By JIM BALSILLIE
Saturday, September 21, 2019
Jim Balsillie is a former chairman and co-CEO of Research In Motion, where he commercialized Canadian intellectual property in more than 150 countries.
It's impossible to pinpoint exactly when Canada's policy makers started positioning the country as a cheap labour pool for multibillion-dollar foreign tech companies, but we know when it went mainstream: Amazon HQ2.
In September, 2017, Seattle-based Amazon launched a continentwide search for a city to host its second headquarters. From Wilmington, Del., to Winnipeg to Woonsocket, R.I., towns, cities and regions from across North America - a staggering 238 to be exact, though unsurprising given the stakes - put forward proposals outlining why they should become the e-commerce giant's new home.
While competing U.S. cities offered tax breaks, infrastructure improvements and expedited construction approvals, Ed Clark, Canada's official representative, openly boasted that "our competitive advantage" is "software programmers that cost 34 per cent to 38 per cent less than in U.S.," vowing "that's an edge the government is determined not only to maintain but to sharpen." In the end, 20 finalists were announced, with Toronto being the lone - and ultimately unsuccessful - Canadian contender.
Two years later, this goal to actively drive down wages still lives in a dubious inventory of taxpayer-funded marketing materials that repeatedly highlight the cheap salaries of "high-quality" Canadian tech talent and the substantial tax incentives or taxpayer subsidies that await those who open tech branch plants in Canada.
These misguided strategies put our tech workers in a global wage race to the bottom, competing on cost with the salaries in Poland, Ukraine and India. They are just daring our top talent to leave the country - and research shows that 66 per cent of our software engineering graduates have already left.
Frequent talk of innovation by the federal government over the past four years did not result in a national innovation strategy - a road map of policies to see Canada prosper in knowledge-based and data-driven economies. An ad hoc set of programs was introduced to appease the domestic tech sector, but a strategic approach to commercializing Canadian intellectual property (IP) and data were pushed aside in favour of a feverish pursuit of jobs, establishing Canada as a prominent hub of highly skilled but cheap tech branch plants.
The Trudeau government is not the first to misunderstand what an innovation strategy actually entails. The previous Conservative government under Stephen Harper confused a science and technology strategy with an innovation strategy, as they principally focused on building research capacity at universities by increasing their research and development funding.
Both governments inherited a bureaucracy that missed the shift from a 20th-century production-based economy of manufacturing tangible goods to the 21stcentury knowledge-driven economy based on generating, owning and controlling intangible assets, namely IP and data. Yet our current government decided to keep the outdated approaches, opting not to bring a single innovation-policy expert or even an economist with an innovation background inside its brain trust. They simply assumed the innovation economy is a continuation of the traditional production economy and that the aggressive use of a traditional policy tool called foreign direct investment (FDI) applied to the tech sector would manifest the same spillovers it does in the traditional economy.
Not only are the outcomes different, but projects such as Sidewalk Toronto - Canada's highest-profile FDI project to date - are currently demonstrating that outdated thinking could have devastating, colonial-style effects on our prosperity, security and sovereignty. But it's not too late to change course. With the federal election campaign now under way, and with candidates outlining how they will advance Canada's economy and wellbeing, there is no better time to revisit failed approaches and find a better way to support our world-class innovators.
THE NEW ECONOMY IS NOT LIKE THE OLD ECONOMY The knowledge-based and datadriven economies of intangibles are fundamentally different from the traditional economy of tangible goods in many critical ways.
In the production-based industrial economy of tangibles, the ability to produce goods efficiently at scale and sell them at a lower price than competitors enables the capture of markets, which underpins profitability. In the traditional economy, trade agreements open foreign markets so producers can gain greater access to production economies of scale. Competition naturally emerges from the removal of restrictions on commerce, as companies compete by lowering production costs. Efficiency is based on the optimization of supply chains on a global scale.
The win-win dynamics of global trade spreads the benefits.
Foreign direct investment takes the form of a business in one country possessing controlling ownership of what is commonly known as a "foreign branch plant" in another. Traditional foreign branch plants provided Canada with capital, advanced infrastructure, thousands of good jobs, valuable knowledge and skills development - and, critically, the creation of local supply chains.
Economists generally quantify the additional positive economic gains from these foreign branch plants as three times the actual FDI investment. For example, if a car or oil company spent $100-million on a major plant upgrade, the overall economic benefit to Canada would be around $300-million. These branch plants were tides that rose all boats and they were an essential part of growing Canada's prosperity throughout the 20th century.
The most valuable S&P companies in 1976 included U.S. Steel and Bethlehem Steel, Sears and Woolworth's, as well as American Can and American Tobacco.
According to IP merchant bank Ocean Tomo, virtually all their value was in tangible assets.
But in the 1980s, the production economy started to be eclipsed by the knowledgebased economy of intangibles, where the strategic focus of companies shifted to generating and controlling IP assets. As a result, the five most valuable companies in the world today - Apple, Amazon, Alphabet (Google's parent company), Microsoft and Facebook - have 95 per cent of their total value in intangible assets.
Over the past 10 years, the intangibles economy expanded to include the data-driven economy, where the strategic focus turned to generating, exploiting and controlling data assets, such as the application of machine learning to produce artificial intelligence as well as generating valuable IP developed on top of data assets.
The knowledge-based and data-driven economies are not a continuation of the traditional, production-based economy.
Foundationally different in structure, the intangible economies feature unprecedented economies of both scale, scope and network externalities. Utilizing highly centralized executive structures, a company's objective has turned to generating as much IP and data assets as possible and deploying these assets to extract economic rents.
Critically, those two types of intangible assets can be scaled globally with near-zero marginal production costs, which results in "winner take all" economics.
The dominant effect of FDI in the innovation sector is not to create new jobs but to repatriate the top IP, data and talent in a host country and reduce Canada to what economist Dan Ciuriak calls the "mediocre middle."
"EMBRACE, EXTEND AND EXTINGUISH" The impact of big tech on smaller firms is the stuff of tech sector legend, starting with Microsoft's strategy to "embrace, extend and extinguish" smaller companies.
Sometimes small companies are acquired for the complementary technology, but often they are bought to shut them down and pre-empt possible competition.
"Today's giants are much more ruthless and introspective. They will eat their own children to live another day," a prominent venture capitalist Matt Ocko told The Economist last year. Instead of growing local technology ecosystems, they access highly skilled talent, intimidate startups into terms of a sale, threatening to put young entrepreneurs out of business by making their service a free feature. They have tons of data to identify emerging rivals faster than ever before. Venture capitalists dodge spaces where tech giants might step, referring to "kill zones" as sectors where they won't invest because tech giants have entered that space.
Facebook famously copied Snap's features. Amazon labels many startups as "partners," only to copy their functionality and offer them for free. Google purchased a promising Waterloobased startup called BufferBox, took its IP to HQ and then shut it down. Huawei creates "partnerships" with Canadian universities where taxpayer-funded professors transfer critical IP to China. The French firm Ubisoft gets up to 37 per cent of its Montreal work-force costs subsidized by Quebec taxpayers. Branch plant engineers register IP created in Canada and transfer it to HQ, where it's commercialized and taxed.
While predatory on the surface, these actions reflect the nature and structure of the IP and data-driven economy, where markets are prone to concentration. High rent extraction and abusive power are rampant because global value chains are inherently unstable.
Yet Canada is courting big tech giants inside our largest incubators, creating our very own "kill zones" and turning them against our promising young companies. Using outdated, traditional economy logic, the federal government's Advisory Council of Economic Growth recommended that we double down on courting foreign tech with incentives and cheap labour in the hope of spurring Canadian innovation - without any evidence of how this would affect domestic innovators or develop Canada's competitiveness.
The discussion of big tech - and in particular, the role of foreign branch plants in Canada - is full of non-economic concepts rampaging through economic realms or anecdotal and foggy transplanting of the economic logic of the traditional economy to the intangible economy. Proponents of tech colonialism naively pin their hopes on big tech to do in Canada what it doesn't do anywhere else: help increase domestic innovation.
Despite mounting evidence of data colonialism and big tech's anti-competitive behaviour, some harbour illusions that tech branch plants bolster local supply chains the way production economies once did and that knowledge spillovers function as they do in the traditional economy.
What is Facebook's supply chain for social media? Google's for search? Big data and machine learning are highly centralized activities that do not draw on suppliers and sub-suppliers, as is the case in traditional sectors such as autos and aerospace. The free services of the platform firms do not generate meaningful local tax revenue but allow them to capture valuable economic rents. Uber's local supply chains consist of poorly paid, precariously employed drivers, while the value accrues to the IP and data assets located in the cloud.
One of the common arguments made by proponents of big tech as the driver of domestic innovation is that branch plants serve as training grounds for talent that can then build globally scaling companies. This logic worked in the traditional economy, where knowledge spillovers occurred in large foreign production facilities or with highly autonomous research labs. But as research by economists from the National Bureau of Economic Research, Harvard University, the Journal of International Business Studies, the London School of Economics and others shows, when multinationals locate their research facilities at incubators or universities today, the intent is to absorb knowledge back to the head office rather than introduce it into the hub.
This feature of inward investment in the technology sector is most evident by the IP filings from foreign branch plants in Canada. Knowledge generated in Canada is assigned to the company headquarters, where it's commercialized, generates value for its shareholders and is taxed to the benefit of foreign government coffers.
There is some hope that in the recessed corners of Ottawa, awareness of the role of foreign branch plants is catching on. A recent internal government report prepared by Canadian Heritage suggests that some policy leaders know there is a major problem with statements such as "Canada's tech sector is falling behind the rest of the world" and a key challenge is "the outsized role of subsidiaries of foreign companies in Canada's tech sector." Yet these observations are not followed by strategic action.
Contrary to popular narratives, the choice Canada's tech sector faces is not between openness and protectionism, immigration and xenophobia or economic growth and stagnation.
These are all false dichotomies.
The prominence of these myths in our public discourse on innovation illustrates just how misguided Canada's policy ideas are in the digital era.
SMART COUNTRIES, SMART APPROACHES There is a global race under way by large firms and smart nationstates to own critical IP, especially for emerging areas such as blockchain, AI and machine learning. Smart countries understand that knowledge-based and data-driven global value chains are inherently unstable, because IP is a legal monopoly and data asymmetries break markets.
The pace of the intangible economy is accelerating, as evidenced by the rapid rise in IP filings in the past two decades, especially in AI patents, and in the data generated through the rapid adoption of connected devices. The scale of this transformation in the source of market value recalls the shift in the source of wealth from land to capital that started with the industrial revolution and marked the transition from the feudal to the capitalist system. While equally profound in terms of its implications for the organization and governance of economies and societies, the current shift is unprecedented in terms of its velocity and demands sophisticated policy approaches.
While remaining open for business, smart countries are applying a new policy lens to FDI, advancing strategies for technological sovereignty and combatting anti-competitive behaviour with a series of antitrust measures. After years of study, in 2017 the EU presented a new foreign investment screening regulation for greater scrutiny of foreign interests and protection of its strategic IP and data assets. At a national level, countries such as Britain, France and Germany each created new rules for FDI in order to safeguard their strategic IP, data and AI assets. Germany even blocked the hiring of its top AI scientist by a foreign firm.
To confront the new and distinctive properties of the IP and data-driven economy, the U.S.
government passed the Foreign Investment Risk Review Modernization Act in 2018, which included widely expanded restrictions on foreign possession of "critical technologies" and the "data of U.S. citizens." President Barack Obama's former treasury undersecretary for international affairs, Nathan Sheets, said he was originally skeptical of any efforts to restrain foreign investment but by 2017 he was convinced of the need to fight back, stating: "As I left the Treasury I was quite concerned about where this was heading."
In advanced policy circles, the competition to dominate IP and data assets, especially AI, has been characterized as "AI Nationalism" or "techno-nationalism." These are not xenophobic or close-minded reactionary moves. They simply reflect the fact that AI is a general-purpose technology that affects all sectors and parts of society and that the power of data's network effects dramatically enhances economic, technological and military powers.
As new policies and strategies are being rolled out across the globe to shrewdly deal with FDI, Canada's outdated pursuit of it has most recently culminated in a project called Sidewalk Toronto. Originally championed by all three levels of government, this "smart city" envisioned the technically complex role of managing strategic IP and data-intensive infrastructure via a traditional land redevelopment agency called Waterfront Toronto.
Predictably, its mismanagement became a global media controversy and a revealing parable of what happens when Canada's policy makers insist that traditional-economy skill sets are suitable for the innovation economy. Even the Auditor General had to point out that a real estate agency has no expertise or capacity for IP and data projects.
Sidewalk Toronto - which I have been on the record as opposing - not only raises questions about the economic benefits for Canada, but poses serious new urban, civic and political governance risks. Harvard Business School professor Shoshana Zuboff put it best in Ottawa in May, testifying at the International Grand Committee on Big Data, Privacy and Democracy: "The real prize is the smart city itself. And right now, the war is being waged in Toronto. If Canada gives Google Toronto, a blow will be struck against the future possibilities of democratic societies in the 21st century."
It's hard to believe that our politicians would willingly undermine Canada in such a direct fashion or even aim for a legacy of destroying democratic governance. More likely, this is a byproduct of succumbing to "lobbynomics," a set of policies shaped by the regulatory capture by big foreign tech multinationals that, as The Logic recently reported, were given unprecedented access to the Prime Minister's Office. Such approaches to policymaking help explain why Canada caved on the IP and data sections of the U.S.-Mexico-Canada trade agreement (USMCA), which were championed by Silicon Valley and Big Pharma.
The most disheartening aspect of these misguided approaches is how unbecoming they are of a country with extraordinary potential for the intangible economy of IP and data.
In the modern era, the size of the country and its military prowess take a back seat to the ingenuity of its population. Canada developed foundational AI technology only to watch it be transferred to Google without an economic or national-security review. Our universities consistently rank at the top in the world for peer-reviewed research in critical sectors such as clean tech, health and agriculture.
Canada's innovators export world-beating technologies and create quality technical and nontechnical jobs. They are the ones who create all the attendant wealth effects of the innovation economy: enormous private and public wealth, new tax revenues, venture capital and philanthropy. As Louis Têtu, the CEO of Coveo, Canada's most successful AI company, said: "We're not going to pay for the next hospital by luring Amazon to Canada or giving away our engineers to Google. We're going to build the next hospital by creating wealth here."
Our federal government - no matter the outcome of this fall's election - must stop operating in an outdated paradigm to deal with today's economy. Our policy makers need to distinguish between foreign portfolio investment (FPI), which increases our entrepreneurs' access to capital while allowing Canadians to both retain control and accrue the subsequent prosperity effects, and FDI, with its attendant loss of control and economic gains.
As the economic research shows, FDI in the IP and data-driven economy has immensely negative spillover effects, while FPI creates positive spillover effects, such as recent investments into Coveo, Shopify, Wattpad, Lightspeed, PointClickCare and other bright domestic stars.
It is precisely those and other domestic scale-ups - firms with $100-million to $2-billion in annual revenue - that hold the most promise for Canada's longterm prosperity. Research shows such firms generate the most well-paying jobs, including executive non-technical jobs that reside only in HQ. A BDC study shows that scale-ups with 100 to 499 employees represented just 1 per cent of domestic companies but had the strongest revenue growth in Canada from 2001 to 2013 and contributed 12 per cent of GDP and 16 per cent of Canadian jobs.
If we want prosperity for our country in the 21st-century economy, it will help to have politicians, policy makers and pundits who understand how value capture in the knowledge-based and data-driven economy works.
Canada's continued weak performance in innovation is not peripheral, not just some rough edges that can be sanded down with a tweak, rebranding or more hype. The 21st-century economic development strategy for Canada needs modern strategies for trade, research funding, sectoral investments and a Competition Bureau that understands the intangible economy alongside a comprehensive national data strategy, because data is the critical input into AI.
We equally need a dramatically updated analytical framework for the effects of contemporary FDI, as well as innovation experts designing and implementing our innovation strategy. We also need politicians who believe our country's innovation potential goes beyond being a nation of cheap tech talent with taxpayer subsidies.
Canada will not address its productivity challenges, prosperity gap with the United States and other leading innovation countries - essentially the longterm threat to our quality of life - without paying serious attention to the fact today's economy is fundamentally different than that of the past. It's not what our entrenched policy makers want, but it's what we urgently need.
ILLUSTRATION BY MICHAEL HADDAD
ILLUSTRATION BY MICHAEL HADDAD