In the early 2000s, Finland was the darling of industrial and employment policy analysts everywhere. This small country with a population of 5.5 million and a GDP roughly equal to the state of Oregon experienced what looked like a high tech-led productivity revolution. Real GDP per capita in local currency terms rose 55 percent from 1995 to 2007—nearly double the US increase and close to the pinnacle of the twenty-one richest OECD industrialized economies.
Yet spectacular growth abruptly halted after 2008. GDP continued to rise with population growth, but from 2008 to 2019 real Finnish per capita income declined. The European Central Bank’s dilatory response to the eurozone crisis, and the austerity policies that followed, undoubtedly explain part of this abysmal performance. But an equally large part is due to Finland having many of its growth eggs in a single basket: Nokia. Nokia’s handsets and related telephony equipment accounted for 20 percent of Finnish exports at peak, driving Finland’s current account surplus to nearly 7 percent of GDP.1 When the Apple iPhone launched in 2007, Nokia’s handset market collapsed, exports fell by half, Finland’s current account swung into deficit, and a decade plus of economic stagnation began.
Finland is not the only economy facing “Nokia risk.” A larger group of seven countries—all of them relatively small, rich, and with stable governments—are similarly exposed. In Denmark, Israel, South Korea, Sweden, Switzerland, and Taiwan a handful of firms account for a hugely disproportionate share of both profits and R&D spending. The firms which dominate these seven economies have all been extraordinarily successful in the knowledge economy of the past three decades: Samsung Electronics in Korea, Taiwan Semiconductor Manufacturing Co. in Taiwan, Novo Nordisk (pharmaceuticals) in Denmark, and Roche and Novartis (pharmaceuticals) in Switzerland. For the decade of 2011–2020, these firms have had large shares of cumulative profits both domestically and abroad. It is largely thanks to these profits that these small countries have such a significant share of global profit (larger than their share of global GDP) and, in turn, a relatively high per capita income.
Table 1 shows the ratio of the share of cumulative global profit and of R&D expenditures to country share of global GDP, with some comparisons to the larger rich OECD countries. The figures are based on 21,580 global ultimate owners (headquarters firms) with annual sales averaging at least $10 million between 2011 and 2020, and which have enough comprehensive financial data to make comparison possible. While this calculation excludes a big chunk of small and medium sized industries (or SMEs, firms with fewer than 250 employees), these 21,580 are the biggest among the 41 million in the Bureau van Dijk Orbis database. The handful of firms discussed here is thus both mathematically and macroeconomically significant, as the very long tail of SMEs typically depends on those larger firms for their revenue. For example, Danish pharmaceutical giant Novo Nordisk alone accounts for a quarter of cumulative profits for the 200 plus Danish firms in Table 1; its 0.2 percent share of cumulative profits for all 21,580 firms is almost 50 times the average for that group.
Table 1: Ratio of share of cumulative Profit and R&D spending to share of Global GDP, by country, 2011–2020
Profit | R&D | # of firms | |
Taiwan | 2.83 | 4.07 | 591 |
Sweden | 2.18 | 2.24 | 360 |
Switzerland | 2.00 | 4.89 | 131 |
Denmark | 1.94 | 1.05 | 222 |
Korea | 1.63 | 2.10 | 706 |
Finland | 1.15 | 3.05 | 185 |
Israel | 0.53 | 1.07 | 101 |
For comparison: | |||
Norway | 1.80 | 0.24 | 253 |
Japan | 1.58 | 3.07 | 2694 |
USA | 1.45 | 1.75 | 2524 |
Germany | 0.97 | 1.70 | 1274 |
In the coming years, these firms and the national economies that house them will confront two big challenges. On the economic front, the fifth Schumpeterian growth wave, which was built on information and communication technologies (ICT) as well as traditional and first-generation bio-engineered pharmaceuticals, has exhausted much of its growth impulse. A potential sixth wave built on artificial intelligence (AI), machine learning (ML), and second-generation biotech (CRISPR) as general-purpose technologies threatens these firms’ existing production systems and markets. At the same time, governments everywhere are ramping up antitrust and other attacks on visible monopolies, and edging towards more nationalistic industrial and trade policy. In light of the central role of these firms that account for disproportionate shares of their home economy exports, failure to navigate these challenges will result in a major economic shock for their small, rich countries.
Schumpetarian waves and industrial transitions
Joseph Schumpeter’s analysis of dynamic change in capitalist economies—what he termed “creative destruction”— sheds light on the economic risks facing the core firms in these seven economies. Schumpeter argued that the central puzzle in economics was explaining the sources of dynamic growth. In an economy that actually embodied the starting assumptions of mainstream economics—small, competitive firms with no barriers to entry and no pricing power—profits would fall to the cost of capital plus some managerial wage for owner-operators. In essence, profits would only cover depreciation. Consequently, extensive and intensive growth would slow to the rate of population growth plus some gains from the normal productivity creep that incremental innovation produced for firms—in short, the kind of economic slump that prompted the secular stagnation debate of the 2010s. This circular economy, as Schumpeter called it, would never produce the periodic eruptions of rapid technological change and growth that he observed in the two centuries after the industrial revolution. As he put it, in these circumstances, you could add as many stagecoaches as you wanted to the economy but never get to a railroad.
Dynamic growth required radical innovation in five key, interconnected areas: new modes of transportation, new energy sources, new consumer goods, new general purpose production technologies, and, though he underemphasized this, new legal organization or governance structures for firms. Writing on the eve of the Second World War, Schumpeter identified four such revolutions that had so far taken place. The first initiated the industrial revolution, centering on canals, water mills, textiles, and other household nondurables, with small owner-operated factories collecting handicraft producers under one roof. The second, in the mid-1800s, focused on steam power, railroads, and iron goods, as well as larger but still owner-operated factories using custom made machinery. The third, towards the end of the 1800s, emerged from electricity, steel steamships, urban trams, bicycles and chemicals, and saw the rise of large corporations which began to separate ownership and management. The fourth wave centered on internal combustion engines for land and air transport, petroleum, mass consumption of consumer durables like vehicles, continuous flow-assembly line production, and vertically integrated and often multidivisional firms with full separation of ownership and management. This wave began in the United States, most famously with Ford’s Model T, and spread to Europe and Japan. The transition periods between each of these growth spurts all saw increased domestic or international conflict and decreased investment in ageing growth sectors. Thus the transition from wave three to wave four saw the beginning of industrial unions and coincided with rising interimperial conflict that eventually produced World War I.
The fifth wave began in the US in the 1960s with the development of the semiconductor chip, and soon spread globally. It is based on connectivity via electronics (ICT), negative energy consumption via digitalization, pharmaceuticals and first-generation biotechnologies, software and semiconductors, global supply chains, and de jure vertically disintegrated firms with de facto control by lead firms. This era’s iconic product, the smartphone, embodies the whole range of electronics products developed from the 1940s onward in a compact and relatively cheap form.
There are many reasons to think this fifth wave is nearing exhaustion. Smartphone sales levelled off in 2018 and then declined somewhat, signalling replacement sales rather than growth. Despite the annual iPhone launch hype, most of the improvements to smartphones in recent years are largely marginal. Roughly 80 percent of the world’s population has 4G access—if they can pay for it. The entire global electronics industry is linked to personal computers and smartphones, which account for roughly half of global chip sales. Similarly, new pharmaceutical discovery levelled off in the 2000s, with most new drugs being copies or modifications of older drugs.2
But sales volumes simply reflect more important trends. Several key technologies and industries have already emerged in the energy and transportation space as well as, albeit with less clarity, in production software and pharmaceuticals.3 Alternative and renewable energy sources are clearly replacing fossil fuels as the basis for electricity generation and equipment power, including the critical transportation sector. And AI and ML appear to be general purpose production technologies reshaping both biotechnology (along with the new CRISPR-Cas gene editing technology) and the organization of machinery production. Table 2 summarizes the economic and political risks this sixth wave poses for core firms in these seven economies.
Table 2: Economic and political risks
Country | Firm | Sector | Political Risk | Economic Risk |
Denmark | Novo Nordisk | Pharmaceuticals | Regulation of Insulin prices | AI based genomics shifts value to personalized drugs |
H. Lundbeck | Pharmaceuticals | Regulation of Insulin prices | AI based genomics | |
Vestas | Energy – windmills | National champions in other countries | Growth potential is high | |
Finland | Nokia | Telecom equipment | 5G becomes a national security issue | AI & ML shift value to software |
Israel | Teva | Pharmaceuticals | Price regulation in general | AI based genomics |
Korea | Samsung | Memory chips; phones | Policy driven on-shoring by US and EU | AI design shifts more value to systems on a chip |
Hyundai | Vehicles | Electrification | AI shifts value to battery management software & driver aids | |
Sweden | Ericsson | Telecom equipment | 5G becomes a national security issue | Software-based switching systems replace hardware-based systems |
Volvo (Trucks) | Vehicles | Electrification | AI shifts value to battery management software & driver aids | |
ABB, Assa Abloy, Sandvik | Machinery Firms | Deglobalization, EV effects on German car industry | AI & ML shift value to software; EV requires fewer parts | |
Switzerland | Novartis | Pharmaceuticals | Price regulation in general | AI based genomics |
Roche | Pharmaceuticals | Price regulation in general | AI based genomics | |
Taiwan | TSMC | Semiconductor fabrication | Policy driven on-shoring by US and EU; war | Subsidized national champions |
Hon Hai Precision | Electronics assembly | US-China tensions | AI design shifts more value to systems on a chip, simplifying assembly |
Calculating risk
Creating the huge reticulation networks that made new energy and transport sources useful required equally huge investments. One mile of railroad was pointless, 100 miles was revolutionary. Schumpeter, in his earlier works, argued that only entrepreneurs hyping potential monopoly profits could induce bankers to finance these huge investments. He called this the Mark I model, in which small start-up firms run by visionaries upend existing incumbents. Many of the current software giants fit this pattern, but it also characterized the earliest days of Ford. Later, Schumpeter noted that high-profit corporations could channel their monopoly profits to dedicated internal research labs and generate the same kind of revolutionary innovation—his Mark II innovation model. Here, think of ATT’s Bell Labs, which invented the transistor.
Software aside, we largely live in a world that combines both Mark I and Mark II innovation in a complex web that mostly favors larger firms. Typically, states promote radical innovation, often by funding basic research in university labs and their small firms spin-offs—classic Mark I innovation. But larger firms then typically provide those small firms with more funding to develop a commercializable product that their own Mark II R&D teams will perfect. While these smaller firms often get the publication glory, the larger firms usually get the bulk of patents and thus profits. Apple’s SIRI assistant, initially developed by a US-wide research team organized through the Stanford University’s Research Institute on behalf of the Defense Advanced Research Projects Agency (DARPA), was absorbed by Apple into its own software ecosystem.
If profits and R&D were evenly distributed across firms in each of our seven economies, these new technologies might not pose so big threat. But under the present combination of Mark I and Mark II innovation, a handful of firms account for the bulk of profits and R&D. For example, insulin accounts for roughly 7 percent of cumulative Danish goods exports for the period 2002–2021, with most of this coming from the pharmaceutical giant Novo Nordisk. 80 percent of Novo Nordisk’s revenue comes from sales of insulin and diabetes-related products, but 80 percent of Denmark’s workforce is in the service sector, which apart from Maersk-Møller and Legoland largely does not generate exports. Thus, the fate of Novo Nordisk has the potential to impact one of the country’s only significant export sectors.
Table 3 shows the degree to which this very narrow slice of firms dominates both profitability and R&D in our group of seven economies. The table thus captures the degree of vulnerability around Mark II innovation. Comprehensive data on Mark I innovation—which ranges from a handful of people tinkering in their parents’ garage to modestly sized start-ups with no revenue—is not available. As noted above, Schumpeter’s two pathways to radical innovation are more intermingled today than in the past, when vertically integrated firms were rather sealed off from both universities and small firms. These days, much Mark I innovation is often captured by the larger firms through acqui-hires, acquisition, or copycat innovation and litigation. Equally so, accurate, comparable data on the relative share of national profits by firm size is not available. While the $10 million annual operating revenue metric is biased towards fairly large firms, it does reach into the upper end of the long tail of SMEs that constitute the majority of firms in most countries. Moreover, SMEs typically have labor productivity levels one-third below larger firms.
Table 3: Firm share of cumulative national R&D and profits for high R&D firms, 2004–2020
Company | R&D share | Profit share | |
Denmark | Novo Nordisk | 34.6% | 35% |
Finland | Nokia | 77.6% | 19% |
Israel | Teva Pharmaceutical | 43.9% | 7.6% |
Elbit Systems | 9.1% | 8.3% | |
Korea | Samsung Electronics | 47.1% | 36.6% |
Hyundai Motor | 8.3% | 7.4% | |
Sweden | Ericsson | 38.7% | 7.9% |
Volvo (trucks) | 19.8% | 10% | |
Switzerland | Roche | 33.4% | 23.7% |
Novartis | 29.7% | 16% | |
Taiwan | Taiwan Semiconductor | 12.6% | 29.1% |
Hon Hai Precision Industry | 12% | 14.2% |
These large firms dominate exports, and thus generate the foreign exchange needed to buy the considerable volume of goods and services that the seven economies do not produce locally. Given their small size, these economies cannot attain a domestic division of labor that generates the bulk of their consumption—hence their need for export. In all but one of the countries, firms with more than 250 employees account for more than half of exports. Denmark is the exception; there the share is so close to half that it makes no difference (data for Taiwan are not available, but as noted above TSMC’s large share suggests the pattern of large firm dominance is not much different). Table 3 understates the importance of the larger firms, as these typically account for fewer than 1 percent of the total count of firms, while firms with fewer than 10 employees account for over 90 percent.
As noted above, these few dominant firms largely explain why these societies capture a larger share of global profit than their share of global GDP (Table 1), which in turn explains why they are relatively rich societies. Success in generating high-value exports and their associated profits permits these societies to exchange a small volume of high-value exports for a much larger volume of lower-value imports. This disproportionate profit share reflects both pluck—local ability in the form of prior investments in R&D and human capital—as well as the broader dynamics of the modern economy.
These firms rely on their investments in R&D to maintain their competitiveness. Their investments in turn explain the above average share of R&D in their economies as compared with the larger economies of the G7. Table 4 provides four relevant pieces of information. It contrasts R&D spending and the number of full-time researchers in the population relative to the G7 countries. It also shows the high share of total R&D by business, as well as the disproportionate share of total R&D spending accounted for by both the largest and the fifth largest firms and thus their even greater share of business R&D (the picture is somewhat different in the case of Taiwan). Table 4 thus shows the degree to which future innovation—at least in terms of the Mark II innovation that turns Mark I innovation into a global product—rests on a very narrow foundation. We could construct an index of risk by multiplying the share of business R&D by the share of the top one or five firms, but this would give a false sense of precision. The central point here—looking at the last two columns—is one of amazing concentration of research spending.
The overlap of high profitability and profit share, export share and R&D share, is not accidental. It indicates past competitiveness and near monopoly or dominant positions in world markets. Profits fund the R&D that enables dominance and thus continued above average shares of global profits; those profits fund high levels of per capita income—among others, all those researcher jobs. And they fund, in some cases, extensive welfare states or at least state-education funding that generates the human capital those researchers possess, and which is the basis for past and potentially future dominance in high-tech sectors. Thus, for example, between 1991 and 2000, Swedish education spending increased by 2.1 percentage points of GDP to 7.4 percent, with half of that going to tertiary education. The Nordics have significantly higher literacy scores in standardized international tests than the rest of the rich countries, though Sweden has recently slipped back.4
Table 4: R&D intensity of economy, various measures, percent, average as indicated, ranked by column 1
R&D percent of GDP | FTE researchers per 1000 workers | Business percent of R&D | Top 1 share of Business R&D | Top 5 share of Business R&D | |
2010-2019 | 2010-2019 | 2010-2019 | 2005-2020 | 2005-2020 | |
Israel | 4.36 | — | 85.7 | 43.9 | 69.5 |
Korea | 4.02 | 15.9 | 78.1 | 47.1 | 72.2 |
Sweden | 3.25 | 15.2 | 69.5 | 38.7 | 68.8 |
Switzerland | 3.17 | 8.6 | 71.2 | 33.4 | 78.3* |
Finland | 3.1 | 15 | 67.4 | 77.6 | 85.8 |
Taiwan | 3.08 | 13.8 | 76.6 | 12.6 | 39.3 |
Denmark | 2.99 | 14.1 | 64.7 | 34.6 | 59.2 |
Unweighted G7 Average | 2.27 | 8.71 | 65.4 | — | — |
Source: Columns 1–3 OECD Main Science and Technology Indicators https://www.oecd.org/sti/msti.htm; 4–5 EU data https://iri.jrc.ec.europa.eu/data
The franchise economy
The shift from the fourth (mass production) to the fifth (ICT) Schumpeterian wave involved changes in corporate strategy and structure that had significant knock-on effects. Chief among them, it boosted income inequality and increased the degree to which firms’ profitability depended on the legal regime around intellectual property (IPRs—patent, copyright, brand, and trademark). In the Fordist era, corporate strategy aimed at monopoly or oligopoly profits through control over large masses of physical capital arranged into continuous flow, assembly line systems. Profitability rested on running those systems at something close to their full capacity. This pushed firms to vertically integrate and negotiate peace with their typically unionized labor forces, which in turn reduced income inequality and funded internal research labs.
But as more and more firms adopted this vertically integrated, unionized structure, profits began to decline. Workers revolted against the monotony and pace of assembly line production, and decolonization enabled raw-materials producers to push up prices, disrupting energy and metals markets. Put simply, once everyone adopted a Fordist product and production structure, the world ran out of cheap oil and docile semi-skilled assembly-line workers.
Firms reacted to this militance by changing their corporate structure. They shrank their labor forces and opted to subcontract or offshore their low-wage, low-skill workforce. Similarly, they expelled physical assets—machines—used for producing undifferentiated goods into spin-off firms, as when GM and Ford created Delphi (now Aptiv) and Visteon as parts producers. At the same time, they began seeking more durable monopolies based on IPRs produced by a labor force high in human capital and supported by an army of subcontractors. This shift, which both coincided with and enabled the emergence of the fifth Schumpeterian wave, produced what I call a franchise structure.
In the franchise economy, lead firms with lots of human capital, few actual employees, and substantial intellectual property portfolios outsource much of production to two other generic types of subordinated firms. Second layer firms are typically more capital-intensive firms with some barrier to entry for their production processes. Third layer firms are labor intensive firms producing undifferentiated goods and services. The lead firm orchestrates almost everything in its value chain without bearing any of the risk of holding that physical capital or dealing with masses of workers. In our seven economies, the lead firms are all among the country’s largest, albeit sometimes hybrids of the top two layers. Profitability for those firms derives from their control of intellectual property rights. Nokia, for example, lives on as a producer of network hardware and software based on earlier and ongoing R&D that generated more than 10,000 US patents from 2000 to 2019. Unlike the barrier to entry posed by capital intensive production, patents are vulnerable to radical technological change and also radical change in the legal regime surrounding the patent.
While the shift to a franchise structure was good for firms with robust IPR portfolios and, by extension, for the high human-capital intensity of the workforce of our seven economies, it was less good for workers and firms producing undifferentiated goods and services. Downsizing meant shifting relatively well-paying jobs to low-wage countries, hollowing out the middle of the income distribution. Between 1998 and 2016, for example, Sweden’s industrial giants cut their workforce from 80,000 to 49,000 people. While younger labor market entrants into more IPR-based firms partly compensated for this, the loss of so many mid-level jobs has had broad socioeconomic effects, arguably contributing to the rise of anti-immigrant parties in the country. Nativist policies, in turn, threaten the ability of firms to attract the global talent they need to sustain R&D.
Coping with risk
Danish industrial policy with regard to pharmaceuticals illustrates the tensions around both Mark I and Mark II types of innovation strategy. Recent Danish governments have promoted Denmark as an “innovation country” and formed a “Disruption Council” intended to preserve the country’s economic position. Danish R&D is highly concentrated in the pharmaceutical and biomedical sector (Vestas’ windmill production and servicing provides a side bet on alternative energy, but Chinese competition squeezes profitability). A government-run public goods strategy supports the emergence of Mark I type firms in biotechnology, thus providing a “feedstock” for Mark II oriented Novo Nordisk. This strategy continues a long tradition of “extension” type services designed to bring new technology and best practices from universities and high performing firms to more traditional firms including manufacturing SMEs. It is supplemented with venture capital from the Danish Growth Fund (Vækstfonden). The policy has helped keep Danish firms largely at the technology frontier, producing a narrower gap between leading and lagging firms than in the rest of the OECD. But the challenge in pharmaceuticals is anticipating the bio-genomics revolution rather than adapting to it.
In cooperation with Novo Nordisk, the Danish government runs a cluster of data collection and dissemination organizations: the Danish National Genome Center (Nationalt Genom Center), the National Biobank (Danmarks Nationale Biobank under the State Serum Institute), and the overarching Royal Bio and Genome Bank. These centers leverage the comprehensive and diachronic data the state collects from all hospitals and practitioners on a large range of conditions and patient variables. For example, the Biobank currently has serum samples from nearly 1 million people and 25 million samples in toto. The Genome Center was able to establish a genomic sequence database for distinguishing Covid-19 variants within five days in March 2020. Quite apart from the value to Novo Nordisk (which accounts for 90 percent of Danish pharmaceutical employment and profits), the databanks also enable smaller biotech firms to access large volumes of data on relatively smaller (in terms of incidence) diseases and thus help them overcome one major research hurdle. Put simply, they generate and hold the data that machine-learning and AI driven R&D projects need to function. Novo Nordisk meanwhile acts as a central contractor with many smaller research-oriented firms—though these contracts generally choke off growth from Mark I innovation in favor of pre-emption by the lead firm. Overall, however, Denmark lacks indigenous AI and ML firms that might supply expertise to the entire sector, as compared with either Israel or Sweden.
Those countries display different strategies with different weaknesses. Israel’s huge defense related investment in software and sensor capacity created a vibrant Mark I-type tech sector. But Israelis doing that Mark I innovation typically sell their firms or technology to US Mark II type firms, as when Google bought the navigation firm Waze or when Intel acquired Mobileye, which develops autonomous driving technologies. The relative absence of big domestic tech firms explains Israel’s deviation from the broader pattern of profit share being above GDP share (Table 1). It also explains why so many Israelis—an estimated 100,000—simply migrate to Silicon Valley even though Israel’s so-called Silicon Wadi usually houses more start-up firms per capita than any other country. Both trends potentially inhibit a response to the AI and ML revolution in software. The Israeli state, meanwhile, has relaxed the stringent employment and production requirements it put in place to stimulate Silicon Wadi in the first place.
By contrast, Sweden has an abundance of large, mature Mark II type firms. Yet these have gradually hollowed out actual production in favor of simply generating intellectual property. The extreme case here is Volvo. Its old automobile capacity was sold first to Ford and then to the Chinese holding company Zhejiang Geely.5 Geely relies on Swedish engineering talent to design the new Volvo electric vehicle line (including Polestar), but production has largely shifted outside of Sweden. Similarly, Swedish pharmaceutical giants Pharmacia and Astra respectively ended up controlled by Pfizer (USA) and AstraZeneca (UK). As noted above, the big Swedish manufacturing firms have steadily reduced employment. Like Israel, Sweden is increasingly a hunting ground for foreign multinational firms looking for talented individuals. This sustains high-wage employment—at least for some. But it also means the profits end up somewhere else, and large Mark II firms that might anchor a research network are harder to form.
High tides
The sixth Schumpeterian wave, should it indeed appear, poses serious risks for the largest, export-focused firms of our seven economies. Presently, a narrow set of IPR-based firms in the Mark II model does the forward-looking investment in R&D that enables the transformation and scaling up of Mark I innovation required to catch that wave. It also generates both the jobs and the revenues needed to sustain a politically acceptable level of imports, employment, and growth in general. The potential inability of the big, highly profitable firms that anchor local research ecosystems to transition from their current production model to the novel production models emerging will have serious consequences.
This risk extends beyond the “innovator’s dilemma.” Domestically, the loss of core manufacturing jobs in the second layer of the franchise economy has provoked populist backlashes. In both Israel and Sweden, this has empowered parties hostile to state-led industrial policy favoring highly paid knowledge workers. Externally, growing geopolitical tension between the United States and China has prompted efforts to reshore or “near-shore” the ICT sector, particularly semiconductor production. All told, this probably tilts the global playing field towards firms from the larger and more geopolitically powerful countries. As noted above, those firms already fish in the human-capital pools of the seven economies of our study, potentially undermining the survival or emergence of Mark II firms that anchor local R&D and production ecosystems. If the winner-take-all nature of the franchise economy continues, these older firms must run fast simply to stay in place.
For the pawns of the global economy—smaller economies without national champions like Nokia or Samsung, and without oil-fund assets as in Norway—these challenges are even more pronounced. They enter this race with greater headwinds, weighed down by external debt, relatively untrained workers, and, in the worst cases, an over-reliance on unprocessed raw materials exports.
All data on exports are from the UN International Trade Center. Data on R&D are from the European Commission database on the 2500 largest R&D spenders in any given year from 2004 through 2019. This cumulates to 5303 firms over 56 countries and territorial units.
↩That said—as we will note below—a rising share of drug approvals concern bio-similars rather than the traditional small molecule pharmaceuticals.
↩The US state is already organizing policy around this. See remarks by National Security Advisor Jake Sullivan in September 2022.
↩Data supplied by John Stephens, University of North Carolina.
↩Not to be confused with AB Volvo, now largely a heavy truck manufacturer after spinning out Volvo Cars.
↩
Filed Under