Overview
Over the past 25 years, information and communication technology (ICT) has transformed both the economy and society. More than 5 billion people, or over two thirds of the global population, have access to the Internet,https://unsdg.un.org/latest/stories/wired-good-digital-connectivity-sustainable-future while three-quarters of the world now own a mobile phone.https://www.itu.int/itu-d/reports/statistics/2022/11/24/ff22-mobile-phone-ownership/
This vast, dynamic digital ecosystem has given birth to new industries, like social media, and revolutionised existing ones, such as finance, via innovations including digital banking and payments.
ICT's march won't end there. Continued advances in fundamental technologies, in particular semiconductors and networking infrastructure, are now giving rise to breakthroughs in artificial intelligence (AI). This will enable the USD23 trillion ICT industry to become even more influential, affecting sectors as diverse as retail, healthcare and industrials.The IT and Communication Services sectors of the MSCI All Country World Index, data as of 30.11.2023
This is inevitably transforming investment portfolios, too.
Technology companies account for an increasingly large part of investors’ holdings, irrespective of whether those assets are passively or actively managed. To those old enough to have experienced the dot.com boom and bust, this may stir uncomfortable memories. Yet there are arguably more problematic aspects to the tech sector's rise. Chief among them is the idea that tech companies also adhere to the highest environmental, social and governance (ESG) standards.
Indeed, tech stocks represent an even bigger slice of ESG-aligned portfolios: the S&P 500 index’s tech sector makes up the largest allocation in many popular ESG funds.https://www.cnbc.com/2021/05/18/alphabet-was-the-most-widely-held-stock-for-the-largest-esg-funds.html
At first glance, the ICT sector's strong ESG credentials are easy to rationalise. Many large companies in the industry have low carbon footprints while the sector is also home to firms developing products and services that contribute to the building of a more sustainable economy.
But that's not to say ESG investors can afford to be complacent. The tech sector's dynamism means that its ESG credentials will also be in flux.
All of which means it is more important than ever for investors to keep abreast of the ESG risks and opportunities emerging within the ICT industry at large.
ESG considerations for tech companies are diverse and complex. They fall into two broad categories:
- Material financial risks and opportunities for the tech sector, or issues that can have an impact on the bottom line. These issues can adversely affect companies if managed poorly or be turned into a competitive advantage for those taking a leading role in addressing them.
- Positive impact of the tech sector on the environment or society through its products and services.
By understanding the evolution of ESG within tech, we believe investors can better distinguish leaders from laggards, allocate capital more appropriately, and engage with tech companies to achieve greater positive outcomes.
Integrity and safety of products and services
It is hard to overstate the scale and the impact of certain gigantic online platforms.
Today, Apple has an installed base of more than 2 billion iPhones and iPads globally. More than 3 billion people use one of Meta’s social media products (Facebook, Instagram or WhatsApp) daily, while Google is estimated to perform over 8 billion searches every day.Apple Sept 23 earnings call, Meta Sept 23 earnings slides, https://blog.hubspot.com/marketing/google-search-statistics
From e-commerce to social media and beyond, the emergence of these huge digital platforms over the last three decades, as well as the reliance by companies on increasingly sophisticated AI algorithms, means that product and platform design decisions can have a profound impact on consumers, users and even society at large.
This also applies to non-consumer facing companies like software firms SAP, Salesforce, Adobe and ServiceNow whose cloud software platforms power hundreds of thousands of businesses globally. Small and start-up companies, in particular, benefit from access to leading-edge technologies through the cloud at prices which they would not have been able to afford traditionally thus lowering spending on internal IT infrastructure.
Material financial risks and opportunities: As the scrutiny of consumers and policymakers intensifies, so too do the regulatory and legal risks ICT companies face.
Social media has the potential to build communities and improve social interaction, but excessive reliance on digital devices and online platforms can result in issues such as excessive screen time, addiction and cyberbullying.
For example, in the US, parent groups, school districts and individual states are suing a number of social media platforms over the mental health struggles of children. While Section 230 of the 1996 US Communications Decency Act shields them from legal liability for user-generated content, the claimants argue the platforms and their algorithms are akin to a faulty product.https://www.theverge.com/2023/11/14/23960956/meta-google-tiktok-snap-social-media-addiction-lawsuits
Legal matters aside, platforms run reputational risk if they are perceived to breach ethical standards. While regulations are not comprehensive in addressing these concerns, self-regulation by companies is crucial.
Companies operating in social media must take proactive steps to protect their business and reputation by implementing measures that mitigate addiction risks, prioritise online safety, establish reporting mechanisms for harmful content, cooperate with authorities to address abuse and support mental well-being.
Meta, for example, has developed more than 30 features to support teenagers and families on its applications.
Ensuring physical safety is another priority. In 2016, Samsung lost an estimated USD1 billion after having to replace 2.5 million devices it recalled because of the risk of exploding batteries.
Yet these risks pale in comparison to the potential threats presented by unchecked development of artificial intelligence (AI).
The rapid emergence and exponential adoption of generative AI applications generate a variety of opportunities but also new risks, many of which are not fully appreciated.
Generative AI offers the advantages of creative content generation, automation of tasks, personalised user experiences and improved decision-making. But it also makes it easier to create malware or deepfakes that could affect a company’s operations or its product integrity.
Governments are starting to introduce rules concerning the provision of AI technology, with the European Union’s AI Act being the most comprehensivewww.europarl.europa.eu/news/en/headlines/society/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence, and US President Joe Biden’s Executive Order on AI the most recent.www.whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence
Tech companies developing AI solutions must beware of this shifting regulatory landscape. While regulations may contribute to increasing trust in AI among the public, they can also create the risk of financial and reputational damage in case of rule breaches and non-compliance.
We believe establishing best practices in the responsible development and deployment of AI is crucial for companies in the IT sector to maintain leadership.
In July, a group of seven companies, including Alphabet, Meta and Microsoft, established best practices in the field of AI. They made a commitment to share relevant information among themselves and with governments and researchers. Their principles encompass prioritising security, ensuring transparency with the public and conducting thorough internal product testing prior to public release.
Data security and privacy
Material financial risks and opportunities: With the average cost of “megabreaches” – incidents that involve the loss of least 10 million records – hitting USD400 million per breach, data security and increasing regulation are among the important ESG-related risks facing the tech industry.www.ibm.com/reports/data-breach
Publicly reported data compromises* by US companies
*data compromises include data breaches, data exposures and data leaks. Source: Identity Theft Resource Center, 2022 Data Breach Report.
Tech companies that rely heavily on user or consumer data are particularly vulnerable to data breaches, as they could damage the firm’s reputation and lead to a loss of market share.
Take the 2017 data breach at Equifax, a leading consumer credit reporting company. While the firm may not technically belong to the ICT sector, it has a business model that relies heavily on data, demonstrating how serious the effects can be. After its data on almost 150 million US consumers was leaked publicly online, Equifax was fined USD275 million by the Federal Trade Commission and ordered to set up a USD300 million fund for victim compensation.www.ftc.gov/news-events/news/press-releases/2019/07/equifax-pay-575-million-part-settlement-ftc-cfpb-states-related-2017-data-breach
Similarly, a breach suffered by Yahoo in 2013 almost derailed the Internet group’s takeover by Verizon, which ended up cutting USD350 million from its original offer.https://www.nytimes.com/2017/10/03/technology/yahoo-hack-3-billion-users.html
Data privacy has also become an increasingly important ESG issue for tech companies, and not just for the firms that implement more stringent privacy practices.
The fact is that changes in one company's policies to protect customers can have a material impact on other businesses within the supply chain. Take Apple. It changed its policy in 2021 to require consumers to opt in for data tracking by third-party apps on its iOS platform. This subsequently reduced advertising sales for a number of major companies, including many in tech and communications.
Positive impact: Certain tech companies can help other businesses mitigate cyber security risks. For example, AI-enabled smart security intelligence and automation management solutions such as those provided by Palo Alto Networks reduce the time needed to detect and counter cyber attacks to hours and minutes from days or weeks. That such services will be in increasing demand is not in doubt. In the third quarter of 2023, the frequency of ransomware attacks nearly doubled over the same period last year.https://www.corvusinsurance.com/blog/q3-ransomware-report?utm_campaign=FY23-Q4-Quarterly%20Ransomware%20Report&utm_source=ransomware%20blog&utm_medium=press The US Securities and Exchange Commission, meanwhile, has unveiled a new rule requiring public companies to issue periodic risk management, strategy and governance reports on cybersecurity risks, as well as to disclose material security breach incidents within four business days.
Corporate governance
Ownership structure
Material financial risks and opportunities: A common corporate governance red flag among tech companies is the existence of dual-class shares. In general, one class of shares, often exclusive to the founders of a company or a select group chosen early on, carries greater voting rights than the class of shares available to subsequent shareholders.
The rationale for having multiple classes of shares is to allow founders, executives, or other key stakeholders to retain control and make strategic decisions without being overly influenced by short-term shareholder interests. This structure provides stability and continuity, allowing companies to focus on their long-term goals and vision despite market pressures.
However, such structures can also create a wedge between ownership and voting rights and could prevent active oversight of the company’s founder or chief executive officer.
Dual-class shares exist at tech giants such as Meta, and were blamed by investors for the company's lack of strategic direction. A decline in the firm's stock price and public criticism of its leadership by activist shareholders finally led Meta’s to refocus on profitability. This was rewarded by investors, with its share price having recovered.
The Council for Institutional Investors has adopted a nuanced position on dual-class stock.Council of Institutional Investors, Dual-Class Stock (2021); https://www.cii.org/dualclass_stock While recognising that it can offer some stability during a company's early stages, they recommend unifying the two classes within a maximum of seven years. This recommendation follows academic evidence indicating that dual-class stock, while valuable in the initial period post-IPO, typically has a negative performance impact beyond a certain timeframe.Cremers, Lauterbach and Pajuste (2018): https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3062895
Business ethics
Material financial risks and opportunities: The tech industry has always been prone to business ethics controversies. Take instances of anti-competitive behaviour. In perhaps the most famous tech-related antitrust case, a US court ruled in 2001 that Microsoft had illegally maintained a monopoly in the personal computer market, which forced the company to modify its business practices.
In Europe, it was ordered to pay a fine of more than USD600 million for similar charges. Once all the investigations were finished, Microsoft had to share its software interfaces with third-party companies.
Regulators are also taking a proactive stance to prevent anti-competitive behaviours and foster competition in the tech and communications sector. The EU’s new Digital Markets Act, for example, aims to prevent large tech platforms from abusing their market power.
Bribery and corruption are business ethics topics that also remain relevant in the sector. For instance, telecom giant Ericsson paid a fine of over USD1 billion in 2019 after admitting to bribery and other wrongdoing in multiple countries. In 2022, the company was forced to pay a further penalty of USD200 million for breaching the 2019 prosecution deal on cooperation and disclosure provisions.
Environmental footprint vs handprint
Negative environmental impact
The tech sector is responsible for less than 3 per cent of global carbon emissions. Its footprint is smaller than that of more heavily polluting sectors, such as electricity, transport and manufacturing.
Yet with tech accounting for a growing share of the economy, its environmental footprint is expanding too. The number of Internet users worldwide has doubled and global digital traffic has grown 12-fold since 2010.UNEP
The digital system contribution to global emissions is estimated to increase to 14 per cent by 2040 – equivalent to that of the United States – unless any changes are made.
The industry is encountering a challenge where gains in efficiency, resulting in reduced resource usage per unit of production, may be counterbalanced by increased consumption of the same product. This phenomenon is commonly referred to as the Jevons paradox.www.oecd-forum.org/posts/the-jevons-paradox-and-rebound-effect-are-we-implementing-the-right-energy-and-climate-change-policies
As a result, the increase in resource usage by tech companies has to be viewed in the context of resource savings in aggregate across the entire value chain as new technology is adopted.
"Tech companies are not blind to their environmental footprint. Indeed, multiple efforts are underway to strengthen their positive handprint by reducing emissions and the use of natural resources."
The challenge also goes beyond greenhouse gas emissions. For example, data centres which power cloud computing require not only vast amounts of electricity but also large quantities of natural resources such as water.
They typically require about 3-5 million gallons of water per day, the same amount a city of 30,000-50,000 needs, to operate smoothly.https://www.washingtonpost.com/climate-environment/2023/04/25/data-centers-drought-water-use/
Similarly, water represents a key restrictive factor for sustainably expanding the semiconductor industry.
A study of 27 semiconductor companies shows it takes an average of 2.58m3 of water and 361.3 kwh of energy, and emissions of 263.9 kg of CO2 equivalent to make a single 8-inch wafer.Wang, Q. et al (2023) Environmental data and facts in the semiconductor manufacturing industry: An unexpected high water and energy consumption situation, https://doi.org/10.1016/j.watcyc.2023.01.004
All of this is not to say that tech companies are blind to their environmental footprint. Indeed, multiple efforts are underway to strengthen their positive handprint by reducing emissions and the use of natural resources.
Many large tech companies are also investing in carbon reduction initiatives such as carbon capture projects.
In 2022, Alphabet, Meta, Stripe, Shopify and McKinsey Sustainability partnered and committed to investing nearly USD1 billion to accelerate the development of technological carbon removal.
The data centre industry, for its part, is trying to capitalise on economies of scale. Computing workloads in “hyperscale” data centres – which typically have at least 5,000 servers and 10,000 square feet – are almost five times more energy efficient and six times more water efficient than traditional data centres.https://iopscience.iop.org/article/10.1088/1748-9326/abfba1
Telecom firms, for their part, are turning to AI to modernise and optimise all their network nodes to improve performance and achieve efficiency savings, especially as energy generation and provision can account for as much as 60 per cent of their annual operating expenses.https://www.gsma.com/futurenetworks/wiki/energy-efficiency-2/
Positive impact: Tech firms play a crucial role in accelerating the clean energy transition and improving energy and resource efficiency. Without the tech sector’s expertise and products, other energy-intensive industries would not be able to decarbonise.
Semiconductors, for example, are key components for clean technologies and the contribution of microelectronics to energy transition is hard to overstate.
From solar panels to wind turbines, smart grid to electric vehicles and industrial automation, these tiny chips harness, convert, transfer and store renewable energy as electricity, and help it move through the electric grid with minimum loss of power.
How semiconductors power the energy transition
Source: Pictet Asset Management
The number of power semiconductors used in the global renewable energy market is expected to grow at a compound annual rate of 8-10 per cent in the five years to 2027.https://www.reportlinker.com/p06244226/Power-Semiconductors-in-the-Global-Renewable-Energy-Market-Report-Trends-Forecast-and-Competitive-Analysis.html?utm_source=GNW
Specifically, semiconductors that are designed to optimise power conversion for maximum energy efficiency not only reduce heat production and cooling requirements but also help lower energy wastage and CO2 emissions.
Almost half of the semiconductors and equipment companies surveyed by MSCI generated revenue from products and services that contribute to at least one clean tech activities, such as renewable energy or energy efficiency.MSCI, Industry Report, Semiconductors & Semiconductor Equipment, December 2022.
Tech firms are helping reduce our reliance on natural resources in other ways.
Those that develop 3D design and product management software help businesses reduce resource consumption in areas ranging from construction and manufacturing to services.
Hardware and software solutions for e-commerce also yield positive environmental benefits – e-commerce is estimated to cut the carbon footprint of retailers by nearly three times and reduce traffic by a factor of 4-9 times compared with brick-and-mortar counterparts.https://www.oliverwyman.de/content/dam/oliver-wyman/v2/publications/2021/apr/is-ecommerce-good-for-europe.pdf
Another source of resource efficiency gains arises from the adoption of cloud computing.
As organisations shift from running their IT systems on their own on-site servers to using the large data centres provided by cloud service providers, improvements in energy efficiency can be considerable.
A 2018 study by WSP and Microsoft found efficiency gains ranging from 22 to 93 per cent, with smaller organisations estimated to benefit the most.https://www.wsp.com/en-us/insights/microsoft-cloud-computing-environmental-benefit-study Once the reliance on renewable energy used to power data centres was factored in, the life-cycle CO2 emissions reduction ranged from 72 to 98 per cent vs on-premise computing.
Technology in society
Human capital
For tech companies, human capital considerations are business critical. They are essential to attract and retain talented employees in intermediate to high-paying jobs within the industry. It is also an area where tech companies can deliver value by helping customers deal with labour supply shortages in other segments of the economy through automation.
Labour shortage indicator in the EU (%)
Source: European Commission (2023), “Employment and Social Developments in Europe 2023”
Material financial risks and opportunities: In recent years, several flagship tech companies have been involved in controversies, which had financial and reputational consequences.
In June 2022, after five years of litigation, Alphabet, Google’s parent company, paid USD118 million to settle a class-action lawsuit that alleged it had discriminated against female employees and violated California’s Equal Pay Act.
Similarly, Uber paid USD10 million to settle a class-action lawsuit with 400 female and minority software engineers in 2018.
While the financial impact of such settlements may be small for a company like Google or Uber, the reputational damage on it as an employer may be more significant in industries that are among the most affected by labour and skills shortages.
The importance of talented employees at technology firms means that companies have to be mindful of employees’ preferences and ethical perspective when determining corporate strategy.
For example, in 2018, Alphabet’s employees objected to the company’s involvement in “Project Maven”, a US Department of Defense programme to develop and deploy AI technologies, because of concerns about the potential use of AI in warfare.
After several protests and employee resignations, the company abandoned the contract and set up a new set of ethical principles regarding AI, which are now viewed as industry leading.https://ai.google/responsibility/principles/
Positive impact: Technology firms can also help other industries deal with the changing world of work. Labour markets have become tighter in recent years as a result of aging in developed countries, low labour market participation of certain population groups and the reshoring of production.
This has been compounded by a surprising and rapid withdrawal of people from segments of the workforce since the global pandemic, or otherwise known as the great resignation.https://ec.europa.eu/commission/presscorner/detail/en/ip_23_3704
Tech companies - such as those involved in automation, robotics or AI - are well placed to alleviate the problem as their products can help raising the productivity of the existing human workforce.
Human rights
Material financial risks and opportunities: The ICT industry typically subcontracts production of components. This practice shifts risks from companies’ own operations to their supply chain.
With a significant amount of subcontracting occurring in the Asia-Pacific region, issues such as labour conditions at subcontractors come to the fore. For example, Hon Hai Precision Industry – the Taiwanese assembler of iPhone and the world’s biggest contract electronics manufacturer also known as Foxconn, has been under fire for labour conditions at its Chinese factories for over a decade. Areas of concern range from employee welfare, factory unrest, violations of employment laws, as well as accidents.
This may well have contributed to the underperformance of Foxconn shares relative to regional and global technology indices since 2010.Covering period 30.06.2010 and 30.06.2023
Positive impact: At the same time, the ICT industry can offer solutions to better manage human rights risks in companies’ supply chains. Supply chain sustainability software is emerging as a promising growth area in managing supply chain risks as they allow users to onboard and track suppliers more effectively, better manage certification-related processes and set compliance requirements.
Access to knowledge and health
Material financial risk and opportunities: Digital inclusion brings new risks and challenges brought on by the ease of access to and rapid dissemination of data. Technological products have the potential to contain and contribute to biased information or even misinformation leading to overall societal harms.
AI algorithms can unintentionally amplify false information by reinforcing biases in source datasets, aligning with specific agendas, and prioritizing sensational content for user engagement.
In some instances, Tech giants, in particular social media platforms, have also been criticised for accommodating government pressure seeking to censor dissenting views. To address these risks, governments and leading companies are implementing regulations and ethical practices.
Measures such as transparency requirements, algorithmic audits, and responsible AI development guidelines are being adopted to mitigate the impact of misinformation and bias.
However, regulations are not necessarily consistent globally such as the US Communications Decency Act which does not hold any internet media company liable for the content posted on its site compared to the recent EU Digital Markets Act which does for large internet platforms. This puts the onus, therefore, back on companies on how they wish to address the issue. Prominent tech companies like Meta, Google, and Microsoft have voluntarily agreed to safeguards to address these risks.
Transparency also plays a crucial role in AI models. Lack of transparency hinders other businesses from determining if they can safely build applications relying on commercial foundation models. It also affects academics relying on these models for research, policymakers creating effective regulations, and consumers understanding model limitations or seeking to redress for any harms caused.
The Foundation Model AI Index evaluates the transparency of these models and can ultimately assess their leadership.https://hai.stanford.edu/news/introducing-foundation-model-transparency-index Leading foundational AI model providers will have to provide customers with greater assurances around data quality and governance. Google, for example, has now committed to indemnifying customers against copyright infringement claims when using its generative AI products.https://cloud.google.com/blog/products/ai-machine-learning/protecting-customers-with-generative-ai-indemnification
Positive impact: Notwithstanding legitimate concerns around product safety - including privacy, security and mental health impacts that may arise - it is undeniable that innovative technologies can also make a valuable societal contribution by removing or alleviating obstacles that limit access to basic types of infrastructure.
Three distinctive areas stand out:
Plugging the digital infrastructure gap. Access to telecommunication and the internet can be a game changer in fighting poverty and promoting economic empowerment. It should help bridge the infrastructure gap, estimated at USD15 trillion by 2040.
Education. Information technology is already contributing to democratising education. The advent of online education, in particular massive open online courses (MOOCs), has opened academic content to a much broader audience - DIY access to knowledge across many fields is now possible for individuals without access to formal education channels. The impact can be substantial particularly in developing countries where the levels of “learning poverty” -- the number of 10-year old children who cannot read and understand a simple story by the end of primary school – stands at 53 per cent for low- and middle-income countries.https://elearningindustry.com/how-can-developing-countries-leverage-elearning-platforms
Advancing health outcomes. Health records and treatment monitoring have been simplified through the use of connected devices such as wearables and virtual clinical tests. AI tools help recognise anomalies and improve diagnoses and are also expected to accelerate drug development. In 2021, more than 100 drug and biologic applications submitted to the US Food and Drug Agency included AI/machine learning components, compared with 14 in 2020.Liu Q. et all (2022), “Landscape Analysis of the Application of Artificial Intelligence and Machine Learning in Regulatory Submissions for Drug Development From 2016 to 2021”, Clinical Pharmacology & Therapeutics, Volume 113, Issue 4, 771-774. https://ascpt.onlinelibrary.wiley.com/doi/10.1002/cpt.2668 Robotics’ innovations have enabled surgeons to perform complex procedures with enhanced precision and accuracy, leading to better surgical outcomes while reducing healthcare costs.For instance, a recent study reported that a robotic surgery reduces the chance of patient readmission by half and revealed a 80 per cent decline in the prevalence of blood clots, which is a significant cause of health decline and morbidity. https://www.news-medical.net/news/20220515/Robotic-surgery-reduces-readmissions-benefits-patients-overall-recovery.aspx
Conclusion
As tech permeates every aspect of our lives and grows in its economic relevance, the ICT industry has developed an increasingly complex interaction with environmental and social challenges.
We believe that by better understanding these ESG dynamics -- both risks and opportunities – and incorporating such insights beyond traditional financial analysis can investors capitalise on a diverse set of opportunities from the tech transformation.
"By better understanding these ESG dynamics -- both risks and opportunities – and incorporating such insights beyond traditional financial analysis can investors capitalise on a diverse set of opportunities from the tech transformation."
