Shareholder engagement, defined as various strategies investors can use to influence the operational practices of the companies they hold, plays a critical role in driving corporate accountability and sustainable practices. While research suggests no direct link between governance practices and financial returns (see our last note on the topic, “Governance, Are the Efforts All in Vain?”, July 21, 2022), these practices improve the corporate environment by reducing ineffective boards, fostering better decision-making, and curbing overly dominant CEOs. Institutional investors have an outsized influence on corporate governance action. According to Broadridge, in 2022, U.S. retail investors voted only 29% of the shares they owned whereas institutional investors voted 82% of theirs.
In this week’s note, we provide a preliminary analysis of the 2024 proxy voting season results, on compensation, directors’ elections and main trends in shareholder proposals, with a focus on North America.
Paychecks Under the Microscope
Top executive compensation is assessed by shareholders in a say-on-pay vote, a process where shareholders have a non-binding vote on the compensation of top executives, typically occurring annually during a company’s annual general meeting.
Proxy advisors provide research and recommendations on how shareholders should vote on various corporate governance issues, including executive compensation and director elections. Their objective is to guide institutional investors in making informed voting decisions that align with best practices and shareholder interests. The biggest proxy advisors in the world, Institutional Shareholder Services (ISS) and Glass Lewis, significantly influence voting outcomes with negative recommendations leading to substantial declines in support on say-on-pay. For instance, a negative recommendation from ISS can lead to a 33% decrease in support for Say-on-Pay proposals, while Glass Lewis’s negative recommendations result in a 12% decline. This year, both advisors focused their decisions on issues such as non-rigorous performance goals, lack of transparency, and the use of discretion in bonus plans. These advisors’ compensation analyses scrutinize performance targets linked to remuneration rather than just looking at the total compensation figures, emphasizing the importance of aligning executive pay with company performance and long-term shareholder value.
A preliminary analysis of the 2024 proxy season on Say-on-Pay voting among S&P 500 and Russell 3000 companies shows higher rates of approval (over 50% votes for) and increased support compared to previous years. No company with a compensation proposition that failed in 2023 failed again in 2024, indicating satisfactory changes to last year’s failed proposals.
The table below highlights the increased support for executive compensation proposals among S&P 500 and Russell 3000 companies from FY 2023 to May 2024.
In corporate governance standards, a Say-on-pay support under the 70% threshold is often seen as a clear signal to the board that the current compensation plan may need reassessment. Low support levels can lead to increased scrutiny from investors and proxy advisors. Recent voting results show that these phenomena remain low, below 5%.
One compensation package that caused quite a stir this season was that of Tesla CEO Elon Musk. Back in January of this year, the Delaware Court struck down the $55 billion pay package for Tesla CEO Elon Musk proposed back in 2018. Even though the package was approved by 73% of unaffiliated Tesla shareholders at that time, the court found that Musk, as a controlling shareholder, had too much influence over the Board and that conflicts of interest were not fully disclosed. As a result, the court decided the package was unfair and rescinded it. Tesla decided to once again submit this package for shareholder approval at the 2024 AGM in June. Despite a campaign from the New York City Comptroller’s Office and other investors urging shareholders to reject Musk’s pay (one which significantly impacted discussions and was noted in reports by ISS and Glass Lewis), the proposal was approved by shareholders. And it came in addition to approval to change Tesla’s state of incorporation from Delaware to Texas.
The season highlights that transparent and performance-aligned executive compensation remains crucial for shareholder approval, underscoring the influential role of proxy advisors in guiding voting decisions.
Boardroom Battles
Support for director nominations remained high in the U.S. in 2024, with average support rising from 95.9% to 96.4%. The season saw significant impacts from the Universal Proxy Card (UPC) rule, which mandates the inclusion of all director nominees on a single ballot when there is a proxy fight and a shareholder want to elect other directors than proposed by the company. This change makes it easier for shareholders to vote across slates, referring to the lists of candidates nominated by different parties. Previously, shareholders had to choose between separate ballots for the company's and the dissident's candidates, but the new rule allows for a mix-and-match approach, simplifying the process and enhancing shareholder choice in proxy contests.
Notable proxy contests were seen this year at Crown Castle Inc. and The Walt Disney Company, where management achieved clean sweeps despite proxy advisor recommendations for some dissident nominees. Meanwhile, at Norfolk Southern Corporation three dissident nominees were elected.
Gildan’s case was the major event of the Canadian proxy season: in May 2024, Browning West, a California-based hedge fund manager, led a historic and successful proxy campaign at Gildan Activewear, resulting in the complete replacement of Gildan’s board and the reinstatement of co-founder Glenn Chamandy as CEO. This campaign was supported by unanimous recommendations from major proxy advisory firms and led to the resignation of Gildan’s entire board and CEO just days before the annual meeting. The full timeline can be found in the table below:
This year's proxy season demonstrates the importance of shareholder choice and robust governance practices, with the Universal Proxy Card rule enhancing the ability of shareholders to influence board composition effectively.
Shareholders’ Frontline
United States
In the 2024 proxy voting season, significant trends emerged regarding shareholder proposals. According to Ceres, a nonprofit organization advocating for sustainability leadership, there was a noticeable support for climate transition plans, particularly within the banking sector. This indicates growing investor demand for more detailed disclosures on how companies plan to mitigate and adapt to climate change. However, overall support for environmental proposals, which include a range of issues from biodiversity protection to sustainable resource use, declined, highlighting varying levels of investor interest in different aspects of environmental sustainability.
Diversity, equity, and inclusion (DEI) proposals faced increased scrutiny and declining support, reflecting a more critical view from investors. This underscores the importance of carefully framing and justifying ESG initiatives to garner investor backing. According to the Groupe Investissement Responsable Inc. (GIR), a Canadian leading proxy advisor, among Fortune 250 companies, five shareholder proposals received majority votes. These proposals were predominantly governance-focused, including measures to adopt simple majority voting, the right to call a special meeting, and the repeal of rules enacted without shareholder approval. The GIR also mentions that shareholder proposals (excluding anti-ESG proposals) garnered an average support of 19.94%. The ten most frequently filed environmental and social (E&S) proposal categories are presented below:
Tech companies were heavily targeted, with fifteen shareholder proposals for Amazon, twelve for Alphabet, and eleven for Meta Platforms. The E&S proposals that received the highest support are presented in the next table, as highlighted by the GIR:
Canada
In Canada, the GIR highlighted notable shareholder proposals that received majority approval at BMO, CIBC, Air Transat, and Metro. These proposals advocate for in-person shareholder meetings supplemented by online options, reversing the trend of virtual meetings necessitated by the COVID-19 pandemic. The Movement for the Education and Defense of Shareholders (MEDAC), known for its advocacy efforts, particularly with major banks, saw relative success in support on its environmental proposals, reflecting a certain level of influence for its shareholder initiatives.
Key proposals that garnered substantial support in Canada are included in the table below:
Anti-ESG Shareholder Proposals
In the first five months of 2024, the number of anti-ESG proposals saw a significant increase. Despite this surge, they remain unpopular among investors, with average support declining slightly to 1.6%. The increased number of these proposals is partly driven by recent regulatory initiatives in some red states. However, their arguments have not gained traction with larger institutional investors. Overall, while anti-ESG factions are more vocal, their efforts have yet to translate into substantial support. According to GIR, anti-ESG proposals in Canada showed similar levels of support.
This season's proxy voting outcomes on shareholder proposals underscore the ongoing tension between pro- and anti-ESG perspectives within the investment community. While anti-ESG proposals continue to receive minimal support, the substantial backing for governance reforms suggests that investors prioritize strong governance structures, which they view as foundational to effectively addressing environmental and social issues. This indicates a nuanced stance where investors may support ESG initiatives, but with a particular emphasis on robust governance as a critical component of overall corporate responsibility.
Since its launch in November 2022, ChatGPT has become hugely popular and it’s no wonder: with an IQ of above 120 (placing it among the top 10% of humans), the generative AI tool can do things that many humans can’t.
But ChatGPT and the like are just a small sliver of what artificial intelligence can and will be able to do.
General purpose technology (GPT), which artificial intelligence is presumed to be, is defined as ‘a single generic technology, which initially has much scope for improvement and eventually comes to be widely used and to have many spillover effects. Examples of GPTs include electricity and the personal computer. In other words, a technology that is pervasive and significantly impacts aggregate productivity growth.
However, the pervasiveness and productivity growth associated with GPTs often take very long to develop. A popular example is electricity: the big impact of electricity came in the 1920s, about 40 years after Thomas Edison first distributed electrical power in New York in 1882.
The long lag time between invention and widespread application occurs because GPTs tend to entail major investments, learning costs, and complementary innovations. Often, the net benefits of the new technology itself are low early for the earliest adopters.
Prior to artificial intelligence, the most recent example of a GPT is the personal computer. The first desktop computers available for the mass market were introduced in the 1970s, but it took two decades before a significant share of households had adopted them. In 1993, still only 23% of American households owned a computer. The share of workers using computers in their jobs also grew relatively slowly before becoming ubiquitous in the 2000s.
This widespread adoption of computers was followed by an increase in U.S. productivity growth. Between 1991 and 2004, total factor productivity (TFP) grew 1% per year on average, almost twice as fast as its average growth rate in the 1980s (about 0.6%).
Like other GPTs, it is highly likely that artificial intelligence will spur a strong, sustained increase in productivity growth across whole economies that could last decades. Like past GPT related productivity booms, however, these advances could surprise by how long they take to be truly transformative.
An early indicator of enhanced penetration of AI into the overall economy should be the observation of a wave of increased capex, IT and training expense. This has clearly started, but remains timid in most areas.
Capex is currently significantly greater for Nasdaq listed (i.e. tech-focused) companies compared to indexes that reflect the broader economy, as shown in the chart below. This suggests that so far, much of the acceleration of IT spending has been fairly concentrated.
A sign of AI pervasiveness could be when the capex efforts of financials, health care or other sectors match levels in the IT space.
In the meantime, we examine what the overall effects (and roadblocks) of artificial intelligence might be beyond sheer productivity gains.
Artificial Intelligence Will Create More Jobs Than It Destroys
AI, like all new technologies, has been accompanied by a raft of concerns about how its usage will affect certain aspects of society. The first is that job automation will lead to significantly higher unemployment rates. In our view, this fear is widely exaggerated.
The periods of greatest unemployment tend to occur during economic recessions, typically brought on by financial crises and spikes in oil prices.
In fact, studies have shown that technology tends to create more jobs than it destroys. Deloitte has recently published a convincing set of charts across a number of industries where jobs were created/and destroyed over the past 140 years of innovation. No doubt the debate suffers from a common cognitive bias, loss aversion. Analysts tend to focus on the job-destroying effects of technological change (because they are more easily observed) rather than its creative aspects.
Moreover, there is already evidence that the impact of artificial intelligence on the job market is playing out exactly as described above: census bureau data of U.S. firms show that adoption of AI (and other technologies like robotics) so far varies substantially across industries, but is concentrated in larger firms. Adopters have higher labor productivity and greater demand for skilled labor, but the effects on overall employment levels at these firms were limited or ambiguous. This is also borne out by survey data, as highlighted in the chart below.
Importantly, while general purpose technologies prior to the personal computer most severely impacted “blue collar jobs”, it is most likely that service sector employment, such as IT and finance, will see their operations being the most impacted by AI in the coming years.
The bottom line is that there is no precedent that suggests the introduction of artificial intelligence will be different than past GPTs: job losses will occur - but will be offset by job growth in new sectors, This topline approach, however, does mask important weaknesses in the labor market, most notably income inequality.
Artificial intelligence Will Exacerbate Income Inequality
A more serious potential concern regarding AI adoption is a rise in socio-economic inequality.
There is a great deal of historical evidence to suggest that general purpose technologies have increased economic inequalities and there is potential – though not a foregone conclusion – that artificial intelligence will accentuate this trend.
Previous technological advances have been skill-biased, meaning that high-skill workers experienced wage gains due to productivity improvements while people in lower-skilled jobs, (i.e. more repetitive tasks), experienced both job losses and decreased real wages due to automation. It’s estimated that between 50% and 70% of changes in the U.S. wage structure (among the most flexible job markets in the world) over the past four decades are due to the relative wage declines of workers specializing in routine tasks in industries that experienced rapid automation.
This phenomenon, unsurprisingly, bears heavily on the financial outcomes of those with low levels of educational attainment. In the U.S., real earnings of individuals without a high-school degree are now 15% lower than they were in 1980, whereas real wages of workers with a post-graduate degree rose significantly.
The next phase of technological innovation is no different. In the words of technologist and investor Kai-Fu Lee, “corporate profits will explode, showering wealth on the elite executives and engineers lucky enough to get in on the action….”.
The remedy to income inequality stemming from general purpose technologies is already known and well documented. It requires a redistribution of wealth via taxation and assurance of a social safety net that includes access to quality health care and education. Differences in Gini coefficients among the most technologically advanced countries underscores that the tendency toward technologically-induced income inequality can be counteracted through policy. In this sense, although it is a foregone conclusion that increasing automation means greater dispersion in the price of labor according to skill, it is not inevitable that artificial intelligence will lead to greater social inequalities.
Contrary to income inequality within nations, income inequality between countries has been falling in recent decades as faster-growing emerging economies like China and India have narrowed the income gap with advanced economies. This trend is in danger of reversing, given that AI is developing in a monopolistic fashion that risks concentrating economic power within a handful of actors.
Left Unchecked, Artificial Intelligence Will Concentrate Economic Power Even More
A common tendency among general purpose technologies is that they tend to create monopolies. Anti-trust laws exist and are enforced to prevent customers from punitively high prices and to ensure that economic power does not translate into egregious political power (witness the U.S. oil industry). In a review of general purpose technologies, the U.S. Department of Justice concludes that in all cases, “dominant firms have accumulated far more monopoly power than is necessary to motivate and sustain the most rapid and beneficial rate of technological progress”.
Thus, fears about monopolistic power concentrated in the hands of artificial intelligence “first movers” is warranted, based on the historical precedent of previous GPTs.
Aspects of the AI industry are already acting as a monopoly. The table below shows the major links in the artificial intelligence supply chain (more commonly known as the “AI stack”) with the degree of market concentration at each step.
Currently, it is the early inputs that are most monopolistic. For example, TSMC, the semiconductor manufacturer, is known to prioritize its contracts with Apple over other chip consumers during periods of shortages. Note that the companies in the “AI supply chain” with the most monopolistic power (NVDA, TSMC, ASML, AMZN, MSFT, GOOG) are also the stock prices that have been among the biggest gainers in dollar terms since the launch of ChatGPT.
The stakes of monopolistic behaviour are arguably far higher with artificial intelligence relative to past GPTs. Concentration in the number of foundation models, i.e. the number of artificial intelligence models on which data is trained and uses self-supervision, means that model providers can have an outsized influence on information and can shape the public discourse in profound ways. The current concerns about the outsized influence of media platforms owned by Meta would pale in comparison to the control possible by AI models in the hands of a select few.
The bottom line is that the artificial intelligence industry currently displays monopolistic tendencies. Similar to income inequality, there are policy prescriptions to counteract this, but left unchecked, historical precedents suggest that artificial intelligence will facilitate a higher concentration of economic power.
Artificial Intelligence’s Biggest Roadblock: Less Energy Inefficient Than the Status Quo
By status quo, we mean the human brain.
Artificial intelligence is different from previous general purpose technologies in that it is - in its current form – highly energy inefficient compared to a human. Note that in all other general purpose technologies, the energy needs of the technology to complete a task were vastly inferior to the preceding method of task completion. Consider the steam engine: the very first steam engine trains are estimated to have the power of approximately 300 horses and while steam engine powered locomotives are not energy efficient by today’s standards, they are vastly superior to the horse.
Artificial intelligence is already more powerful (i.e. smarter) than the human brain to solve problems with defined scope. However, to complete tasks, artificial intelligence today is less energy efficient than the human brain. The chart below shows the energy requirements of a mouse and human brain compared to computers. To pass from artificial intelligence to general intelligence, i.e. to move from solving defined tasks to solving problems in various domains (i.e. like a human being can) would take almost unimaginable quantities of energy.
Indeed, according to the Frontiers in Artificial Intelligence, researchers estimate that biological computing is about 9 × 108 times more energy efficient than artificial computing architecture. The authors uncover that to compete with the human brain across an economy, artificial intelligence’s energy use would need to be orders of magnitude higher than that of a large, highly industrialized nation. In other words, the authors conclude that “the principles of thermodynamics of computation pose a significant, if not insurmountable barrier, to the emergence of any artificial general intelligence system”. Of course, AI energy efficiency will no doubt see gains in the coming years, but it seems that humans have an important head start.
Heightened awareness of the environmental and social issues associated with global economic activities is driving demand for greater transparency and prompting governments to take regulatory action. Greater consumer scrutiny and a more stringent regulatory backdrop add to reputational risks that investors must consider, but also create opportunities: companies that provide the testing and certifications needed to comply with these evolving standards will benefit from increased demand. In this note, we explore investment opportunities in the testing, inspection and certification (TIC) market.
The Growing Testing, Inspection and Certification Market
Despite recent backlash on ESG, companies appear to recognize its merits. The need to comply with government regulations and meet expectations of various stakeholders (customers, board, senior management, and shareholders) were identified among the top (“very significant”) reasons for having a sustainability strategy in a recent Morgan Stanley survey where 85% of respondents see sustainability as a value creation opportunity. Both factors translate to greater demand for testing and certification. The market is expected to achieve a cumulative annual growth rate of 4.5% over the next decade.
1) Government Regulations
Over the past two decades, there has been a notable tightening of environmental policies globally (see chart below). Though data for the Environmental Policy Stringency index is only available up to 2020, the trend continues as protection of the environment protection remains a key public issue.
For example, in the U.S., the Biden Administration has sought to pass a slew of environmental policies. The EPA has published new standards addressing vehicle tail-pipe emissions, acceptable air, water and waste levels from fossil-fuel power plants, mandatory independent audits for chemical plants and utilities’ responsibility in monitoring and treating PFAs in the water supply. Efforts have been made to try and safeguard these laws from a change in administration. Though a Trump presidency creates a more challenging context for passing environmental policy, reversing pollution controls (like tracking PFAs in water) could be unpopular with the public. This suggests that more stringent standards are likely to continue as public awareness of environmental health risks grows.
Europe is recognized as the leader when it comes to environmental policy, and the EU continues to introduce new policies, most recently the ecodesign regulation that sets sustainability requirements for a broad array of goods sold in the EU.
While new, more stringent environmental regulations are found mostly in richer countries, the environmental standards in developing nations are likely to tighten as these countries attain greater economic wealth. We can see this in the progress of both China and India’s Environmental Policy Stringency numbers in the past ten years, shown in the chart. As developing countries continue to improve environmental policy, the demand for testing and certification will also increase in these regions.
2) Green Credibility
In addition to adhering to new laws and regulations, companies are also striving to live up to consumers’ higher standards. Environmental criteria are growing in importance in purchasing decisions. Although this trend has given rise to the issue of greenwashing - where companies make misleading claims about environmental benefits of their products with the hopes of capitalizing on demand - growing awareness is driving a push for standardized metrics and disclosures. Companies can gain credibility by providing specific environmental data points and disclosures and obtaining third-party verifications. Moreover, the pressure for companies to disclose on a growing number of data points including water usage, biodiversity impacts, worker and product safety and supply chain traceability will broaden the testing, measuring and certification needed to meet higher standards.
The Testing, Inspection and Certification Companies
The TIC market is involved in a wide variety of industries ranging from health care and food to mining and automotive. There are a few exceptions like Eurofins Scientific, that focus almost exclusively on bio-analytical testing. In today’s note, we focus on the companies that are most highly levered to the trends described above, i.e. increasing regulation and sustainability adherence, and that are exposed to a wide array of industries.
The TIC market is considered to be highly fragmented, with only a few large global players competing against many highly specialized local entities. To this end, it is an industry that is ripe for M&A activity. The large, publicly traded TIC companies tend to be long-standing, established companies with entrenched relationships. It should not be a surprise that companies operating in the quality assurance and certification space have impeccable ESG records; after all, they are paid to ensure the quality of others! All five of the names that we have identified rank above the 90th percentile for ESG practices, according to CSRHub scores. They are also all headquartered in Europe, except for ALS in Australia.
The stock prices of the five companies we analysed – Intertek, Bureau Veritas, Applus, ALS, and SGS – tend to be highly correlated and have a beta to the global equity market of well under one. This is no doubt because many of the companies derive significant revenue from the public sector and because it is difficult to reduce expenditure on testing and assurance, even during economic downturns. All of the companies’ annual reports cite the importance of the two megatrends we described above – sustainability and regulation. Note that we include Applus in our analysis in order to have a larger company sample size, despite the fact that the company will (finally) be taken private: last month Spanish regulators approved the takeover by two private equity firms, making it no longer investable for public equity participants.
- Bureau Veritas serves a very diverse customer base including marine, Oil & Gas, electricity, transportation, manufacturing, agri-food and commodities, buildings & infrastructure, and consumer products. By revenue, it is the second largest TIC company, founded over a century ago. In an industry where relationships and trust are crucial, Bureau Veritas has a high moat.
- SGS SA operates a highly diverse TIC business, with operations across industries including electrical and electronic goods testing, food, crop and health science, cosmetics & hygiene, ag commodities, metallurgy, etc. One third of their business is generated in Europe, one-third in Asia, with the remainder derived from North America, Africa and the Middle East. Similar to Bureau Veritas, SGS is one of the longest standing global TIC companies that also benefit from a high moat related to reputation and relationships.
- Intertek is a UK-listed company, though only 6% of revenues are generated in that country. About a third of revenue is derived from the U.S. and almost 20% from China. Indeed, of the five major TIC companies, Intertek is the one most exposed to both the U.S. and China. We see this as an advantage, though there may be some turbulence should Trump be elected (and reverse some of Biden’s environmental regulatory progress). Intertek serves an extremely diverse customer base, including health & safety (food and health care industries), consumer products, building & construction and the energy industry.
- ALS’s most substantive revenue stream (42%) is related to the commodities industry, ranging from mine site analysis to metallurgical testing. Another one-third of revenues is generated from environmental testing including water analysis. Food and pharmaceutical testing accounts for the final 25% of sales. Of the five TIC companies, this Australian-based company has more cyclical stock returns, owing to its higher exposure to the mining industry (the stock has outperformed its peers over the past year thanks to the bull market in commodities).
Current Healthcare Challenges
Healthcare systems globally are facing unprecedented challenges, such as overwhelming patient loads, strained resources, and logistical hurdles, all of which were starkly highlighted during the COVID-19 pandemic. The main concerns worldwide about healthcare systems are reflected in the chart below:
AI's Emerging Role in Healthcare
As discussed in our recent notes ("Waste & Water, Not Your Typical AI Investments," April 18, 2024, and "From Gridlock to Grid Smart: AI’s Role in Electricity Efficiency," April 4, 2024), AI has the potential to disrupt many themes important to sustainability-minded investors. This week, we focus on AI’s applications in healthcare.
Artificial Intelligence offers potential solutions to some of these issues. The exponential growth of healthcare data, advancements in AI algorithms, and the potential for improved diagnostic accuracy and treatment outcomes are driving AI adoption in healthcare. The surge in healthcare data is fueled by electronic health records (EHRs), wearable devices, and genomic sequencing. Advances in AI, especially machine learning and deep learning, enhance the ability to analyze complex medical data and provide actionable insights. Moreover, the medical industry continuously strives for more accurate and personalized diagnostics and treatments to improve patient outcomes and operational efficiency, driven by demographic changes and technological advancements.
Relative to the healthcare sector’s weight in the index, its share of AI mentions lags Technology, Financials and Industrials.
However, there is evidence of that the potential for AI usage is gradually growing within the sector, as indicated by the rapid increase in AI mentions in the earnings calls of healthcare companies, especially over the past two quarters (see chart below).
Research by The Economist suggests that more than 50% of health AI vendors were established since 2017, indicating a rapid expansion and interest in this field. However, this advancement is accompanied by significant concerns ranging from ethics, safety and regulatory challenges. These will need to be addressed to ensure the responsible and equitable deployment of AI technologies.
Ethical Considerations in AI Adoption
One major concern is bias. Bias in AI within healthcare risks exacerbating existing health disparities. When AI systems are trained on data that lack diversity or are otherwise unrepresentative of the broad population, they can produce biased outcomes that disproportionately affect marginalized groups. Some studies have shown that healthcare algorithms often underperform for racial and ethnic minorities compared to white populations, leading to unequal treatment and care recommendations.
The following chart illustrates the complex perspectives of Americans regarding racial and ethnic bias in health and medicine issues. The data reveals that 64% of Black adults identify racial and ethnic bias as a major problem in healthcare, compared to 27% of White adults.
Addressing this issue requires a multi-faceted approach. Healthcare companies need to ensure that their AI systems are developed with diverse, representative datasets. Accessing large, diverse sets of medical data is hard. Training health-oriented AI requires large volumes of high-quality data. Many companies generate this data themselves as part of their drug development programs, ensuring it is comprehensive by including data from various demographic groups and continuously updating datasets to reflect the current population.
Moreover, transparency in the development and deployment of AI models is crucial. This involves making the data sources, decision-making processes, and potential biases of these models clear to both users and regulatory bodies. Furthermore, involving stakeholders from diverse backgrounds, including patients and community representatives, throughout the AI development lifecycle can help identify and mitigate potential biases early on.
Beyond the issue of bias, several other significant concerns arise with the integration of AI in healthcare. The following list is derived from a 2021 KPMG study, where Americans were surveyed about their concerns regarding AI in healthcare:
- Safety Issues: Ensuring rigorous testing and validation of AI systems is crucial, as evidenced by a 2021 study in Nature Medicine where an AI system misdiagnosed pneumonia, potentially delaying treatments. Machine errors are less acceptable than doctors' errors, similar to the scrutiny faced by autonomous cars compared to human drivers.
- Privacy and Security: Healthcare data is a prime target for cyberattacks, as seen in the 2017 WannaCry ransomware attack on the UK's NHS, underscoring the need for robust cybersecurity measures.
- Malicious Takeover: AI systems could be hijacked by malicious actors, leading to consequences like altered patient records or disrupted services.
- Loss of Personal Interaction: Numerous studies stress the importance of human elements in patient care, which AI cannot fully replicate. Although interestingly, people seem to find it easier to disclose their intimate history to a chatbot than to a nurse, highlighting the potential of AI to nevertheless facilitate sensitive healthcare conversations.
- Economic Inequality: Dominance by large tech companies in AI healthcare could stifle innovation and marginalize smaller players, exacerbating economic inequalities.
- Job Displacement: Although a concern, integration of AI should complement, not replace, healthcare workers, as highlighted in studies emphasizing reskilling opportunities to ensure job transformation rather than elimination.
Navigating AI Regulations
Despite these issues, regulation for AI in healthcare has begun, but likely has much further to go to ensure that the above concerns are appropriately safeguarded against. As of 2024, several key developments indicate a trend towards more robust regulatory frameworks:
There is currently little comprehensive regulation surrounding AI in healthcare, but we expect this to become crucial over time. There is a global consensus on the need for regulation to keep pace with AI advancements. These elements underscore the necessity for a balanced approach to AI adoption in healthcare, ensuring technological advancements are ethically sound and beneficial to patient well-being. The companies that will be best positioned for the future are those that proactively engage with regulators to shape effective and fair AI policies, ensuring compliance with emerging standards and participating in ongoing dialogues about regulatory developments.
Investment Opportunities in AI-Driven Healthcare
The healthcare sector holds the most promise for positive disruption from AI, showing potential to outperform humans in tasks like disease diagnosis from medical images. However, the risk of amplifying existing biases and regulatory uncertainty impacts the risk/reward profile for AI investments. Thus, healthcare companies that effectively integrate AI while addressing these ethical concerns and navigating regulatory changes will emerge as winners.
To identify these companies, we take the following points into account:
- Certain healthcare subsectors are more likely to be impacted by AI than others. In a previous note on healthcare ("High Impact: Surviving a Recession With Care," June 2, 2022), we categorized healthcare subsectors within our two sub-themes:
#1 - Healthcare Innovation: Advances in science are enabling research on medical treatments and therapies. In addition, technological innovation is improving outcomes for patients with a range of diseases. This progress helps to extend life expectancy and improve quality of life for individuals suffering from illness.
This subtheme comprises the subsectors Research, Solutions to Critical Diseases and Health Technologies.
#2 - Access to Healthcare: Rising demand for quality healthcare from developing nations and aging populations of the developed world will require an expansion of existing products and services. Companies that leverage improved productivity to provide diagnostics, medication, products for wellbeing and patient care services will contribute to improving healthcare access.
This subtheme comprises Diagnostics, Drug Production, Wellness, Health Insurance Providers, Care Facilities & Services, Healthcare Distributors & Services and Equipment & Devices.
While AI can support improvements in all of these subsectors, we highlight in the table below the areas where we think AI will have the greatest impact.
These are the 5 subsectors stand out as most likely to benefit and/or be disrupted from AI: Research, Solutions to Critical Diseases, Health Technologies, Diagnostics, and Equipment & Devices.
- To find the best companies within these subsectors that fully leverage AI, we conduct the following steps:
- Identified healthcare companies with the most AI mentions in their transcripts over the past five years, selecting the top 100 and filtering out companies with a market capitalization under US$ 1 billion.
- Analyzed the remaining 60 companies by subsector, keeping only those in the five subsectors poised to benefit the most from AI, narrowing it down to 50 companies.
- Evaluated operational performance (ESG) and controversies, giving preference to companies free from controversy and scoring highly on environmental, social, and governance factors using CSRHub scores, reducing the list to 20 companies.
- Conducted deeper research on AI integration in product/service development, selecting 15 companies that have well-integrated AI in their offerings and/or plans.
The curated list of companies is in the table below.
These companies, across the subsectors of Diagnostics, Research, Solutions to Critical Diseases, Health Technologies, and Equipment & Devices, are at the forefront of leveraging AI in healthcare. They exhibit high operational performance, minimal controversy, and robust AI integration in their product development. As such, they should be considered as an entry point to gaining exposure to healthcare companies that are leveraging AI.
The opening months of 2024 have brought four macro developments worth monitoring that will impact investment outcomes for sustainability portfolios for the remainder of the year. In this installment of our quarterly update on Macro Meets Sustainability, we address the following:
- Fed interest rate expectations: what comes next and what does it mean for interest-rate sensitive plays like renewable energy?
- Implications on sustainability-focused investment strategies should Trump win the U.S. elections in November;
- Electrification and the copper rally; and
- China dumping and what is next for the EU.
A Neutralized, Neutral Fed
Whenever big gaps form between the Fed’s expectations and market participants’ expectations of interest rate policy, there is opportunity (and risk) for investors. That is NOT the current environment. Recall that as recently as January, the bond market was predicting a less than 5% chance that the Fed policy rate would still be at 525 basis points at this point in the year, fully expecting that the Fed would cut interest rates up to six times in 2024. Expectations have now fallen to a rather benign three rate cuts, beginning in June or later, which is incidentally, what the FOMC also expects of themselves (as per their March summary of projections). Although we wouldn’t be surprised to see even fewer rate hikes this year (due to sticky inflation), a Fed on hold with market participants that agree makes for a reasonably docile rate backdrop.
This should be viewed as especially good news for the “pain trade” in the renewable energy sector. In this sector especially, monetary policy had become incredibly restrictive. As we wrote in a note last week (“A Second Wind: What Peak Interest Rates Mean for Renewables”), simply removing the rate headwind improves the prospects of many renewable energy projects, since most of the operational costs associated with running utility-scale wind and solar projects are derived from financing costs (rates). To be sure, without interest rates coming down, the operating costs of running renewable projects do not improve. But with credit spreads in this sector suggestive of a low risk of default, we believe that “bottom fishing” in renewable energy production – though not equipment manufacturers due to the risks related to China as described below – is timely given the incredible long-term prospects.
A Trump World?
Although there are still seven months before the next U.S. president will be decided, the enormous gap between the two parties’ platforms suggests that investors of all stripes should consider how their portfolio will stand up to a change of guard.
President Biden recently released his proposed 2025 budget: the trajectory is one of very tepid deficit reduction, albeit a hefty increase in corporate tax rates is included. In contrast, Trump has indicated that if elected he would push to extend for another decade all the individual tax cuts that are set to expire at the end of this year. On the face of it, a Trump government appears likely be more “inflationary” than a Biden government.
More notably, the “flavor” of fiscal policy in the U.S. will certainly change under Trump. In 2022, President Biden’s government signed into law the Inflation Reduction Act (IRA) as well as the Chips and Science Act. Both acts represent an Industrial Policy approach to fiscal expansion: they target specific industries that are in the national interest. We expect that industrial policy will increasingly become the norm across the globe (please see our note from July 21, 2021, “Macro Goes ESG”). But the problem is that Trump does not share the Biden administration’s view of which issues are in the national interest – ignorance about climate change being the most glaring one.
That said, Biden and Trump do agree that re-shoring manufacturing and protecting American jobs are priorities. In this sense, many “green industries” may not suffer much under a Trump government. After all, many climate-focused companies truly are helping to make America great again.
Our preliminary work suggests that from a sector perspective, outside of the fossil fuel industry and a possible knee-jerk election-day reaction, a changing of the guard may not be overly deleterious for sustainability portfolios; we intend to delve further into this topic in an upcoming research note.
Dr. Copper
Copper - the "metal of electrification" — is vital to all energy transition plans. Demand for the metal is expected to double between now and 2035, but the supply side is constrained on several fronts (underinvestment over the past decade, environmental backlash and bans, cost overruns, lower ore grades, etc).
The logical outcome of this demand/supply imbalance is a rise in copper prices. The breakout in copper above 4$/lb in March is just a small sign of things to come. However, before investors pile into the long-term copper trade, it’s worthwhile reviewing recent dynamics (after all, copper as part of the industrial metal complex, is subject to significant boom/bust cycles):
- Most of the demand for copper over the past year has been due to China’s massive ramp up of renewable energy production. Solar capacity increased by nearly 150% last year, from an ever-larger base. It is reasonable to expect that demand for copper from this sector will slow in the year ahead. After all, some of the largest solar producers in China, such as Longi, have admitted that supply glut has now formed in China and will force a significant slowdown in production.
- The more conventional source of demand for copper, China’s residential sector, is still in the doldrums and is unlikely to be revived anytime soon. Indeed, the Chinese government has shown little impetus to use previous tactics to re-ignite the construction sector.
- Part of the March rise in copper prices is due to a concerted effort among Chinese copper smelters to refine less copper at several loss-making plants. The implication of this move is that global copper inventories have begun building again and are no longer precariously close to all-time lows.
The bottom line is that we remain structural bulls on copper due to the enormous demands from electrification, however a more sustained rally and/or “copper mania” may only take place once some of China’s power glut is re-absorbed. Note that copper is a highly recyclable commodity; sustainability investors should first consider companies with strong potential in the recycling space. However, given the enormous amounts of copper required, high-quality miners with superior ESG standards can also be considered. Since most of the rest of the industrial metal complex does not share copper’s bullish dynamics, we suggest investors focus on miners that are producing exclusively (or nearly exclusively) on copper. Freeport-McMoRan (FCX US) and Ero Copper (ERO TSE) are two companies that fit this bill.
China Dumping: No Lift Off For Europe
As mentioned above, China currently appears stuck with overcapacity in solar panel production, but also a whole host of other industries from steel to electric vehicles. Indeed, China has (re)begun flooding its export markets with deeply price discounted goods to help support its domestic manufacturing sector.
But China’s goods dumping is perhaps nowhere more contentious than in the “green goods trade”. After all, cheap accessible green products should be a godsend to global decarbonisation. Unfortunately, both U.S. and EU efforts to address climate change have been wrapped around another goal: to reshore manufacturing capacity and create jobs. Thus, while China solar panel production (as well as wind power equipment, batteries and EVs) may be good for the global energy transition, they are problematic for competing nations’ goals.
For now, it appears that the EU is unlikely to impose new anti-dumping tariffs on Chinese solar panels or other renewable energy equipment, opting instead to find more ways to subsidize local production. We reiterate our view that the odds of other regions catching up to China are very low (see our note March 8, 2023 “China’s Solar Power and Deglobalization”). Investors keen to gain exposure to renewables should avoid solar equipment manufacturers and focus on renewable energy operators.
On September 10, 2020, around the first time when an ounce of gold traded at 2,000, we wrote a note: “Finding a place for gold in an ESG portfolio” whereby we identified sustainable companies involved in gold recycling and mining equipment. While the price of gold has traded sideways for the last three years, with the recent gold breakout to new all-time highs, we are revisiting the topic.
Gold is a scarce commodity. A cube of gold
with an edge measuring only 22 m would contain all the gold that has ever been mined! Other than its distinctive color, gold has attractive chemical properties: it is inert (does not corrode or tarnish), highly malleable and ductile (it can be stretched very thin), has high conductivity (electricity and heat) and is nontoxic (compatible with humans in medical devices or dental work).
What Is Gold Used For?
Due to the above properties, gold has been used as a medium of exchange and store of value (currency), in art (jewelry, decoration, craftmanship) and in technology (telecom, electronics, aerospace, medical devices, dentistry).
Despite having other useful applications, the main sources of demand for gold are jewelry and finance.
The image below, from the World Gold Council, shows the interaction between annual supply and demand for gold over the last 10 years. Gold supply has come mainly from new mining (75%) with the remainder sourced from recycling (25%). It is worth noting that the average annual amount devoted to functional use in technology (8%: 329 tonnes) is lower than the supply from recycling (25%: 1,172 tonnes). In other words, the quantity of gold needed for functional applications could be entirely derived from recycling.

What Are the Factors Determining the Price of Gold?
Let’s look at the supply and demand picture:
Supply: there is not a lot of gold left to mine, according to the World Gold Council, approximately 80% of identified gold reserves have already been mined leaving just 20% of tougher to access reserves, where the concentration of gold is lower. Meanwhile, recycling could bring a lot of existing gold back towards more useful endeavours. The chart below shows the main cost differences between mining and recycling. In mining, costs are associated with exploration, geology, and mining methods. Recycling involves expenses related to collection, sorting, and processing. There are other economic factors that determine the marginal cost of gold like interest rates, currency and inflation. Even with the price of gold at its historical high, gold supply is expected to experience only modest growth over the next decade given declining ore grade, higher extraction costs and stricter environmental regulations.
Demand: Gold experiences demand from various sectors, with jewelry, industry, financial products, and central banks are the primary sources.
- Jewelry: Gold holds cultural significance and serves as a symbol of wealth and fashion.
- Industry: In electronics and renewable energy, gold is prized for its conductivity and corrosion resistance. Medical devices like pacemakers, implants and electrodes utilize gold for its biocompatibility and inertness. Aerospace applications such as satellites benefit from gold's ability to reflect infrared radiation and protect from heat. Gold nanoparticles are used in chemical processes, while the telecom and auto industries also rely on gold for various applications.
- Financial Products: Gold is often sought after as an investment safe haven during times of economic uncertainty and also serves as a hedge against inflation. It is also considered as a form of currency, functioning as a medium of exchange, store of value, and unit of fungibility. Exchange-traded funds (ETFs), futures contracts, certificates, structured products, and gold bars offer investors ways to access the gold market.
- Central Banks: Significant amounts of gold are held by central banks as part of their reserves. They view it as a means of portfolio diversification that provides stability amidst economic fluctuations. Gold acts as a store of value, a reference point for currency valuation, and a hedge against inflation and currency devaluation.
While new supply of gold is expected to be limited, demand could increase at a faster pace from greater jewelry demand and increased industrial use. Financial product demand is a wild card; the advent of Bitcoin ETFs has taken financial flows that would normally have gone to gold (even though the gold price increased in 2023, physical gold ETF saw outflows). The most promising source of demand lies in central bank reserves that are reacting to the confiscation of Russian assets. The chart below highlights that for central banks surveyed, they expect the portion of gold in their reserves to increase (63% expect it will increase compared to 9% that believe it will decrease from the current 15%).
How to Gain Exposure to Gold?
Since the price of gold broke out to all-time highs, should a sustainable investor gain exposure through gold miners, gold royalties, gold recyclers or some form of gold bullion ownership?
Gold Miners:
The practices of gold miners have improved significantly over the years, especially for the larger firms, but many issues still exist. From an environmental standpoint, deforestation, water and air pollution, soil degradation and energy needs are all major concerns. From a social standpoint, human rights violations, health and safety, and other issues such as worker displacement, labor exploitation, and cultural disruption are worrisome. Finally, in terms of governance, challenges include corruption, lack of transparency, weak regulations, and the abuse of land and indigenous rights.
For all of these reasons, it is difficult to find long investment candidates in the mining sector from a sustainability perspective. We find it much easier to find companies to potentially short sell. Gold miners are levered to the price of gold because they have fixed costs and benefit disproportionately from higher prices. Moreover, the value of their reserves increases with the price of gold. As a result, their beta to the underlying global gold price is generally much higher than one.
Gold Royalties:
The role of royalty companies is to provide upfront financing to support initial mining costs in exchange for a percentage of the revenue generated from the sale of gold. They are exposed to the same issues as miners: instead of being the operator, they finance gold mining activities. Royalties have a lower gold beta than miners because they have lower fixed costs.
Gold Recyclers:
Considering the environmental and social damage from gold extraction, recycling offers a more sustainable option. Gold can be recycled without losing its purity, it is much less energy intensive to recycle and recycling avoids the bulk of the environmental and social risks highlighted above. Recyclers have an even lower beta to gold than royalties (less than 1).
Gold Bullion:
The sustainability argument for gold bullion is that it has already been mined. This is a weak argument in our opinion because gold mining is hard on the environment and society but some investment products, like Royal Mint’s Responsibly Sourced Physical Gold ETC, try to improve the sustainability of gold bullion by having a high percentage of recycled gold. Holding gold bullion or investment products linked to gold is not as damaging as miners or royalties, but to us, recyclers are still the most sustainable.
Which Companies Caught Our Attention?
The table below highlights gold miners with weak ESG scores versus recyclers with strong ESG scores, using CSRHub data. Barrick, B2B, Anglo and Kirkland Lake stand out as strong potential gold miner short candidates, while Sims, Umicore, Are Holdings and Johnson Matthey are potential long recyclers.
Next, from a financial perspective, the table below highlights a similarity in revenue growth over the past five years for both miners and recyclers. Valuation remains attractive for both groups, though it is worth noting that miners currently exhibit more favorable profit margins than recyclers. However, as highlighted above, 80% of identified reserves have already been mined. Given that the remaining 20% will be harder to access and yield lower concentrations of gold, miners will need to inject significant capital to explore and increase production. This will inevitably be reflected in higher capital expenditures and interest costs, thereby impacting the overall profitability of mining companies.
How Has the Gold Complex Behaved Recently?
The chart below shows an index of the four miners compared to an index of the four recyclers and the price of gold. Even if the price of gold has increased steadily since the beginning of the decade, gold miners and recyclers have struggled.
How to Position?
The resilience of the price of gold to much higher interest rates, falling inflation and reduced financial market volatility has been surprising. As we discussed earlier, physical gold ETFs saw outflows in 2023 but the main reason for gold’s resilience lies in increased central bank demand. As we have shown, most central banks expect to continue to accumulate gold in the next 5 years. Recyclers that have not yet followed the price of bullion higher should be the investment of choice for sustainable investors looking for gold exposure.
The share of consumers’ budgets dedicated to spending on food has made new multi-decade highs; in the U.S., households have not spent this much of their income on food since the 1980s. Is the long era of cheap and plentiful food behind us? We think so.
Although cyclical forces such as a tight labor market are also contributing to higher food prices, much of the increase in food costs are due to the headwinds to agricultural productivity from climate change, biodiversity loss and other environmental and social factors at a time when the global population is still growing. Indeed, these headwinds are affecting a multitude of food supply chains. Work at Ceres reveals that within each of the eight major food commodity supply chains, there exists at least one environmental or social factor considered a ‘high’ risk. This includes factors that pose threats to food production itself, as well as instances where food production contributes to environmental and social harm.
These risks are not simply theoretical and/or something to consider in the future. Take cocoa for example; West Africa produces most of the world’s cocoa, but two years of hot and dry weather have severely impacted crops; both current yields and the viability of farming in the future are at risk. U.S. cocoa futures reached a nearly fifty-year high in February!
From an environmental perspective, no food industry is more problematic than beef. Risks are rated as “high” or “medium” across all major environmental and social factors according to Ceres.
If the status quo prevails, it is likely to lead to levels of price inflation that can not be fully borne by the consumer. Put simply, industry profit margins are at risk from the environmental and social factors in the table above. In today’s note, we analyze the food supply chain of beef and cocoa. We explore how investors can both protect their capital and improve environmental and social outcomes.
What’s The Beef About Beef
The U.S. and Brazil are the largest producers of beef, accounting for one-third of global production. The main difference is that beef raised in the U.S. is primarily nourished on grains (corn, soybeans, and alfalfa), while beef produced in Brazil is mostly grassfed. In the U.S., the beef value chain has five major links:
- Animal Feed Suppliers: Producers of corn, soybean and alfalfa. Three companies produce nearly 40 million metric tons of feed annually.
- Cow-calf (ranching) Operations: calves are birthed, raised, and weaned on site. This stage of production is extremely fragmented, with over 700,000 farm operations across the U.S.
- Cattle feeding (feedlots): Cattle spend the last six months being fed a high-energy diet. Until recently, JBS, Tyson and Cargill owned about 95% of the feedlot capacity. However, since around 2020 most of the U.S. feedlot capacity is owned by private market firms.
- Meat packing and processing: There are only four companies that slaughter, cut and pack about 85% of all U.S. cattle.
- Food Retailers & Restaurants: About two-thirds of U.S. beef supply goes into food service—restaurants and cafeterias—while one-third is sold in supermarkets.
For each stage of the beef value chain, we use public company reports and government data to estimate profit margins. The table below summarizes each stage of production and relevant characteristics.
There are two key takeaways from the table above:
- With the exception of feed suppliers, each stage of production tends to deliver healthy profit margins, albeit it is the cow-calf operations (ranchers) that have the most variability (uncertainty) in profits.
- At least two stages in the value chain can be considered extremely concentrated. Four companies – JBS, Tyson, Cargill and NFC – control virtually all of the meatpacking in the U.S. and these companies tend to have the highest margins in the value chain.
Cracking the Cocoa Conundrum
Cocoa production is heavily concentrated in Ivory Coast and Ghana, which collectively account for nearly 60% of the world's production. The cocoa supply chain involves four distinct stages:
- Production: Cocoa is cultivated in plantations, then collected by collector members of cooperatives and processed at fermentation and drying centers to produce cocoa beans. Millions of farmers participate in this highly fragmented part of the chain.
- Processing, transport, trade and distribution: Beans are refined and ground to produce cocoa butter, cocoa liquor, and cocoa powder. Seven traders handle 62% of cocoa export volumes, making it a concentrated market.
- Manufacturing: Cocoa products are used by food manufacturers to produce chocolate confectionery and by industries such as cosmetics and pharmaceuticals. Mars, Hershey, Cargill, and Blommer—some of the world’s biggest chocolate manufacturers and processors—are headquartered in the U.S.
- Retail and consumption: Chocolate products are sold through food retailers and consumer outlets (grocery stores, wholesalers, and online).
As with beef, in the table below, we use public company reports and government data to estimate profit margins and we summarize each stage of production of cocoa and provide relevant characteristics.
Again, there are two key takeaways from the table above:
- While farmers receive a disproportionately small share of profits, manufacturers and retailers command a significant portion.
- Market concentration is exceptionally high in processing, trade, and manufacturing, with only a few major players dominating these sectors, while small-scale farmers bear the brunt of the risks at the production level. This concentration means that processing, trade, and manufacturing entities have the ability to diversify some of the risks they face, contrasting sharply with the vulnerability of small-scale farmers.
Investment Implications
So far, our analysis has shown that in the two foodstuffs we analyzed, there is at least one stage in the value chain that is completely dominated by a limited number of companies. In the cocoa chain, Mondelez, Olam, Cargill, Barry Callebaut and Nestle dominate. For beef, JBS, Tyson, and Cargill dominate meat processing, while ADM, Cargill and Bunge have the stronghold on animal feed production.
The table below revisits the initial eight foodstuffs analyzed by CERES and shows that there are a handful of companies involved in ALMOST EVERY food value chain.

In other words, there are select public companies that are likely to participate in virtually every meal of the typical western diet!
At the early stages in the value chain, ADM, Cargill, and Bunge are involved in all eight foodstuffs. At the transformation stage, Nestle, Mondelez and Kellogg are each involved in at least 50% of the foodstuffs.
From a consumers’ perspective, the realization that a mere handful of companies have such dominance in what we eat can be unsettling. However, from an investor’s perspective, particularly one that is concerned with sustainability, it is clear that influencing the operations of just a handful of companies can potentially have enormous impacts throughout our food system. In other words, food is an industry whereby aggressive investor engagement, when done at key links in the value chain, can be effective by concentrating efforts on just a short list of companies.
Engagement at the Agricultural Production Stage: ADM, Bunge and Cargill
ADM, Bunge and Cargill together are the main agribusiness companies that buy, trade, process and ship agricultural commodities such as corn, soybean, palm, wheat, etc. ADM and Bunge are both listed companies while Cargill is private. Another major difference is that ADM and Bunge have similar, elevated ESG scores across all material SASB categories (energy management, water management, product quality, employee safety, supply management, materials sourcing) while Cargill is severely lagging (in 2019, it was named the worst company in the world by Mighty Earth!)
That said, all three companies have a multitude of controversies; ADM scores a D- according to Refinitiv Eikon and Bunge is not much better with a C-. Controversies range from animal testing, workplace conditions, child labor, deforestation, shareholder rights, anticompetition and more.
Engagement at the Processing & Transformation Stage: Nestlé, Olam, Mondelez and JBS
Nestlé: Nestlé demonstrates comprehensive disclosures and proactive approaches to water management, energy efficiency, and product safety, and is considered better than its peers on these topics. However, they lack disclosures on corrective audits in supply chains, GMO products, and ingredient sourcing in regions with high water stress. Furthermore, Nestlé has faced a series of recent controversies (scoring a D according to Refinitiv Eikon), including lawsuits over anti-competition practices (infant formula), allegations of illegal water bottling (California), labor disputes and working condition controversies. They have long been accused of child labor in their cocoa supply chain and following stakeholders’ pressure, they announced a US$1.4 billion investment plan in 2022 to address the issue, even though there are limitations in their ability to effectively tackle this pervasive issue.
Olam Group Ltd: Olam provides operational disclosures that are slightly behind their peers, although the good news is this company is relatively free of controversies. However, they acknowledge that child labor remains a persistent issue in cocoa production. Olam relies on a vast network of suppliers spanning over 60 countries, with a direct and indirect supply base predominantly farmed by small-scale farmers in emerging markets. While the company is aware of child labor, their efforts remain paltry: they provide training on the topic to 240,000 farmers which represents a mere 5% of all farmers they work with.
Mondelez International: Mondelez stands out not only for its operational performance but also for its product portfolio, which primarily consists of highly-processed and sugary snacks. Unlike other companies mentioned, Mondelez specializes in indulgent treats that often contain high levels of sugar, unhealthy fats, and artificial additives. This focus contributes to potential health risks such as weight gain, obesity, diabetes, and heart disease associated with excessive consumption of these products. In addition to these concerns, Mondelez, like its peers, faces regular controversies, scoring a C- according to Refinitiv Eikon, regarding product recalls, data breaches, advertisement norms violations, and child labor in its supply chain.
JBS: JBS has been long-considered a major laggard on most ESG-related issues, with particularly low scores related to working conditions, human rights and its supply chain. But the company’s controversies go even beyond these issues and include environmental scandals, child labor issues and anti-competitive behaviour (Refinitiv Eikon ratings is a D). All of these areas are ripe for investor engagement.
Our hypothesis at the end of 2023 that the artificial intelligence hype had become extreme with respect to related stock valuations has not panned out so far in 2024. Not only did Nvidia’s stock price soar on last week’s earnings announcement, but the company’s CEO Jensen Huang has further fuelled excitement related to generative AI with his optimistic guidance, “[AI] Demand is surging worldwide across companies and nations… Auto, health, robotics, financial services… multibillion-dollar industries are embracing our generative AI”.
Although the valuations of many “AI companies” make our eyes water (NVIDIA is trading at 66 times earnings), there is certainly much to be excited about from a productivity perspective. After all, as we highlighted in our April 2023 note, “Finding Real Returns in Artificial Intelligence” ChatGPT displays an IQ of above 120, placing it among the top 10% of humans. This is quite the progress, given that less than a decade ago, researchers ranked Google’s AI IQ at 47, just shy of the average IQ (55) of a human 6-year-old. That said, for AI to make its way into all spheres of the economy like Internet did in the late 1990s, a massive economy-wide CAPEX effort will be needed – and as of now, that has yet to happen.
And what about from a sustainability perspective? There has been healthy debate (although perhaps not enough action) surrounding ethical aspects of AI. Questions have been raised about the energy intensity of AI tools, about data biases that are leading to discriminatory outcomes, about potential job losses, etc. On the flip side, AI promises to address challenges across industries, including resource optimization and medical breakthroughs. In short, artificial intelligence’s influence on our future is complex and heterogeneous. For investors looking to put capital to work beyond the semiconductor companies that have supplied AI with the chips to flourish, we note the key Sustainable Market Strategies’ themes where leveraging AI should produce superior real-world outcomes alongside superior returns.
Productivity with Purpose
With labor shortages persisting in many developed countries and an ever-increasing need to make better use of physical resources, the first major allure of artificial intelligence is its potential to increase productivity, either by replacing less efficient technologies (and/or labor) or via optimisation of resources.
Software companies were one of the “early movers” with respect to excitement in the application of AI tools. Indeed, four software companies with strong ESG credentials that have been widely lauded for their AI integration – Salesforce, ServiceNow, Adobe and Workday – have seen their stock prices jump between 70- 80% in the past year: a less stellar performance than Nvidia perhaps, but nonetheless extraordinary relative to the S&P500.
That said, investors would be justified if concerned that last year’s hype will not be repeated in 2024: ServiceNow and Workday both reported quarterly earnings in February and the stock market reaction was negative. Both companies disappointed on revenue growth. Investors seemed particularly underwhelmed by Workday’s announcement of the acquisition of an “AI talent solution company”. What is more, Workday has recently come under pressure for facilitating discriminatory biases with its existing AI tools. At valuations still sky high (the lowest P/E of these companies is Adobe at 47 times earnings!), it makes sense to look elsewhere for productive and sustainable applications of AI.
Beyond software, one area that we find exciting as an AI sustainability opportunity is in industrial automation and digital design. In our note last year, we screened for companies that were best positioned to leverage AI advances for purposeful applications, i.e. supporting decarbonization, optimizing physical resource use and improving safety in industries like energy and food production.
The companies we identified at the time were already making good use of artificial intelligence tools and this has continued, with several companies like Schneider Electric and Rockwell Automation announcing new partnerships with Microsoft to leverage their OpenAI tools.
- Rockwell Automation (ROK) provides automation and software solutions to the manufacturing sector, primarily in the U.S. Like many parts of the U.S. manufacturing sector, the company struggled over the past year with an inventory bulge and supply chain constraints, which have been reflected in the company’s muted stock price performance. With the U.S. manufacturing sector showing signs of revival (e.g. the ISM manufacturing survey has bottomed), companies like Rockwell offer an interesting entry point.
- Schneider Electric SE is an automation and energy management company, with about three-quarters of revenue generated from energy management and one-quarter from industrial automation. This makes the company particularly well-positioned to leverage AI advances for purposeful applications as well as to capitalize on the structural buildout of electricity grids (see last section below).
Leave No Person Behind
At Sustainable Market Strategies, investing to reduce social inequalities centers around three key ideas: improving health outcomes through greater access to care and medical breakthroughs, improving education and increasing labor fairness. Of these three sub-themes, we believe education is the area where artificial intelligence is already playing out in stock prices and where sustainability investors can leverage near-term trends.
Artificial intelligence apps, such as ChatGPT, are increasingly part of an educational toolbox just like calculators, and simple web searches. Artificial intelligence employs adaptive learning, intelligent content curation, and smart monitoring and reporting to teach skills: dozens of learning apps powered by AI have sprung up in recent months. One stand-out is Duolingo. The company’s main focus is on language learning, which is in high demand due to ever-increasing international worker flows and remote work. In a note last June “Education Stocks: Will ChatGPT Decide Who Passes or Fails”, we contended that Duolingo would be an AI winner. Since then, the company has added personalized language lessons fuelled by AI-powered feedback and has reached all-time high engagement rates, as measured by the activity of daily and monthly users. The stock price is up over 40% since our June note and Q4 earnings that were reported earlier this week confirm that the company is able attract paid subscribers to their products.
As for health care, this is the sector that holds the most promise for positive disruption from AI. Already, there are several research studies showing that AI can perform as well as or better than humans at some health care tasks. Studies show, for example, that patients often prefer responses to medical questions from chatbots over physicians, giving the chatbot higher ratings on both quality and empathy!
But the best medical use cases of AI involve settings where large data sets (especially images) that can be applied to well-defined problems are available. Diagnosing diseases, by reviewing X-rays for example, is a very effective use of AI. Companies like Quest Diagnostics, which offers access to diagnostic testing services for cancer, cardiovascular disease, neurological disorders, already leverage AI in their diagnostic tools, most notably for dementia.
A major risk, however, is that current medical AI tools tend to amplify existing discriminative biases in health care outcomes: since AI-based health care technology depends on the data that it is fed, a lack of representative data risks leading to biased models that ultimately produce incorrect health assessments especially according to race. Moreover, although the health care sector is heavily regulated, no regulations directly target the use of AI in medical devices. This is likely to change and will significantly alter the risk/reward profile of health care companies actively investing in AI tools.
Finally, in terms of labor fairness, there remain many challenges to ensure that AI tools do not accentuate existing cognitive biases. In addition, at a macro level, there are good odds that artificial intelligence will impact income inequality in ways similar to previous technology booms: if AI is levered mainly to increase automation, then a widening in the gap between capital and labor and between the professional class and other workers is likely.
A recent Pew survey in the U.S. is illustrative: the survey showed that workers with a university education are twice as likely as those without to be exposed to AI in their jobs. These same workers were far more likely to say that AI will help them more than hurt them in their jobs. On the flipside, retail trade, construction and accommodation and food services are all industries where workers are unlikely to see much benefit from AI tools and are also at risk of being replaced by technology. These are industries that tend to be lower wage jobs with low levels of unionization or other worker protections. In other words, there is some evidence to suggest that gains from artificial intelligence will accrue in already high-paying sectors and largely bypass and/or harm employment in lower paying industries. It is challenging to find AI focused investment ideas related to alleviating income/labor fairness, particularly since, as described above, some of the best company software tools (Salesforce, Workday) allegedly create increased discriminatory biases.
Achieving The Energy Transition
Artificial intelligence is a voracious consumer of energy. An oft-quoted study published in late 2023 projects that based on current and future sales of Nvidia microchips specifically for AI, the amount of electricity required to power the world’s data centers could jump by 50% by 2027, thanks to AI alone. According to the study, that’s about the power consumption of the Netherlands, and would require the entire output of about 15 average-size nuclear power plants.
Many companies are working on smaller, less power-hungry AI models that, but even if AI-related energy is half of the current estimate, it will require substantial increases in power production. Most problematic is that the energy “needs” of AI are arriving at a time when renewable energy supply is still falling short of overall energy demand: without the complicating factor of AI, the energy transition is already a massive challenge.
To be sure, artificial intelligence can be used to make the grid smarter and optimize energy usage. But it’s likely that over the next five years, demand from the production of data across industries will far outpace energy savings from optimization within the energy sector.
Crucially, it is the demands on distribution and transmission of electricity (i.e. the grid) that is particularly challenged by artificial intelligence demands. Transporting electricity from where it is produced to where data centres are desired requires a buildout of the electricity grid.
This problem is particularly acute in North America, where grid planners have recently nearly doubled their 5-year load growth forecast; in the U.S. electricity demand is now expected to increase by 4.7% over the next five years. By way of comparison, over the past decade, grid planners have been forecasting a mere 0.5% annual growth rate.
We see this as a major bottleneck for artificial intelligence usage and a major opportunity for companies involved in the buildout of energy transmission. Schneider Electric, Eaton Corporation, ABB, Nexans, Prysmian should become household names for investors that want exposure to the energy transition.
Less Talk, More Action
Eradicating forced labor and ending modern slavery is a target of the UN’s Sustainable Development Goal number eight: Promote sustained, inclusive and sustainable economic growth, full and productive employment and decent work for all.
Human Rights Violations are the result of a few factors:
- Complex subcontracting means that working conditions that are unacceptable in consumer countries remain largely “hidden” or detached from the brands we buy. Supply chain opacity hinders accountability, allowing ongoing exploitation and abuse.
- Weak labor laws and enforcement in outsourcing countries allow corporations to cut costs at workers’ expense.
- Relentless competition to secure the lowest pricing can create a race to the bottom, leading to exploitation and poor treatment of workers as companies prioritize profit margins over ethical considerations.
The risk of forced labor in global supply chains remains one of the most significant human rights issues for companies and their investors. At least 25 million people are currently estimated to be victims of forced labor in supply chains supporting many of the world’s favorite brands.
Awareness on these issues continues to grow since the Rana Plaza tragedy in 2013, as dissemination of information regarding supply chain practices becomes more accessible thanks to the global digital age. Consumers, advocacy groups, and governments are increasingly demanding transparency, accountability, and ethical practices in supply chains. However, tangible progress from corporations is sporadic. Many firms have formally pledged to enhance their supply chain oversight, but these commitments rarely translate into substantive change. This gap between promise and practice is starkly illustrated by a December 2023 study that shed light on 39 European clothing brands (including Prada, Levi’s and Inditex), at probability of using Uyghur forced labor in their supply chains. This is far from being the first report on the topic: issues of forced labor of the detained Uyghur population in China’s Xinjiang region, and its presence in supply chains of major companies have been known since 2009!
A significant percentage of companies assessed by the Corporate Human Rights Benchmark claim high standards for supplier conduct, with 85% stating prioritizing human rights in their supplier contracts. However, there is a clear shortfall in transparency and practical support. Only 27% of these companies offer evidence of aiding their suppliers in upholding these human rights standards, indicating a disconnect between stated policies and actions fostering responsible supply chains. Additionally, the Fashion Transparency Index reveals that none of the companies it has assessed have fully disclosed their first or second-tier supplier lists.
Moreover, placing the responsibility for upholding human rights solely on suppliers risks falling short of objectives to create decent work for all. Improvements like reduced overtime and on-site healthcare in factories, frequently showcased during audits, are often not maintained. Companies operating in Asian countries often lack a thorough understanding of their supply chains, making it challenging to ensure they are not involved in human rights abuses. Western companies seeking to verify the integrity of their suppliers in Xinjiang face potential backlash from the Chinese government.
This persists despite company claims that they are responding to the issue.
The Regulatory Mirage
Understanding the gaps between corporate promises and supply chain realities means examining regulatory responses. The table below highlights many recent regulations in various jurisdictions that aim to transform voluntary principles into enforceable laws. The impact of these new regulations is highly uncertain due to the challenges of interpreting supply chain disclosures. Regulations tend to expand disclosure requirements, but can result in a deluge of data that investors find challenging to navigate. Moreover, regulations frequently lack a clear mandate for companies to take concrete corrective actions.
Unfortunately, recent regulations have not always been effective in preventing violations. A high-profile example occurred in 2020: despite the UK Modern Slavery Act being in effect, fashion retailer Boohoo came under intense scrutiny for perpetuating modern slavery conditions within its supply chain. Reports emerged detailing workers in Leicester's garment factories, which supplied Boohoo, being paid significantly below the minimum wage, with some wages as low as £3.50 an hour, alongside working in unsafe and exploitative conditions. The repercussions were immediate and severe for Boohoo. Following these allegations, Boohoo's share price fell by more than 20% in two days after major online retailers such as ASOS, Zalando, and Next dropped the brand from their platforms. It ultimately wiped £1.1bn off the value of the business.
As for the consequences of non-compliance, companies may face substantial fines. Under the proposed Corporate Sustainability Due Diligence Directive (CSDDD) in the EU, whether such sanctions truly deter companies may hinge on the level of enforcement by member states. It remains to be seen whether this will have a significant impact on corporate behavior.
Unveiling Delusions
Moving beyond regulatory efforts, it is crucial to consider the interplay of corporate advocacy and accountability. Some entities actively lobby against more stringent measures designed to ensure accountability and protect human rights. These tactics highlight the tension between corporate interests and the push for greater corporate responsibility.
There is further evidence of this disconnect in stakeholder actions, where companies’ ethical commitments are questioned - yet go un-penalized in the market. The following table examines the market's response—or lack thereof—to significant human rights controversies involving major corporations.
The observed market data underscores a concerning trend: stock prices remain largely unaffected by the disclosure of human rights violations. This indifference signals a disconnect between ethical corporate behavior and financial consequences, indicating the market's challenges in serving as a corrective force. Boohoo's stock decline, as stated above, was an exception, attributed to the company’s lack of Human Rights Due Diligence (HRDD) systems. This contrasts sharply with other industry giants, which have purportedly demonstrated commitment through the implementation of more effective HRDD systems.
The Investor as a Catalyst
As an investor committed to responsibility, it is crucial to understand that regulatory measures alone are currently insufficient and market responses can be unpredictable. Human rights concerns, while recognized, are not yet a priority for the majority of investors.
This highlights an opportunity for investors to lead the way in elevating human rights. It is crucial for investors to conduct thorough research. Frameworks exist to guide investors in this path.
The OECD Human Rights Due Diligence (HRDD) framework guides companies in identifying, preventing, and mitigating their impacts on human rights throughout their operations and supply chains. The framework offers a structured approach to assess risks, integrate responsible practices, track progress, and communicate actions, aligning business conduct with international human rights standards, as shown in the chart below.

While the HRDD Guidance is designed for companies to ensure responsible business conduct, these key principles are equally critical for investors to apply in their assessment and engagement strategies.
To operationalize these principles, there are a variety of tools investors can access. However, investors must be cautious when navigating these resources as some have faced recent controversies. One such example is the Higg Index, a system developed by major fashion brands to gauge environmental impact. This tool has been criticized for favorably rating synthetic materials, derived from fossil fuels, over natural ones. Critics argue that some of the underlying data is industry-funded and not fully open to independent verification, potentially misleading consumers about the sustainability of products.
Acknowledging these complexities, we have compiled a list of our top tools and resources for investors’ evaluation of human rights practices.
Investors can also play a role in advocating for human rights through proxy voting and shareholder proposals. Recent trends show progress in this area. In 2023, the number of shareholder proposals in the U.S. on human rights saw an uptick to 37 (compared to 16 in 2022). Despite typically not gaining majority support, average approval rates have improved, reaching a median support of 31.4% (compared to a 24% median support between 2020 and 2022). This signals growing acknowledgment of human rights issues in corporate governance. Investors leveraging these mechanisms are contributing to a heightened corporate focus on human rights matters.
Internet access has become an indispensable element of modern life. Having the basic proficiency to use the internet is a crucial skill: beyond simply obtaining information, essential services like recruitment, education, healthcare, banking, and commerce are increasingly offered through online platforms. The COVID-19 pandemic helped accelerate online service use, but also magnified the consequences of access disparity. Individuals with limited or no internet do not experience the same accessibility to these essential services, which creates a form of social inequality. This is referred to as the digital divide. Disparities in internet use not only exist between urban and rural residents, but also gender and even race.
We see telecom companies as providers of an essential service given the important role the internet plays in daily life. On this basis, telecom companies are potential candidates for sustainability portfolios. Of course, not all telecom organizations are created equal; there are certain characteristics investors can look out for to identify internet providers that are contributing to positive social and environmental outcomes. For example, telecom providers that offer services in EM have more potential to reach underserved communities – especially in countries where access to finance mostly happens through efficient telecom networks. Product impact, business ethics and companies’ engagement with local communities are also relevant. This is especially important as the large telecom providers benefit from owning the infrastructure resulting in a market with limited competition.
When to Invest in Telecoms
Telecom may have become an “essential service”, but the industry is hardly viewed as essential in investment portfolios. Telecom companies have been known for paying stable dividends (more on this below), but despite this advantage, the telecom industry has been unloved by investors for years. The unpopularity of telecoms is perhaps with good reason: high debt levels, complex corporate structures, heavy-handed regulation and costly capex are just some of the reasons that made telecom unattractive over the past decade, with only intermittent bouts of outperformance (see the chart below).
This may seem surprising, given that the demand for their services continue to rise. As the chart below shows, global data consumption is expected to keep on its exponential rise higher, driven especially by video content. Unfortunately, however, telecom companies have very little pricing power; in the U.S. for example, internet services inflation, as measured by the consumer price index, did not keep pace with overall service price inflation during the past ten years. In other words, the demand for connectivity continues to rise but telecom companies’ ability to get paid to provide it has not: connectivity and data services have largely been commoditized.
Meanwhile, telecom companies are constantly under pressure to make significant investments in the costly infrastructure that enables them to serve customers. In the current context, just as the transition to 5G networks is becoming well-advanced, newer technologies are on the horizon that will once again require large investments and potentially higher debt levels. Telecoms have had a difficult time generating return on expensive, large-scale investments due to the constant “need” to upgrade infrastructure.
So, what do telecoms have going for them for an investment perspective?
As we highlighted above, telecoms tend to provide stable dividends which boost their appeal during phases of equity market (and economic) growth slowdowns. Demand for telecom services suffer little fluctuation during economic booms and busts and this revenue stability tends to be sought out by equity investors during recessions. With most global equity markets at or near all-time highs, forward-looking investors that are getting nervous about the timing of the end of the global bull market can look to sustainable telecom plays to add a more defensive stance to portfolios.

How to identify potential winners?
We consider both telecom infrastructure companies and service providers. For the infrastructure companies with at least 10 billion USD in market cap, we identify three that take top ranks in terms of sustainability: Cellnex Telecom, Infrastrutture Wireless Italiane and American Tower Corp. Both American Tower Corp and Cellnex serve emerging and frontier markets 31.5% and 14.7% of revenues respectively.
Identifying the most impactful wireless service providers is challenging. Among a group of 15 of the largest companies (market cap greater than 10 billion USD) with a significant portion of revenues from wireless services, each has a CSRHub overall sustainability rank falling between the 87th and 98th percentile. By contrast, the range for the six wireless infrastructure companies with a similar market cap was spread 38 points and only two companies had overall ranks above the 85th percentile.
To pinpoint the standout wireless providers, we eliminate companies with revenues from media ownership. Large companies that own media outlets (Verizon, Comcast and Bell Canada respectively own CNN, NBC and CTV) can be viewed as a conflict of interest. The concentration of power can result in unfair competition where companies prioritize their affiliated businesses and potentially limit consumer choice. Furthermore, these legacy media companies face growing competition from new entrants and streaming services. From a more impactful product perspective, European companies Orange (France, OR FP), Telefonica (Spain, TEF SM), and UK play Vodafone Group (VOD LN) earn at least 15% of their revenues from servicing emerging markets. Telus Corp in Canada stands out for its health service offerings, accounting for close to 10% of year-to-date revenues as of Q3 2023 and has the largest exposure to EMs among North American companies. Additionally, Telus consistently ranks at the top for material sustainability topics.
From a financial perspective, telecom companies in general are known for their high dividends. Drilling down, infrastructure telecom plays (shown in blue) tend to have lower dividends, albeit stronger sales growth (consistent with their higher valuations) compared to wireless service providers. Of the wireless service providers, Telus has shown the strongest sales but despite this, it has underperformed the S&P TSX over the last two years. The reverse is true for Telefonica which is also more leveraged than its peers.
In the chart below, we highlight the European telecom companies that we believe have attractive sustainability and financial characteristics. These stocks have outperformed the Stoxx 600 during the past year as the European economy has slowed, underscoring the defensive characteristics of the telecom industry. We expect the outperformance bout to continue until clear signs of an economic upturn prevail.
Investors can also look to subsidiaries of the larger companies that have their own independent listing on the stock exchange. Telefonica Brazil (parent: Telefonica), Vodafone Idea (parent: Vodacom), TIM SA (sister to INW-IT, spun out from parent company TIT-IT) earn all of their revenues in emerging and frontier markets. This can provide investors with stronger exposure to the growth in these markets (1-year sales growth of at least 9%). These subsidiaries have similar equity performance and rankings on material topics as their developed country parent companies.