01 August 2020

Schengen Border Technologies

The Statewatch report Automated suspicion: The EU's new travel surveillance initiatives comments 
 
This report examines, explains and critiques a number of large-scale EU information systems currently being planned or built that will significantly extend the collection and use of biometric and biographic data taken from visitors to the Schengen area, made up of 26 EU member states as well as Iceland, Liechtenstein, Norway and Switzerland. In particular, it examines new systems being introduced to track, analyse and assess the potential security, immigration or public health risks posed by non-EU citizens who have to apply for either a short-stay visa or a travel authorisation – primarily the Visa Information System (VIS), which is being upgraded, and the European Travel Information and Authorisation System (ETIAS), which is currently under construction.
 
The visa obligation has existed for years. The forthcoming travel authorisation obligation, which will cover citizens of non-EU states who do not require a visa, is new and will massively expand the amount of data the EU holds on non-citizens. It is the EU’s equivalent of the USA’s ESTA, Canada’s eTA and Australia’s ETA. These schemes represent a form of “government permission to travel,” to borrow the words of Edward Hasbrouck, and they rely on the extensive processing of personal data.
 
Data will be gathered on travellers themselves as well as their families, education, occupation and criminal convictions. Fingerprints and photographs will be taken from all travellers, including from millions of children from the age of six onwards. This data will not just be used to assess an individual’s application, but to feed data mining and profiling algorithms. It will be stored in large-scale databases accessible to hundreds of thousands of individuals working for hundreds of different public authorities.
 
Much of this data will also be used to feed an enormous new database holding the ‘identity data’ – fingerprints, photographs, names, nationalities and travel document data – of non-EU citizens. This system, the Common Identity Repository (CIR), is being introduced as part of the EU’s complex ‘interoperability’ initiative and aims to facilitate an increase in police identity checks within the EU. It will only hold the data of non-EU citizens and, with only weak anti-discrimination safeguards in the legislation, raises the risk of further entrenching racial profiling in police work.
 
The remote monitoring and control of travellers is also being extended through the VIS upgrade and the introduction of ETIAS. Travel companies are already obliged to check, prior to an individual boarding a plane, coach or train, whether they have the visa required to enter the Schengen area. This obligation will be extended to include travel authorisations, with travel companies able to use the central databases of the VIS and ETIAS to verify whether a person’s paperwork is in order or not. When people arrive at the Schengen border, when they are within the Schengen area and long after they leave, their personal data will remain stored in these systems and be available for a multitude of further uses.
 
These new systems and tools have been presented by EU institutions as necessary to keep EU citizens safe. However, the idea that more personal data gathering will automatically lead to greater security is a highly questionable claim, given that the authorities already have problems dealing with the data they hold now.
 
Furthermore, a key part of the ‘interoperability’ agenda is the cross-matching and combination of data on tens of millions of people from a host of different databases. Given that the EU’s databases are already-known to be strewn with errors, this massively increases the risks of mistakes in decision making in a policy field – immigration – that already involves a high degree of discretion and which has profound implications for peoples’ lives.
 
These new systems have been presented by their proponents as almost-inevitable technological developments. This is a misleading idea which masks the political and ethical judgments that lie behind the introduction of any new technology. It would be fairer to say that EU lawmakers have chosen to introduce unproven, experimental technologies – in particular, automated profiling – for use on non-EU citizens, who have no choice in the matter and are likely to face difficulties in exercising their rights.
 
Finally, the introduction of new databases designed to hold data on tens of millions of non-citizens rests on the idea that our public authorities can be trusted to comply with the rules and will not abuse the new troves of data to which they are being given access. Granting access to more data to more people inevitably increases the risk of individual abuses. Furthermore, the last decade has seen numerous states across the EU turn their back on fundamental rights and democratic standards, with migrants frequently used as scapegoats for society’s ills. In a climate of increased xenophobia and social hostility to foreigners, it is extremely dangerous to assert that intrusive data-gathering will counterbalance a supposed threat posed by non-citizens. ... If all non-citizens are to be treated as potential risks and assessed, analysed, monitored and tracked accordingly, it may not be long before citizens come under the same veil of suspicion.

Robots and Routine Job Replacement

'The rise of robots and the fall of routine jobs' by Gaaitzen J. de Vries, Elisabetta Gentile, Sébastien Miroudot and Konstantin M. Wacker in (2020) 66 Labour Economics comments 

This paper examines the impact of industrial robots on jobs. We combine data on robot adoption and occupations by industry in thirty-seven countries for the period from 2005 to 2015. We exploit differences across industries in technical feasibility – defined as the industry’s share of tasks replaceable by robots – to identify the impact of robot usage on employment. The data allow us to differentiate effects by the routine-intensity of employment. We find that a rise in robot adoption relates significantly to a fall in the employment share of routine manual task-intensive jobs. This relation is observed in high-income countries, but not in emerging market and transition economies. 

The authors state 
Rapid improvements in robot capabilities have fuelled concerns about the implications of robot adoption for jobs. While the creation of autonomous robots with flexible 3D movement continues to be a major challenge to engineers, rapid progress is being made. Robots can now perform a variety of tasks, such as sealing, assembling, and handling tools. As robot capabilities continue to expand and unit prices fall, firms are intensifying investment in robots ( Frey and Osborne, 2017 ; Graetz and Michaels, 2018; Acemoglu and Restrepo, 2020). What is the impact of robot adoption on labour demand? Do robots substitute for tasks previously performed by workers? The main contribution of this paper is to empirically study the impact of industrial robots on the occupational structure of the workforce across industries in a set of high-income as well as Emerging Market and Transition Economies (EMTEs). We combine a large and detailed occupations database with data on industrial robot deliveries from the International Federation of Robotics. The database on occupational employment from Reijnders and de Vries (2018) allows us to examine the share of employment in occupations with a high content of routine tasks – i.e. tasks that can be performed by following a well-defined set of procedures. We delineate occupations along two dimensions of the characteristics of tasks performed, namely ‘analytic’ versus ‘manual’, and ‘routine’ versus ‘non-routine’. We thus distinguish four key occupational groupings, namely routine manual, routine analytic, non-routine manual, and non-routine analytic task-intensive occupations (as in Autor et al. 2003 ; Reijnders and de Vries 2018 ; Cortes et al. 2020). We follow Graetz and Michaels (2018) in constructing measures of robot adoption by country-industry pairs and relate these to changes in occupational employment shares. Our sample covers 19 industries for 37 countries at varying levels of development from 2005 to 2015, and includes major users of industrial robots, such as the Peoples Republic of China (PRC), Japan, South Korea, Germany, and the United States. Our main finding is that country-industry pairs that saw a more rapid increase in robot adoption experienced larger reductions in the employment share of routine manual jobs.
 
Our approach is motivated by the following economic considerations. Firms produce a variety of products using a continuum of tasks (Acemoglu and Autor, 2011), and these products differ in the number of tasks that can be performed by robots (Graetz and Michaels, 2018). For example, the share of replaceable tasks by robots differs between apparel and automotive and appears larger in the latter.  This gives rise to differences across industries in the technical feasibility of robots substituting tasks previously performed by humans. Advances in machine capabilities expand the set of tasks carried out by machines ( Acemoglu and Restrepo, 2018 ). Firms will adopt robots if it is technically feasible and the profit gains exceed the costs of purchasing and installing robots. Given higher wages in advanced countries, the technical constraints to robots replacing tasks are more likely to bind for firms in these countries. Hence, improvements in robot capabilities would result in a larger employment response in advanced countries compared to developing countries.
 
We use these economic insights in our analysis. In particular, the technical feasibility of adopting robots guides our instrumental variables (IV) strategy to identify the causal relation between robots and labour demand. Economic feasibility motivates our distinction of the impact of robot adoption between advanced and developing countries. Using two-stage least squares (2SLS) estimation, we find that robot adoption lowers the employment share of routine manual occupations. This relation is observed in high-income countries, but not in emerging market and transition economies.
 
This paper relates to recent studies that examine the impact of robot adoption on socio-economic outcomes. Graetz and Michaels (2018) find that robot adoption contributed to an increase in productivity growth across industries in high-income countries between 1993 and 2007. Their findings suggest that robot adoption did not reduce employment, which is corroborated in this paper. This is also observed by Dauth et al. (2019) , but not by Acemoglu and Restrepo (2020) , who examine geographic variation in robot adoption across the United States and find that robots are labour replacing. Dauth et al. (2019) use detailed linked employer-employee data for Germany to show that displacement effects are cancelled out by reallocation effects, such that in the aggregate no employment effects from robot adoption are observed. Data availability did not allow Graetz and Michaels (2018) to examine the impact of robots on workers that perform different tasks. Yet, Autor (2015) emphasizes that workers with routine task-intensive occupations are most likely to be affected by automation. This paper aims to contribute to our understanding of the impact of robots on such occupational shifts. 
The remainder of this paper is organized as follows. Section 2 reviews the key theoretical mechanisms between automation and labour demand. Section 3 describes the methodology and instrumental variables. Section 4 documents patterns in the occupational structure of the workforce and robot adoption. Section 5 empirically studies the impact of robot adoption on the task content of labour demand. Section 6 concludes.

Tracing Apps

'Digital contact tracing, the Apple/Google API and big tech’s newfound role as global health policy makers' by Tamar Sharon in (2020) Ethics and Information Technology 1–13 comments

Since the outbreak of COVID-19, governments have turned their attention to digital contact tracing. In many countries, public debate has focused on the risks this technology poses to privacy, with advocates and experts sounding alarm bells about surveillance and mission creep reminiscent of the post 9/11 era. Yet, when Apple and Google launched their contact tracing API in April 2020, some of the world’s leading privacy experts applauded this initiative for its privacy-preserving technical specifications. In an interesting twist, the tech giants came to be portrayed as greater champions of privacy than some democratic governments. This article proposes to view the Apple/Google API in terms of a broader phenomenon whereby tech corporations are encroaching into ever new spheres of social life. From this perspective, the (legitimate) advantage these actors have accrued in the sphere of the production of digital goods provides them with (illegitimate) access to the spheres of health and medicine, and more worrisome, to the sphere of politics. These sphere transgressions raise numerous risks that are not captured by the focus on privacy harms. Namely, a crowding out of essential spherical expertise, new dependencies on corporate actors for the delivery of essential, public goods, the shaping of (global) public policy by non-representative, private actors and ultimately, the accumulation of decision-making power across multiple spheres. While privacy is certainly an important value, its centrality in the debate on digital contact tracing may blind us to these broader societal harms and unwittingly pave the way for ever more sphere transgressions. 

 In discussing 'contact tracing and the automation of a public health practice' Sharon states 

Since the outbreak of the COVID-19 pandemic in early 2020, governments and health authorities around the world have attempted to mobilize digital technologies to address this novel threat, including the use of tracker wristbands, smartphone applications, thermal cameras, facial recognition and drones (The Economist 2020). In the prolonged anticipation of more permanent solutions like a vaccine, contact tracing apps in particular are being explored as tools to help contain the spread of the virus (EC 2020; WHO 2020). Contact tracing is a time-tested method that has been successfully used to fight infectious disease outbreaks including syphilis, measles, HIV and Ebola. It involves identifying infected individuals and informing the people they have been in contact with that they are at risk, through a meticulous process of retracing where and with whom an infected individual has been in proximity. Automated contact tracing offers several advantages over traditional contact tracing in the case of the COVID-19 pandemic (CDC 2020a; Ferreti et al. 2020). First, it seeks to automate a labor-intensive practice in a situation where there is a scarcity of human contact tracers. Moreover, it may offer more accuracy where human memories are fallible—particularly in the case of COVID-19, where infection can be asymptomatic for up to two weeks. The speed of contagion of the COVID-19 virus, finally, requires swift contact tracing in order to be effective. Digital contact tracing seeks to address these limitations, by providing speed, scale and accuracy. 
 
As with many attempts at automation, numerous obstacles impede the path to smooth, seamless digital contact tracing. It is not at all clear, in the first instance, that these contact tracing apps will be effective (Ada Lovelace Institute 2020). In countries where digital contact tracing was first deployed, such as China, Singapore and South Korea, the actual role of this technique in controlling the spread of infections is ambiguous (Frieden 2020). Accuracy is another major concern here. Bluetooth, currently the preferred technology for digital contact tracing, can result in high amounts of false positives, by picking up “contacts” that are not epidemiologically significant (Lee 2020; Vaughn 2020). Effective digital contact tracing also relies on a high level of uptake by the population, which will be difficult to ensure if these systems are to be voluntary (Hinch et al. 2020). This issue is complicated by the question of who can participate in digital contact tracing. Not everyone has access to a smartphone, even in wealthier nations. And of those who do, an estimated 1.5 billion people globally still use basic phones that do not have the necessary technical requirements, such as Bluetooth “low energy” chips, that are being used in many contact tracing apps (Counterpoint 2020). Importantly, these populations tend to be lower socio-economic groups and older people, exactly those people who are also among the most vulnerable to the virus (O’Neil 2020). While these limitations have dampened some of the initial enthusiasm around digital contact tracing as an easy solution to curbing the spread of the virus, the technology is still seen as an important complement in national post-lockdown strategies. At the time of writing, at least 80 contact tracing systems have been launched or are in development around the world,1 and supra-national bodies like the European Commission and the WHO are publishing guidelines for app development, or developing their own (EC 2020; Dave 2020).
 
Beyond these more practical questions, one of the major points of contention in the implementation of digital contact tracing has been its potential to cause harm through privacy breaches (Ienca and Vayena 2020; McGee et al. 2020). In Europe and the United States, in particular, where public awareness on the use of digital surveillance for public interests has gained a heightened sensitivity to privacy issues since the Snowden revelations, this triggered a vigorous public debate on the need to develop privacy-friendly digital contact tracing. Yet, when Apple and Google—corporations whose data practices are typically the focus of ethical debate—launched their contact tracing API in April 2020, some of the world’s leading privacy experts applauded this initiative for its privacy-preserving technical specifications. In an interesting twist, the tech giants came to be portrayed as greater champions of privacy than some European governments.
 
This article explores what else is at stake when two of the world’s most powerful corporations move into the field of pandemic response management, even when this is done in a privacy-preserving manner. Drawing on Michael Walzer’s (1983) theory of justice, and the autonomy of spheres of social life as a precondition for equality and justice, I propose to view the Apple/Google API in terms of a broader phenomenon of powerful tech corporations encroaching into ever new spheres, by virtue of the fact that their digital expertise has become a coveted currency in almost all spheres of life. From this perspective, the (legitimate) advantage that tech companies have accrued in the sphere of the production of digital goods provides them with (illegitimate) access to the spheres of health and medicine, and more worrisome, to the sphere of politics. Each of these transgressions, I explain, poses specific risks that are not captured by the focus on privacy harms. Encroachment into the sphere of health and medicine can lead to a crowding out of significant traditional sectorial expertise and the reorganization of health and medicine in line with the values and interests of corporate actors. Encroachment into the sphere of politics can lead to new dependencies on corporate actors for the delivery of essential public goods, often underpinned by relationships of charity and gratitude rather than justice and duty, and ultimately to the shaping of public policy by non-elected, unaccountable actors. The overall risk is an accrual of advantage and power across spheres—what Walzer calls tyranny. While privacy is an important intrinsic and instrumental value, the centrality that it has received in the debate on digital contact tracing, and arguably in other debates on digitalization, may blind us to these broader societal harms and unwittingly pave the way for ever more sphere transgressions.

30 July 2020

WEF, Genomics and Precision Medicine

Time for a spot of Davos? The World Economic Forum (WEF) promotes its Genomic Data Policy Framework and Ethical Tensions White Paper with the rubric

Genomic data collection is accelerating in historically understudied and excluded populations. The information will fill knowledge gaps, spur medical discoveries and lead to more targeted and appropriate care; however, it comes with significant risk if ungoverned. Insufficient genomic data policies expose communities to the risk that certain actors will extract information from their population and use it for their own benefit. These risks have been realized in the past and will occur again. To address them and prompt action, the World Economic Forum collaborated with 30 global leaders to develop a forward looking, scalable policy framework and set of six ethical tensions, as well as a companion guidance document, that policy makers, business leaders, researchers, community members and others can leverage and modify for use in a local context.

The paper states 

The Leapfrogging with Precision Medicine Project, which is part of the Precision Medicine Portfolio of the World Economic Forum, focuses on co- designing and piloting policy, governance and business frameworks that enable healthcare leaders in emerging economies to prepare for and integrate precision medicine approaches into their health ecosystems. Leapfrogging with Genomic Data is one workstream within this project.
 
Genomic and genetic data – the digitized record of a person’s DNA – is an especially sensitive form of human health data, and its collection and use support the scientific research and improved diagnostics and treatments that underscore precision medicine. Genomic and genetic data collection is accelerating, including in low- and middle-income countries (LMICs) and emerging economies, to fill critical gaps in the understanding of populations not traditionally included in research and to support more precise clinical care. Without future-looking policies that address genomic and genetic data collection and use in research, countries face two main risks: 1) their data does not inform scientific research that may lead to more population-relevant screening guidelines, diagnostics and treatments; or 2) their data may be used by and primarily benefit outside parties. Without future-looking policies that address genomic and genetic data collection and use in clinical care, countries may inadvertently slow adoption of advancing healthcare approaches that can improve patient care pathways.
 
This white paper aims to begin addressing the need for new or modified policies by proposing a genomic data policy framework and corresponding set of ethical tensions for policy-makers, business leaders, researchers, patients and others to consider before taking actions that affect or involve the collection and use of human genomic and genetic data for research and clinical use. Generally, genomics refers to all genes and their interrelationships and genetics focuses on a single or set of genes. The distinction is important and nuanced in several fields; however, for the purposes of this paper, the authors will refer to genomics when our thinking applies to data about all or some genes.
 
This paper is written from a “future of healthcare” perspective, and with a focus on LMICs and emerging economies. This is not to imply that there should be a different standard among countries, but to ensure consideration of the differing perspectives and needs informed by these countries’ diverse historical, societal and cultural contexts.
 
This work also aims to keep the interests of research participants and patients at the forefront of policy and ethical considerations, and we hope that is reflected in the following pages. Advancement in genomics would not be possible without those who provide their data. Inclusion of citizens, patients or community representatives in the discussion and development of approaches to genomic data collection and use would likely prevent numerous blind spots, conflicts and sources of mistrust, while fostering understanding and better outcomes.
 
The areas of focus for the policy framework are four foundational elements germane to human genomic data collection and use: consent, data privacy, data access and benefit sharing. These are followed in the paper by six ethical tensions that underpin these elements. The paper focuses on principles and guidelines, not the implementation or application of these elements. As such, topics including data security and infrastructure, while critically important to data privacy and access, are not addressed here. Gene-editing research and testing is out of scope, too. Forms of health data beyond human genomic data are also out of scope for this paper, though a future document may consider expanding the ideas to other -omic data. Finally, this document focuses on activity within the medical and scientific establishment and not the direct-to-consumer genetic testing market or non-healthcare fields such as law enforcement or surveillance, though we have drawn from developments in those spaces.
 
How to use
 
At its core, this white paper is intended to provide a basis for discussion and decision-making primarily by policy-makers, but also by researchers, clinicians, patients and business leaders who engage in the collection and use of genomic data.
 
This white paper begins with a high-level, forward- looking, scalable policy framework. In this section, core terms are defined, and key policy principles are outlined. The goal of the policy framework is to set forth principles that may inform the development of corresponding policies, regulations or guidance, modified as appropriate for local context.
 
The next section explores a set of six ethical tensions that permeate the four foundational elements explored in the policy framework. Corresponding ethical questions will help facilitate discussion and prompt awareness of gaps or barriers when developing a genomic data policy that attends to ethical concerns. Working through the questions will help in projecting how various people who participate in genomic data collection and use may grapple with ethical issues before, during and after data collection, and in carrying ethical considerations into policy, research design and clinical applications.
 
The policy framework and ethical tensions are meant to be complementary. Policy should reflect a society’s ethical positions on issues, and too often the two are divorced until a conflict arises.
 
The framework and tensions reflect a distillation of critical elements of genomic data policy and ethics. As such, the tools in this document are presented as a starting point to develop or refine a set of guiding principles and ethical standards as you craft or revise genomic data policy and regulation, or best practices in your government, organization or institution.
 
They are intended for customization and use in a local context.
 
We suggest exploring the framework, ethical tensions and questions through multistakeholder working sessions. Including stakeholders – research participants, patients, researchers, physicians, nurses, business leaders and others – who are affected differently by genomic data collection and use will expose issues that may otherwise be overlooked and uneven power dynamics that often complicate ethical positions and corresponding actions. A multistakeholder approach will help cultivate a comprehensive understanding of policy principles and ethical dynamics and create a sound path forward within your jurisdiction. A companion ethical tensions guidance document, a mini-guide to running scenario vignette-based workshops and a set of sample scenario vignettes are available to guide these activities (see Appendix).
 
Methodology
 
The content of this white paper was developed through desk research, expert interviews and multistakeholder workshops. Desk research began by collating a Genomic Data Policy Resource Guide, which offers a review of prevailing laws, regulations, guidelines and commentary addressing consent, data privacy, data access and benefit sharing of health and genomic data, drawn from General Data Protection Regulation (GDPR), Health Insurance Portability and Accountability Act (HIPAA), Genetic Information Nondiscrimination Act (GINA) documents, as well as organizations including the Global Alliance for Genomics and Health (GA4GH), Human Heredity and Health in Africa (H3Africa), the Organisation for Economic Co-operation and Development (OECD), the World Health Organization (WHO) and other sources. Additional research included the collection and review of real-world examples of tensions in genomics and genetics, corresponding expert commentary, professional organization guidelines and other existing frameworks for data collection and use.
 
A World Economic Forum Leapfrogging with Genomic Data workshop in San Francisco, USA, in July 2019 used various scenario vignettes, representing opportunities and risks germane to LMICs, to explore and validate the four foundational elements of genomic data policies discussed below: consent, data privacy, data access and benefit sharing. Each vignette focused on one of the four elements of genomic data policy and presented fictionalized short stories oriented five to 10 years in the future. The vignettes included two main personas, often representing perspectives of the Global North and South. The personas helped illuminate the ethical tensions underpinning each policy issue. Thirty experts in policy, industry, research and civil society from Africa, the Middle East, Latin America, Asia and North America worked through these vignettes in small groups, with each group adopting one of the personas as they identified the cause of the conflict, underlying considerations and possible solutions. Groups then returned to plenary to address the group that adopted the counter persona. These presentations offered an opportunity to question each other and discuss conflicts from a solution-oriented perspective.
 
A World Economic Forum Roundtable on Ethical Tensions, held in November 2019 with a select group of experts in bioethics, helped refine the ethical tensions informed by the July scenario vignette workshop and develop intuitive and accessible tools for the application of these tensions to policy development. Additionally, Elissa Prichep co-led two events: 1) the Genomic Data Policy Consultative Session with the Rwanda Ministry of Health; and 2) the Roundtable on Governance of Human Genome Sequencing with the Dubai Future Foundation to begin transitioning this framework to government officials and translating it into policy approaches within a local context.
 
Throughout, the authors conducted dozens of interviews with thought leaders in government, academia, research, medicine, civil society and industry who are located in and working across emerging economies. Those interviewed educated us, answered our questions, provided feedback on our work and expanded the expert network engaged in this project.

Medicines Access and global trade

The Directors-General of the World Trade Organisation, World Health Organization and World Intellectual Property Organization have released a new edition of the Trilateral Study on Access to Medical Technologies and Innovation which 

seeks to strengthen the understanding of the interplay between the distinct policy domains of health, trade and intellectual property (IP), and how they affect innovation and access to medical technologies, such as medicines, vaccines and medical devices. The second edition provides an improved, evidence-based foundation for policy debate and informed decision-making at a critical time for global health. 

The Study
  
discusses key factors determining access to medical technologies and innovation, including regarding medicines, vaccines and other medical technologies, such as medical devices and diagnostics. The second edition draws practical lessons from experiences regarding the intersections between public health, IP and trade within the broader perspectives established by the human rights dimension of health and the Sustainable Development Goals (SDGs). The study records numerous significant developments since 2013 when the first edition was launched. Among the new topics covered are antimicrobial resistance and cutting-edge health technologies. The revised edition provides updated data on health, innovation trends in the pharmaceutical sector, and trade and tariffs regarding medical products. It also includes an updated overview of access to medical technologies globally and key provisions in regional trade agreements. In addition, it takes account of developments in IP legislation and jurisprudence.
 
A COVID-19 section at the start of the publication provides a factual overview of the developments and measures taken to address this extraordinary public health crisis, which began after the work on the second edition of the study had been completed. The section guides the reader to parts of the study that are of direct relevance to the issues that have been raised during the pandemic.

It states 

Public health is inherently a global challenge and thus assumes high priority for international cooperation. The World Health Organization (WHO) is the directing and coordinating authority for health, but the interaction between health issues and other policy domains – human rights, development policy, intellectual property (IP) and international trade – creates a strong rationale for cooperation and coordination between the WHO and other international organizations, in particular the World Intellectual Property Organization (WIPO) and the World Trade Organization (WTO). This study and its updated and reviewed second edition have emerged from an ongoing programme of trilateral cooperation among these agencies. It responds to an increasing demand, particularly in developing countries, for strengthened capacity for informed policy-making in areas of intersection between health, trade and IP, focusing on access to and innovation of medicines and other medical technologies. The need for cooperation and coherence at the international level has intensified over the past decades, as successive multilateral decisions have confirmed. 

 The study is set in an evolving health policy context. An integrated approach can reinforce a dynamic, positive interplay between the measures that promote innovation and those that ensure access to vital medical technologies. The aim of the technical cooperation activities of the WHO, WIPO and the WTO is to facilitate understanding of the full range of options and their operational context. This study draws together the materials used in technical cooperation and addresses needs for information in an accessible, systematic format to support ongoing collaborative efforts. 
 
Navigating the study 

 The study has been prepared as a capacity-building resource for policy-makers. The study is structured so as to enable users to grasp the policy essentials, and then to look more deeply into areas of particular interest. After explaining the need for policy coherence and the role of each of the cooperating agencies to address the global disease burden and health risks (see Chapter I), the study lays out a general panorama of the policy landscape (see Chapter II), so that all interrelated elements can be seen in context. It then provides more detailed accounts of issues specifically connected with innovation (see Chapter III) and access (see Chapter IV). The contents reflect the multilateral policy debate over the past two decades, recognizing that innovation and access are inevitably intertwined – both are indispensable ingredients to meeting an evolving global disease burden. 

 ƒƒChapter I presents the general background to health policy relating to medical technologies and to international cooperation in this field, sets out the distinct roles and mandates of the three cooperating agencies, and outlines the global disease burden that defines the essential challenge for health policy.
 
ƒƒChapter II outlines the essential elements of the international framework – health policy, IP and trade policy, including regulatory issues, as well as technical barriers to trade, sanitary and phytosanitary measures, health services and procurement rules. It lays the basis for the following, more detailed analysis of the innovation and access dimensions in Chapters III and IV. It outlines the key insights of economics for medical technology innovation and access. A final section reviews the policy issues associated with traditional medical knowledge and access to genetic resources, in view of its significance for national health systems and as an input to medical research. 

 ƒƒChapter III provides a more detailed overview of policy issues concerning the innovation dimension of medical technologies. The historical pattern of medical research and development (R&D) provides a backdrop for analysing the current R&D landscape. The chapter looks at challenges with overcoming market failures in medical product R&D in areas such as neglected diseases and antimicrobial resistance. It then outlines alternative and complementary instruments to incentivizing and financing R&D. It outlines the role of IP rights in the innovation cycle, including issues relating to IP management in health research and selected pre- and post-grant patent issues. A final section looks at influenza vaccines as a distinct example of innovation management and product development to address a specific global health need. 

ƒƒChapter IV deals with key aspects of the access dimension, describing the context for access to health technologies, with more detailed case studies on access in respect of HIV/AIDS, hepatitis C, tuberculosis (TB), non-communicable diseases (NCDs), and vaccines. It sets out the key determinants of access related to health systems, IP and trade, and it analyses access to health products in specific areas. It reviews in particular pricing policies, transparency across the value chain of medicines and health products, taxes and mark-ups, and procurement mechanisms, as well as regulatory aspects and initiatives to transfer technology and boost local production, patent quality and review procedures, compulsory and voluntary licences, free trade agreements and international investment agreements, tariffs and competition policy. 

 As access and innovation issues are increasingly considered across a broader range of policy areas, a more diverse set of stakeholders, values, experience, expertise and empirical data now shapes and informs policy debates, through: ƒƒgreater diversity of policy voices, creating opportunities for cross-fertilization between traditionally distinct policy domains ƒƒenhanced possibilities for harvesting the practical lessons of a far wider range of innovation and access initiatives ƒƒimproved global inclusiveness, quality and availability of empirical data on a range of interconnected factors, including the global health burden, access and pricing of medicines, regulatory and trade policy settings, and national IP systems. 

 The cross-cutting character of these policy domains means that some themes are introduced in Chapter II, in the course of sketching out the general policy framework, and are later elaborated in Chapter III and/or Chapter IV, which look in more detail at how these elements have bearing on innovation and access, respectively. For example, the general elements and principles of IP policy are set out in Chapter II, while Chapter III elaborates aspects of IP policy, law and practice that bear particularly on innovation of medical technologies, and Chapter IV considers how specific aspects of IP impact access to technologies. Similarly, the broad rationale for regulation of medical technologies is set out in Chapter II, and Chapters III and IV deal with the implications of product regulation, respectively, for the innovation process and for access to medical technologies. Regarding trade policy, Chapter II sets out the main elements, and Chapter IV considers the impact of trade and trade policy settings on access to medicines and other medical technologies.
 
The global burden of disease necessitates dynamic responses
 
The global burden of disease is in transition. Populations are ageing due to progress in preventing and treating infectious diseases. But the burden of NCDs in low- and middle-income countries (LMICs) is rising, leadingtoadoubleburdenofdisease(seeChapterI, section C). While preventive measures with respect to lifestyle, physical inactivity, tobacco use and harmful use of alcohol, nutrition and environmental factors are key, the innovation system has to adjust to these changes in the global disease burden. The focus on access to medicines – which, in the past, has been on communicable diseases such as HIV/AIDS, TB and malaria – has broadened. Access to treatments for NCDs, including expensive cancer treatments in middle- income countries, will be the challenge of the future and the focus of the access debate (see Chapter IV, section B.4)
 
Access to medicines and the right to health
 
Access to medicines and health services is an element of the fulfilment of the right of everyone to the enjoyment of the highest attainable standard of health. Furthering access to medicines is also part of the United Nations Sustainable Development Goals (SDGs) (see Chapter II, section A.1–3). Lack of access to health technologies is rarely due to a single factor. The “value chain” of medicines and health products (see Figure 4.3) includes R&D, regulation, selection, procurement and supply, distribution, prescribing of medicines and diagnostics, dispensing, and responsible use (see Chapter IV, section A.2). Selection of the medications requires a health system to identify which medicines are most important to address the national burden of disease. This selection can be guided by the WHO Model List of Essential Medicines. Political commitment to adequate and sustainable funding is a basic condition for effective and sustainable access. Universal health coverage (UHC) has crystallized as a key aim of the SDGs (see Chapter IV, section A.1). Affordable prices are a critical determinant of access to medicines, especially in countries where the public health sector is weak and a large part of the population pays for medicines out of pocket. Generic medicine policies are key interventions to control health budgets and make medicines and other health products and services more affordable. Yet even generic medicines can still be unaffordable to health systems. A substantial part of the global population cannot access even the most basic medicines (see Chapter IV, section A.3). The overarching condition for providing access to needed medical technologies and health services is a functioning national health-care system (see Chapter IV, section A.4–12.).
 
Efforts to scale up treatment coverage for HIV/AIDS have become a major focus for policy-makers since the turn of the millennium. Low prices for generic antiretroviral treatments have helped governments and donor agencies strive to end the AIDS epidemic by 2030, as set out in target 3.3 of the SDGs (see Chapter IV, section B.1). In the area of antimicrobial resistance (AMR), there is a need to simultaneously secure wide availability of core antimicrobials, while also ensuring good stewardship (appropriate antimicrobial use to improve patient outcomes and minimize the development and spread of resistance) and the research in, and development of, new antimicrobials (see Chapter II, section A.5, Chapter III, section C.2 and Chapter IV, section B.2).
 
While most cases of TB can be successfully treated with medicines that have been available for many decades and are low cost, there has been growing concern about drug-resistant TB. Three new medicines were approved between 2012 and 2019 to treat drug-resistant TB, but access to them has been limited for reasons including limited clinical data, lack of national registration, high prices and a lag in implementing new treatment guidelines (see Chapter IV, section B.3).
 
NCDs put an enormous and continuous financial strain on household budgets, and major gaps in access to both originator and generic medicines for NCDs persist. Shortcomings in access have been highlighted, for example, for newer cancer treatments and insulin for diabetes. For all countries, the cost of inaction far outweighs the cost of taking action on NCDs (see Chapter IV, section B.4). Health systems, including in high-income countries, face rising launch prices, in particular for cancer and “orphan” medicines.
 
Hepatitis C has seen treatment breakthroughs, but these new treatments entered the market at very high prices, leading to treatment being unavailable, rationed or delayed in numerous countries. Thanks to the conclusion of licensing agreements for some of the treatments, generics are available at relatively low prices in most LMICs (see Chapter IV, section B.5). National immunization programmes are a highly effective public health tool for the prevention of illness and the spread of infectious diseases. Distinct market conditions and know-how requirements create a different landscape for the development and dissemination of vaccines (see Chapter III, section B.4(e) and Chapter IV, section B.7; see also Chapter III, section E). Other areas addressed by the study are access to paediatric formulations and medical devices (see Chapter IV, sections B.6 and B.8).
 
Measures to contain costs and increase access
 
Governments employ many different means to contain costs for medical technologies. Policies aimed at increasing access concern areas such as procurement, pricing and IP (see Chapter IV, sections A and C), and they increasingly use health technology assessments to control costs (see Chapter IV, section A.4). Import tariffs (see Chapter IV, section D.1), various taxes (see Chapter IV, section A.5) and mark-ups along the supply chain (see Chapter IV, section A.6) can increase consumer prices and constrain access, and can also be targeted by cost- containment policies, which must, however, ensure sustainable margins for commercial suppliers in order to be economically viable.
 
Differential pricing applied by companies can be a complementary tool to increase access. Price differentials may exist across different geographical areas or according to differences in purchasing power and socio- economic segments (see Chapter IV, section A.4(g)). Another strategy for enhanced access to medicines is to promote the development of local production capacity and leverage technology transfer. Policy coherence associated with local production is crucial to achieving sustainable public health and industrial development benefits (see Chapter IV, section A.10).
 
The WTO Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement) makes available to WTO members flexibilities in implementing access policies, such as patentability criteria and patent review procedures, and regulatory review exceptions (see inter alia Chapter II, section B.1 and Chapter IV, section C.3). With regard to access to patented products, these flexibilities include the use of compulsory or government-use licensing, wherein generic versions of the patented product can be locally manufactured or imported without the authorization of the patent holder.
 
Regulation of health technologies
 
Regulation of health technologies addresses essential health policy objectives: products must be safe, efficacious and of adequate quality. It also shapes the landscape for access and innovation. Regulatory review processes affect the time and cost it takes to bring new products to market and may delay market entry of new products (see Chapter II, section A.6).
 
Clinical trials are research studies in which groups of human participants are enrolled to evaluate the safety and/or effectiveness of new health technologies. The registration and publication of clinical trials are important for public health. The WHO considers registration of clinical trials a scientific and ethical responsibility and maintains the International Clinical Trials Registry Platform. From the perspective of public health policy, clinical trial results should be publicly available, so that researchers and other interested groups can assess the efficacy and potential side effects of new products (see Chapter III, section B.7). The emergence of biotherapeutic medicines has raised challenges for regulatory systems, notably with regard to regulating similar biotherapeutic (also termed biosimilar) products (see Chapter II, section A.6). Another challenge for regulatory systems is posed by substandard and falsified (SF) medical products, which are found in all parts of the world but are typically a much greater problem in regions where the regulatory and enforcement systems are weak. To effectively combat SF medical products, regulatory intervention may be required, whereastheapproachtofalsifiedorcounterfeitmedical products may involve criminal investigation (see Chapter II, section B.1(f) and Chapter IV, sections A.12 and C.3(h)). WHO prequalification has contributed substantially to improving access to quality medical products in developing countries through ensuring compliance with quality standards (see Chapter IV, section A.11(a)).
 
Innovation in medical technologies: the evolving policy landscape
 
Innovation in medical technologies requires a complex mix of private- and public-sector inputs. It differs from innovation more generally due to the ethical dimension of health research, a rigorous regulatory framework, liability questions and the high cost and high risk of failure. Economic, commercial, technological and regulatory factors have precipitated rapid change in the current landscape for R&D, involving more diverse innovation models and a wider range of active players. Providing adequate incentives to absorb the high cost and associated risks and liabilities is a central policy challenge; this has been the historic role of the patent system, in particular as applied to pharmaceuticals. While estimates vary of the actual cost of medical research and product development, innovation is undoubtedly costly and time consuming. The risk and uncertainty of innovation increases R&D costs in this sector, which include the development costs of the vast majority of inventions that fail before reaching the market (see Chapter III, section B.3). Rising expenditure for medical research has not been matched by a proportionate increase in new products entering the market, sparking a debate about research productivity and a quest for new models of innovation and for financing R&D. Many initiatives are exploring new strategies for product development, thus informing a rich debate about how to improve and diversify innovation structures to address unmet health needs. Current policy discussions have identified possibilities for open innovation structures, and a range of “push and pull” incentives, including schemes such as prize funds that would delink the price of products from the cost of R&D (see Chapter III, section C.5). The WHO Consultative Expert Working Group on Research and Development: Financing and Coordination recommended some of these options, including beginning negotiations on a globally binding convention or treaty on R&D (see Chapter III, sections C.4 and C.5(i)).
 
New thinking on industry’s role and structure and the public/ private divide
 
The evolving innovation landscape is driving change in the pharmaceutical industry. Driving factors include rising global spending on prescription drugs, increasing payer scrutiny of prescription drug prices in high-income markets, the progress of non-profit initiatives engaged in medical research and product development, new research tools and platform technologies, increased industry focus on personalized medicines, and the greater share of global demand from large middle- income-country markets. The historic industry model of vertically integrated in-house R&D is opening up to more diverse and collaborative structures, with major industry players developing products by integrating technologies that are licensed in or acquired through mergers and integration of smaller firms. Originator firms have also invested in generic production capacity. An increasing proportion of new medicines are for orphan indications. At the same time, most large pharmaceutical companies have withdrawn from antimicrobial research in light of the poor potential for investment returns.
 
The role of public research and academic institutions, increasingly, also in developing countries, has come under the spotlight as those institutions seek to reconcile public-interest responsibilities with the capital and product development capacity offered by private-sector partnerships (see Chapter II, section C; Chapter III, sections A and B; and Chapter IV, section D.5(d)).
 
Research and innovation gaps in neglected diseases and other areas: a policy challenge generating practical initiatives
 
For diseases that predominantly affect people living in poorer countries, the innovation cycle is not self-sustaining and fails to address their health needs, due to low potential for revenue, underfunded health services and generally weak upstream research capacity. A similar situation arises where sales are likely to be low, for example, in antibiotics and treatments or vaccines for emerging pathogens. In this type of environment, market-based incentives, such as patent protection, cannot by themselves address the health needs of developing countries. The landscape of health research for these diseases has evolved. Product development partnerships (PDPs) have been a significant development over the past decade, drawing together not-for-profit entities and industry players, with major philanthropic funding, significantly increasing the number of products in development for neglected diseases, and identifying pathways regarding existing research gaps (see Chapter III, section C.6). Originator pharmaceutical companies also engage increasingly in philanthropic research. Several companies have established dedicated research institutes to research diseases disproportionately affecting developing countries or participated in cooperative projects to share assets and knowledge, such as WIPO Re:Search, which has been developed to make better use of IP-protected assets and improve access (see Chapter III, section C.6–8). However, much more needs to be done by the international community in this area. AMR has been recognized as a global threat, and is addressed by many countries in national action plans and by a WHO Global Action Plan on Antimicrobial Resistance. Private investments are insufficient to fill current R&D gaps. New non-profit initiatives have been established by a range of actors to reinvigorate the pipeline of drug candidates.
 
The IP system at the centre of debate on innovation and access
 
Apart from the patent system and test data protection, other relevant IP rights include trademarks, for example, the relationship with international non-proprietary names (INNs), and copyright, for example, covering the package insert of medicines (see Chapter II, section B.1(d)–(e)). The patent system has been widely used for health technologies, especially by the pharmaceutical sector. Indeed, the pharmaceutical sector stands out in terms of its dependence on patents to capture returns to R&D, but its role in innovation and how to enhance its effectiveness are matters of continuing debate (see Chapter III, section B). The rationale for having patents is to make investment in innovation attractive and to offer a mechanism that ensures that the knowledge contained in the patent documents is accessible. Patents can function to structure, define and build innovation partnerships. The role of intellectual property rights (IPRs) in the innovation cycle is addressed in Chapter III, section D. The impact of patents on access is complex and an area of particular focus. IP policy, the laws that embody the policy, and the administration and enforcement of those laws each aim to balance and accommodate a range of legitimate interests in a way that promotes overall public welfare (see Chapter II, section B.1).
 
The global IP framework is defined in particular by the treaties administered by WIPO and the TRIPS Agreement, which forms part of the WTO legal system and in turn incorporates the substantive provisions of several WIPO treaties, including the Paris Convention. The TRIPS Agreement sets minimum standards for IP protection and enforcement. For example, patents must be available for any innovation in all fields of technology, provided they are new, involve an inventive step (or are non-obvious) and are capable of industrial application (or are useful). Substantive patent examination leads to a higher degree of legal certainty regarding the validity of granted patents. Where search and examination are of low quality, this can have an adverse effect because it may raise false expectations in respect of the patent’s validity. Review procedures allow courts and other review bodies to correct erroneous grant of patents and give relief where necessary, in order to ensure that the patent system, as a whole, functions as a public-interest policy tool. Strict patentability criteria and strict patent examination supported by patenting examination guidelines contribute to preventing strategies employed to delay the entry of generic competition, such as “evergreening” (see Chapter III, section D.4(b) and Chapter IV, section C.1).
 
Integral to the patent system is the requirement to disclose the innovation described in patent documents, thus creating an extensive knowledge base. The resultant patent information serves as a tool for charting freedom to operate, potential technology partnerships, and procurement options, as well as giving policy-makers insights into patterns of innovation (see Chapter II, section B.1(b)(viii)–(xi)). While patent information has become more accessible, coverage of data for many LMICs remains a challenge. Recent trends show a growth in patent applications on health technologies from key upper-middle-income economies (see Chapter III, section A.5).
 
The protection of clinical trial data also illustrates the complex relationship between the IP system and innovation and access. Protecting these data against unfair commercial use is important given the considerable efforts made to generate these data, which are needed to bring new medicines to the market. For this purpose, in some jurisdictions, newly approved medicines are protected by periods of regulatory exclusivity, such as data exclusivity and market exclusivity, during which the medicines regulatory authority may not accept a submission for approval of a generic and/or may not approve a generic for marketing. The TRIPS Agreement requires protection of test data but does not specify the exact form it should take, and national authorities have taken diverse approaches (see Chapter II, section B.1(c)).
 
How IP is managed can determine its impact on public health
 
Appropriate licensing of patents can help build partnerships and enable innovation through cooperation to bring new health technologies to fruition. Private- sector licensing strategies typically aim at commercial objectives, but public-sector entities can use patents to leverage public health outcomes. New models of socially responsible licensing protect IP while ensuring that new health technologies are available and affordable. Public– private partnerships have resulted in creative licensing agreements that forgo profit maximization in favour of providing essential technologies to poorer countries at affordable prices. Voluntary licences also form part of corporate social responsibility programmes, especially for HIV/AIDS treatments.
 
The Medicines Patent Pool has reinforced the trend towards voluntary licensing programmes that increase access to medicines by enabling new formulations and enhancing provision of cheaper generic medicines for developing countries (see Chapter IV, section C.3(b)).
 
Policy options and IP flexibilities also impact on public health
 
A wide range of policy options and flexibilities are built into the international IP regime and can be used to pursue public health objectives. Action is needed at the regional and domestic levels to determine how best to implement such flexibilities, so that the IP regime responds to each country’s individual needs and policy objectives. Key options include transition periods for least-developed countries (LDCs) (see Chapter II, section B.1), differing IP exhaustion regimes, refining the criteria for grant of a patent, making available pre-grant and post- grant review procedures, exclusions from patentability and exceptions and limitations to patent rights once granted, including regulatory review exception (“Bolar” exception) to facilitate market entry of generics, as well as compulsory licences and government-use licences. Countries have used one or more of these instruments to improve access to medicines for both communicable and noncommunicable diseases (see Chapter IV, section C.1–3). WTO members amended the TRIPS Agreement to permit wider use of compulsory licensing. The additional flexibility enables members that need to import medicines because of insufficient or no local manufacturing capacity to seek supply from generic manufacturers in other countries where the medicines are patent protected. For this purpose, potential exporting members can grant special compulsory licences exclusively for export under what is termed the “Special Compulsory Licensing System” (see Chapter IV, section C.3 and Annex III). While the legal scope for flexibilities is now clearer, thanks also to the Doha Declaration, and some flexibilities are widely implemented (such as “Bolar” exceptions), policy debate continues on the use of measures such as compulsory licensing.
 
International trade is an essential avenue to access
 
International trade is vital for access to medicines and other medical technologies, markedly so for smaller and less-resourced countries. Trade stimulates competition, which, in turn, reduces prices and offers a wider range of suppliers, improving security and predictability of supply. Trade policy settings – such as tariffs on medicines, pharmaceutical ingredients and medical technologies – therefore directly affect the accessibility of such products (see Chapter II, section B.3–5 and Chapter IV, section D). Trade policy and the economics of global production systems are also key factors in strategic plans to build domestic production capacity in medical products. Non- discriminatory domestic regulations founded on sound health policy principles are also important for a stable supply of quality health products. Access to foreign trade opportunities can create economies of scale to support the costs and uncertainties of medical research and product development processes. Developed countries have dominated trade in health- related products, but India and China have emerged as leading global exporters of pharmaceutical and chemical inputs and, in the case of China, of medical devices, and some other developing countries have shown strong export growth recently. Countries’ imports of health-related products differ considerably according to their level of development, illustrating substantial and widening gaps in access: in 2016, a small number of countries (China, European Union member states, Japan and the United States) accounted for the majority of imports. Some new players are emerging from developing countries, while LDC imports have grown least, starting from a low base.
 
Import tariffs on health-related products can affect access: since they increase cost early in the value chain, their impact on price may be magnified. Developed countries have largely eliminated such tariffs, in line with the WTO Pharmaceutical Agreement of 1994. Other countries have reduced tariffs significantly, but the picture is still mixed: some developing countries structure tariffs to promote local production, while LDCs apply lower tariffs (see Chapter IV, section D.1).
 
Competition policy promotes effective innovation and supports access
 
Competition policy is relevant to all stages in the process of supplying health technologies to patients, from their development to their sale and delivery. The creation of sound, competitive market structures through competition law and enforcement thus has an important role to play in both enhancing access to health technologies and fostering innovation in the pharmaceutical sector. It can serve as a corrective tool if IP rights hinder competition and thus constitute a potential barrier to innovation and access. Competition authorities in several jurisdictions have taken action to address anti-competitive practices in the pharmaceutical sector, including some patent settlements, certain licensing practices and pricing policies. Competition policy also has an important role to play in preventing collusion among suppliers of medical technology participating in procurement processes (see Chapter II, section B.2 and Chapter IV, section D.2).
 
Access to medical technologies through more effective government procurement
 
In many countries, access to medical technologies largely results from government procurement, with pharmaceuticals made available through public funds or subsidies. Procurement systems aim to obtain medicines and other medical products of good quality, at the right time, in the required quantities and at favourable costs. These principles are particularly important in the health sector, given the large expenditures, health impact of value for money and quality issues, with some programmes reportedly paying considerably more than necessary for medicines (see Chapter IV, section A.8).
 
Procurement policies favouring open and competitive tendering, coupled with the rational use of medicines, become all the more important in ensuring continued access in a fiscal climate in which national budgets are under pressure and philanthropic programmes face funding constraints. Good governance in procurement is consistent with increasing access to medical technologies through lower prices and uninterrupted supply. The WTO’s plurilateral Agreement on Government Procurement provides an international framework of rules to promote efficiency and good governance in public procurement, with particular application to procurement of medicines, promoting transparency, fair competition and improved value for public expenditure (see Chapter II, section B.4).
 
Free trade agreements have increasing relevance to access
 
The international policy and legal framework has been made more complex by the growth of free trade agreements (FTAs) and international investment agreements, outside the established multilateral fora (see Chapter II, section B.5 and Chapter IV, section C.5). Policy debate in this context has focused on IP, such as patent term extensions, regulatory exclusivities and other measures, such as patent linkage, as well as pharmaceutical regulation provisions n these agreements, and their impact on access to medicines. The later generation of FTAs often includes side letters or provisions confirming the Doha Declaration and, in particular, the right of WTO members to take measures to protect public health. These agreements also set standards in other policy areas with implications for access, notably, standards established on government procurement and competition policy, as well as preferential tariffs on pharmaceuticals, inputs and other health products. FTAs usually require implementation in domestic laws, which, in turn, can directly affect access to, and innovation in, medicines and medical technologies.

Government Access to Vehicle Data

The Australian National Transport Commission (NTC) discussion paper Government access to vehicle-generated data comments
 Vehicles are increasingly capturing a range of useful data about the road environment, the vehicle itself and the way it is used. The vehicle industry is also rapidly expanding the capability of vehicles to connect and share data. This could provide a new opportunity for governments to improve their transportation systems through access to and use of this new data. If government access is not considered in a nationally consistent way, governments risk creating a fragmented, overly burdensome or low-access data environment. 
Australia’s transport agencies have identified that this new data will be important for operating more dynamic and responsive transportation systems. This new vehicle-generated data has the potential to improve road safety, optimise the road network and better inform network planning. For this report we have defined vehicle-generated data as any data generated by a vehicle that produces information about the vehicle, the environment around the vehicle or the use of the vehicle. 
The purpose of this project is to develop policy options for government access to and use of vehicle-generated data for the purposes of road safety, network operations, investment, maintenance and planning. 
The purpose of this discussion paper is to:
  • ▪ discuss our understanding of key issues and challenges arising from government access to vehicle-generated data 
  • seek views on the opportunity statement and problem statements contained in this paper  
  • seek views on options that could address these problems. 
What are the opportunities and benefits? 
Improved road safety has been identified as a key need that could be addressed through greater access to vehicle-generated data. For example, sharing and reporting of traffic safety events could use vehicles to detect and warn occupants about dangerous road conditions, allowing transport agencies to respond more rapidly to incidents. It is also the area with the highest willingness among vehicle manufacturers to share data. Unlike other types of data such as vehicle movement data, this data is unique to the vehicle and cannot be as easily replicated from other data sources. 
To better understand the needs of transport agencies and industry, we hosted several co- design workshops. The workshops generated 23 use cases identifying different potential uses for vehicle-generated data. Transport agencies identified significant potential benefits for road safety, network planning and optimisation. Transport agencies saw that this data could better inform decision making and reduce road trauma. However, the detailed benefits and costs of these use cases are still unknown. We also found that further data requirement and business case development is needed on priority uses for vehicle-generated data. Further collaboration between industry and government to better understand the potential benefits and costs would be highly beneficial to achieve this and is strongly supported by stakeholders. 
Australia does not lag significantly behind international jurisdictions in government access to vehicle-generated data; however, there are early international collaboration efforts that Australia can learn from. Key among these is the European Union’s memorandum of understanding between government and industry on the exchange of vehicle-generated data to support eight safety-related use cases. 
What are the barriers and gaps? 
Vehicle-generated data can be costly to generate, carry, store and use, and can reveal sensitive information about users and businesses. Much of this data is not stored, broadcast or shared. There is currently a low market penetration of vehicles that can share this data. The key barriers to government access include:
  • There is no compelling reason or incentive for generators of vehicle-generated data to provide this information to transport agencies (with the exception of the road access, safety and productivity benefits provided to heavy vehicle operators through regulatory telematics). 
  • There are trust, cost and operational barriers to the exchange of vehicle-generated data and, outside of heavy vehicles, there is no data access framework to address these issues. 
  • In comparison with international markets, there are currently fewer vehicles capable of capturing and communicating vehicle-generated data on Australia’s roads, with only market-based mechanisms to encourage uptake. 
What are the opportunities and problems? 
We have identified one key opportunity for government to access vehicle-generated data: 1. There is an opportunity for stakeholder collaboration on exchange or sharing of vehicle-data for road safety purposes to understand: – what vehicle-generated data can be used to support road safety in Australia – what an appropriate framework and forum might look like to support such an exchange. 
We have identified three problems that will we need to overcome to create wider government access:
1. Vehicle-generated data is currently not provided to transport agencies for purposes that may have publicly beneficial outcomes. This could be due to current vehicle capabilities and/or a lack of incentive or reason for industry and road users to provide the data (the exception to this being heavy vehicles enrolled in a current regulatory access or compliance scheme under the Heavy Vehicle National Law). 
2. There is a lack of a data access framework to provide the necessary trust, data exchange systems, data standards/definitions, understanding of data needs and governance to establish data access and use (the exception to this being heavy vehicles enrolled in a current regulatory access or compliance scheme). 
3. The level of uptake and penetration of connectivity across the Australian vehicle fleet may delay the benefits of vehicle-generated data, particularly related to safety-critical events.
What are we proposing to address the opportunity and problems? 
Recognising that vehicle-generated data is still at the nascent stage of development in Australia and that stakeholders remain unclear on priorities, there is an opportunity for governments to adopt a new policy approach. We propose that a new collaboration between industry and governments begin with a focus on road safety data as the priority and common mission. This approach is in line with the European Union’s approach and has early consensus from industry and government. We propose: For future development on government access to vehicle-generated data, road safety is the priority for exchanging vehicle-generated data between industry and government. Industry and government should collaborate on identifying opportunities for exchanging road safety data and adopt a principle of non-commercial sharing or exchange. 
We have identified three options to address problems 1 and 2, which are: –
Option 1: Rely on existing arrangements between government and industry, with no changes to existing legislation or frameworks. 
Option 2: Establish a data exchange partnership between industry and government that will identify opportunities for exchanging vehicle-generated data as well as develop standards and consider proof of concept. 
Option 3: Introduce new legislation requiring industry to collect, store and retain vehicle-generated data while providing access to government.
The NTC’s preliminary preferred option is option 2. We believe this option can provide the best opportunity for government to better understand how to maximise the potential benefits and opportunities of vehicle-generated data while actively collaborating with industry. This option has received general early support from government and industry. 
To address problem 3 – a lack of stimulus to bring forward vehicle connectivity – we are proposing that the Commonwealth considers the costs, benefits and system requirements to require vehicles to send automated crash notification system messages and have these received and actioned. Europe has achieved this through introducing its eCall system. This would bind all vehicles to a capability to send data messages over private networks. This proposal could be enacted through the Commonwealth government considering adoption of international standards into an Australian Design Rule (with consequential amendments to the relevant state and territory in-service vehicle standards) and infrastructure and capability to receive and use emergency notification messages. This would result in a significant increase in the fleet penetration of connected vehicles in Australia. 
List of questions 
1 Do our problem and opportunity statements accurately define the key problems to be addressed, and do they capture the breadth of problems that would need to be addressed? 
2 In our table, have we accurately captured all the regulatory and legislative mechanisms government could currently use to access vehicle-generated data?
3 Are there other major local or international jurisdictional developments providing further access powers or arrangements for vehicle-generated data?
4 Do you agree with our assumptions on the currently low uptake and limited availability of technology that supports the generation of vehicle data and that there are few and limited current government access arrangements for vehicle-generated data?
5 What issues do you believe will be created if ExVe is adopted and that would need to be considered in Australia? 
6 Is there value in establishing a national data aggregator or trust broker? Could good data definitions, practices and cooperation between entities achieve the same outcome? 
7 Can you provide us with more information on either the costs or benefits for government access to vehicle-generated data for the use cases listed in Appendix B? 
8 Are there relevant international standards that should be adopted for vehicle- generated data? Are there any standards that could be locally developed? . 
9 Have we accurately described the key barriers to accessing vehicle-generated data? Are there additional barriers? 
10 Do you agree that road safety data should be considered the priority purpose for which we seek to exchange data with industry? 
11 What are the key data needs of transport agencies beyond those already identified? 
12 What further benefits from vehicle-generated data should be considered? 
13 We contend that a prioritised starting point should be established from which data for other purposes can be further developed. Are there other approaches that could achieve this? 
14 Do you agree with the analysis presented in Table 7? What other opportunities are there for vehicle-generated data, and why? 
15 Have priorities changed for land transport policy and for data access from vehicles with the onset of COVID-19? 
16 Should road safety be adopted as the priority for developing use cases for government use of vehicle-generated data? If not, what other approach should Australia take? 
17 Can data other than for the purposes of road safety be exchanged on non- commercial terms? 
18 Does the NTC’s preferred approach (option 2) best address the problems we have identified? If not, what approach would better address these problems? 
19 Does the NTC’s proposed approach best address the problems we have identified? If not, what approach would better address these problems? .

29 July 2020

Antiquities and AML/CTF

Tracking and Disrupting the Illicit Antiquities Trade with Open Source Data by Matthew Sargent, James V Marrone, Alexandra Evans, Bilyana Lilly, Erik Nemeth and Stephen Dalzell (Homeland Security Operational Analysis Center) considers the following research questions
  • What do the actors, networks, and markets that enable the looting, trafficking, and sale of antiquities look like? 
  • What data sources can be used to assess the structure and transaction volume of the illicit antiquities market? 
  • What are the potential strategies and data sources that would guide more-effective enforcement? 
The authors comment
The sale of illicit goods provides an important funding source for terrorist organizations, organized crime, and rogue states. Therefore, tracking and disrupting these networks is an important national security goal, but it is often difficult to accomplish because of the clandestine nature of these transactions. ... The illicit antiquities market has become an area of concern for policymakers. It is fueled by a well-documented rise in looting at archaeological sites and a fear that the proceeds of such looting may be financing terrorism or rogue states. Efforts to craft effective policy responses are hindered by the lack of data and evidence on two fronts: the size of the market and the network structure of participants. In lieu of reliable evidence on these two fronts, the conversation has been dominated by speculation and hypotheses and has generated some widely accepted theories of how the illicit antiquities market operates. 
In this report, the authors compile evidence from numerous open sources to outline the major policy-relevant characteristics of that market and to propose the way forward for developing policies intended to disrupt illicit networks. The approach uses multiple methods and data sources, with the understanding that no single piece of evidence can provide a complete picture of the market and that only by cross-referencing and triangulating among various sources can salient market characteristics be illuminated.
Their key finding are that
The market size is smaller than often reported. Market structure varies widely, but it often appears ad hoc and opportunistic. The West is not the only end market for looted antiquities. Technology used in the looted antiquities trade is mostly unsophisticated. Policy responses should
  • address a decentralized network that relies heavily on trust and communications between buyers and sellers who do not have an ongoing personal relationship. 
  • Increase fear of law enforcement by sharing stories that highlight the risks of illicit trades or accounts of sting operations. 
  • Increase skepticism about fakes and replicas by highlighting the lack of specialists in the network and noting recent examples of doubt about the authenticity of high-profile antiquities. 
  • Undermine trust by increasing the perceived threat of surveillance on messaging and transaction platforms. The methods demonstrated in this report should be applied and validated in other research contexts.
We could, of course, hold the major auction houses to their claims about vigilantly opposing illegal  traffic in antiquities ... something that is clearly not the case.

The authors state

Our evidence implies that the size and structure of the illicit market is at odds with the conventional wisdom espoused by some journalists and researchers. Commonly quoted estimates of the size of the illicit market range from hundreds of millions to billions of dollars, while news stories frame antiquities trafficking as the domain of well-organized criminal networks. In contrast to these assertions, our research points to a market that looks smaller and is less organized.
• The market size is smaller than often reported. Our data indicate that the illicit market is likely much smaller than the multibillion-dollar claims that are commonly quoted. We find no evidence that illegal sales are occurring in large or even steady quantities on deep web platforms, such as Facebook or Telegram, and we find virtually no evidence of antiquities being traded on the dark web. As for the “visible” market of auction houses, dealers, eBay, and other online outlets, we find that the entire market (looted or not) is not likely to be larger than a few hundred million dollars each year. Even this is likely an overestimate, given that our appraisals contain data with fakes, replicas, and other nonantiquities.
• Market structure varies widely, but it often appears ad hoc and opportunistic. Organized crime actively participates in the illicit antiquities market in some regions, including Bulgaria, Greece, and Italy, but we found a second type of network that involves some activity in Iraq and Turkey. This network is best characterized as an organic, ad hoc system of smugglers, middlemen, and brokers whose trade in antiquities is opportunistic and sometimes sporadic. In addition to characterizing these often-discussed aspects of the market, the report illuminates other characteristics that are relevant for both policymakers and for future research.
• The end market for looted antiquities is not only the West. It is often assumed that the end market for looted antiquities is the Europe and the United States. But our informant interviews suggest that antiquities from the Middle East and Levant are ending up in Iran, Turkey, the Persian Gulf States, and other nearby countries. Furthermore, the prices being paid to middlemen for low- or medium-quality goods are on par with what would be paid in the end market in Europe or North America. These observations are contrary to the assumption that middlemen are paid just a fraction of the final price as they help in moving antiquities to Europe or the United States. The implication of this finding is that it is important to tailor policy responses to the situation if, indeed, the supply chains for certain goods are different from was has been assumed.
• Technology used in the looted antiquities trade is mostly unsophisticated. Studies of other illegal goods suggest that the dark web would be a natural place to sell looted antiquities. However, our analysis of dark web platforms finds virtually no evidence of antiquities sales. On deep web platforms, such as Telegram and Facebook, we also find little evidence of antiquities sales, although other secure messaging applications, such as WhatsApp and Viber, are used to coordinate sales and streamline communications within existing networks. We agree with other researchers who have concluded that the low level of enforcement and the existence of low-risk sales channels reduces the need for dark or deep web sales of antiquities. ...
In the absence of reliable statistical data, researchers have relied instead on historical and empirical evidence gathered primarily from field research, interviews,legal records, and government sources. Yet research designs in the field are often poor, producing a fragmented body of case studies and an overreliance on anecdotal data. In particular, five case studies dominate the published research:

• 9/11 attacker Mohammed Atta’s unsuccessful effort in 1999 to sell Afghan relics to raise funds for the attack
• an Irish Republican Army cell’s efforts in the 1990s to trade stolen art for heroin after failed efforts to find an underground buyer
• the 1999 discovery of an international art smuggling ring that sought to trade an estimated $35 million in stolen masterpieces for cocaine
• the discovery of antiquities among insurgent weapon caches captured in Iraq from 2003 to 2007
• the Islamic State’s organized trade in Iraq and Syria.

Although evocative, these cases are frequently presented without sufficient explanation to support the authors’ accompanying assertion that terrorist groups draw significant revenue from antiquities. With the exception of the Islamic State example, the published analysis of these cases can demonstrate only the existence of intermittent or opportunistic efforts to profit from antiquities trafficking, but they and cannot illuminate the frequency, scale, or organization of these illicit transactions.

In the absence of grounded data, journalists, researchers, and policy experts regularly inflate the importance of antiquities trafficking in funding for international terrorism and organized crime. Linking cultural property crimes to these high-profile law enforcement issues offers a means to bring funding and political attention to what has traditionally been an underrecognized issue. However, the facts and figures used to support these arguments are often misinterpreted or overstated. Unsubstantiated claims about the relationship among looting, weapons, drugs, and money laundering are common in both expert and popular publications, and inaccurate or exaggerated estimates of stolen items’ value abound.23 The absence of a comprehensive effort to quantify the trade has also encouraged the spread of misleading or inaccurate statistics. The Guardian’s June 2014 publication of an anonymous intelligence official’s claim that the Islamic State may have plundered $36 million worth of goods—including weapons, equipment, and natural resources—from the al-Nabuk region of Syria has been widely misrepresented as either evidence that the organization pilfered $36 million in antiquities from one site alone or as a low-range estimate of the organization’s total trafficking revenue. Similarly, an unsubstantiated claim that art crime, cultural property trafficking, or the illicit antiquities trade (the variation itself a demonstration of the literature’s definitional challenges) represents the third highest-grossing criminal trade, behind drugs and weapons, has proliferated through popular, governmental, and academic publications. Although INTERPOL issued a rare public disclaimer underscoring that despite the absence of “any figures which would enable [it] to” determine such a ranking, the estimate continues to be “frequently mentioned at international conferences and in the media.”

The fundamental problem, as Deborah Lehr of the Antiquities Coalition has observed, is that, as with most illicit markets, “there’s no real information or statistics on the size of this illegal trade.” However, this lack of data has not diminished the spread of wildly varying estimates about the size of the antiquities market. An article in the Wall Street Journal included estimates from Michael Danti, an archaeologist and academic director of ASOR CHI, placing the value of the Islamic State’s antiquities trafficking in the “low tens of millions” annually, while a French security official placed the figure at $100 million. Lehr, of the Antiquities Coalition, estimated that “[c]oming out of Syria, it is $2 billion” and “[w]ith Egypt, it is probably $3–10 billion, globally it has to be a much more significant number.”

There has been little apparent effort in the field to ground these estimates in data or to understand the size of the market for antiquities. By way of comparison, the volume of all sales of Greek, Roman, and Egyptian antiquities by the major auction houses—Bonham’s, Sotheby’s and Christie’s—in 2015 amounted to $41 million. Among these sales were artifacts whose provenance could be traced back as far as 1732, and only $326,000 of these sales were objects whose provenance could not be established before 2000. Moreover, 25 percent of all the items offered at auction were not sold either because there was no bidder or because the reserve price was not met. This reality that antiquities auctions represent a small market that is not always able to find buyers in well-advertised sales is at odds with the media’s assumption that there is a booming unmet demand for these goods that is capable of supporting a billion-dollar black market.

The US Senate Permanent Subcommittee on Investigations Committee on Homeland Security and Governmental Affairs report on The Art Industry and US Policies that Undermines Sanctions gives a fascinating view of AML/CTF.

The report states
The United States government imposes economic sanctions on foreign adversaries in attempt to change their behavior. In theory, sanctions are simple. U.S. persons and companies are prohibited from doing business with sanctioned persons and entities. This prohibition should bar access to the world’s largest economy. The United States imposes sanctions for a wide range of reasons. For example, the United States has imposed sanctions on Russia for election interference, human rights abuses, providing support to Venezuela and Syria, but mainly in response to Russia’s invasion of Ukraine. 
This report focuses, in particular, on a case study documenting how certain Russian oligarchs appear to have used transactions involving high-value art to evade sanctions imposed on them by the United States on March 20, 2014 in response to Russia’s invasion of Ukraine and annexation of Crimea. 
Specifically, the Subcommittee traced purchases of high-value art back to anonymous shell companies linked to sanctioned individuals Arkady and Boris Rotenberg, two Russian oligarchs, and Arkady’s son, Igor. It appears the Rotenbergs continued actively participating in the U.S. art market by purchasing over $18 million in art in the months following the imposition of sanctions on March 20, 2014. Shell companies linked to the Rotenbergs also transferred over $120 million to Russia during a four-day window between President Obama’s March 16, 2014 executive order stating that the U.S. would be sanctioning certain Russian individuals and the Treasury Department specifically naming the Rotenbergs as sanctioned on March 20, 2014. In addition, certain Rotenberg-linked shell companies continued transacting in the U.S. financial system long after Arkady and Boris Rotenberg were sanctioned. The Subcommittee determined these Rotenberg- linked shell companies engaged in over $91 million in transactions post-sanctions. While Russia-related sanctions, including those against the Rotenbergs, were set to expire in March 2020, President Trump extended them for another year. The effectiveness of these sanctions, however, is in question. To date, Russia has not withdrawn from Crimea and has even expanded its military operations in surrounding waters. The Subcommittee sought to understand why the sanctions have not been more effective and, after reviewing a number of suspect transactions, launched a narrow investigation into high-value art. If wealthy Russian oligarchs can purchase millions in art for personal investment or enjoyment while under sanction, it follows that their businesses or hidden resources could also continue accessing the U.S. financial system. 
The Subcommittee’s investigation uncovered a complex set of facts involving shell companies with hidden owners, intermediaries who mask purchasers and sellers, and lax money laundering safeguards in the U.S. art industry. 
The art industry is largely unregulated. The art industry is considered the largest, legal unregulated industry in the United States. Unlike financial institutions, the art industry is not subject to Bank Secrecy Act’s (“BSA”) requirements, which mandate detailed procedures to prevent money laundering and to verify a customer’s identity. While the BSA does not apply to art transactions by art dealers and auction houses, sanctions do. No U.S. person or entity is allowed to do business with a sanctioned individual or entity. 
The art industry has been enjoying a boom. According to the 2019 Art Basel and UBS Global Art Market Report, world-wide art sales hit $64.1 billion in 2019. That report found the United States was the world’s largest art market comprising 44 percent of global sales, or around $28.3 billion. The art industry is generally divided into sales at public auctions and by private dealers. In 2019, sales at auction houses made up 42 percent of total art sales, while the remaining 58 percent of sales were through private dealers. The four biggest auction houses by sales—Sotheby’s, Christie’s, Phillips, and Bonhams—are selling art for sizeable amounts. In November 2017, Leonardo da Vinci’s Salvator Mundi sold at auction at Christie’s in New York for over $450 million. In May 2019, Christie’s New York sold Jeff Koon’s Rabbit for over $91 million, the highest price ever paid for a piece by a living artist. Even during the COVID-19 pandemic, an online auction at Sotheby’s brought in $234.9 million in total sales, including $84.55 million for Triptych Inspired by the Oresteia of Aeschylus by Francis Bacon. In turn, the auction houses report large annual sale numbers. Sotheby’s reported $4.8 billion in sales for 2019, while Christie’s reported $2.8 billion in sales for just the first six months of 2019. 
Investors have taken notice. Deloitte’s 2019 Art and Finance Report noted that “artnet’s Index for Top 100 Artists produced an 8 percent Compound Annual Growth Rate between 2000 and 2018, compared with 3 percent for the S andP 500.” For example, Banksy’s Devolved Parliament sold at Sotheby’s in London on October 3, 2019 for around $12.2 million; the artist’s previous record for a painting sold at auction was $1.87 million for Keep It Spotless in 2008. 
Secrecy is pervasive in the art industry. While the art market is not regulated by the BSA, it is governed by unwritten rules. A large number of art sales happen through intermediaries referred to as “art advisors” who can represent both purchasers and sellers. In a typical transaction, a purchaser may not ask who owns the piece of art they are purchasing; the seller may not ask for whom it is being purchased or the origin of the money. And in general an art advisor would be reluctant to reveal the identity of their client for fear of being cut out of the deal and losing the business. 
Auction houses have voluntary AML polices. Because the art industry is not subject to BSA requirements, when a piece of art is sold, there is no legal requirement for the selling party to confirm the identity of the buyer or that the buyer is not laundering money through the purchase. While the four biggest auction houses have voluntary anti-money laundering (“AML”) programs, the employees who facilitated art purchases in the Subcommittee’s case study said they never asked the art advisor the identity of his client. Instead, the auction houses considered the art advisor the principal purchaser and performed any due diligence on the art advisor, even when it was well-known that the ultimate owner was someone else. With regard to the funds used to purchase art, the auction houses told the Subcommittee they rely on financial institutions to ensure the integrity of the funds, even though the auction houses interact directly with the buyer. But these voluntary AML policies are just for sales through the auction houses. As stated above, the majority of art sales are private transactions. A private dealer interviewed by the Subcommittee stated she had no written AML policies, tries to work with people she knows and trusts, looks for red flags, and relies on her gut. She also explained that her practices have significantly changed over the years and that she also relies on advice from AML lawyers. 
Secrecy, anonymity, and a lack of regulation create an environment ripe for laundering money and evading sanctions. Tracing the ownership of anonymous shell companies, including those involved in high-value art transactions, is difficult. That difficulty continues even though corporate secrecy suffered a blow in the spring of 2016 when the International Consortium of Investigative Journalists (“ICIJ”) shocked the world by releasing information on 214,488 offshore entities from the Panamanian law firm Mossack Fonseca (the “Panama Papers”). One email chain included among the Panama Papers and made public described links between nine offshore companies to the Rotenbergs. The email chain listed Boris Rotenberg as the ultimate beneficial owner (“UBO”) of Highland Ventures Group Limited (“Highland Ventures”) and Arkady Rotenberg’s son Igor as the UBO of Highland Business Group Limited (“Highland Business”). The email copied London-based attorney Mark Omelnitski, who used his firm the Markom Group to establish and maintain shell companies for the Rotenberg family. 
The true ownership of the listed shell companies was not, however, as straightforward as the Panama Papers email chain suggested. For example, based on financial information reviewed by the Subcommittee during its investigation, Arkady Rotenberg appeared to be the UBO of Highland Ventures, not his brother Boris. That information included non-public wire transfers showing multi-million dollar transfers from a company owned by Arkady Rotenberg to Highland Ventures. In 2012 and 2013, that company—Milasi Engineering—transferred over $124 million marked as annual dividends to Highland Ventures. The December 31, 2014 Financial Report for Milasi Engineering listed Arkady Rotenberg as its UBO, making it clear that Highland Ventures received its funding from a company owned by an individual the U.S. would later sanction. Milasi Engineering also held shares in Stroygazmontazh, a gas pipeline company sanctioned in April 2014 by the United States due to its ownership by Arkady Rotenberg. Arkady Rotenberg transferred his business interests to his son, Igor. In July 2014, four months after the United States sanctioned Arkady, Mr. Omelnitski’s firm, the Markom Group, executed paperwork that appeared to transfer Arkady’s interest in Milasi Engineering to his son, Igor, who was not sanctioned at the time. Milasi Engineering was owned by two other holding companies. The Markom Group transferred the ownership of those two companies to Highland Ventures, which it asserted had always been owned by Igor. Therefore, from July 2014 to April 2018, when Igor was finally sanctioned by the United States, Milasi Engineering was owned on paper by an unsanctioned individual. A report by a bank investigator produced to the Subcommittee determined the transfer of Milasi Engineering from Arkady to Igor was done solely to evade sanctions, and the Markom Group “intentionally structured [the ownership of these shell companies] to be opaque in order to hide the identities of true beneficiaries.” In response, the bank closed all accounts associated with the Markom Group. This included accounts held by art advisor Gregory Baltser. Mr. Baltser is a U.S. citizen, who must comply with U.S. sanctions laws. 
Intermediaries played a central role in the Rotenbergs’ art purchases in the United States. As previously explained, Mr. Omelnitski and his company, the Markom Group, established and maintained shell companies for the Rotenbergs to mask their identities. The Rotenbergs also employed art advisor Gregory Baltser, who facilitated the purchase and sale of high-value art both before and after sanctions without disclosing the names of his clients. Mr Baltser's business is based in Moscow. Prior to sanctions, funds Mr. Baltser used to purchase art linked to the Rotenbergs followed a unique and recognizable financial path: Mr. Baltser bid on specific artworks at auction, purchased the art, and then assigned the title to the art to a Belize company named Steamort Limited (“Steamort”). Steamort paid for the art using funds the Subcommittee traced back to Highland Business. Mr. Baltser, however, was not the owner of Steamort; he had a contract with Steamort to serve as a consultant to purchase art on behalf of the company. A copy of that contract was produced to the Subcommittee by Christie’s. Both the contract and financial records showed that Steamort paid Mr. Baltser $9,500 a month for his services. 
In total, between March 2010 and October 2018, financial records show Mr. Baltser received $1,116,000 in fees for his consulting services under the Steamort Agreement. Company documents obtained by the Subcommittee listed Steamort’s only director and shareholder as Jason Hughes. According to a report by ICIJ, Mr. Hughes was associated with over 200 other companies as a nominee director—an individual who masks the true UBO of a shell company. The owner of Steamort remains unknown. In 2012, Christie’s questioned Mr. Baltser about who owned Steamort, and asserted that Mr. Baltser could no longer bid at auctions until he provided Steamort’s UBO. Initially, Mr. Baltser responded that he did not know who owned Steamort. When pressed and threatened with missing the opportunity to bid at an upcoming auction, Mr. Baltser verbally told Christie’s that Steamort was owned by “Luisa Brown.” Christie’s accepted this verbal assertion, conducted AML checks on Ms. Brown, found no derogatory information, and cleared Mr. Baltser to continue bidding at auctions. Mr. Baltser never provided any documentary evidence of Steamort’s ownership by Ms. Brown. The Subcommittee was unable to confirm if an individual named Luisa Brown was the UBO for Steamort, or if she even existed at all. 
Mr. Baltser opened an auction agency and club in Moscow. In late 2012, Mr. Baltser announced he was planning to open BALTZER Auction Agency and Club. The agency would be located in Moscow and its members would be “the leading Moscow and Russian collectors – the active participants of auction biddings at many world marketplaces.” Mr. Baltser proposed to partner with both Christie’s and Sotheby’s. As part of the proposed agreement, Mr. Baltser stated that he would bid at auctions on behalf of his clients under an account in the name of BALTZER. 
This allowed Mr. Baltser to guarantee on his website that “we can give you complete anonymity.” Under the proposed agreement, Mr. Baltser pledged to conduct all AML and sanctions checks on his clients and provide an annual certification to the auction houses that no member of BALTZER was engaged in money laundering. Mr. Omelnitski served as BALTZER’s chief AML officer and represented Mr. Baltser in contract negotiations with the two auction houses. To be clear, Mr. Baltser put the same attorney who established and maintained shell companies to mask the Rotenbergs’ ownership in charge of his new venture’s AML compliance. 
Christie’s partnered with BALTZER. Christie’s accepted Mr. Baltser’s proposal and signed the agreement with BALTZER on February 4, 2014. At the end of 2014, Mr. Omelnitski certified to Christie’s that “despite BALTZER having a significant number of Russian clients there were no transactions, which fall under recent sanctions against Russia.” Mr. Omelnitski failed to provide another such certification for the next three years, despite repeated requests from Christie’s to provide the annual certificate promised in the agreement. In 2018, Christie’s renegotiated its agreement with BALTZER to require client due diligence documents after each purchase. 
A Sotheby’s employee identified Arkady and Boris Rotenberg as Mr. Baltser’s clients. Sotheby’s also considered Mr. Baltser’s business proposal, but ultimately declined. During negotiations, a Sotheby’s employee represented to Sotheby’s management that Mr. Baltser had told her that his clients included Russian oligarchs. In fact, she told Sotheby’s management that Mr. Baltser had identified Arkady and Boris Rotenberg as two of his clients (five months prior to U.S. sanctions). During her Subcommittee interview, however, the same Sotheby’s employee said Mr. Baltser had never told her that Arkady and Boris Rotenberg were his clients. Instead, she asserted she fabricated this information in an effort to convince Sotheby’s to accept BALTZER’s proposal. Despite declining the proposal, Sotheby’s continued to conduct business as usual with Mr. Baltser and his new company, BALTZER, and never questioned whether Arkady and Boris Rotenberg were his clients. The Subcommittee independently traced post-sanction purchases by BALTZER to shell companies linked to the Rotenbergs, suggesting the Sotheby’s employee was not truthful in her Subcommittee interview. Mr. Baltser continued to purchase art with funds linked to the Rotenbergs even after March 2014 sanctions. Following the imposition of sanctions by the United States on Arkady and Boris Rotenberg in March 2014, the funds Mr. Baltser used to purchase works of art at auction houses continued to follow the same general financial path as before sanctions. By this time, BALTZER provided another layer of anonymity for the funds used to purchase art. After Mr. Baltser successfully bid at auction, funds were wired from Highland Ventures to Steamort, just as they had arrived from Highland Business before sanctions. Steamort then wired funds to BALTZER, which paid the auction house and took title of the purchase. All four auction houses considered Mr. Baltser the principal purchaser, rather than an agent for a buyer, and never asked for whom he was purchasing the art. Any client due diligence was performed only on Mr. Baltser and not his undisclosed clients, satisfying the voluntary AML policies at the auction houses. 
Highland Ventures purchased a painting through a private art dealer. The funds used to purchase René Magritte’s La Poitrine for $7.5 million in May 2014 through a private art dealer followed a different path. In this transaction, Highland Ventures took title to the painting and was listed on the invoice as the buyer. Anna Wilkes, an employee of Mr. Omelnitski’s Markom Group, signed on behalf of Highland Ventures as its Director. The funds used to pay for the painting were wired to the private dealer from a company named Advantage Alliance. The Subcommittee traced those funds to a company called Senton Holdings. An investigation by a financial institution–produced to the Subcommittee–determined Senton Holdings was owned by Arkady Rotenberg, linking him through the chain of wire transfers to the purchase of the painting. 
Art was shipped to Germany for storage. La Poitrine, like much of the art traced to companies linked to the Rotenbergs, was shipped to a storage facility in Germany called Hasenkamp. The Subcommittee contacted Hasenkamp and was told the art was originally stored there under the name Highland Business; no individual was named. Later, a company named Taide Connoisseur Selection took over the contract to store the art at Hasenkamp. The only individual named on Taide Connoisseur Selection’s website was Mr. Omelnitski. In August 2019, during the course of the Subcommittee’s investigation, the Taide Connoisseur Selection account at Hasenkamp was closed and all art stored under the account was shipped to Moscow. Art purchases linked to the Rotenberg shell companies totaled millions of dollars. In total, the Subcommittee traced funds for over $18 million in art purchased in the United States from March 2014 to November 2014, both at auction houses and through private sales back to shell companies that appeared to be funded or owned by the Rotenbergs. Sotheby’s agreed to sell Brucke II for Mr. Baltser during the Subcommittee’s investigation. Mr. Baltser also sold paintings owned by his clients. In late 2018, he attempted to sell Lyonel Feininger’s Brucke II. Brucke II was originally purchased through Mr. Baltser on February 4, 2014 at an auction at Christie’s in London. The painting later appeared on a list of 31 paintings sent to Christie’s by a BALTZER employee, who stated that the list represented the collection of one of Mr. Baltser’s clients. The Subcommittee traced 16 paintings on the list purchased in the United States back to suspected Rotenberg shell companies. This suggests that all 31 paintings were owned by the Rotenbergs. 
When Mr. Baltser attempted to sell Brucke II in late 2018, both Christie’s and Sotheby’s expressed interest in having the painting at their auctions. Ultimately, Mr. Baltser’s client chose Sotheby’s to sell Brucke II at auction in February 2019. At the time, the Subcommittee was actively investigating the auction house and Mr. Baltser. Sotheby’s requested Mr. Baltser provide the name of the UBO of the painting, including whether that individual was currently sanctioned. 
Mr. Baltser said Brucke II had been resold since it was purchased at Christie’s in February 2014 and now belonged to a company incorporated in the Marshall Islands and provided a Russian passport for the company’s UBO. The Subcommittee asked Sotheby’s to request the name of the February 2014 purchaser of the painting; Mr. Baltser declined to disclose the name of that purchaser due to a non-disclosure agreement. Sotheby’s ultimately pulled the painting from the 2019 auction due to a lack of interest. 
The Subcommittee asked to interview Mr. Baltser, but through his attorney, he declined the request and stated he was in Moscow and had no plans to return to the United States. Through his attorney, Mr. Baltser stated that: he has never represented or transacted with Arkady or Boris Rotenberg; Highland Business and Highland Ventures were not listed as sanctioned by the Treasury Department; and he did not have access to the Panama Papers. 
A delay between the 2014 announcement and imposition of sanctions created a window to send U.S. dollars to Russia. On March 16, 2014, President Obama signed an executive order authorizing the Treasury Department to impose sanctions on individuals for Russia’s annexation of Crimea. But the Treasury Department did not name the specific individuals sanctioned under the executive order until March 20, 2014. During this four-day window, Rotenberg-linked shell companies transferred over $120 million through the United States to Russia. On March 18, 2014, Highland Ventures transferred over $39.5 million from its account at The Pictet Group in Switzerland through the U.S. financial system to its account at Gazprombank in Moscow. That same day, Culloden Properties transferred over $82 million from its Pictet Group account in Switzerland through the U.S. financial system to its account in Moscow at the Gazprombank. Both the Panama Papers and documents produced to a financial institution by the Markom Group—and subsequently provided to the Subcommittee—identify Boris Rotenberg as the owner of Culloden Properties. Rotenberg-linked shell companies transacted in U.S. dollars post-sanctions. Shell companies linked to the Rotenbergs continued conducting transactions through the U.S. financial system even after the imposition of sanctions in March 2014. For example, including its art purchases, Highland Ventures was involved in transactions worth over $16 million. Advantage Alliance was involved in transactions worth over $29 million. And while the UBO of Steamort remains hidden, the company served as an intermediary between Rotenberg-linked shell companies and BALTZER in the purchase of art. Following the imposition of sanctions in March 2014, Steamort was a part of transactions totaling over $22 million. In total, the Subcommittee identified over $91 million in transactions by Rotenberg-linked shell companies after sanctions were imposed on the Rotenbergs in March 2014. 
The Subcommittee’s Investigation 
The Subcommittee initiated its investigation after reviewing a number of suspicious transactions that appeared to involve art purchased through auction houses and private dealers. Funds used in these transactions originated at entities linked to the Rotenberg family through the Panama Papers and other public information. As part of its investigation, the Subcommittee reviewed over one million documents from the four major auction houses, a private art dealer, an independent public gallery, and seven financial institutions. The Subcommittee interviewed current and former employees of Sotheby’s, Christie’s, Phillips, and Bonhams. The Subcommittee also interviewed a private dealer based in New York and two art advisors located overseas who were all involved in the same multi- million dollar transaction. All entities cooperated with the Subcommittee’s requests, except for Mr. Baltser. Through his attorney, Mr. Baltser declined to be interviewed by the Subcommittee and stated he was in Moscow with no plans to return to the United States. 
Findings of Fact 
(1) The art market is the largest legal, unregulated market in the United States. The art industry is not subject to the BSA and is not required under U.S. law to maintain AML and anti-terrorism financing controls for transactions. However, all U.S. persons and entities are prohibited from transacting with sanctioned individuals or entities as determined by the U.S. Treasury Department Office of Foreign Asset Control (“OFAC”).  
(2) Sotheby’s, Christie’s, Phillips, and Bonhams all have voluntary AML controls in place. Despite no legal requirement to do so, the four auction houses reviewed by the Subcommittee had established voluntary AML policies. 
(3) Private art dealers are not subject to AML requirements. One private dealer told the Subcommittee she had no written AML or sanctions policies and instead relied on her gut and worked with people she knew. She also explained that questioning the identity of the buyer and the source of funds in an art transaction was not done in the art industry, nor would the dealer for the purchaser want to provide that information. During an interview with the Subcommittee, she explained that her practices have changed over the years and that she relies on the advice of lawyers with expertise in AML and related areas and looks for potential red flags in transactions. 
(4) The auction houses treated an art agent or dealer as the principal purchaser of art, even if they had reason to believe they were working with an undisclosed client. This practice enables the auction house to perform due diligence on the art agent or dealer instead of identifying and evaluating the undisclosed client, creating a significant AML vulnerability. 
(5) The United States sanctioned members of the Rotenberg family in March 2014. On March 16, 2014, President Obama signed Executive Order 13661 imposing sanctions on Russia due to its annexation of Crimea. Arkady and Boris Rotenberg were among the Russian citizens specifically sanctioned on March 20, 2014 due to their close ties to Russian President Vladimir Putin, which included awards of large government contracts to companies they owned. The U.S. Treasury Department later sanctioned specific Rotenberg-owned companies (on April 28, 2014), Boris Rotenberg’s son Roman (on June 30, 2016), and Arkady Rotenberg’s son Igor (on April 6, 2018). Both Roman and Igor Rotenberg were sanctioned due to their financial ties to their sanctioned fathers. 
(6) Information released in 2016 from the law firm of Mossack Fonseca— known as the “Panama Papers”—linked the Rotenbergs to certain shell companies involved in high-value art purchases reviewed by the Subcommittee. The Panama Papers included an email chain made public that listed nine shell companies in the British Virgin Islands (“BVI”) linked to Arkady, Boris, and Igor Rotenberg. That email copied attorney Mark Omelnitski and identified Igor Rotenberg as the UBO for Highland Business Group Limited (“Highland Business”) and Boris Rotenberg as the UBO for Highland Ventures Group Limited (“Highland Ventures”). 
(7) Mark Omelnitski is a London-based attorney linked to the Rotenbergs and art advisor Gregory Baltser. Mr. Omelnitski—through his company the Markom Group—assisted the Rotenbergs in establishing and maintaining shell companies. He also assisted art advisor Gregory Baltser in establishing his art agency BALTZER in Moscow. In discussions Mr. Baltser had with Sotheby’s and Christie’s about partnering with BALTZER, Mr. Omelnitski served as Mr. Baltser’s attorney and also represented that he administered Mr. Baltser’s AML and sanctions policies as his Money Laundering Reporting Officer. 
(8) Both Highland Business and Highland Ventures received funding from companies linked to Arkady Rotenberg. Highland Business received funding from Advantage Alliance Ltd (“Advantage Alliance”), which an internal bank investigation linked to Arkady Rotenberg. Highland Ventures received over $124 million in funding from Milasi Engineering Limited. The 2014 Milasi Engineering Financial Statement listed Arkady Rotenberg as the ultimate beneficial owner of the company. 
(9) Gregory Baltser is a Moscow-based art advisor who facilitated art purchases linked to the Rotenbergs. Mr. Baltser purchased art in the United States with funds that the Subcommittee traced back to Highland Business and Highland Ventures. 
(10) Prior to the implementation of sanctions on Arkady and Boris Rotenberg in March 2014, the funds Mr. Baltser used to purchase certain art followed a pattern: Highland Business wired the funds to purchase the art to Steamort Ltd (“Steamort”). Steamort then wired the funds from its bank account in Estonia to the auction house and took title to the art. Steamort’s UBO remains unknown. In 2012, Christie’s questioned Mr. Baltser about the ownership of Steamort, a company formed in Belize. Mr. Baltser told Christie’s that he did not know and could not provide the name of the owner. After Christie’s threatened Mr. Baltser would be unable to bid at an auction unless he identified the owner of Steamort, Mr. Baltser told Christie’s the UBO for Steamort was “Luisa Brown.” Christie’s accepted this verbal assertion and cleared Mr. Baltser to bid in the auction. Mr. Baltser never provided any documentation that Luisa Brown was the owner of Steamort. The Subcommittee was unable to determine Steamort’s UBO or if Ms. Brown existed. (11) Mr. Baltser established a private art agency and club called BALTZER in Moscow in 2013. Mr. Baltser used BALTZER to take title and purchase art for his clients. This change altered the payment pattern outlined above to include BALTZER as the entity paying the auction houses for art Mr. Baltser purchased. In addition, following the imposition of sanctions on the Rotenbergs by the United States in March 2014, the Subcommittee traced funds to purchase art to Highland Ventures. Highland Ventures would wire the funds to Steamort; Steamort would wire the funds to BALTZER; and BALTZER would wire funds to the auction house and take title for the art. 
(12) Christie’s partnered with BALTZER, including allowing Mr. Omelnitski to conduct AML and sanctions checks on BALTZER clients. In the February 2014 agreement, Christie’s agreed to rely on BALTZER to conduct due diligence on clients and provide an annual AML compliance certification. Mr. Omelnitski provided the first AML compliance report in December 2014 and confirmed that no BALTZER clients were sanctioned. Mr. Omelnitski did not provide another report until October 2017 when he emailed the amounts associated with BALTZER art purchases; he provided no certification regarding AML or sanctions compliance. Christie’s renegotiated the agreement in March 2018 and required BALTZER to produce customer due diligence to the auction house within 10 days of every purchase. 
(13) Sotheby’s declined the BALTZER proposal, but continued business as usual. While considering Mr. Baltser’s proposal, the Sotheby’s Baltser Account Representative told Sotheby’s management that Mr. Baltser’s clients included Russian oligarchs, specifically Arkady and Boris Rotenberg before they were sanctioned by the United States. During her Subcommittee interview, the Baltser Account Representative stated that Mr. Baltser never told her that Arkady and Boris Rotenberg were his clients. Instead, she fabricated this information to convince Sotheby’s to agree to the proposal with BALTZER. 
(14) Despite having voluntary AML and sanctions policies, auction houses failed to ask basic questions of Mr. Baltser, including for whom he purchased art. This allowed Mr. Baltser to continue to purchase art despite the imposition of sanctions by the United States on the Rotenbergs, completely undermining any action taken by the auction houses to block transactions by sanctioned individuals. 
(15) Mr. Baltser purchased over $18 million in art from May to November 2014 using funds traced to Rotenberg-linked shell companies. These transactions included a $7.5 million private sale of René Magritte’s La Poitrine in which Highland Ventures took title to the painting, and Advantage Alliance wired the purchasing funds. The Subcommittee traced those funds from Advantage Alliance to Senton Holdings Ltd, which one financial institution determined was owned by Arkady Rotenberg. An employee of Mr. Omelnitski’s signed the contract for sale on behalf of Highland Ventures. Mr. Baltser sought to sell art linked to the Rotenbergs. In August 2015, an employee of BALTZER sent Christie’s a list of 31 artworks, including 16 works the Subcommittee linked to Rotenberg-related shell companies. The BALTZER employee indicated the 31 pieces belonged to the same collection and sought “any opportunities to promote these works or to make this collection more valuable.” That list included René Magritte’s La Poitrine and Lyonel Feininger’s Brucke II. 
(16) In February 2019, Sotheby’s and Christie’s competed to sell Lyonel Feininger’s Brucke II; Sotheby’s was selected to sell the painting. Both Sotheby’s and Christie’s provided enhanced deal terms to sell the painting at auction. Sotheby’s planned to auction the painting at its February 26, 2019 Modern Art Evening Sale in London. The painting was estimated to sell for between £4 million and £6 million. Prior to the auction, however, Sotheby’s stated it pulled the painting due to a lack of interest. 
(17) When questioned by Sotheby’s, Mr. Baltser declined to provide the February 2014 purchaser of Brucke II. Brucke II was purchased by Mr. Baltser on behalf of an undisclosed client at a Christie’s auction in February 2014. In April 2019, Sotheby’s asked Mr. Baltser to reveal the name of the February 2014 buyer. Mr. Baltser did not reveal the name, but asserted the February 2014 buyer no longer owned Brucke II. Instead, he stated the painting now belonged to a Marshall Islands company and provided a Russian passport for the UBO of the company. 
(18) During the course of the Subcommittee’s investigation, Sotheby’s, Christie’s, and Phillips stopped transacting with Mr. Baltser and BALTZER. Rotenberg-linked companies continued to move at least $91 million through the U.S. financial system following the imposition of U.S. sanctions in March 2014. The Subcommittee determined that companies linked to the Rotenbergs continued to have access to the U.S. dollar and the U.S. financial system despite the imposition of sanctions against Arkady and Boris Rotenberg. 
Recommendations 
(1) Congress should amend the Bank Secrecy Act to add businesses handling transactions involving high-value art. The art industry is currently not subject to AML requirements under the BSA. The European Union recently required businesses handling art transactions valued at €10,000 or more to comply with AML laws, including verification of the identity of the seller, buyer, and UBO of the art. 
(2) Congress should require the Treasury Department to collect beneficial ownership information for companies formed or registered to do business in the United States. This information should be available to law enforcement for investigatory purposes. Beneficial owner information maintained by the Treasury Department should include appropriate privacy and security protections. 
(3) When imposing sanctions on an individual, the Treasury Department should consider routinely imposing sanctions on the individual’s immediate family members. While the U.S. sanctioned Arkady and Boris Rotenberg in March 2014, for example, it did not sanction the brothers’ children until later dates. The Treasury Department stated it imposed sanctions on Igor Rotenberg in 2018 because he “acquired significant assets from his father, Arkady Rotenberg, after OFAC designated [Arkady] in March 2014.” This allowed Arkady and Boris Rotenberg to evade U.S. sanctions by transferring their interests in companies to their children while maintaining operational control. 
(4) The Treasury Department should implement and announce sanctions concurrently. While President Obama announced sanctions for Russia’s annexation of Crimea on March 16, 2014, the Treasury Department did not officially impose sanctions on specific individuals and entities until March 20, 2014. During this four-day window, millions of dollars were transferred through the United States and back to Russia. The Treasury Department should take necessary actions to both announce and implement sanctions to avoid creating a window of opportunity for individuals to evade sanctions. 
(5) The Treasury Department should lower or remove the ownership threshold for blocking companies owned by sanctioned individuals. According to guidance by the Treasury Department, a company is blocked if it is majority owned by a sanctioned individual. If the sanctioned individual has a minority ownership in a company, that company is not blocked, even if the sanctioned individual owns 49 percent of the company. 
(6) The Treasury Department should maximize its use of suspicious activity reports (“SARs”) filed by financial institutions. Under the BSA, financial institutions are required to file SARs with the Treasury Department’s Financial Crimes Enforcement Network. These reports document financial transactions that appear to involve money laundering or terrorist financing, among other illicit activities. The Treasury Department should more effectively mine SARs for information related to Specially Designated Nationals and add these entities to the Specially Designated Nationals and Blocked Persons List or alert other financial institutions of the risks of transacting with the entities. This would increase the effectiveness of imposing sanctions. 
(7) OFAC should issue comprehensive guidance on the steps auction houses and art dealers should take to ensure they are not doing business with sanctioned individuals or entities. That guidance should clarify what steps auction houses and art dealers should take to determine whether a person is the principal seller or purchaser of art or is acting on behalf of an undisclosed client, and which person should be subject to a due diligence review. 
(8) OFAC should issue guidance interpreting the informational exception to the International Emergency Economic Powers Act related to “artworks.” That guidance should interpret the artworks exception narrowly to encompass matters with informational content, while excluding typical works of art such as paintings, etchings, and sculpture.