agriculture or forestry [A01B], climate change adaptation [Y02A], climate change mitigation in buildings [Y02B], etc.). This database helps to compare climate change-related licensing deals with other technological classes.
Both ktMINE databases used in the analysis include mainly publicly disclosed licensing deals. The analysis was restricted to the deals that disclosed a patent number for which a CPC code could be identified. CPC data need to distinguish between climate and non-climate-related deals. Since disclosure is not compulsory for licensing deals, and those convey potentially sensitive information, the representativity of the sample may be limited. In addition, while the datasets collect information up to 2021, the number of observations for recent years is very low, a drawback for several patent-based metrics.
Source: Dussaux, Angnelli and Es-Sadki (2023[21]), “Exploring new metrics to measure environmental innovation”, https://doi.org/10.1787/e57a8a13-en.
International collaboration in environment-related patents
As a conduit for knowledge flows and coproduction, international collaboration is an important and complementary factor in fostering technological advancement. This is especially relevant in the context of technologies that promote global environmental public goods. This applies particularly to technologies that address environmental impacts extending beyond national boundaries, such as pollution emissions (e.g. carbon dioxide [CO2] or sulfur oxides [Sox]) or resource flows (e.g. freshwater) (Haščič and Migotto, 2015[30]). While the propensity to file a patent with at least one foreign co-inventor varies widely across countries, there appears to be no decisive difference between environment-related patents and all patents with respect to this particular measure (Figure 3.17). In some countries, environment-related Patent Cooperation Treaty (PCT) patents exhibit a higher share of international co-invention, but in others, this is lower than the average across all technologies. Among emerging economies, India engages in environmental technology patenting collaborations particularly frequently, though not more than in all technologies. By contrast, in China, international co-invention appears to play a relatively minor role in the reference period.
PCT patents with at least one foreign co-inventor, as a percentage of all patents within each economy
Note: Data refer to patent applications filed under the Patent Cooperation Treaty (PCT) by priority date and inventor’s location, using simple counts. Only OECD and G20 economies with more than 100 environment-related patents (ENV-TECH definition) are included.
Source: OECD (n.d.[27]), STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats (accessed March 2025).
Tracing ideas from the lab to the market is key for understanding the long-term impact of science, technology, as well as of innovation policies. Some of the limitations of patents as indicators of innovation output may be at least partially addressed by tracing the products that stem from patents, providing the linkages between patents and products can be established in a robust manner.
IProduct is a recent research effort led by Professor Gaétan de Rassenfosse that links products to patents using virtual patent marking. The construction of the database exploits the virtual patent marking (VPM) statute introduced in the 2011 Leahy-Smith America Invents Act. Under this act, patentees may give notice to the public that their product is patented. Web-crawling software identifies VPM sections of websites and parses the product-patent links to populate the database, which is then linked to the PATSTAT worldwide patent database, among others. Data collection and enrichment are done through user contributions.
While the beta version of the database that is currently available does not allow for the construction of robust, internationally-comparable indicators to gauge the contribution of science and innovation to environmental sustainability, it does provide helpful illustrative examples of the knowledge basis of products relevant to environmental protection. The following three products are linked to patents with the Y02 classification codes, which identify patents relevant to climate change mitigation and adaptation:
Industrial equipment Power Wave® welding platform, protected by US8581147B2 (Y02P80/10 Efficient use of energy, e.g. using compressed air or pressurised fluid as an energy carrier).
Tablet Computer Google Nexus 9, protected by US7672219B2 (Y02D30/50 Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate).
Crop Vistive® Gold Soybeans, protected by US7566813B2 (Y02E50/10 Biofuels, e.g. biodiesel).
A smaller, comprehensive subset of the database focused on medical devices, biotechnology and pharmaceutical companies has been used to study the impact of US federal funding for science. Scientific publications cited in the patents protecting the commercial products have been extracted and their funding sources identified. The findings suggest that 50 to 60% of products by biotechnology and pharmaceutical firms exploit NIH-funded research, and about 12% for medical devices. The project also found that some products available today embed science that was published about 25 years ago (de Rassenfosse and Thursby, 2019[31]).
There is some evidence, however, that not all patents are equally likely to be virtually marked. For example, De Rassenfosse (2018[32]) finds that firms are more likely to mark their products if they have a higher chance of being infringed, if they pursue an active branding strategy and if they need larger external financing. The linking of products to patents also does not fully capture application mechanisms via industrial processes not covered by product databases. Any analysis must also take into account potential variability in time lags between intellectual property protection and product launch date and allow for a minimum lag before assessing use.
Source: Authors, based on Dussaux, Angnelli and Es-Sadki (2023[21]), “Exploring new metrics to measure environmental innovation”, https://doi.org/10.1787/e57a8a13-en and IProduct dataset descriptions at IPRoduct™ | Welcome to IPRODUCT - The plateform connecting patents to products
Trademark measures of green technology market introduction
A trademark is a symbol, word or combination of signs that uniquely identifies and differentiates the goods or services of one business from those of others. Unlike R&D investments and patents, trademarks are directly linked to the commercialisation of products, as they serve to market new goods and services (Zolas, Lybbert and Bhattacharyya, 2017[33]). They also provide a way to track non-technological innovations that patents or R&D data may not capture. Trademarks are widely used across nearly all industries, including the service sector. By examining trademark trends in green technologies, policymakers, investors, and researchers can better understand how environmental innovations are moving from R&D to mainstream adoption and which markets are becoming key in environmental technology adoption.
Trademarks play an important role in marketing eco-friendly products and engendering consumer trust. Many firms are leveraging trademarks to communicate social and environmental commitments by using green, sustainable and environmentally friendly language, i.e. “reduce waste”, “save the planet”, and “circular economy”. Trademark offices have been careful to assess some of the terminology firms use to avoid greenwashing (Delmas and Burbano, 2011[34]). For example, the USPTO refuses registration of a mark that contains the terms “organic”, “natural”, or “sustainable” as these terms are considered too vague. This is because the usage of the term can affect a consumer’s purchasing decision by misdescribing the product or service in an environmentally friendly context. The use of trademark data as a basis for indicators requires a rigorous classification methodology, which determines the relevance of each trademark to environmental sustainability as well as to key technology categories (Box 3.12).
The methodology of identifying environment-related trademarks builds on two principles: 1) defining environment-related activities by building a suitable and representative climate change vocabulary with terms that represent these activities; and 2) continuously iterating and reviewing environment-related trademarks to ensure that the majority of the population is identified. The string-searching algorithm is similar to the strategy used by Nanda et al. (2015[35]), whereby the trademark description is used to search for the keywords and expressions within these strings. An example of a green trademark is one that includes the keywords “lithium” + “battery” or “low emission” + “car” in its description.
Using trademarks as an indicator of innovation also comes with challenges. Trademarks make it difficult to identify narrowly defined technologies, especially processes, as they cover bundles of products and services; some trademark descriptions are, therefore, relatively vague. For instance, a description containing recycling waste may not indicate the type of recycling technologies registered under the trademark. Another illustration is that a single trademark can cover waste-to-fuel equipment and mechanical recycling equipment. Trademarks also capture products that do not represent a technical novelty. The only criterion for registering a new trademark is the novelty of the sign itself, which must not be similar to any already registered (Millot, 2009[36]).
Source: Dechezleprêtre, Dernis and Mulligan (forthcoming[37]), Identification of Green Trademarks.
Trademark data from the European Union Intellectual Property Office (EUIPO), the United States Patent and Trademark Office (USPTO) and the Japan Patent Office (JPO) reveal that the share of trademarks covering environment-related goods and services has grown relatively steadily since 2000. The proportion has more than tripled in the United States (from 1% to 4%) and Europe (from 2.6% to 9.7%) and has more than doubled in Japan (from 2.7% to 6%) (Figure 3.18).
Environment-related trademarks as a share of total trademarks filed and environment-related PCT patent filings as a share of total PCT filings
Note: For trademark classification methodology see Box 3.12.
Source: OECD (n.d.[27]), STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats (accessed December 2022).
EUIPO consistently displays the largest share of climate-related trademarks from the three offices, with over 10 000 registered in 2021. USPTO has a lower average share of climate trademarks across the period but has the most registrations, accounting for over 17 000 trademarks in the same year.
As for patents, a decrease in environment-related trademark activity was observed around 2011‑12. However, growth appears to have picked up again in the most recent years for which data are available, while patents have not. It is difficult to assess whether this is related to changes in private R&D expenditure since, as explained in Chapter 2, private sector measures of R&D expenditure for energy and environmental goals are not widely available. Time series on R&D expenditures on energy and environment for US businesses might provide some overall proxy for OECD-wide trends. Estimates in Figure 3.19 suggest that energy-related R&D performed by companies and internally funded – including all types of energy technologies – remained constant in real terms from 2012 until 2017, after increasing from 2010 to 2012. From 2017, the trend seems to be upward but not sufficient to prevent an overall decline in the share of energy R&D over total R&D expenditure. Business R&D expenditure on environmental protection has remained flat throughout the entire period.
Source: OECD, based on the US National Center for Science and Engineering Statistics and Census Bureau, Business Enterprise Research and Development Survey (multiple years), https://www.census.gov/programs-surveys/brds.html. The energy and environmental protection application areas statistics were produced annually from 2010 - 2019. There onwards, they are produced for odd-numbered years.
The fact that diffusion and commercialisation efforts kept increasing despite lacklustre growth in R&D and patenting suggests that companies are working with available technologies or focusing their efforts on non-technological aspects of innovation.
Broken down by sector, the clean energy and low-carbon mobility sectors attract the highest shares of trademarks across all three intellectual property offices (IPOs) Figure 3.20). JPO leads in clean energy, with a substantial increase of trademark share in the sector between 2008-11 and 2018-21, while the EUIPO leads the low-carbon mobility category. While attracting the highest trademark shares in these two sectors, the USPTO has marginally larger shares than other IPOs in the pollution control sector.
Between 2018 and 2021, trademarks registered at the JPO increased rapidly in the energy storage and clean energy sectors. In both instances the share of trademarks attributed to these sectors increased by more than 10 percentage points. While the overall share of JPO trademarks in low-carbon mobility remains strong at 18.3%, a substantial decrease from the 27.7% was seen in the previous period. This is driven by an overall increase in green trademarks rather than a drop-off in the registration of low-carbon mobility trademarks. The EUIPO saw an increase in low-carbon mobility registrations between the two periods, both in share and volume of registrations. The share increased from 18.1% to almost 28%.
As a share of all environment-related trademarks
Note: For classification methodology see Box 3.12
Source: OECD (n.d.[27]), STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats (accessed December 2022).
Measurement rationale
Innovative start-ups are major contributors to the development of radically new technologies and services that can help transform production and consumption patterns (Criscuolo and Menon, 2014[38]; Andrews, Criscuolo and Menon, 2014[39]; Akcigit and Goldschlag, 2024[40]). As such, they are likely to play an important role in enabling the transition to a more resource-efficient economy, as is the venture capital ecosystem that typically fuels their growth (Dalla Fontana and Nanda, 2023[41]; Croce et al., 2024[42]). It is generally understood that venture capital has significant limitations with respect to the type of technologies it will support, which potentially constrains its role in fostering radical technological change (Lerner and Nanda, 2020[43]). Nevertheless, the role of innovative start-ups and venture capital in enabling the sustainability transition is likely to grow going forward against the backdrop of digitalisation (Greenstein, Lerner and Stern, 2013[44]) and sector-specific developments, such as increasing reliance on modular technologies in the energy sector (Popp et al., 2020[45]).
Indicators of environment-related entrepreneurship and venture capital financing
A recent OECD working paper examines the contribution of start-ups and venture capital to environmental sustainability on the basis of a new dataset, which merges two databases from private providers for maximum global coverage (Dechezleprêtre and Kelly, 2025[46]). To identify which start‑ups are relevant to environmental sustainability, most studies rely only on sectoral classification by the data providers themselves. This is potentially problematic as some industry categories (e.g. energy) are too broad for precise disambiguation. This method may also be less effective for capturing start-ups developing technologies relevant to multiple sectors. Sustainability-relevant start-ups are identified using pre-existing industry classification and company tags used by the original data providers, where these were defined in sufficiently granular and non-ambiguous terms, as well as specific keywords and expressions in the firm’s description (Box 3.13).
A new OECD Start-up Database pools information from two leading private data providers on venture capital (VC) funding, Crunchbase and Dealroom. The consolidated database provides data on start-ups and venture capital financing of approximately 1.7 million companies worldwide. Start-ups are classified into environmental or “green” activity areas based on a methodology that combines industry classification searches with text analyses of the firm’s description.
While start-ups are generally understood to be young firms that are dynamic, innovative and entrepreneurial, what constitutes an “entrepreneurial” or “innovative” firm is notoriously difficult to delineate adequately through available data. Some studies merely refer to a firm’s age, while others use criteria such as sector of activity, participation in a support scheme (e.g. incubator) or involvement in a VC deal, which suggests a confirmed potential to deliver scalable innovations, a prerequisite for securing venture capital (Criscuolo and Menon, 2014[38]; Gompers and Lerner, 2001[47]). In this case, start-ups are defined as firms established after 2000 and feature in the source datasets, implying that they have either sought or secured venture capital.
The OECD methodology to identify environment-related start-ups uses a two-pronged approach. It draws on industry classification and company tags from the source providers when these are defined in sufficiently granular and unambiguous terms. A complementary method uses a string-searching algorithm targeting specific keywords and expressions in the firm’s description, similar to Nanda et al. (2015[35]). This identified companies relevant to environmental sustainability that are not captured by the first step – i.e. not associated with the listed industries or tags – that may be offering relevant goods or services. An example would be a firm in the digital sector using artificial intelligence (AI) methods to offer energy efficiency services.
This list of “green” application activity areas includes: 1) energy storage (start-ups developing an energy storage solution, such as lithium-ion batteries); 2) energy efficiency (start-ups developing an energy-saving solution, such as smart energy management); 3) low-carbon mobility (start-ups developing low-carbon vehicles and transportation infrastructure such as electric vehicles); 4) clean energy (start-ups developing low-carbon energy generation solutions, such as solar panels); 5) sustainable food and agriculture (start-ups developing low lifecycle emissions food products or agricultural processes such as plant-based protein); 6) pollution abatement (start-ups developing processes reducing pollution, such as bioremediation); 7) circular economy (start-ups developing processes to reduce waste, repurpose, recycle and reuse, such as recycling); and 8) other or general (start-ups classed as “green” that do not fit in any of the categories 1 to 7). The list is based on classifications used in recent academic literature and relies on industries, tags, and keywords provided by the original data provider. Each identifier (industry, tag or keyword) is linked to one environmental activity. Since several identifiers can be associated with one start-up, it can belong to multiple green activities. Estimates apply fractional counting, with each activity area having an equal weight. Some alternative approaches rely on the use of surveys in order to determine which startups are relevant to environmental sustainability (European Commission, 2024[48]), however, such an approach would not be feasible in the context of a large and continuously evolving multi-country database.
The source data are originally assembled to serve primarily venture capitalists’ and other investors’ needs. Consequently, the underlying data does not include the entire population of new start-ups created. The collection methods rely on online postings, news, and public announcements about investment deals. Therefore, firms that never receive funding are probably much less likely to be included in the database, as are start-ups that fail compared to successful start-ups and those frequently covered by the media. Any “stealth” funding rounds would naturally also not be captured in these databases. Lastly, data collection efforts are more likely to focus on the countries where the data providers’ clients concentrate, so start-ups established in VC hubs are more likely to get captured.
Source: Dechezleprêtre and Kelly (2025[49]), Venture Capital, Innovation and Business Success in Cleantech Start-ups: The New Green Economy. https://doi.org/10.1787/ba73f647-en
Looking at the change in the green start-up share between the first half of the sample (2010-16) and the second half (2017-22), the majority of the surveyed countries show an increase, including large economies such as China, India and Indonesia (Figure 3.21). Sweden, Latvia, and Norway experienced the largest increases, while Mexico, Costa Rica, and Iceland experienced the largest decreases. As shown in the figure, it is important to note that countries have a small population of green start-ups.
Percentage point difference between periods (2010-16 to 2017-22) for the share of green start-ups within each country’s start-up population
Note: For classification methodology see Box 3.13.
Source: Dechezleprêtre and Kelly (2025[49]), Venture Capital, Innovation and Business Success in Cleantech Start-ups: The New Green Economy. https://doi.org/10.1787/ba73f647-en based on OECD Start-up Database, STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats, March 2025.
Smaller European countries lead in terms of the share of venture capital channelled into sustainability-relevant start-ups, which was close to 60% of the total VC volume between 2018 and 2022 in Sweden and Estonia (Figure 3.22). The two largest countries in terms of total VC investment, the United States and China, were close to the OECD average; respectively, 12% and 18% of venture capital was channelled into green start-ups during this period. Other large economies like Indonesia perform higher than the OECD average on this metric, while Brazil and India were significantly below it in the surveyed period.
With regard to the overall volume of venture capital channelled into green start-ups, there was a substantial overall increase in recent years, peaking at approximately EUR 74 billion in 2021 across all countries included in the dataset. While the last year on record (2022) defies this trend due to worsening macroeconomic conditions that affected all sectors, the share of “green” VC continued to increase, recovering after the bursting of the “cleantech bubble” around 2010. The absolute volume of sustainability-relevant venture capital has grown within the OECD, the European Union, the United States and China. However, as evident, the share going to sustainability-related start-ups has grown only in China and the European Union, with growth being particularly steep in China, from just over 5% in 2010 to over 30% in 2022.
As a share of total venture capital
Note: For classification methodology see Box 3.13.
Source: Dechezleprêtre and Kelly (2025[49]), Venture Capital, Innovation and Business Success in Cleantech Start-ups: The New Green Economy. https://doi.org/10.1787/ba73f647-en based on OECD Start-up Database, STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats, March 2025.
Green venture capital in EUR millions (left axis) and as a share of total venture capital (right axis)
Note: The figure shows only deals classified as seed, early series and late series. These aggregated deal categories were constructed as follows: 1) seed deal type includes deals classified as angel, pre-seed, seed and equity-crowdfunding in Crunchbase and angel and seed deals in Dealroom; 2) early series deal type includes series A and B deals in both Crunchbase and Dealroom; and 3) late series includes deals classified as series C-J in Crunchbase and series C-I in Dealroom. Where deal types in the original databases did not clearly align with this classification (e.g. convertible note, corporate round or unknown deals), additional information, including funding volumes, country, deal year, sector and information source, was used to predict the aggregated funding category using a random forest model. Deals for which any of this information is missing were not classified.
Source: Dechezleprêtre and Kelly (2025[49]), Venture Capital, Innovation and Business Success in Cleantech Start-ups: The New Green Economy. https://doi.org/10.1787/ba73f647-en based on OECD Start-up Database, STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats, March 2025.
The population of green start-ups is further categorised by the environmental activity area they contribute to, based on the methodology described briefly in Box 3.13 and in more detail in Dechezleprêtre and Kelly (2025[49]). The decreasing share of energy efficiency and clean energy over time is particularly noteworthy, although the latter has picked up again in terms of share in 2021-22. Meanwhile, the circular economy, sustainable food and agriculture, and low-carbon mobility categories have slightly expanded in recent years. Energy storage has mainly remained stable.
As share of all environment-related startups
Note: Start-ups have been allocated to the above activity areas using the methodology described briefly in Box 3.13 on the basis of fractional counts.
Source: Dechezleprêtre and Kelly (2025[49]), Venture Capital, Innovation and Business Success in Cleantech Start-ups: The New Green Economy. https://doi.org/10.1787/ba73f647-en based on the OECD Start-up Database, STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats, March 2025.
The low-carbon mobility activity area has been particularly dynamic, with steep growth in VC investment, particularly in China and the United States, and comparable levels of investment in both countries in terms of volume. Both countries have also seen a major decline in new investments between 2017 and 2019 as the electric mobility sector consolidated. In China, however, this activity area continued to grow as the share of total venture capital increased from 10% in 2018 to 12% in 2022. In contrast, in the United States and the European Union, it declined in terms of share of total venture capital from 7% to 3% and from 11% to 7%, respectively.
VC investments in clean energy have been higher in the United States compared to the European Union and China in terms of total investment volumes. In Japan, there was a significant spike in the share of total venture capital due to one particularly large deal in 2022. China saw, again, an increase in the share of total venture capital, as did the European Union, the United States and the OECD. Other sustainability-related sectors have seen much lower volumes of VC investment overall. The United States used to dominate with respect to venture capital invested in the sustainable food and agriculture and pollution abatement categories, however, the EU recently took over in both activity areas in terms of share. Japan has again seen a major spike in terms of share of environment-related VC dedicated to pollution abatement in 2022 (Figure 2.25).
As a share of total venture capital
Note: OECD and non-OECD countries. Start-ups have been allocated to the above activity areas using the methodology described briefly in Box 3.13 and in more detail in Dechezleprêtre and Kelly (2025[49]), Venture Capital, Innovation and Business Success in Cleantech Start-ups: The New Green Economy. https://doi.org/10.1787/ba73f647-en. Deal volumes have been allocated fractionally, as one start-up can belong to multiple activity areas.
Source: OECD Start-up Database, STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats, March 2025.
References
[40] Akcigit, U. and N. Goldschlag (2024), “Understanding the innovation puzzle: Firm size, inventors, and industrial policy”, University of Chicago Working Paper.
[39] Andrews, D., C. Criscuolo and C. Menon (2014), “Do resources flow to patenting firms?: Cross-country evidence from firm level data”, OECD Economics Department Working Papers, No. 1127, OECD Publishing, Paris, https://doi.org/10.1787/5jz2lpmk0gs6-en.
[17] Block, J. et al. (2025), “Green patents and green trademarks as indicators of green innovation”, Research Policy, Vol. 54/1, p. 105138, https://doi.org/10.1016/j.respol.2024.105138.
[38] Criscuolo, C. and C. Menon (2014), “Environmental policies and risk finance in the green sector: Cross-country evidence”, OECD Science, Technology and Industry Working Papers, No. 2014/1, OECD Publishing, Paris, https://doi.org/10.1787/5jz6wn918j37-en.
[42] Croce, A. et al. (2024), “Cleantech and policy framework in Europe: A machine learning approach”, Energy Policy, Vol. 186, p. 114006, https://doi.org/10.1016/j.enpol.2024.114006.
[29] Cunningham, C., F. Ederer and S. Ma (2021), “Killer Acquisitions”, Journal of Political Economy, Vol. 129/3, pp. 649-702, https://doi.org/10.1086/712506.
[41] Dalla Fontana, S. and R. Nanda (2023), “Innovating to net zero: Can venture capital and start-ups play a meaningful role?”, Entrepreneurship and Innovation Policy and the Economy, Vol. 2, pp. 79-105, https://doi.org/10.1086/723236.
[32] de Rassenfosse, G. (2018), Notice failure revisited: Evidence on the use of virtual patent marking, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w24288.
[31] de Rassenfosse, G. and M. Thursby (2019), EAGER: IPRoduct: A Database of Linked Products-intellectual Property Rights.
[37] Dechezleprêtre, A., H. Dernis and N. Mulligan (forthcoming), Identification of Green Trademarks.
[22] Dechezleprêtre, A. et al. (2020), Invention and Global Diffusion of Technologies for Climate Change Adaptation: A Patent Analysis, World Bank Group, Washington, DC, http://documents.worldbank.org/curated/en/648341591630145546.
[49] Dechezleprêtre, A. and P. Kelly (2025), The New Green Economy: Venture Capital, Innovation and Business Success in Cleantech Start-ups, OECD Publishing.
[46] Dechezleprêtre, A. and P. Kelly (2025), “Venture capital, innovation and business success in cleantech startups: The new green economy”, OECD Science, Technology and Industry Working Papers, No. 2025/13, OECD Publishing, Paris, https://doi.org/10.1787/ba73f647-en.
[34] Delmas, M. and V. Burbano (2011), “The drivers of greenwashing”, California Management Review, Vol. 54/1, pp. 64-87, https://doi.org/10.1525/CMR.2011.54.1.64.
[21] Dussaux, D., A. Agnelli and N. Es-Sadki (2023), “Exploring new metrics to measure environmental innovation”, OECD Environment Working Papers, No. 221, OECD Publishing, Paris, https://doi.org/10.1787/e57a8a13-en.
[20] Dussaux, D., A. Agnelli and N. Es-Sadki (2023), “Exploring new metrics to measure environmental innovation”, OECD Environment Working Papers, No. 221, OECD Publishing, Paris, https://doi.org/10.1787/e57a8a13-en.
[48] European Commission (2024), Science, research and innovation performance of the EU, 2024 – A competitive Europe for a sustainable future, https://data.europa.eu/doi/10.2777/965670.
[4] Eurostat (2023), Community Innovation Survey 2020 (CIS2020), https://ec.europa.eu/eurostat/cache/metadata/en/inn_cis12_esms.htm.
[47] Gompers, P. and J. Lerner (2001), “The venture capital revolution”, Journal of Economic Perspectives, Vol. 15/2, pp. 145-168, https://doi.org/10.1257/JEP.15.2.145.
[28] Graham, S., A. Marco and A. Myers (2018), “Patent transactions in the marketplace: Lessons from the USPTO Patent Assignment Dataset”, Journal of Economics & Management Strategy, Vol. 27/3, pp. 343-371, https://doi.org/10.1111/jems.12262.
[44] Greenstein, S., J. Lerner and S. Stern (2013), “Digitization, innovation, and copyright: What is the agenda?”, Strategic Organization, Vol. 11/1, pp. 110-121, https://doi.org/10.1177/1476127012460940.
[30] Haščič, I. and M. Migotto (2015), “Measuring environmental innovation using patent data”, OECD Environment Working Papers, No. 89, OECD Publishing, Paris, https://doi.org/10.1787/5js009kf48xw-en.
[5] Horbach, J., C. Rammer and K. Rennings (2011), “Determinants of eco-innovations by type of environmental impact: The role of regulatory push/pull, technology push and market pull”, SSRN Electronic Journal, https://doi.org/10.2139/ssrn.1805765.
[16] ISO (n.d.), ISO Survey 2023, https://www.iso.org/the-iso-survey.html (accessed on 17 June 2025).
[7] Kemp, R. et al. (2019), Maastricht Manual on Measuring Eco-Innovation.
[10] Kinne, J. and J. Axenbeck (2020), Web Mining for Innovation Ecosystem Mapping: A Framework and a Large-scale Pilot Study,.
[43] Lerner, J. and R. Nanda (2020), “Venture capital’s role in financing innovation: What we know and how much we still need to learn”, Journal of Economic Perspectives, Vol. 34/3, pp. 237-261, https://doi.org/10.1257/jep.34.3.237.
[25] Marco, A. et al. (2017), “Global markets for technology: Evidence from patent transactions”, Research Policy, Vol. 46/9, pp. 1644-1654, https://doi.org/10.1016/j.respol.2017.07.015.
[36] Millot, V. (2009), “Trademarks as an indicator of product and marketing innovations”, OECD Science, Technology and Industry Working Papers, No. 2009/6, OECD Publishing, Paris, https://doi.org/10.1787/224428874418.
[35] Nanda, R., K. Younge and L. Fleming (2015), “Innovation and Entrepreneurship in Renewable Energy”, in The Changing Frontier, University of Chicago Press, https://doi.org/10.7208/chicago/9780226286860.003.0008.
[14] OECD (2025), “Fostering convergence in SME sustainability reporting”, OECD SME and Entrepreneurship Papers, No. 66, OECD Publishing, Paris, https://doi.org/10.1787/ffbf16fb-en.
[3] OECD (2023), OECD Business Innovation Indicators: Statistical Highlights, https://www.oecd-ilibrary.org/content/dam/oecd/en/data/datasets/business-innovation-statistics-and-indicators/business-innovation-indicators-2023-highlights.pdf.
[23] OECD (2022), Patents on environment technologies, https://www.oecd.org/en/data/indicators/patents-on-environment-technologies.html.
[18] OECD (2010), Measuring Innovation: A New Perspective.
[27] OECD (n.d.), STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats.
[1] OECD/Eurostat (2018), Oslo Manual 2018: Guidelines for Collecting, Reporting and Using Data on Innovation, 4th Edition, The Measurement of Scientific, Technological and Innovation Activities, OECD Publishing, Paris/Eurostat, Luxembourg, https://doi.org/10.1787/9789264304604-en.
[2] OECD/Eurostat (2018), Oslo Manual 2018: Guidelines for Collecting, Reporting and Using Data on Innovation, 4th Edition, The Measurement of Scientific, Technological and Innovation Activities, OECD Publishing, Paris/Eurostat, Luxembourg, https://doi.org/10.1787/9789264304604-en.
[45] Popp, D. et al. (2020), Innovation and Entrepreneurship in the Energy Sector, University of Chicago Press, http://www.nber.org/chapters/c14375.
[19] Probst, B. et al. (2021), “Global trends in the invention and diffusion of climate change mitigation technologies”, Nature Energy, Vol. 6/11, pp. 1077-1086, https://doi.org/10.1038/s41560-021-00931-5.
[11] Rammer, C. and N. Es-Sadki (2023), “Using big data for generating firm-level innovation indicators - a literature review”, Technological Forecasting and Social Change, Vol. 197, p. 122874, https://doi.org/10.1016/j.techfore.2023.122874.
[24] Serrano, C. (2010), “The dynamics of the transfer and renewal of patents”, The RAND Journal of Economics, Vol. 41/4, pp. 686-708, https://doi.org/10.1111/j.1756-2171.2010.00117.x.
[6] Siedschlag, I., S. Meneto and M. Tong Koeckling (2019), “Determinants of green innovations: Firm-level evidence”, SSRN Electronic Journal, https://doi.org/10.2139/ssrn.3495884.
[15] Sousa Lira, J., E. Salgado and L. Beijo (2019), “Which factors does the diffusion of ISO 50001 in different regions of the world is influenced?”, Journal of Cleaner Production, Vol. 226, pp. 759-767, https://doi.org/10.1016/j.jclepro.2019.04.127.
[9] Statistics Canada (2023), Adoption of clean technologies, by industry and enterprise size, https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=2710036201.
[8] Statistics Canada (2022), Survey of Advanced Technology, 2022, https://www.statcan.gc.ca/en/statistical-programs/instrument/4223_Q1_V1.
[26] USPTO (n.d.), Patent Assignment Dataset, https://www.uspto.gov/ip-policy/economic-research/research-datasets/patent-assignment-dataset.
[13] Vollebergh, H. and E. van der Werf (2014), “The role of standards in eco-innovation: Lessons for policymakers”, Review of Environmental Economics and Policy, Vol. 8/2, pp. 230-248, https://doi.org/10.1093/reep/reu004.
[12] Wildnerova, L. et al. (2024), “Which SMEs are greening?: Cross-country evidence from one million websites”, OECD SME and Entrepreneurship Papers, No. 60, OECD Publishing, Paris, https://doi.org/10.1787/ddd00999-en.
[33] Zolas, N., T. Lybbert and P. Bhattacharyya (2017), “An ‘algorithmic links with probabilities’ concordance for trademarks with an application towards bilateral IP flows”, World Economy, Vol. 40/6, pp. 1184-1213, https://doi.org/10.1111/TWEC.12382.
← 1. Similarly, the UK Innovation Diffusion and Adoption Survey conducted by the Department of Science, Innovation and Technology was launched in August 2024 to understand the diffusion and adoption activities of businesses. The questions target diffusion and adoption with respect to the 20 technologies and practices, including some relevant to the environment, namely: 1) low-carbon energy, heating and propulsion technologies (including, but not limited to, nuclear, renewables); 2) recycling and waste technology (including, but not limited to, CCUS); 3) battery and energy storage technologies; and 4) sustainable business practices.