Data Inclusivity: Creating Win-Win Digital Opportunities
[This whitepaper was also presented to the Pacific Basin Economic Council November 20, 2017. It's republished here with pictures.]
Free trade mattered in the 20th century which fostered multilateral and bilateral efforts in reducing or eliminating trade barriers in the form of trade subsidies, dumping, tariffs and import quotas. As a result of these efforts, millions of consumers have been freed from the chains of poverty over the past 30 years. Free import-export trade practices certainly mattered to these people. As global economies head into the 21st century, trade barriers are rearing their heads in new forms blocking the creation of new value driven business models and ideas in the digital import-export trade economy.
Emerging technologies like artificial intelligence (“AI”), smart cities, self-driving vehicles, fintech, cloud computing, 5G and robotics have been disrupting traditional business models and will shape the wealth of nations for the next 100 years. The key to success underlying all of these new technologies is data which has become the gold standard for defining 21st century corporate success. Global data flows grew 45 times from 2005 to 2014, faster than international trade or financial flows (see page 2 here). Fully 50% of all global trade in services now depends on access to cross-border data flows (page 16 here). We are in an age in which data has become more valuable than oil. Data has become the new asset class.
Regardless of the technological sophistication of nations or private enterprises, in one way or another, the data they are generating through the use of smart devices, mobile computing and high-speed internet connectivity is increasingly becoming a source of trade-related conflict and tensions throughout the world economies. This is because efforts to address trade barriers to the free flow of data have been lagging behind the complexities of the digital economy.
Part I of this whitepaper explores the major trade barriers, such as data localization and AI subsidies, that are or will endanger the successful mass adoption of emerging technologies on a free trade basis.
Part II discusses the opportunities created by data-driven enterprises whose success depend on the free flow of data.
Part III discusses the importance of harmonizing national laws related to the free flow of data so as to create a level playing field of market players competing on the principle of free trade.
Part IV advocates the urgent need to create an international or regional NGO (or leverage the resources of an existing NGO like PBEC) to serve as a neutral platform to connect, host, and facilitate discussions among private enterprises with national governments so as to lower and eliminate barriers to the free flow of data.
Part I: Reactionary Trade Barriers
Some nations and private enterprises have been benefiting from the new opportunities created by emerging technologies. However, most nations are being excluded from such opportunities due to a variety of structural reasons such as low capital investments in technology or inexperienced political leadership. When nations see that their citizens are being excluded from participating meaningfully in data-driven opportunities, they fall victim to protectionists policies that have plagued 20th century trade. Instead of enacting policies that restrict the flow of physical goods, the trade barriers of the 21st century restrict the free flow of data (invisible intangibles) across national borders. For example, the following trade barriers have already or will become increasingly pervasive region-wide. Their tightening grip on the global free flow of data will endanger the creation of wealth, increased living standards and longevity usually associated with free trade.
A) Data Localization
Data localization undermines the free flow of data because it imposes artificial barriers on the conduct of data intensive industries like cloud computing and fintech. (See in general for this discussion: Data localization laws: trade barriers or legitimate responses to cybersecurity risks, or both?, John Selby, Oxford International Journal of Law and Information Technology, 2017, 00, 1–20). Governments have been passing data localization laws that restrict the transfer of various types of personal information over national borders. U.S. industry representatives have noted that data localization measures pose “a significant problem for U.S. firms doing business across borders, due to the importance of free flowing data for digital trade (page 16 here).”
According to a private survey in 2017, about 120 countries have enacted some form of data protection laws that have the potential of restricting the free flow of data.
There are two major motivations for adopting these laws. The first is security related: the domestic government is keen to prevent its citizens from being the target of foreign surveillance. The second is more protectionist in nature: local governments want to protect the internet industries within its respective domestic economies.
There are two kinds of data localization. Many governments have passed laws that require global internet firms to store local client data on servers physically located within the domestic borders of the relevant government. This is known as “localized data hosting”. The other type is known as “localized data routing” which requires internet service providers to route data packets required by local Internet users across networks located only in that jurisdiction. Both forms of data localization threatens to undercut innovations in Big Data and would render the use of the general Internet slower and less convenient.
For example, due to tightening rules on foreign data and cloud services implemented by a regional major power such as implementing new surveillance measures and increasing scrutiny of cross-border data transfers, a large e-commerce firm is selling off its hardware from its public cloud business in such jurisdiction. This is because data localization rules prevents data companies from pooling user and consumer data in cloud storage databases across the world. Forcing private enterprises to store data on local servers adds to increased business cost and compliance burdens. In June 2017, a major economic power has asked its biggest trading partner not to enforce its new cybersecurity law alleging that it could damage global trade by requiring local and overseas firms to submit to security checks and store user data within the country.
The U.S. International Trade Commission (“USITC”) was asked by the U.S. Trade Representative in January 2017 to conduct three investigations to examine the impact of barriers to digital trade (defined as commerce in products and services delivered via the Internet) on the competitiveness of U.S. firms in international markets. The USITC issued its report for the first of these investigations in August 2017 entitled Global Digital Trade I: Market Opportunities and Key Foreign Trade Restrictions which examined regulatory measures such as data localization that impede free digital trade. According to this report, data localization laws negatively affect the commercial development of internet communication services, cloud-based data processing, e-commerce and the internet of things. (See page 278 of Report here).
B) AI Related Subsidies
The world’s top economies have been spending time, money and resources in researching and developing commercial uses of robotics and the AI software that would help automate industries. Both the private and public sectors are involved in these initiatives. The three countries leading the race in robotics and AI research are the U.S., China and India. Unlike the U.S. and India however, the biggest sponsor of research and development in China is its government.
China in particular is set to make huge advances in automated manufacturing technology. In July 2017, China’s State Council announced its intention of becoming the world leader in AI by 2030 by earmarking over US$150 billion for its domestic AI industry. Almost three years ago, China announced its “Made in China 2025” initiative to reboot its manufacturing sector to focus on smart collaborative robots powered by AI systems. These efforts are backed by government incentives and investments. The China Machinery Industry Foundation noted that the country is set to expand sales of domestically produced industrial robots to 100,000 by 2020. It is estimated that there are over 800 start-up robotics companies in China that are nurtured by government incentives.
To what extent will government incentives and investments effectively amount to trade subsidies in the 21st century market place where machines, robots and AI systems are set to replace or augment human labor? In the robotics and AI sphere, controversial public subsidies drastically exacerbate the human-replacement effects that these two technologies are expected to create.
1) Effects of Publicly-Funded Subsidies & Incentives
Government support of any domestic industry has two major effects. First, it reduces the cost of doing business for the industry being assisted by government incentives such as favorable local tax breaks, land lease arrangements, interest-free loans, grants and cash investments. For example, public incentives in robotics would reduce the capital expenditure costs for building factories in which robots will work. Second, these incentives cause goods produced by such assisted industry to be cheaper than in foreign markets. The economic effect of these public assistance is a lower domestic price for the goods produced by the assisted industry. Controversial subsidies therefore raise the price of foreign goods and services compared to domestic ones that in turn reduce the domestic economy’s reliance on imports.
2) Subsidized Robotics & AI Compounds Harms
Special concerns are implicated when public subsidies are applied to robotics and AI systems technologies to incorporate them into domestic manufacturing operations. The ensuing effect compounds the problems caused by subsidies (as described above) and the human-labor replacement issues presented by robotics and AI systems.
First, the point of heavy public investments into robotics is to make better and faster robots that can work cheaper than human workers. Imagine a factory fully automated by robots and AI systems subsidized by public incentives. Public subsidies helped reduce the cost of building this factory and researching and developing the robotics and AI systems used to automate it. A factory that needs significantly fewer workers enjoys another competitive advantage: cheaper labor cost because robots do not need retirement, medical and social insurance payments nor holiday bonuses, leave times and sick days. For example, large e-commerce warehouses invests in machines that are able to lift and stack boxes 24 hours a day with just a 30 minute break to re-charge, and have reduced manpower by 80 percent. A subsidized smart factory which enjoys cheaper labor costs would create even cheaper goods than in foreign markets and significantly raise the price of foreign goods within the domestic market. This is the classic harm which trade policies seek to redress.
Second, there have been many recent studies on the human-replacement effects of advanced robotics and AI manufacturing systems. According to a study recently released in October 2017, Redwood Software and Sapio Research found that IT leaders believe automation could impact 60% of businesses by 2022. The accounting firm PwC also released a recent study which showed that workers may need to practice and learn new skills or face redundancy in the age of manufacturing robotics. PwC’s report surveyed 10,000 people across China, India, Germany, the UK, and the U.S. and nearly 37% believe AI and robotics will endanger their jobs.
The human-replacement effects of robotics are particularly acute in the manufacturing sector of emerging economies. Why? Economists believe that emerging countries like India need to create a wide manufacturing base in order to lift millions of their respective citizens out of poverty (or keep them from falling into poverty). But they cannot do so if manufacturing jobs will be reduced by domestic robotics investments and cheap imports made by overseas automated factories. In India, automation is taking a toll on both unskilled and skilled workers. For example, automation will reduce 30 percent of India’s coveted IT workforce. An emerging country like India simply cannot afford to see large swathes of its population (many of whom are subsisting near the poverty line) be rendered redundant by emerging technologies adopted both domestically and internationally. For example, a commentator noted:
“[i]f ever a country faced the prospect of a dystopian future marked by gangs of unemployed youth in the millions wandering about creating mayhem, pillaging and plundering for lack of anything else to do, it's India -- unless it is able and willing to do something about it very quickly and efficiently.”
At the current rate of research and development of advanced manufacturing technology driven by robotics and AI software systems, many emerging nations will face the same social issues encountered in India. This will be a very negative development for free trade. How so? It is axiomatic that a nation facing severe domestic social problems will enact protectionist trade policies of the type that followed the great depression of the early 1930s. Enacting protectionist trade barriers may be short-sighted, but governments often are not rational when confronted with mounting social challenges caused by poverty, that more often times than not, have led to terrible bloodshed, revolutions and societal chaos. A region plagued with trade barriers will create a lose-lose situation for all.
C) Intellectual Property Laws Not Barrier to Free Flow of Ideas
Every commercial economy grants patent rights to novel and useful inventions. However, the laws of these nations usually do not allow naturally occurring phenomenons, mathematical formulas and most importantly, genetically modified humans to be patented (see U.S. Patent & Trademark Office, Manual of Patent Examining Procedure, Ninth Edition, Revision 07.2015, section 2105.) Fundamental subject matter like the discovery of a new mineral or human beings should not be patented because these concepts represent the “basic tools of scientific and technological work”. (See U.S. Supreme Court decision Mayo Collaborative Servs. v. Prometheus Labs., Inc., 566 U.S. 66, 71, 132 S.Ct. 1289, 182 L.Ed.2d 321 (2012)). If fundamental concepts are patented then the public no longer has the right to use them for research and academic purposes without paying a license to the patent owner. The patent ineligibility of these concepts help ensure the free flow of academic and scientific ideas which is just as important as preserving the free flow of world trade in goods and services.
The threat now is that modern medical advances in genetically modified human embryos could be patented in nations that do not share such views on subject matters that are patent ineligible. If such nations grant patent rights on inventions related to human beings then any further medical research into humans that involves using such inventions could only be done by paying license fees to the corresponding patent owner. This would undercut the growth of further medical breakthroughs and decrease global living standards. Academic ideas, like international trade, should flow freely.
D) Privacy Legal Regimes New Trade Barrier?
More and more nations are adopting comprehensive privacy protection laws. For example, as of May 2017, there are over 120 different national data privacy laws with another 31 countries proposing to adopt them. The European Union has developed a comprehensive approach to data protection and privacy through its General Data Protection Regulation (“GDPR”), which will be implemented in May 2018 by individual EU members. The GDPR views privacy as a "fundamental right". It is therefore constitutional in nature on par with a human right. Essentially, the GDPR provides three major rights to an individual to protect his/her privacy: (i) the right to consent (see Article 6), the right to be forgotten (Article 17) and right to access (Article 15). The U.S. Chamber of Commerce views the implementation of GDPR as an “immense regulatory burden” that affects U.S. businesses operating in the EU, as well as broader EU competitiveness in the digital economy (see footnote 1055, page 379 here).
But to what extent are privacy protection laws a barrier to free trade? According to the World Trade Organization’s General Agreement on Trade in Services (“GATS”), cross-border trade restrictions are expressly permitted if they enable:
“the protection of the privacy of individuals in relation to the processing and dissemination of personal data and the protection of confidentiality of individual records and accounts” (see Article XIV(c)(ii), General Exceptions).
However GATS also expressly prohibit trade restrictive privacy measures if they are:
“applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where like conditions prevail, or a disguised restriction on trade in services” (see Article XIV, General Exceptions).
A good faith argument may be made that data protection laws like the GDPR are unjustifiably discriminatory because they create economic barriers (often insurmountable) to entry for developing countries. In its 2016 report on data protection regulations and international data flows, the United Nations Conference on Trade and Development (“UNCTAD”) noted that 30% of countries do not have any data protection policies in place. UNCTAD also noted that many developing countries are having problems adopting and/or implementing privacy laws because they lack the time, resources and understanding to do so (see page xii here). Companies doing business in developing countries that are unable to adopt or implement effective privacy laws will be at a competitive disadvantage compared to their foreign peers. The ensuing lack of clarity causes business uncertainty that can reduce investments.
According to the USITC, U.S. industry representatives have reported that privacy laws may impede digital trade (page 13 here). Many U.S. companies and trade associations are particularly concerned about the GDPR since U.S. firms’ do extensive trading with the EU. These concerns are shared by many Asian companies as well. Critics of the GDPR allege that it places “significant restrictions on the data and is expected to raise the cost of data storage and processing” (see page 16 here).
Part II: Potential Data Opportunities Threatened
No one can doubt that AI, machine learning and smart technologies are creating new business opportunities in a world that is increasingly becoming more interconnected digitally. The World Bank estimates that Internet users tripled since 2005 to 3.2 billion in 2015, representing 60% of people globally. The Organization for Economic Cooperation and Development (OECD) reports that in 2014, on average 95% of enterprises in OECD countries had a broadband connection and 76% had a website or homepage. For example in Asia, the rise of mobile devices has created a market with a combined population of over 560 million in Singapore, Malaysia, Indonesia, Vietnam, Thailand, Philippines, and Hong Kong. As more people throughout the world are becoming digitally connected, the data they generate while online will fuel the rise of data-driven business models.
The think-tank Gartner Inc. issued a report noting that the following emerging technologies have the most potential to create new business models in an increasingly interconnected world. First, transparently immersive technologies enable more human-centric experiences that would increase transparency amongst people, business and things. Second, growing computational power and expansive data sets allow the growth of smart machines that are able to adapt to changing commercial conditions. Third, platform technologies will evolve new business models that will link human–technology experiences.
The various trade barriers of the 21st century (some of which are discussed in Part I above) threaten the creation of new business models in digital trade because they impede the flow of data. If these trade barriers are not addressed, they risk holding back overall global growth and development.
Part III: Harmonizing Regional Law & Practice on Data Flow
The treatment of data relevant law and practice, such as on data localization and privacy, is disparate and varied. Therefore, it is important for businesses and forward-thinking government leaders to work toward the harmonization of national laws related to the free flow of data so as to create a level playing field of market players competing on the principle of free trade.
The lack of compatibility of data relevant law and practice amongst nations increase business uncertainty and decrease investments. It also undercuts the adoption and proliferation of emerging technological developments and reduces potential accompanying societal benefits such as job creation.
In its 2016 report (page 8) on data protection regulations and international data flows, the United Nations Conference on Trade and Development identified some areas where coordinated action is particularly needed:
- addressing new technologies;
- managing cross-border data transfers; and
- managing the compliance burden.
These areas are particularly relevant to efforts to reduce or eliminate trade barriers that inhibit the free flow of data.
Part IV: Neutral NGO Platform: Bridging Businesses & Governments
The trade barriers discussed in Part I threaten to undercut the development of new business models and value (as discussed in Part II) in the digital trade economy of the 21st century. Without effective efforts to harmonize disparate national treatment of data relevant law and practice, the flow of data would remain interrupted and impeded.
To bring urgent global attention to the problem of trade barriers to the free flow of data in the complex digital trade economy, this Whitepaper advocates creating an international or region non-government organization (“NGO”) or leveraging the resources of an existing NGO (like the Pacific Basin Economic Council) to serve as a neutral platform to connect, host, and facilitate discussions among private enterprises with national governments so as to lower and eliminate barriers to the free flow of data.
The NGO-led neutral platform benefits both private businesses and national governments and creates win-win opportunities for both.
Many companies have departments that handle government relations and which can potentially be used to address these issues. However governments are usually hesitant to adopt legal changes in response to the lobbying efforts of a single company. This is a matter of perception since governments may be accused of favoritism, or worse, if they are perceived to be changing law and practice to disproportionately benefit the privileged corporate few. It is also inefficient to approach governments on an individual basis as this wastes corporate resources with no guarantee of success. Instead, businesses are better served if they unite and negotiate with governments under the auspices of a single NGO such as PBEC that is able to attract public leaders under a neutral forum to discuss digital trade barriers on a multinational basis.
In return for digital trade-friendly law and practices, private companies may provide investments in worker retraining programs that will help reallocate employees displaced by the vicissitudes of robotic manufacturing as well as other incentives. It has been reported that the academic foundations of emerging countries’ colleges and universities are poorly funded with outdated curriculum and under-paid, sub-standard teachers. Private businesses, especially those in the technology industry, have the resources to donate or provide state of the art training (for example in computer programming languages) in the native tongue of the domestic host country. Private companies can help ensure the relevancy of a developing economies’ citizenry in a digital age rife with labor and other social changes. This in turn leads to increased national stability and prosperity which reduce the likelihood of the host country adopting barriers to digital trade.
As an example of the potential for positive social changes that large corporations may accomplish, the Gates Foundation has just announced an open-source software to create interoperable payments platforms that connect all digital financial providers and customers. It is hoped that this new platform will expand financial inclusion to those whom Mr. Gates call the "unbanked" and help them gain access to financial services hitherto unavailable. The platform is dubbed "Mojaloop" inspired from the Swahili word for “one”. The Gates Foundation was able to accomplish this magnificent feat on its own. Imagine the nature and scope of positive social changes that a resourceful NGO with an agenda to connect businesses and governments to address data trade barriers can achieve.
Conclusion: Data Inclusiveness, a Win-Win Opportunity
Commentators discuss the term “singularity” as some moment in the future when advances in AI intelligence make possible exponential technological advances, resulting in massive changes to human civilization. But why wait for some indefinite future to make a difference when most businesses have the technologies now to help create positive changes for developing economies. The rise of the modern internet, mobile connectivity, food sciences and cheap energy can potentially enable emerging countries to benefit from emerging technologies. Making the world data inclusive prevents developing countries from adopting trade barriers to the free flow of data crucial to digital trade. Helping developing nations benefit from opportunities created by data will create ripples of positive knock-on effects for the global economy. Under the auspices of a NGO, private businesses and governments can address trade barriers to the free flow of data in a neutral forum for the mutual benefit of an inclusive data-driven economy.