January 16th, 2021

Supercomputer

—   Ella Coon

Supercomputer

—   Ella Coon

 

CDC 6600 Supercomputer, Computer History Museum, 1964.

 
 

In March 1976, Deputy Secretary of the Department of Defense (DOD), William “Bill” Clements invited William “Bill” C. Norris, CEO and Chairman of the supercomputer producer Control Data Corporation (CDC), to a closed-door meeting at the Pentagon. Secretaries and undersecretaries from the United States Army, Navy, and Air Force were to attend, as well as a selection of spokespersons from the public university system and private sector. Clements requested Norris come prepared with “any important aspect of Defense management or posture that… warrants perspective” and to be candid in his comments.

Preparing for the meeting, Norris wrote a note. The subject was “East-West trade.” The DOD was not giving enough attention to export administration, the CEO wrote. The Department of Defense’s “inconsistent” approach to reviewing export applications for computer technology to Central and Eastern European countries, Norris continued, was fostering an “unhealthy,” “adversarial” relationship between the military and private industry.

A rival of IBM, CDC was a leading supercomputer producer in the mid-1960s and 70s. In 1976, the firm had a capitalization of \$1.82 billion, annual revenue of \$1.21 billion, \$41.5 million in annual profits, and over 100 subsidiaries stretching from Australia, to South Africa, and across Europe. Due to the high valuation of its machines, CDC was ranked eighty-fifth on the Fortune 500 list for assets in 1976, just above Kraft, Coca-Cola, Singer, and Boeing.

Since the dawning of the US-Soviet détente, CDC had internationalized some manufacturing operations behind the Iron Curtain. The firm entered into licensing agreements with communist states to lease out the rights to produce computer peripherals, such as disk packs and printers. By 1976, CDC opened a subsidiary for this purpose. ROM Control Data, in the Socialist Republic of Romania, was one of the first US subsidiaries in the Eastern Bloc, or Comecon (The Council of Mutual Economic Assistance).

The DOD’s resistance to CDC operations in the Eastern Bloc was relatively new. Following the so-called Halloween Massacre in November 1975, in which Gerald Ford reorganized his cabinet to appease the ascendant neoconservative wing of the Republican Party, the upper echelons of the US government began to challenge business engagements in Central and Eastern Europe. In particular, Donald Rumsfeld’s appointment to lead the DOD ushered in a new era in defense. Adopting a “hardliner” position towards the Soviet Union, détente under Rumsfeld was reconfigured to mean increased defense expenditure, arms buildup, and a cordoning off of the flow of strategic technologies to the Soviet Union and Eastern Bloc. In other words, the DOD’s new position reprised containment1 and foreshadowed Reagan’s Star Wars, while explicitly clamping down on the flow of potentially strategic technical knowledge.

Two incompatible American visions of managing the flow of technical knowledge around the globe emerged. CDC built its export strategy around ad hoc licensing agreements and the maintenance of intellectual monopolies. The DOD, on the other hand, increasingly sought to police, but not stop, the flow of potentially strategic knowledge in its own bid to legitimize itself in the lull between the Vietnam War and Star Wars. A struggle ensued between the firm and the federal agency over who would be the final arbiter of global technical knowledge flows. Their struggle speaks not simply to the domestic dynamics informing the United States’ trade relations in the mid-1970s, but also the political-economic motivations undergirding globalization more broadly.

“Eastern European Strategy”

CDC’s Eastern European strategy sought to transform the Eastern Bloc into a regional factory producing computer peripherals—like disk drives, tape readers, and printers—to be sold on the world market.

The project had two phases. The first—called “limiting indigenous development”—consisted of obtaining restrictive licensing agreements with communist states, confining their ability to sign contracts with other computer firms for peripherals as well as discouraging independent research and design (R&D) and manufacture of these products. The goal was to position the company as the most favored supplier of peripheral technologies (and later, of computers) to these states.

The second, never fully-realized phase planned to push these products out of the so-called Second World and into the Third, thereby creating global dependencies on CDC hardware. The firm’s licensing agreements were intended to be scaled to produce a surplus that would exceed domestic need. From there, in an almost Hobsonian supposition, it was thought that Central and Eastern European countries would sell the excess to buyers beyond their borders, particularly in the Third World.

The company saw the balance of payments as directing this circulation somewhat automatically. In particular, Comecon states lacked hard currency—i.e. US dollars. Central and Eastern Europe underwent economic stagnation and currency crises in the 1970s, a situation compounded by the 1973 Oil Crisis. Given the lack of convertibility for currencies in the Eastern bloc, it was difficult to pay the firm for licensing fees. Thus, CDC was banking on the prospect that communist states would sell the computer peripherals in the Third World to generate revenue, a portion of which would then flow back to the parent company in US dollars.

This elaborate tactical-strategic commercial system was conceivable because the Eastern Bloc’s balance of payments with the Third World was positive. Moreover, the US’s continued suppression of nationalist movements and support for anti-democratic strongmen in the Global South meant that many countries in the Third World did not want to be seen dealing with a company that could be construed as part of the US national security apparatus. As the CEO Norris said, “Presumably, some of the Third World markets may be better accessed by our Socialist partner[s].”

But CDC’s “Eastern European strategy” was overall a failed project. Although the company was successful in exporting a portion of its manufacturing abroad to Central and Eastern European states, like Romania, the grander vision of harnessing the global economic system to proliferate its trademarked products never came to fruition. This failure was in no small part due to the heavy-handed and blinkered manner in which the firm had carved up the world and its dealings. Central and Eastern Europe was not the monolith the firm supposed.

Romania, for example, primarily dealt with CDC not to gain technical capacity to produce peripherals, such as low-grade disk drives and printers, and sell them on the world market, but to build a working relationship with the firm. The Romanian government thought this relationship would help the state eventually buy a newer, powerful supercomputer (a CDC CYBER 80), to be used in oil exploration, thus hopefully giving the country greater autonomy from Moscow. (Neither the Department of Commerce nor State Department, however, would get behind the sale of a CYBER 80 to the state, to say nothing of the DOD.) Marketing peripherals to the Third World was not a top priority for the subsidiary. As a State Department assessed ROM Control Data’s status in 1977, the joint venture was “proving to be a money loser for the firm.” The cable continued: “Costs at the Romanian operation were relatively high… and Romanians had proved also totally unable to sustain their own marketing effort.” Unable, meaning, unwilling.

But the hostility on the part of the Defense establishment also contributed to stymieing CDC’s plan. Although the DOD was not wholly successful in gaining a stranglehold over US export controls, they were successful in shifting the terms of the discussion. Critically, the Export Administration Act of 1979 adopted the suggestion that export applications be reviewed on the basis not of the technology’s intended use abroad or availability on the world market, but its potential strategic applications, a key position supported by the DOD. As a result, CDC’s export license applications, and in particular its applications to transfer technology for low-capacity disk drives, came under increased scrutiny.

The clincher for CDC’s “Eastern European strategy,” however, was the breakdown of CDC itself. In 1978, the Department of Justice (DOJ) brought litigation against the company for mail fraud and illegal foreign transaction reporting two years earlier. In 1979, the firm was sued by its shareholders for failing to disclose these corrupt practices, paying out over \$1.3 million in damages. Furthermore, the firm was over-leveraged. By 1986, it was selling off many of its ventures. The Eastern European Strategy was a market-access strategy, and an unprofitable one at that. It was abandoned soon after.

CDC, proto-global value chains, and a first wave of globalization

What then is to be learned from CDC’s failed Cold War-era venture in Central and Eastern Europe? Although CDC’s dealings with Comecon states in the late 1960s and 70s were certainly particular, they were not singular. Looking at the supercomputer producer’s relations with Central and Eastern European states during this period one can find features of globalization—meaning, narrowly, global financial integration and the internationalization of production—in the era between the first and Second Cold War.

The standard narrative of globalization is that in the early 1990s, with the breakdown of the Soviet Union, the international system was restructured as a global system to support the liberalization of both trade and finance. This was seen in the expansion of the International Monetary Fund (IMF), the World Bank, Organization for Economic Cooperation and Development (OECD), multilateral (‘free’) trade agreements like North American Free Trade Agreement (NAFTA), and the making of the World Trade Organization (WTO). Capitalism won out. History had ended.

There have been many correctives posited to this narrative. Quinn Slobodian argues that the desire to construct a supranational system to protect international capital against nationalist economic interests can be seen as early as the interwar period. Others have sought to push the timeline of globalization back from the Washington Consensus of 1990s to the 1970s, before the final days of the Cold War, to acknowledge the earlier growth of international finance. This was seen through the rise of international capital markets and subsequent debt crises seen in Eastern Europe, Sub-Saharan Africa, and Latin America in the 1980s, and the early decades of the OECD, whose membership hinged on the lifting of capital controls, allowing for hotter money flows.

The case of CDC aligns with the lattermost studies, demonstrating that neoliberal globalization was quite tangible as early as the 1970s. But CDC’s activities in Central and Eastern Europe reveal the other side of globalization—not global financial integration, but the internationalization of production. The development of globalization ran through not only the realms of ideas and finance, but also through Romania’s national computer industry.

A study of CDC’s ventures in the Eastern Bloc helps us consider the economic imbalances inherent to internationalized production. In fact, the geographic dispersal of the firm’s commercial relations abroad could be seen as a kind of proto-global value chain (GVC). The supply chain CDC was hoping to construct for computer peripherals in Central and Eastern Europe adhered to a smile curve—meaning, the so-called higher value stages of production (i.e. R&D, design, and trademarking) were largely located in the United States, while the lower ones (i.e. parts and components and assembly) were outsourced to low-wage countries, such as Romania. The higher value stages of production were effectively governed by an intellectual monopoly: technical knowledge was enclosed and thereafter leveraged to create global commercial dependencies.

CDC’s activities in the Eastern Bloc challenge the image of a standard global value chain; the firm was not explicitly scouting cheaper labor, but more distinctively market access. The firm was, thus, advancing intellectual property rights for computer technology in tandem with the geography of the Cold War to gain a global market advantage with political ramifications, ones that the State Department found highly favorable during the Nixinger-era and less so under Ford. This Cold War context helps to magnify the geopolitical embeddedness of internationalized production—a dynamic with new relevance today. Just look at Trump’s “war” on Huawei. Arguably, there isn’t a more striking contemporary example of the geo-politicization of technology transfers.

In turn, returning to the contest between CDC and the DOD, this competition throws the open secret about globalization into sharper relief: globalization is just as much a political project as an economic one. This is perhaps somewhat obvious regarding the DOD, whose attempts to insert itself as a supervisory power over “intangible,” high-tech global knowledge flows proves inextricable from the ambitions of a newly emboldened Cold War national security state. However, CDC’s efforts to globalize production likewise look eerily imperial in its efforts to leverage the ownership of technical knowledge to create global dependencies.

In other words, CDC’s ventures behind the Iron Curtain reveal that globalization was never really about making ‘free’ markets. The fight over trade terms was, and increasingly is, an embittered one, inextricable from the explicit interests of ad hoc international actors—ranging from firms and domestic agencies to states and economic blocs. Technology transfers, although often framed as a mechanism to share or ‘democratize’ information, instead should be understood as fine-tuned instruments deployed exclusively to benefit the arbiter.


  1. John Lewis Gaddis, George F. Kennan: An American Life, Penguin Random House, 2012.  

Title Supercomputer
Authors Ella Coon
Date 2021-01-16
Collection Analysis
Filed Under
In Series
 
 

Sign up for the JFI Sources, a weekly digest of compelling research across the social sciences.