Digital technologies have been a key element of economic resilience during the COVID-19 crisis. This chapter focuses on several types of digital technologies that help firms and workers stay productive through the pandemic, and beyond. It first examines the adoption of telework, and discusses several enabling factors for its uptake, including pre-crisis experience with remote work, availability and quality of communications infrastructure, and the digital skillset of the population. The chapter then highlights other digital technologies and tools that became crucial in the pandemic, such as those related to e-commerce or cloud computing. It also discusses digital divides along several dimensions, looking at how they can potentially be exacerbated through the rapid uptake of digital technologies, including increasing gaps between leading and laggard firms, urban and rural populations, and the varying digital skill levels of workers.
Strengthening Economic Resilience Following the COVID-19 Crisis
5. Supporting productivity through digital technologies
Abstract
Key findings
The pandemic contributed to rapid growth in digitalisation and information and communication technology (ICT) use. The COVID-19 crisis has been a catalyst for the adoption of digital technologies, leading many countries to increase the use of digital tools such as telework up to nearly their estimated potential.
Digital technologies hold the key to resilience during the crisis and beyond. Telework, and other digital technologies such as e-commerce, were crucial factors in firms’ and workers’ ability to maintain production during the crisis. Factors enabling the use of such technology, such as high-capacity communications infrastructure, digital skills and data security, are crucial to building resilient economies and firms. Governments should ensure that long-term policies, for the recovery period and beyond, foster and embrace the adoption of digital technology to ensure its benefits are spread widely.
Governments must address digital divides in the long term. While the rapid digital uptake has been key to staying productive and connected for many, it can also exacerbate pre-existing divides in digital skills and use. These divides exist across populations (e.g. between rural and urban areas), workers (e.g. those with differing digital skill levels), and firms (e.g. small and large firms). Addressing these growing gaps is crucial in ensuring a strong and inclusive recovery, and in combatting inequalities over the long term.
Introduction
Digital technologies have been crucial in enabling people to communicate, work, shop, learn, entertain themselves and even stay physically active from their homes throughout the COVID-19 pandemic. Digital tools for remote work have been particularly important during the crisis, as they allowed firms and workers – especially those in non-essential industries – to continue operating in the face of lockdowns and social distancing restrictions. Digital technologies can help to reduce physical contact over prolonged time periods, as “living with the virus” has become the new normal.
In many ways, the COVID-19 crisis has accelerated already existing digitalisation trends, while at the same time fostering new uses of digital technologies and the creation of new digitally-enabled business models. The need to reduce physical contact and mobility led workers, firms and consumers to adopt digital technologies much faster and to a much larger extent than they would have otherwise, as recent evidence shows (OECD, 2021[1]; OECD, 2020[2]; OECD, 2020[3]; Riom and Valero, 2020[4]). OECD countries have seen a tremendous increase in Internet traffic in 2020, with average Internet bandwidth growing by 58% between December 2019 and December 2020.1 For some countries the growth was dramatic; for example, Chile and Mexico experienced bandwidth growth of 160% and 135%, respectively, over the same period.
This development has been paralleled by a surge in telework. While levels of telework uptake before the crisis had remained well below their estimated potential, they reached unprecedented new heights through the pandemic across countries (OECD, forthcoming[5]). Recent data for Italy shows that, during the crisis, this gap between actual and potential extent of teleworking narrowed substantially – with telework even temporarily reaching its full estimated potential during the first lockdown (OECD, 2020[3]). Telework has the capacity to increase the resilience of economies in case of future health crises or other shocks restricting mobility, but more widespread telework also promises broader potential benefits for countries and workers, including higher productivity, better work-life balance, and a reduction of carbon emissions.
From a firm perspective, the ability to shift to remote working depends on a number of factors in addition to the types of tasks on the job (varying to a great degree across industries, as discussed in Chapter 3), such as the availability of high-quality communication infrastructure (i.e. broadband), the skills of workers and managers, and prior experience with, or trust in, digital solutions and tools. These factors are relevant not only to the adoption of telework-based working models, but are also crucial for the efficient adoption of other digital technologies, such as cloud computing, online sales, and even artificial intelligence (AI). How they vary at the country level, and how they relate to observed levels of telework, is explored in the following sections.
As the economy recovers, digitally-enabled changes to business functioning may also help reduce carbon emissions and mitigate climate change. Relevant changes include shifts in work modes (such as increased teleworking and teleconferencing), as well as changes to business models (e.g. the rise of e-commerce). Expanding telework by removing barriers to uptake and encouraging companies and workers to adopt new, digital modes of working and interacting can aid in achieving environmental objectives while also building resilience.
During the crisis, the observed increase in the speed and breadth of the adoption of digital technologies has nevertheless revealed the potential to widen digital divides (i.e. differences in the use and benefits that accrue to digital-savvy parts of the population as compared to those less able to leverage the possibilities offered by the digital transformation). This divide concerns a range of outcomes, as digital capabilities are no longer only relevant for workplace or leisure choices, but have become invaluable for managing daily life and staying healthy during the pandemic. This is a concern particularly for older population groups, who typically exhibit relatively lower engagement and skills related to the effective and safe use of digital tools (OECD, 2020[3]; 2020[6]), and have also been among those most at-risk from the health crisis. With many new digital solutions likely to remain in use at least to some extent after the crisis, bridging the digital divide becomes all the more urgent and is likely to remain high on the policy agenda for an inclusive, green and speedy recovery.
Moreover, as many of the societal and economic changes that have been triggered or accelerated by the COVID-19 crisis will likely persist, addressing the needs and challenges related to a widespread diffusion and use of digital technologies will remain important when the health crisis abates (OECD, 2020[6]). Indications that the shift to digital tools represents a permanent change in preferences can be found in several areas. For example, a recent survey finds that changes in online shopping behaviour are likely to last beyond the pandemic (UNCTAD; NetCommSuisse, 2020[7]). Levels of telework are also likely to remain well above pre-crisis levels permanently. Due to prolonged remote working during the crisis, worker and employer preferences appear to have shifted in favour of more frequent telework: workers who experienced telework during the crisis report a clear preference for continuing to do so at least a few days per week (Eurofound, 2020[8]). Recent evidence supports the notion that workers, as well as managers, now consider ideal levels of telework to be much higher than pre-crisis levels, and that the crisis stimulated investments in associated tangible and intangible capital, suggesting that some of the stigma previously associated with telework has broken down (OECD, forthcoming[5]; Bloom, Mizen and Taneja, 2021[9]). In addition, even when the use of remote tools will no longer be a necessity due to the crisis, digital readiness will remain important not only for participation in the labour market but also in many aspects of everyday life. This includes purchasing goods and services; taking part in recreational activities; finding and accessing services such as health care and e-government services; and even interacting with AI technology (Nachtigall and Squicciarini, forthcoming[10]).
The rapid spread of digital technologies will require not only high levels of digital skills, but also sound cognitive and socio-emotional skills for the wider population to navigate change and thrive in the digital era. These skills not only allow individuals to reap the benefits of the digital transformation, but at the same time foster growth and productivity while preventing pre-existing gaps – across several dimensions – from widening. As analysed in more detail in Chapter 7, workers who have better digital skills appear to be in a better position to keep working through the health crisis, and are likely to be less impacted by the ensuing economic recession. This highlights the importance of having the required skills for succeeding in the digital age, but it also shows the potential for widening digital divides, which might create new disparities. The potential for increased polarisation also exists along a number of dimensions that already exhibit gaps in digital technology access and uptake. Evaluating the impact of policies on these divides, and investing in communications infrastructure and digital skills for those lagging behind, will be crucial to prevent a deepening of pre-existing gaps (OECD, 2020[11]). Pre-existing divergences in adoption may be exacerbated not only between people, but also between firms. Indeed, not all businesses develop, adopt or use digital technologies in the same way, nor do they benefit equally from the digital transformation, as documented in a series of work in the OECD’s Going Digital project (Andrews, Criscuolo and Gal, 2016[12]; Gal et al., 2019[13]; Sorbe et al., 2019[14]).
The first part of this chapter is devoted to the analysis of telework, which has been crucial not only to sustain production and employment while maintaining social distancing restrictions, but might also remain the “new normal” after the health crisis. At the country level, there are factors that can limit or accommodate remote work. How these interact with industry (discussed in detail in Chapter 3) and firm characteristics affecting telework uptake are part of the analysis in the first part of this chapter.
The second part of this chapter focuses on other types of digital technologies, which have contributed to upholding not only production, but also supply. Digital modes of performing transactions, such as online sales, have played a major role in retail, but also in other industries which shifted to e-commerce during the crisis (OECD, 2020[2]). The section then discusses more advanced digital technologies that are relevant to firms across industries, such as cloud computing. This section also provides an outlook of potential consequences of the accelerated and uneven uptake of these technologies, focusing on divides between firms in different industries.
A third part of this chapter discusses how the various changes brought about by the crisis can affect productivity, focusing in particular on changes related to digital technologies. Despite the many adverse effects of restrictions on economic activity, some positive developments have emerged, and new opportunities for businesses have arisen that have the potential to translate into long-term productivity gains. The last component of this part of the chapter discusses how pre-existing digital divides should be addressed, to avoid adverse consequences through increased concentration that potentially diminish the productivity-enhancing effects of the accelerated digital transformation.
Adoption of telework
Pre-crisis experience with working from home
Experience with telework prior to the crisis is an important determinant and indicator of how easily and successfully it could be adopted in the pandemic to ensure social distancing at work by limiting staff presence. Besides the different task-related factors determining the potential for telework within and across industries discussed in Chapter 3, actual telework adoption is also shaped by a range of country-level factors.2 These can be of a technical nature (e.g. the availability of broadband infrastructure), or more “soft” factors such as social norms (e.g. a culture of physical presence in the office; the use of output-based management styles instead of input-based styles, such as assessing performance based on hours worked; or trust between managers and workers). Societal acceptance and public support for flexible work arrangements also play a role (e.g. policies implemented in Finland allowing employees to choose their working hours3).
Figure 5.1 presents data on the pre-crisis prevalence of telework alongside a measure of telework potential, by country (Espinoza and Reznikova, 2020[15]). Experience with telework before COVID-19 varied substantially across countries. While it was particularly high in the Nordic countries and the Netherlands, with 25% to 30% of the workforce reporting regular telework already in 2015, it was relatively low in several southern (Greece, Italy, Spain, and Portugal) and eastern (the Slovak Republic, Latvia) European countries, which had adoption rates of 10% to 15%.
The high rates of telework experience in the Nordic countries and the Netherlands prior to the crisis nearly matched estimated telework potential. A few additional countries (Belgium, Canada, England and Israel) also score high in terms of task-based telework potential. Overall, estimates for telework potential generally range between 20% and 40%, with a few outliers at the lower end (Turkey, Mexico). Many countries that had lower pre-crisis telework adoption rates appear to fall well below their assessed potential (e.g. the Slovak Republic, Latvia, Germany and Spain). Some possible reasons for limited telework uptake are discussed in the following sections.
Figure 5.1. Pre-crisis adoption of telework, by country
Note: Working from home indicator: data refer to 2015 and telework is defined as working from home at least several times a month. Task-based telework ability indicator: data collected between 2011 and 2017, data for the United Kingdom refer to England. Average refers to the unweighted average across the OECD reported countries.
Sources: OECD calculations based on Eurofound (2017[16]), European Working Conditions Survey, 2015 (data collection), http://doi.org/10.5255/UKDA-SN-8098-4 for working from home; Espinoza, R. and L. Reznikova (2020[15]), “Who can log in? The importance of skills for the feasibility of teleworking arrangements across OECD countries”, https://dx.doi.org/10.1787/3f115a10-en for task-based telework ability.
Broadband infrastructure as a necessary condition for connectivity and remote working
In addition to the task- and skills-related reasons that explain why workers and firms in some countries are able to adopt telework more easily, communication infrastructure is an important limiting factor, as working remotely requires that both firms and employees can rely on adequate (i.e. fast and reliable) broadband connections. In addition, economic activities in a remote setting may require symmetrical download and upload broadband speeds for applications such as virtual meetings.Previous evidence shows that the availability of high-speed broadband is also an enabling factor for a number of further digital technologies used by firms, such as back- and front-office management systems (Andrews, Nicoletti and Timiliotis, 2018[17]).
Data from firm surveys in OECD countries show that while – in the vast majority of countries – virtually all firms seemed to have access to broadband in the years preceding the crisis (Figure 5.2), high-speed connectivity is not as pervasive (Figure 5.3). In Denmark and the Netherlands, all enterprises with at least ten employees reported to have a broadband connection with a minimum advertised speed of 256 kilobits per second (Kbps). Greece, Latvia and Hungary are among the countries with relatively lower access rates, with about 78% to 80% of enterprises having broadband connections at 256 Kbps – the minimum advertised speed (Figure 5.2). The gap in broadband access by firm size is also largest in these countries.
Figure 5.2. Enterprises with a broadband connection, as a percentage of enterprises in each employment size class
Note: Businesses employing at least ten employees. Broadband connections refers to fixed line broadband services (i.e. of 256 kilobits per second advertised speed or more) subscriptions purchased by households or businesses. Fixed broadband comprises DSL, cable, fibre-to-the-home (FTTH), fibre-to-the-building (FTTB), satellite, terrestrial fixed wireless and other fixed-wired technologies. For Japan and Korea, data refer to both fixed and mobile broadband. Data refer to 2019 except for Colombia, Japan and Korea (2018), New Zealand 2017/18, Australia 2016/17. For Brazil, broadband is defined by type of connection rather than download speed. Average refers to the unweighted average across the OECD reported countries.
Source: OECD (2021[18]), “ICT Access and Use by Businesses”, OECD Telecommunications and Internet Statistics (database), https://doi.org/10.1787/9d2cb97b-en (accessed on 29 January 2021).
Figure 5.3. Enterprises with broadband connections, by speed tiers
Note: Businesses employing at least ten employees. Broadband includes fixed connections with an advertised download rate of at least 256 Mbps, except Brazil for which broadband is defined by type of connection rather than download speed. Data refer to 2019 except for Australia 2016/17, Colombia 2018 and New Zealand 2017/18. Average refers to the unweighted average across the OECD reported countries.
Source: OECD (2021[18]), “ICT Access and Use by Businesses”, OECD Telecommunications and Internet Statistics (database), https://doi.org/10.1787/9d2cb97b-en (accessed on 29 January 2021).
However, while in most countries the majority of companies have broadband access, there are large differences in the speed of the connection provided (Figure 5.3). In Slovenia, the Czech Republic, France, Estonia and Turkey, the overall share of businesses with a broadband connection is above the OECD average, but more than half of these enterprises have relatively low advertised speed connections (speeds of less than 30 megabits per second [Mbps]), which can make efficient teleworking difficult when workers need to connect to firm systems. This contrasts with Denmark and Sweden, where 60% of firms have high-speed connections of 100 Mbps or more.
Figure 5.4 displays the correlation between observed levels of telework and firms’ uptake of broadband across countries. There is a clear relationship between the two variables (with some outliers at lower levels of telework). In particular, actual levels of telework uptake seem to remain below the technical possibilities afforded by firms’ communications infrastructure in Portugal, Spain, Germany and Lithuania.
Figure 5.4. Telework uptake and firm infrastructure for fast broadband speed
Note: Firms with at least 30 Mbps advertised download speed broadband connection, data for 2019. Refer to Figure 5.3 for more context of broadband speeds. The 2020 definition and data on firms’ broadband connections (with the new definition focusing on fixed line connections) confirms the observed patterns. It should be noted that, as such, several countries have substantially higher levels of connectedness; for example France shifted to more than 70%.
Sources: Eurofound (2017[16]), European Working Conditions Survey, 2015 (data collection), http://doi.org/10.5255/UKDA-SN-8098-4 for telework uptake (for more details, see Chapter 3); OECD (2021[18]), “ICT Access and Use by Businesses”, OECD Telecommunications and Internet Statistics (database), https://doi.org/10.1787/9d2cb97b-en (accessed on 21 January 2021) for broadband infrastructure and speed.
It is clear from Figure 5.4 that a lack of high-quality connectivity represents a key limiting factor to the adoption of telework. However, other factors may play a role as well. Looking at how telework varies across countries with a similar level of broadband adoption by firms in terms of average connection speed can provide some indication for the potential of increasing telework update by tackling other “soft” factors. For instance, Spain exhibits relatively low levels of telework compared to Belgium or Finland, despite similar firm access to fast broadband. This suggests that, to the extent that differences cannot be fully explained in terms of industrial structure, Spain may be able to increase firm telework uptake by addressing factors such as management practices or social norms.
Home Internet connections play a complementary role to those of firms in allowing employees to telework. Figure 5.5 plots levels of telework against the share of the population having subscribed to fast broadband (minimum 25/30 Mbps). The positive relationship confirms the importance of fast broadband as a precondition for telework uptake, on both the firm and worker side, with a similar set of countries that, despite having a high share of fast broadband uptake in the population, display below-average levels of telework (Portugal, Spain and Germany).4 That home and firm connections yield a very similar picture is somewhat unsurprising, given that many employees live close to their physical workplaces, and as such the infrastructure and broadband speed available to both them and their firm is likely to be similar. However, if levels of telework remain high after the pandemic, discrepancies in firm and home Internet connections may arise if workers engaging in permanent telework choose to live further from their workplace, including in rural regions. To reap the full benefits of telework, broadband infrastructure expansions or upgrades might therefore be required.
Figure 5.5. Telework uptake and home infrastructure for fast broadband speed
Note: Fast fixed broadband subscriptions per 100 inhabitants (minimum 25/30 Mbps), based on December 2019 speed tiers. Australia: Data reported for December 2018 and onwards is being collected by a new entity using a different methodology. Figures reported from December 2018 comprise a series break and are incomparable with previous data for any broadband measures. Australia reports to the OECD. Speed tier data are only for services purchased over the National Broadband Network (NBN), which comprise the majority of fixed broadband services in operation. There is no public data available for the speed of non-NBN services. Mexico and Switzerland: Data are preliminary. New Zealand: Speed tiers are for 2018 instead of 2019. Poland: Data are OECD temporary estimates.
Sources: Eurofound (2017[16]), European Working Conditions Survey, 2015 (data collection), http://doi.org/10.5255/UKDA-SN-8098-4 for telework uptake (for more details, see Chapter 3); OECD (2021[19]), Broadband Portal (database), www.oecd.org/sti/broadband/oecdbroadbandportal.htm (accessed on 29 April 2021) for broadband infrastructure.
While average broadband download speeds can be a proxy for broadband performance in urban areas, where most economic activity is centred, there is also important within-country variation in broadband network quality, with rural areas having much slower connections (OECD, 2019[20]). This represents a key barrier for households in rural areas to effectively telework, and make use of other data-intensive digital technologies.5
Figure 5.6 shows that high-speed broadband coverage in rural areas still remains a major challenge for many countries. For example, in Europe in 2019, only 59% of rural households were located in areas with coverage of fixed broadband with an advertised minimum speed of 30 Mbps compared to 86% of households in other areas.6 Indeed, this gap is rather large in several of the countries that appear to be downward outliers in Figure 5.4. Despite high average speeds, Lithuania and Spain both display relatively large gaps between the coverage of high-speed fixed broadband for households living in rural areas compared to all households and there is a discernible gap also in Germany, Finland and Sweden.7
A more permanent uptake of telework can also contribute to reducing greenhouse gas emissions that are associated with commuting, beyond the crisis. However, more permanent teleworking arrangements will only be feasible if high-speed Internet access is widely available. In the recovery period, investment should be steered towards upgrading communications infrastructure, such as universal broadband Internet, and enabling technologies that ensure that communication networks can cope with a sustained increase in traffic from teleworking and teleconferencing.
Figure 5.6. Share of households with coverage of fixed broadband with minimum 30 Mbps, in rural areas and all area
Note: Households in rural and total areas where fixed broadband with a contracted speed of 30 Mbps or more is available, as a percentage of households in each category. Data refer to 2019. For European countries, rural areas are those with a population density less than 100 per square kilometre. For Canada, rural areas are those with a population density less than 400 per square kilometre. For the United States, rural areas are those with a population density less than 1 000 per square mile or 386 people per square kilometre. For European countries, fixed broadband coverage of NGA technologies (VDSL, FTTP and DOCSIS 3.0) capable of delivering at least 30 Mbps download was used. For the United States, coverage of fixed terrestrial broadband capable of delivering 25 Mbps download and 3 Mbps upload services was used.
Sources: Data from CRTC (2019[21]), Communications Monitoring Report, 2019 (Canada), https://crtc.gc.ca/pubs/cmr2019-en.pdf; European Commission (2020[22]), Study on Broadband Coverage in Europe 2019 (European Union), https://digital-strategy.ec.europa.eu/en/library/broadband-coverage-europe-2019 and FCC (2019[23]), 2019 Broadband Deployment Report 12 (United States), https://docs.fcc.gov/public/attachments/FCC-19-44A1.pdf.
Targeted investments in communications infrastructure should therefore be part of any green recovery package, in order to achieve persistent behavioural changes and permanent reductions in emissions (provided measures are also taken to reduce the environmental footprint of digital technologies). Such investments may need to be accompanied by new policies and regulations that facilitate and encourage behavioural changes over the longer term. Regulations could include flexible working arrangements or the right to work from home when feasible, as debated for example in Germany (Reuters, 2020[24]).
Digital skills of the population
Workers’ proficiency in digital environments is another key factor to succeed in the digital transformation. Indeed, worker skills affect both the likelihood that firms adopt digital technologies, and the effectiveness and efficiency of their use. With COVID-19, working remotely has become the norm for many jobs. Everything else equal, it is expected that a digital-ready workforce is more likely to telework efficiently, thereby contributing to supply-side resilience.
The Programme for the International Assessment of Adult Competencies (PIAAC) offers harmonised measures of digital readiness that are comparable across countries. These come from standardised computer-based assessments of the ability of adults to use ICT tools and applications to assess, process, evaluate and analyse information in a goal-oriented way.
Using PIAAC data, Figure 5.7 shows that workers’ digital readiness varies substantially across countries, with over 40% of adults in New Zealand and Sweden positioned in the two highest proficiency levels for problem solving in digital-rich environments, compared to around 30% of adults on average across OECD countries. Higher values indicate a higher proportion of the population that could be expected to work productively in the face of a sudden shift to telework.
Figure 5.7. Percentage of adults at highest levels of proficiency in digital environments
Note: Percentage of adults with high scores in PIAAC's problem solving in technology-rich environments. Problem solving in technology-rich environments is defined as using digital technology, communication tools and networks to acquire and evaluate information, communicate with others and perform practical tasks; it measures both problem-solving and basic computer literacy skills (i.e. the capacity to use ICT tools and applications). Time period: 2012-13 (Round 1); 2014-15 (Round 2); 2017 (Round 3). Country coverage: Australia, Austria, Belgium (Flanders), Canada, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Ireland, Italy, Japan, Korea, the Netherlands, Norway, Poland, the Russian Federation, the Slovak Republic, Spain, Sweden, the United Kingdom (England) (Round 1); Chile, Greece, Indonesia, Israel, Lithuania, New Zealand, Slovenia, Turkey (Round 2); Hungary, Mexico, the United States (Round 3). The sample for the Russian Federation does not include the population of the Moscow municipal area. Average refers to the unweighted average across the reported OECD countries.
Source: OECD calculations based on OECD (2019[25]), Skills Matter: Additional Results from the Survey of Adult Skills, https://doi.org/10.1787/1f029d8f-en.
Figure 5.8. Telework uptake and ICT skills
Note: The ICT skills indicator corresponds to the “proficiency in digital environments” of Figure 5.7; see figure notes for details on the data.
Sources: Eurofound (2017[16]), European Working Conditions Survey, 2015 (data collection), http://doi.org/10.5255/UKDA-SN-8098-4 for telework uptake (for more details on this topic, see Chapter 3); OECD (2019[25]), Skills Matter: Additional Results from the Survey of Adult Skills, https://doi.org/10.1787/1f029d8f-en for ICT skills.
Figure 5.8 shows that countries whose workforce possesses higher levels of digital skills are also more likely to exhibit high levels of telework uptake. A few countries display lower levels of telework uptake than their level of digital skills would suggest – notably, Germany and the Slovak Republic fall short of their potential in terms of workers’ skills. In the Slovak Republic, communications infrastructure might be a limiting factor for higher telework uptake (Figure 5.2). In Germany, industry structure is likely to be a large explanatory factor, with manufacturing – typically having lower telework potential – playing a comparatively large role in the country. Social norms for being present in the office could also contribute.
E-commerce and the pre-existing use of digital tools
E-commerce has been of paramount importance during lockdowns, to help satisfy demand and to sustain economic activity and employment. As brick-and-mortar establishments have been forced to close their doors, consumers have moved rapidly towards online purchasing, greatly impacting the retail sector (see also Box 5.1). As discussed in Chapter 3, there is important variation between industries in the ability to switch to remote modes of sales. This industry variation, which is to a large extent determined by the type of product supplied (with services industries involving direct face-to-face interactions that are difficult to provided remotely), needs to be complemented by a within-industry perspective, where there is evidence of a massive shift towards remote sales in several industries where the types of products or services allow for it. Recent OECD work (OECD, 2020[2]) finds that the COVID-19 crisis has led to online modes of transactions of products which were previously rarely sold online, such as groceries.
Share of firms with online sales
In 2019, on average about 25% of firms in OECD countries participated in e-commerce transactions, with a much higher share in New Zealand and Australia (Figure 5.9).
Figure 5.9. Enterprises engaged in sales via e-commerce by firm size, as a percentage of enterprises in each employment size class
Note: Percentage of businesses (with at least ten employees) receiving orders over computer networks. Small businesses are defined as having 10 to 49 employees, large businesses are those with 250 employees or more. For Japan, data refer to businesses with 100 or more employees instead of ten or more, large firms have 300 or more employees. Data refer to 2019 except for Australia (2016/17), New Zealand (2017/18), Colombia, Iceland, Japan and Korea (2018). Agriculture (except in Australia, Chile and New Zealand) and public administration are excluded. Average refers to the unweighted average across the OECD reported countries.
Source: OECD (2021[18]), “ICT Access and Use by Businesses”, OECD Telecommunications and Internet Statistics (database), https://doi.org/10.1787/9d2cb97b-en (accessed on 29 January 2021).
Box 5.1. COVID-19 and the retail sector: Impact and policy responses
COVID-19 mitigation measures dramatically disrupted the retail sector, with a heterogeneous impact across firms depending on the combined effect of three key business characteristics:
Essential vs. non-essential goods: demand diverged across these two retail categories, particularly in countries most affected by the pandemic (Figure 5.10).
Online vs. brick-and-mortar: mitigation measures mostly affect physical stores, and may accelerate the ongoing shift to online retailing.
Liquidity position: retail is characterised by a large variation in businesses’ access to liquidity buffers and external finance.
Figure 5.10. Change in demand for essential versus non-essential retail goods
Note: This graph reports the unweighted average growth in Google searches for essential and non-essential retail item categories (normalised by the overall volume of searches) between April 2019 and April 2020. Essential item categories include “Consumer electronics”, “Grocery and food retailers” and “Pharmacy”; non-essential categories include “Luxury goods”, “Home appliances”, “Home furnishings”, “Luggage and travel accessories”, “Apparel” and “Gifts and special event items”.
Source: OECD (2020[26]), "COVID-19 and the retail sector: impact and policy responses", https://doi.org/10.1787/371d7599-en.
Governments can shield retail from the effects of the crisis, and enhance the sector’s resilience by:
ensuring that liquidity assistance schemes are accessible to retail firms, irrespective of their size
helping essential retailers deal with labour supply disruptions, in particular by smoothing demand-supply matching for retail jobs and providing guidance on health and safety standards
supporting retail firms to implement social distancing measures
ensuring competition in the sector during and in the aftermath of the crisis
promoting the diversification of firms’ sales channels, in particular by helping small brick-and-mortar retailers go online.
Source: OECD (2020[26]), "COVID-19 and the retail sector: impact and policy responses", https://doi.org/10.1787/371d7599-en.
However, not all businesses are equally likely to buy and sell online. In many countries, there is a large gap between large and small enterprises, especially in Slovenia, Sweden, Denmark, France, Portugal, Austria and Belgium. Firms operating in Construction, Professional, Scientific and technical activities, and Administrative and support services are also significantly less engaged in e-commerce than in other sectors, reflecting the bespoke nature of many products in these industries. Box 5.1 details some of the changes in demand for e-commerce in the retail sector as a result of the pandemic, and additional insights into e-commerce in times of COVID-19 can be found in OECD (2020[2]).
Share of population making online purchases
Uptake of e-commerce has steadily increased over the past decade. The COVID-19 crisis has seemingly accelerated this trend, with changes in consumption patterns that may persist in the long term. In parallel to the increase of the share of firms selling online, the number of individuals shopping on the Internet has risen as well, including through the use of online platforms (OECD, 2021[27]). The role of such platforms has increased to a greater extent in countries where lockdown measures were stricter, according to analysis from the OECD COVID Policy Tracker (Bulman and Koirala, 2020[28]). This uptake has mitigated the fall in consumer spending brought on by distancing measures and in-person retail shutdowns, thereby contributing also to the resilience of these economies in the future. In addition, platform use can positively affect the productivity of firms as it enables a better matching of supply and demand, and by allowing businesses to outsource certain tasks to the platforms (e.g. marketing, logistics) (Bailin Rivares et al., 2019[29]).8 Online platforms that facilitate a more efficient interaction with suppliers and customers have also become relatively more important for the emergence of new businesses, as well as for the survival of existing ones (e.g. the rise in food delivery by restaurants).
Figure 5.11. Share of population using e-commerce by age, as a percentage of individuals in each group
Note: Share of individuals who have purchased online over the last 12 months, as a percentage of Internet users in each age group. Data refer to 2019, except for Canada, Colombia, Costa Rica and Japan (2018); Chile and Israel (2017); and Australia (2016/17). For Australia and Israel, the recall period is three months.
Source: OECD (2021[1]), “ICT Access and Usage by Households and Individuals”, OECD Telecommunications and Internet Statistics (database), https://doi.org/10.1787/b9823565-en (accessed on 29 January 2021).
In 2019, more than 80% of adults in the United Kingdom, Denmark, Norway, Sweden, the Netherlands and Switzerland made online purchases (Figure 5.11). In contrast, online shopping did not appear to be as widespread in Latin America, Turkey, Italy, Portugal and Greece. Uptake further appeared to differ by age, with young adults being more than twice as likely as older adults to have made a purchase online, in many eastern and southern countries in Europe. Differences in uptake across ages are less pronounced in countries with higher overall uptake, which also tend to be countries with higher levels of digital skills in the population as a whole (Figure 5.7). While the overall share of population making online purchases has increased over time in all OECD countries, the absolute size of gaps in terms of age and income has increased in many countries over the last seven years.
Uptake of cloud computing
By definition, cloud services can be used from anywhere, allowing employees to work on the same projects remotely and simultaneously. This technology can therefore play a key role at a time when the uptake of telework is crucial for sustaining economic activity.9 As is the case for most digital technologies, an important precondition for the use of cloud computing is a high-speed broadband connection, which is required to allow large data flows between data centres and users (Andrews, Nicoletti and Timiliotis, 2018[17]). Given that high-quality communications infrastructure is indispensable for the use of all digital technologies (including those that are complementary to cloud computing, such as back- and front-office management systems), infrastructure upgrades can in turn enhance the overall resilience of companies that use cloud computing, by assisting their response to unexpected shocks such as disruptions in their supply chains or changes in consumer preferences.
By 2017, more than half of enterprises with at least ten employees had adopted cloud computing in the Nordic countries, Japan, Brazil, Canada and the United States (Figure 5.12). Even though cloud computing may be an opportunity for small and medium-sized enterprises to reduce IT costs, on average only one third of enterprises with fewer than 50 employees used this technology, compared to 59% of large firms in the OECD. The gap between small and large businesses is highest in Belgium, France and Slovenia. The overall adoption rates of cloud computing were lowest in Turkey, even for large businesses. Data on the use of cloud computing by private (rather than business) Internet users (OECD, 2020[3]) lead to a fairly similar ranking of countries, with average use being highest in northern Europe (over 64% in the Nordic countries, and around 53% or more in the United Kingdom, the Netherlands, and Ireland) and lowest in Turkey and Poland (around 26%) and Mexico (20%).
Figure 5.12. Enterprises using cloud computing, by firm size, as a percentage of enterprises in each employment size class
Note: Percentage of all businesses employing at least ten employees using cloud computing. Small businesses are defined as having 10 to 49 employees, large businesses are those with 250 employees or more. Data refer to 2018 except for Australia (2016/17), Brazil (2019), Canada, Switzerland and the United States (2017). Average refers to the unweighted average across the OECD reported countries.
Source: OECD (2021[18]), “ICT Access and Use by Businesses”, OECD Telecommunications and Internet Statistics (database), https://doi.org/10.1787/9d2cb97b-en (accessed on 29 January 2021).
Recent research shows that cloud services are a particularly efficient option for small and young firms, because they allow the use of storage infrastructure technology without the large investment that in-house infrastructure would require (DeStefano, Kneller and Timmis, 2014[30]). Policies incentivising the adoption of cloud computing could therefore especially benefit young and smaller firms and help them move to remote work. Better understanding the policy incentives for cloud uptake is therefore important. OECD research focusing on Germany and the United Kingdom (Andres et al., 2020[31]) shows that capital incentive policies targeting investments in physical capital have the unintended consequence of reducing the probability of cloud adoption.
Besides knowing whether or not firms are using cloud services, it is crucial to understand how they are using this technology. Recent analyses on cloud use by Italian firms show that around 30% of those that assert they are using a cloud service are in fact only using it to substitute for mailboxes or office suits (Manaresi and Calvino, forthcoming[32]). More disruptive usage of cloud services, such as data storage and analytics, is positively linked to the availability of broadband connection, suggesting that investing in infrastructure may ultimately boost both cloud adoption and its effective use.
Productivity during COVID-19 and beyond
Implications of the COVID-19 crisis for productivity
The many changes brought about by the COVID-19 pandemic have had, and continue to have, various implications for productivity. While some of these changes and consequent drops in productivity are likely to be temporary – and disappear once the pandemic is over and the restrictions on contact and mobility are lifted – others are more long-term or here to stay.
There were some clear negative initial impacts of the crisis on economic activity and productivity. Analyses on UK data show that aggregate total factor productivity fell by up to 5% in the last quarter of 2020 (Bloom et al., 2020[33]). Besides the direct restrictions on activity, changes in consumer behaviour (for instance, precautionary or wait-and-see purchase delays by consumers) have added an additional challenge in the short term. Importantly, interactions among firms and among workers – which are key for knowledge spillovers and to boost productivity – remained reduced for prolonged periods of time, with potential longer-term effects in terms of suppressed productivity through missed opportunities and innovations. Productivity can also remain below its potential if firm financial shortages (discussed in more detail in Chapter 4) translate into longer-term financing constraints, hindering the realisation of productivity-enhancing investments. The massive spike in uncertainty – larger than the one associated with the global financial crisis of 2008-0910 – exacerbates the concerns on limited or delayed investments, and the negative effects on productivity. Further long-term challenges with an initial drag on productivity relate to the need to re-train or upskill workers whose jobs have become redundant through the pandemic and the induced shifts in consumer preferences, automation, or digitalisation (more on this topic is discussed in Chapter 7).
During the recovery phase, alongside the phasing out of government emergency support measures, a wave of business closures may have negative effects on aggregate productivity if productive firms go out of business. What is more, while government support measures have helped prevent widespread bankruptcies in most countries (as shown in more detail in Chapter 4), they may have also helped sustain unviable firms, keeping them solvent them through the crisis and into the recovery period, thereby dragging down average productivity as well.
Against this negative backdrop, there have also been positive developments, many of which are likely to be long lasting – in particular those related to digital technology uptake. The pandemic has induced shifts towards digitalisation, which have been particularly relevant in sectors that have traditionally low productivity (e.g. non-market services), or that have been more reluctant to digitalise. In turn, in the medium and long run, this has the potential to boost productivity in these sectors, with likely positive effects on aggregate productivity.
The economic crisis has also more broadly uncovered and created new business opportunities. Some of these have catered to needs that are specific to the pandemic, but many have also responded to what will likely be longer-term changes in consumer and firm behaviour and demand. Some of these changes may alter the modus operandi of entire industries, inducing persistent shifts in societal norms and consumer habits or needs, which represent valuable business opportunities for start-ups, and chances for radical and disruptive innovation. More specifically, and discussed for example in the sections on telework and e-commerce in this chapter, the COVID-19 outbreak is likely to induce persistent demand for remote work and modes of supply of products and services. If accompanied by the right policies, widespread telework can by itself entail productivity gains (see more detail in Box 5.2).
A further productivity-enhancing impact of the COVID-19 crisis could be its cleansing effect, associated with a reallocation of resources to more efficient uses. The strength of this effect will, of course, depend on whether less productive firms will exit the market in the medium run, and can be enhanced by ensuring that new firms can enter, and radical innovators and more productive firms are able to invest and thrive in the longer run.
Beyond the aggregate effects, the COVID-19 pandemic has had very heterogeneous effects across industries (as discussed in Chapter 3) and firms, many of which relate specifically to pre-existing digital divides, as discussed in detail in the following section. Digital-ready firms have significant advantages, as they are more prepared to cope with the shifts induced by the COVID-19 pandemic. Among them, frontier firms are generally better equipped, given their technological advantages and managerial practices. Large firms have additional advantages along other dimensions, with potential adverse consequences on concentration, as discussed in Chapter 4.
Box 5.2. What may happen to productivity if widespread telework becomes the “new normal”?
The COVID-19 crisis may catalyse a wider use of telework beyond the pandemic, with more widespread telework becoming the “new normal”. Evidence on telework before or during the crisis shows that while telework can raise productivity, this is dependent on circumstances, thus highlighting the role of complementary factors. These factors can be at the firm and worker level, across a spectrum of areas, including management, skills, communications infrastructure and an appropriate working environment at home (Bloom, Mizen and Taneja, 2021[9]; Morikawa, 2021[34]; Bloom et al., 2014[35]; Institut Sapiens, 2021[36]). How this unprecedented shift in working models will affect firm performance after the crisis therefore remains to be seen.
Existing evidence on telework before the crisis offers interesting insights into the mechanisms through which telework can improve or harm productivity in the longer term. Several lessons stand out:
Telework has the potential to raise or harm firm productivity through several, countervailing channels (Figure 5.13).
Firms may be able to reduce costs (e.g. through reduced need for office space), but fewer face-to-face interactions in the workplace can impair communication, limit managerial oversight and reduce knowledge flows. To the extent that workers face fewer distractions and feel more content about their work, their efficiency may increase, but telework can also lower their satisfaction.
Worker satisfaction is key to improve productivity. To realise gains, telework has to improve worker satisfaction and efficiency enough to offset the negative effects (e.g. from fewer face-to-face interactions). Thus, genuine efficiency gains are possible if workers and firms alike benefit from telework.
This means that there is an ideal level of telework – with too much or too little resulting in less desirable effects on productivity. Efficiency gains are driven by satisfaction, which may increase most at low – but suffer at very high – levels of telework, and rely on workers having some ability to choose whether and how much to telework. Efficiency losses arise because fewer opportunities for face-to-face interactions imply fewer opportunities for informal information sharing, co-ordination, and so on. This implies an inversely U-shaped relationship between telework and efficiency with a “sweet spot” at intermediate levels of telework.
These lessons are corroborated by ongoing analysis conducted by the OECD Global Forum on Productivity in co-operation with business associations (Business at OECD [BIAC]) and trade unions (the Trade Union Advisory Committee of the OECD [TUAC]). Preliminary survey results confirm that managers and worker representatives generally expect regular telework to become the norm for most employees post COVID-19, with most respondents regarding intermediate levels of telework (two to three days per week) as ideal. Results also confirm that higher productivity is among the main expected benefits of more widespread telework, while highlighting the need for synchronisation of telework schedules among workers, and for telework to remain voluntary.
This area of research therefore suggests that policies can raise productivity and benefit workers by: 1) promoting an optimal level of voluntary telework; 2) facilitating the dissemination of managerial best practices in support of telework arrangements; 3) providing fast, reliable, and secure communications infrastructure; and 4) assuring an appropriate working environment at home.
Figure 5.13. Countervailing channels for telework affecting productivity
Source: OECD (2020[37]), “Productivity gains from teleworking in the post COVID-19 era: How can public policies make it happen?”, https://doi.org/10.1787/a5d52e99-en.
Pre-existing divides across sectors and firms
Prior to the COVID-19 shock, the digital transformation had spread unevenly across countries, sectors, and firms. Economic agents in different countries have adopted digital technologies at a different pace, with prominent cross-country differences in adoption found in cloud computing, online sales and online purchases (OECD, 2019[20]), as discussed in the previous sections of this chapter. Diffusion of digital technologies also varies widely between sectors of economic activity, as also shown by a recent OECD taxonomy that classifies sectors according to their digital intensity, taking into account different facets of digital transformation (Calvino et al., 2018[38]). Business dynamism – in particular, firm entry – has also declined more strongly in digital and intangible-intensive sectors, reflecting differences in the extent to which new firms are able to adopt and benefit from digital technologies and invest in complementary intangible assets (Calvino and Criscuolo, 2019[39]; Calvino, Criscuolo and Verlhac, 2020[40]). Chapter 4 of this volume provides a broader coverage of the topic in the context of business dynamism.
Difficulties faced by young, small, and less productive firms in adopting and effectively using new and digital technologies can been traced back to the rising importance of complementary intangible assets that entail economies of scale and network effects that generate winner-takes-most dynamics (Corrado et al., forthcoming[41]).11 The COVID-19 shock has the potential to exacerbate these patterns, favouring market leaders who had already embraced the digital transformation more, and may therefore be better able to cushion and withstand the impact of the shock, or even use the COVID-19 pandemic to their advantage, thus improving their performance and competitiveness.
Importantly, there is evidence of an increasing digital gap between firms within sectors. This reflects and reinforces a growing divide between firms at the global productivity frontier in each industry, which continually improve performance, and the rest, which lag behind (Andrews, Criscuolo and Gal, 2016[12]). Evidence of the tight relationship between digitalisation and the divide between frontier and laggard firms is presented in Figure 5.14. The figure plots estimates of the speed at which firms in the bottom of the productivity distribution move towards the productivity frontier, relative to that of firms belonging to the middle of the distribution. This differential speed of “catch-up” towards the frontier is related to different industry characteristics. The figure shows that in more digital- and knowledge-intensive sectors, laggard firms catch up with the frontier at a relatively slower pace, thus indicating barriers to technology and knowledge diffusion (Berlingieri et al., 2020[42]; Calvino et al., 2018[38]).12 Similar patterns emerge when considering intangible assets, such as skills and organisational capital, which are crucial complements to digital technologies and may also widen the divide between frontier (intangible-rich) and laggard firms. Ongoing OECD analysis (Corrado et al., forthcoming[41]) shows that these intangibles contribute also to productivity divergences across countries and sectors.
The COVID-19 shock may exacerbate productivity gaps and further increase industry concentration. Because the most productive firms were better able to develop, adopt and use digital technologies well before the pandemic struck, they are likely to suffer relatively less from the disruptions caused by containment measures. Their pre-existing digitally-intensive functioning allows them to shift more easily towards business models that utilise digital technologies such as e-sales, e-purchases, cloud storage and teleworking.
Moreover, the availability of complementary knowledge-based assets – such as skills and organisational capital – allows the large and productive firms that already use digital technologies more effectively to advance faster, further widening the gap between more productive firms and laggards. This may in turn shape aggregate competition and market concentration dynamics. Increases in industry concentration and mark-ups have been higher in sectors relying on intangibles complementary to ICT technologies (Bajgar, Criscuolo and Timmis, forthcoming[43]; Calligaris, Criscuolo and Marcolin, 2018[44]). Declines in business entry and job reallocation have also been more prominent in digital-intensive sectors, driven by winner-takes-most dynamics and barriers to technology diffusion (Calvino, Criscuolo and Verlhac, 2020[40]).
Figure 5.14. Estimated differences in labour productivity catch-up in industries with different ICT-related characteristics
Note: The figure shows the estimated difference in labour productivity (LP) growth, due to the catch-up effect, between firms at the average level of LP gap in the percentile (0-10) group and firms at the average LP gap in the percentile (10-40) group, in industries with low vs. high values of the indicators considered. The time period covered is 2000-12 and the countries included are Australia, Belgium, Canada, Denmark, Finland, France, Hungary, Ireland, Italy, Norway, Portugal, Sweden and Switzerland. See the original source for further details.
Source: Berlingieri et al. (2020[42]), “Laggard firms, technology diffusion and its structural and policy determinants”, https://doi.org/10.1787/281bd7a9-en.
Improving digital skills also has implications for inequality and inclusiveness, as discussed in Chapter 7. Digital skills have been crucial for workers to get through the crisis, not only in terms of continuing to work, but also for participation in other aspects of life. Re- and up-skilling of workers is also important in view of the recovery period, where trends in automation, affecting primarily workers in the middle of the skill distribution, are likely to be accelerated in view of increased resilience to social distancing through automation. Increasing digital uptake of firms should therefore go hand in hand with improving digital skills of the population, in order not to exacerbate pre-existing divides across workers.
Moving towards more inclusive economies and avoiding a widening of the productivity divide, further decreases in firm entry and increased market concentration requires policy makers to foster and support the digital transformation of the whole population of businesses. The key features of the policy mix suggested for this purpose are listed in the following section.
Conclusions and policy implications
The COVID-19 pandemic demonstrates the crucial role of digital technologies for economic resilience. The ability of firms and workers to quickly shift to telework is particularly important in light of social distancing restrictions. The unprecedented scale of telework during the crisis may further catalyse a permanent shift in working modes, with widespread telework becoming the “new normal” after the crisis as well.13
Teleworking allowed firms to maintain activity and production during the crisis, and may have a broad range of positive long-term outcomes, such as improving productivity and work-life balance, while reducing regional inequalities and lowering greenhouse gas emissions from commuting. However, there are large differences between firms and workers in their ability to telework and reap these benefits. Some of these differences can be addressed by policies removing barriers to uptake, such as a lack of appropriate communications infrastructure, or skill upgrading. However, other reasons for differences in uptake, such as inherent task-related explanations, will persist. Ensuring well-being gains also for workers who are not able to benefit from the increased flexibility associated with telework should therefore also remain on the policy agenda.
The gap in digital technology use between the “best” firms and “the rest” may be exacerbated at a time when shifts to digital modes of production are becoming more and more important. Thus, stimulating technology and knowledge diffusion, and ensuring that the benefits of the digital transformation can be shared across firms and workers, should remain key priorities for policymakers for a swift recovery in the aftermath of the COVID crisis. This is particularly relevant given the substantial resources that are going to be included in recovery packages implemented by governments worldwide.
Governments need to facilitate the use of telework with supportive legislation and regulation on digital security and other pertinent issues (e.g. on health and safety, or the right to telework and right to disconnect), as well as infrastructure investment and skills development. Ensuring inclusiveness would also entail enhancing access to digital tools and infrastructure in rural areas, and endowing all individuals with the skills to live, work and thrive in the digital era.
Communications infrastructure and access to broadband are crucial to improve economic resilience and allow firms and workers to benefit from telework and online production and commerce (DeStefano, Kneller and Timmis, 2014[30]). Measures to stimulate investment and competition in communications infrastructure (OECD, 2020[45]) and bridge the urban-rural connectivity divide (e.g. phasing out slower Internet connections by deploying fibre deeper into the broadband network of providers) can go a long way in this respect. The diffusion of improved ICT infrastructure within firms can also be accelerated through financial support for upgrades aimed at performing a number of functions online, including to implement teleworking (OECD, 2020[46]). Fostering digitally-enabled transactions of trade in goods and services by removing barriers can in itself act to accelerate the digital transformation – including the wider use of telework – by reducing access costs for digital networks and equipment (OECD, 2020[47]).
Ensuring the endowment of fundamental skills, such as numeracy, literacy and management, as well as specific (namely ICT) skills, in an equitable manner – across regions and population groups – is crucial to enhance firms’ digital potential and reduce skills-based income inequality. Improving digital skills may help reduce the digital divide (discussed in more detail in Chapter 7), and allow more workers to telework. This can in turn increase efficiency and productivity, and improve job satisfaction. Investments in relevant skills among workers currently less able to telework will be important for the coming economic recovery, and future resilience. Because many workers who already possess skills necessary for telework (e.g. in knowledge-intensive services) are largely concentrated in urban areas, large gains may be obtained by up-skilling workers in rural areas. Promoting online education is particularly suited to provide training opportunities beyond the reach of large cities (Clancy, 2020[48]).
In addition to measures aimed at improving connectivity and skills, governments can take steps to encourage digital uptake by tackling legal and cultural hurdles to telework and ICT use, and mitigating any potential side effects such as cybercrime or data security concerns (OECD, 2020[49]). Such measures may include updating the legal and regulatory framework, to allow workers to telework or work under flexible conditions.
To assist in reversing the long-term decline of productivity growth, and to support sustainability and inclusiveness, government recovery packages should include measures aimed at supporting the digital transformation of the business sector beyond telework and e-commerce. The analysis in this chapter has highlighted significant differences in the uptake and use of broader digital technologies, across countries, sectors and types of firms. In this context, policy should boost digital technology diffusion at the firm level to ensure an inclusive digital transformation, with benefits spread across firms and workers.
Key policy levers act on both the demand and the supply of technology and knowledge. Firm demand can be fostered by increasing firms’ awareness of new technologies, as well as knowledge about their use and the benefits of adopting them. Tools to achieve this include dedicated instruments (e.g. online information platforms), business advisory support, and initiatives to encourage knowledge sharing network creation (including through participating in global value chains, thereby connecting with other firms and fostering technology transfer). Policies aimed at reducing barriers to diffusion and developing firms’ absorptive capacity and ability to use new technology effectively are also important. This can be done by focusing on skill endowment and upgrade, including through fostering mobility, and providing access to finance for training and investment in research and development (R&D) – especially at a time in which firms are in a fragile financial position. Policies that improve the overall business climate can also increase the demand for technology, as they allow potential adopters to grow and thrive. These include policies aimed at increasing competition and those that improve the efficiency of resource allocation (e.g. by improving insolvency regimes).
Policies aimed at supporting the development of technology and knowledge by leading firms and innovators are complementary to measures focused on technological diffusion. A dynamic and innovative ecosystem is crucial in order to make technological breakthroughs more widely applicable and affordable, so that radical innovations diffuse through the market. Supply-side policies should support both incremental and science-based innovation through strong public research systems and tight science-industry linkages, and should encourage experimentation.
Fostering technology diffusion and experimentation, in particular of digital technologies, not only boosts long-term productivity and economic growth, it but can also help transform the business sector to become more resilient. Measures to enhance digital technologies and use, as well as supporting factors such as skills and infrastructure, would thus bring double dividends for policy makers, as they would also foster other desirable outcomes such as business dynamism and inclusiveness, and can bring environmental improvements by reducing greenhouse gas emissions. The crisis provides a unique window of opportunity to implement a wide range of policies that can simultaneously tackle long-term challenges and support an enduring and inclusive period of economic growth.
References
[31] Andres, R. et al. (2020), “Capital incentive policies in the age of cloud computing: An empirical case study”, OECD Science, Technology and Industry Working Papers, No. 2020/07, OECD Publishing, Paris, https://doi.org/10.1787/4bedeb36-en.
[12] Andrews, D., C. Criscuolo and P. Gal (2016), “The Best versus the Rest: The Global Productivity Slowdown, Divergence across Firms and the Role of Public Policy”, OECD Productivity Working Papers, No. 5, OECD Publishing, Paris, https://doi.org/10.1787/63629cc9-en.
[17] Andrews, D., G. Nicoletti and C. Timiliotis (2018), “Digital technology diffusion: A matter of capabilities, incentives or both?”, OECD Economics Department Working Papers, No. 1476, OECD Publishing, Paris, https://doi.org/10.1787/7c542c16-en.
[29] Bailin Rivares, A. et al. (2019), “Like it or not? The impact of online platforms on the productivity of incumbent service providers”, OECD Economics Department Working Papers, No. 1548, OECD Publishing, Paris, https://doi.org/10.1787/080a17ce-en.
[43] Bajgar, M., C. Criscuolo and J. Timmis (forthcoming), “Supersize Me: Intangibles and Industry Concentration”, OECD Productivity Working Papers, OECD Publishing, Paris.
[42] Berlingieri, G. et al. (2020), “Laggard firms, technology diffusion and its structural and policy determinants”, OECD Science, Technology and Industry Policy Papers, No. 86, OECD Publishing, Paris, https://doi.org/10.1787/281bd7a9-en.
[33] Bloom, N. et al. (2020), “The Impact of Covid-19 on Productivity”, NBER Working Papers, No. 28233, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w28233.
[35] Bloom, N. et al. (2014), “Does working from home work? Evidence from a Chinese experiment”, The Quarterly Journal of Economics, pp. 165-218, https://doi.org/10.1093/qje/qju032.
[9] Bloom, N., P. Mizen and S. Taneja (2021), “Working from home is revolutionising the UK labour market”, VOX, CEPR Policy Portal, https://voxeu.org/article/working-home-revolutionising-uk-labour-market.
[28] Bulman, T. and S. Koirala (2020), “The OECD COVID-19 Policy Tracker: What are governments doing to deal with the COVID-19 pandemic?”, OECD Ecoscope blog post 15 May, https://oecdecoscope.blog/2020/05/15/the-oecd-covid-19-policy-tracker-what-are-governments-doing-to-deal-with-the-covid-19-pandemic/.
[44] Calligaris, S., C. Criscuolo and L. Marcolin (2018), “Mark-ups in the digital era”, OECD Science, Technology and Industry Working Papers, No. 2018/10, OECD Publishing, Paris, https://doi.org/10.1787/4efe2d25-en.
[39] Calvino, F. and C. Criscuolo (2019), “Business dynamics and digitalisation”, OECD Science, Technology and Industry Policy Papers, No. 62, OECD Publishing, Paris, https://doi.org/10.1787/6e0b011a-en.
[38] Calvino, F. et al. (2018), “A taxonomy of digital intensive sectors”, OECD Science, Technology and Industry Working Papers, No. 2018/14, OECD Publishing, Paris, https://doi.org/10.1787/f404736a-en.
[40] Calvino, F., C. Criscuolo and R. Verlhac (2020), “Declining business dynamism: Structural and policy determinants”, OECD Science, Technology and Industry Policy Papers, No. 94, OECD Publishing, Paris, https://doi.org/10.1787/77b92072-en.
[48] Clancy, M. (2020), “The Case for Remote Work”, Economic Working Papers, No. 20007, Iowa State University, Department of Economics, http://lib.dr.iastate.edu/econ_workingpapers/102/.
[41] Corrado, C. et al. (forthcoming), “New Evidence on Intangibles, Diffusion, and Productivity”, OECD Science, Technology and Industry Working Papers, OECD Publishing, Paris.
[21] CRTC (2019), Communications Monitoring Report, 2019, https://crtc.gc.ca/pubs/cmr2019-en.pdf.
[30] DeStefano, T., R. Kneller and J. Timmis (2014), “The (Fuzzy) Digital Divide: The Effect of Broadband Internet Use on UK Firm Performance”, Discussion Papers, No. 14/06, University of Nottingham, School of Economics, https://www.nottingham.ac.uk/economics/documents/discussion-papers/14-06.pdf.
[15] Espinoza, R. and L. Reznikova (2020), “Who can log in? The importance of skills for the feasibility of teleworking arrangements across OECD countries”, OECD Social, Employment and Migration Working Papers, No. 242, OECD Publishing, Paris, https://doi.org/10.1787/3f115a10-en.
[8] Eurofound (2020), “Teleworkability and the COVID-19 crisis: a new digital divide?”, Working Paper, No. WPEF20020, Publications Office of the European Union, Luxembourg, https://www.eurofound.europa.eu/sites/default/files/wpef20020.pdf.
[16] Eurofound (2017), European Working Conditions Survey, 2015 (data collection), UK Data Service. SN: 8098, http://doi.org/10.5255/UKDA-SN-8098-4.
[22] European Commission (2020), Study on Broadband Coverage in Europe 2019, https://digital-strategy.ec.europa.eu/en/library/broadband-coverage-europe-2019.
[23] FCC (2019), 2019 Broadband Deployment Report 12, https://docs.fcc.gov/public/attachments/FCC-19-44A1.pdf.
[13] Gal, P. et al. (2019), “Digitalisation and productivity: In search of the holy grail – Firm-level empirical evidence from EU countries”, OECD Economics Department Working Papers, No. 1533, OECD Publishing, Paris, https://doi.org/10.1787/5080f4b6-en.
[36] Institut Sapiens (2021), “Quel avenir pour le télétravail? Pérenniser et sécuriser une pratique d’avenir”, Note, Mars 2021, https://www.institutsapiens.fr/wp-content/uploads/2021/03/Quel-avenir-pour-le-teletravail.pdf.
[32] Manaresi, F. and F. Calvino (forthcoming), “Closing the Italian Digital Gap: the Role of Skills and Intangible Investments”, OECD Science, Technology and Industry Working Papers, OECD Publishing, Paris.
[34] Morikawa, M. (2021), “The productivity of working from home: Evidence from Japan”, VOX, CEPR Policy Portal, https://voxeu.org/article/productivity-working-home-evidence-japan.
[10] Nachtigall, H. and M. Squicciarini (forthcoming), “Demand for AI Skills in Jobs: Evidence from Online Job Postings”, OECD Science, Technology and Industry Working Papers, OECD Publishing, Paris.
[19] OECD (2021), Broadband Portal (database), https://www.oecd.org/sti/broadband/oecdbroadbandportal.htm (accessed on 29 April 2021).
[50] OECD (2021), “Building a more connected world: How the OECD Recommendation on Broadband Connectivity is setting the pace”, OECD Innovation Blog 28 April, https://oecd-innovation-blog.com/2021/04/28/building-a-more-connected-world-oecd-broadband-connectivity-recommendation/.
[1] OECD (2021), “ICT Access and Usage by Households and Individuals”, OECD Telecommunications and Internet Statistics (database), https://doi.org/10.1787/b9823565-en (accessed on 29 January 2021).
[18] OECD (2021), “ICT Access and Use by Businesses”, OECD Telecommunications and Internet Statistics (database), https://doi.org/10.1787/9d2cb97b-en (accessed on 29 January 2021).
[27] OECD (2021), “The role of online platforms in weathering the COVID-19 shock”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/2a3b8434-en.
[51] OECD (2020), “Capacity for remote working can affect lockdown costs differently across places”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/0e85740e-en.
[11] OECD (2020), “Cities policy responses”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/fd1053ff-en.
[45] OECD (2020), “COVID-19 and the low-carbon transition: Impacts and possible policy responses”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/749738fc-en.
[26] OECD (2020), “COVID-19 and the retail sector: impact and policy responses”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/371d7599-en.
[6] OECD (2020), Digital Transformation in the Age of COVID-19: Building Resilience and Bridging Divides, OECD Publishing, Paris, https://www.oecd.org/digital/digital-economy-outlook-covid.pdf.
[2] OECD (2020), “E-commerce in the times of COVID-19”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://www.oecd.org/coronavirus/policy-responses/e-commerce-in-the-time-of-covid-19-3a2b78e8.
[49] OECD (2020), “Keeping the Internet up and running in times of crisis”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/4017c4c9-en.
[47] OECD (2020), “Leveraging digital trade to fight the consequences of COVID-19”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/f712f404-en.
[3] OECD (2020), OECD Digital Economy Outlook 2020, OECD Publishing, Paris, https://doi.org/10.1787/bb167041-en.
[37] OECD (2020), “Productivity gains from teleworking in the post COVID-19 era: How can public policies make it happen?”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/a5d52e99-en.
[46] OECD (2020), “Supporting people and companies to deal with the COVID-19 virus: Options for an immediate employment and social-policy response”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris, https://doi.org/10.1787/d33dffe6-en.
[52] OECD (2019), Going Digital: Shaping Policies, Improving Lives, OECD Publishing, Paris, https://doi.org/10.1787/9789264312012-en.
[20] OECD (2019), Measuring the Digital Transformation: A Roadmap for the Future, OECD Publishing, Paris, https://doi.org/10.1787/9789264311992-en.
[25] OECD (2019), Skills Matter: Additional Results from the Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/1f029d8f-en.
[5] OECD (forthcoming), Measuring telework in the COVID-19 pandemic.
[24] Reuters (2020), “German labour minister wants to put right to home working in law,” 26 April 2020, https://www.reuters.com/article/us-health-coronavirus-germany-homeworkin/german-labour-minister-wants-to-put-right-to-home-working-in-law-idUSKCN2280MI.
[4] Riom, C. and A. Valero (2020), “The Business Response to Covid-19: the CEP-CBI survey on technology adoption”, CEP Covid-19 Analysis Paper, No. 009, Centre for Economic Performance, London, https://cep.lse.ac.uk/pubs/download/cepcovid-19-009.pdf.
[14] Sorbe, S. et al. (2019), “Digital Dividend: Policies to Harness the Productivity Potential of Digital Technologies“”, OECD Economic Policy Papers, No. 26, OECD Publishing, Paris, https://doi.org/10.1787/273176bc-en.
[7] UNCTAD; NetCommSuisse (2020), COVID-19 and E-commerce, https://unctad.org/system/files/official-document/dtlstictinf2020d1_en.pdf (accessed on 14 December 2020).
Notes
← 1. Growth as measured by Internet Exchange Points. See OECD (2021[50]).
← 2. For a review of telework uptake across several countries by firms and workers during the COVID-19 pandemic, see OECD (forthcoming[5]). For a discussion how telework capacities differ within countries, see OECD (2020[51]).
← 3. For more on this policy, see www.weforum.org/agenda/2019/08/finland-s-doing-something-cool-with-flexible-working/.
← 4. Note that the data on fast broadband presented in Figure 5.5 and Figure 5.4 both refer to actual uptake by individuals and firms, and hence depend not only on the availability of communication infrastructure, but also on the demand for broadband services by individuals and firms. Both indicators are therefore likely to understate the technical possibilities of what communications infrastructure could provide if uptake was universal.
← 5. Note that further correlation analysis (not shown) on 2019 data does not find an additional positive relationship of telework uptake with the share of fibre connections (allowing very fast upload and download speeds) in total fixed broadband, going beyond the positive relationship with fast broadband connections depicted in Figure 5.5.
← 6. Coverage is an indicator provided by broadband suppliers, and refers to the number of fixed broadband subscriptions with contracted speed faster than 25/30 Mbps per 100 inhabitants.
← 7. Note that the gaps are not population weighted, so that the absolute number of households affected are relatively low in sparsely populated countries such as Finland. In these countries, mobile connections are likely to also play a more important role for people to stay connected (OECD, 2019[52]).
← 8. However, these productivity gains are conditional on the market of platforms being contestable, to avoid the risk that certain large players become dominant and lower innovation efforts.
← 9. While the analysis here focuses only on cloud services, there are other technologies, such as digital platforms, that enable firms and workers to store, access, and collaborate on data and projects. It is, however, unlikely that firms using these more sophisticated platforms are not also able to use cloud technologies.
← 10. Major uncertainties related to productivity include: the duration and effectiveness of social distancing; market lockdowns; whether “temporary” government interventions and policies will persist; the extent to which pandemic-induced shifts in consumer spending patterns will persist; and the impact on business survival, new business formation, R&D, and human capital investment. The uncertainty is exacerbated by the lack of close historic parallels to the current crisis.
← 11. Note that some older, large firms may also face difficulties in adoption due to outdated IT systems, or because they are stuck in old business models and lack managerial capabilities (OECD, 2020[37]).
← 12. Similarly, previous work looking at the gains from sector-level digital adoption at various segments of the productivity distribution found that it is mostly the most productive firms that benefit (Gal et al., 2019[13]).
← 13. Reflecting the growing importance of connectivity, which has been demonstrated more clearly than ever by the COVID-19 crisis, see also an amended OECD Recommendation of the Council on Broadband Connectivity, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0322.