«Original citation: Owen, Geoffrey (2012) Industrial policy in Europe since the Second World War: what has been learnt? ECIPE Occasional paper, 1. ...»
In 1945 French industry, after two decades of under-investment, urgently needed modernisation, and the private sector was in no state to finance it. To remedy these weaknesses the first post-war governments used two principal instruments: nationalisation, bringing most of the basic industries, including coal, electricity and gas, into public ownership; and a system of indicative planning through which the government targeted high-priority sectors and ensured that funds were made available, mostly through state-owned banks, to support new investment. The only large manufacturing company to be taken over was Renault. This was not done for industrial or economic reasons, but in retaliation for the owner’s alleged collaboration with the Nazis during the war; the managers of Renault, though appointed by the state, were allowed to run the business with a high degree of autonomy. 42 There were also moves to strengthen the country’s capabilities in scientific research. The Centre National de la Recherche Scientifique (CNRS), which had been set up just before the war, was reorganised and given a mandate to develop and coordinate all French science. New technical agencies were created to undertake research in telecommunications, aeronautics and energy. Research into nuclear power, and later nuclear weapons, was entrusted to the Commissariat à l’Énergie Atomique (CEA).
ECIPE OCCASIONAL PAPERover for twelve years, was to save France. “This meant military independence, without which no state was truly sovereign; economic independence, without which no state was master of its own house; and technological revolution, without which no state could maintain the first two conditions”. 45 On the military side, work on developing an atomic bomb had begun in 1954, and the programme was accelerated after de Gaulle’s return to power. The early years of his presidency saw heavy investment in military aircraft and missiles, creating an American-style militaryindustrial complex which included private sector firms such as Dassault, the principal manufacturer of military aircraft, and Matra, a specialist in defence electronics, as well as several nationalised concerns. Government support took the form of public procurement, subsidised research, subsidised exports and industrial diplomacy, together with long-term research and production contracts from the Ministry of Defence. 46 The French government was no less determined to build up the civil side of the aircraft industry. The state-owned Sud Aviation (later part of Aérospatiale), which had enjoyed some success with the Caravelle jet airliner in the 1950s, was the French partner in the Concorde project. It was also the prime mover, along with Snecma, the state-owned aero-engine manufacturer, in the establishment of the Airbus consortium. Although the Airbus project was to go through many difficulties before it established itself as a credible rival to Boeing, the French government never wavered in its support for civil airliner development. Here, as in other areas, there was a consistency in French industrial policy which was lacking in the UK.
Some of the same companies were involved in the development of space technology, which began in 1961 with the creation of a government research institute, the Centre National d’Études Spatiales (CNES). CNES built a launching base in French Guyana and undertook projects for civilian and military uses in cooperation with Aérospatiale and Matra. Work on the Ariane launcher began shortly after de Gaulle had left the presidency, and although this became part of a collaborative European space programme most of the funding came from France.47 Mastery of space technology was another illustration of de Gaulle’s determination not to be dependent on the US.
Unlike the UK, France had not been in the forefront of computer technology in the 1950s and the authorities had been slow to recognise the importance of the industry. Two events in 1964 prompted a change of policy. 48 The first was the acquisition of Bull, the largest French computer company, by General Electric of the US; the government sought to frustrate the deal by encouraging two French companies to take over Bull, but they declined to do so. The second was the launch by IBM of the System 360 family of computers, a move which threatened to reinforce its already dominant world market position. In the following year the case for an independent French capability was underlined by the US government’s decision to block the sale of two Control Data computers to the French Atomic Energy Commission.
The response was the Plan Calcul, an attempt to foster a French-owned computer company capable of withstanding American competition. 49 Three firms were merged to form Compagnie Internationale pour l’Informatique (CII), which would be supported by subsidies and by preferential procurement on the part of government agencies. These firms were subsidiaries of electrical groups - Compagnie Générale d’Électricité (CGE), CSF and Schneider - all of which became shareholders in the new company. The government also took steps to ensure that CII had access to an indigenous source of electronic components. Subsidies were made available under the Plan Composants to a semiconductor producer, Cosem, which was then part of CSF. Control of CSF was later acquired by Thomson, and the enlarged semiconductor
business, Sescosem, became the national champion in what was seen as a vitally important sector of the electronics industry.50 The principle behind the Plan Calcul, as of other grands projets initiated during de Gaulle’s presidency, was what Élie Cohen has described as offensive protectionism. “The sovereign state creates the means of accumulation of scientific and financial resources. It provides future national champions with grants, secure markets through public procurement policies, and prevents foreign entry.” 51 The effect was to create close ties between the government and a group of favoured companies which depended on the state for a large part of their business.
One of the beneficiaries was CGE, which through a series of mergers and acquisitions lifted itself into the front rank of French industry; the number of its employees rose from 33,000 in 1960 to more than 100,000 in 1971. A key event was the so-called Yalta agreement in 1969 with Thomson, through which CGE acquired control of Alsthom, the principal supplier of power engineering equipment, while ceding to Thomson its interests in defence electronics and domestic appliances. Thomson also agreed not to compete in telecommunications equipment, leaving CGE to dominate this sector through its Alcatel subsidiary. Shortly before the CGE agreement, Thomson had acquired control of CSF, which had been an early leader in radio and TV broadcasting and had later established a leading position in military electronics. CSF was an important supplier to the Ministry of Defence, and the acquisition brought Thomson more closely into the world of grands projets. 52 Typically run by énarques 53 who had the same educational background as the government officials with whom they dealt, CGE and Thomson formed part of the oligopolistic core of French industry. By the end of the 1960s, as a result of mergers and acquisitions, most major industries had become more concentrated, with two or three companies emerging as clear leaders. Some of France’s largest firms, such as Michelin and Peugeot, were family-controlled and had few direct links with the state. Others depended on the patronage of government, as participants in grands projets or as suppliers to public sector agencies.
De Gaulle’s approach to industrial policy was broadly maintained under his successor, Georges Pompidou, who held office from 1969 until his death in 1974, but Pompidou was more pragmatic than his predecessor and less obsessed with French independence vis-à-vis the US. An early decision was to accept the recommendation from the French electricity authority, Électricité de France (EDF), to base France’s nuclear power programme on the light water reactor designed by Westinghouse in the US. This was opposed by the CEA which wanted to stick with France’s indigenous gas-graphite technology, but EDF and its suppliers feared that if France became technologically isolated from the rest of the world the potential export market would rapidly disappear.54 The investment in nuclear power, begun under Pompidou and accelerated after the first oil shock in 1973/74, was an example of a state-led project based on cooperation between two government-owned agencies, CEA and EDF, and two contractors, Framatome (then jointly owned by Schneider and the CEA), and the CGE subsidiary Alsthom. Another French project which started during Pompidou’s presidency was the Train à Grande Vitesse (TGV); here, too, a CGE subsidiary, Alsthom was the main private sector contractor to SNCF, the stateowned railway company. The TGV, which became a symbol of France’s technical and engineering prowess, was typical of the French approach to large-scale infrastructure projects, “driven through by small but highly influential and well-trained technological elite, mostly top graduates of a single institution whose members move smoothly between the political, administrative and corporate worlds”.55 14 No. 1/2012
ECIPE OCCASIONAL PAPERWith both nuclear power and the TGV the customer was the government or a government agency, and it was in industries of this kind where French industrial policy was most effective. In industries where the end-users were dispersed and less influenced by the state, the French approach worked less well. Nowhere was this clearer than in computers. By 1971 the ambitions set for CII, the computer company, were a long way from being achieved. It still had a very small share of the French market, which was dominated by IBM, and it was barely represented in other European countries.
To renew its product range, further government support was necessary, and this was provided under the second Plan Calcul, launched in 1971. At this point a possible European solution emerged. Siemens in Germany had entered the industry with its own computers, but by the early 1960s these machines were becoming obsolete; its share of the German market was not much more than 5 per cent. In 1964 it signed a licensing and supply agreement with RCA in the US, and began selling RCA’s Spectra line of computers under the Siemens name.
However, RCA was struggling to keep pace with IBM, and in 1971, after IBM announced enhancements to its System 360 family, it decided to withdraw from the industry. Siemens then turned to CII as an alternative supplier. It put to the French government a plan for a partnership which was later enlarged to include Philips of the Netherlands. The alliance, known as Unidata, seemed a promising venture in European cooperation, but was dogged from the start by disagreements among the partners about product strategy and about organisation.
One solution might have been to convert the loose alliance into a separately managed, freestanding company, but since the French share of such a company would be not much more than 25 per cent, this was not acceptable to the French authorities. Unidata was dissolved in 1974.56 The French authorities were then able to negotiate what appeared to be a more satisfactory arrangement – a rapprochement with Bull. This company had been owned by General Electric since 1964, but like RCA, GE had found it impossible to compete profitably against IBM, and in 1970 it sold its computer division, including Bull, to another American company, Honeywell. Honeywell was willing to participate in a new French company in which French interests would hold a majority of the shares. Under an agreement signed in 1975 (after Valéry Giscard d’Estaing had taken over the presidency of the Republic) CII-Honeywell-Bull was formed, with Honeywell holding 47 per cent and the balance by the French government and by CGE. (The two other shareholders in CII, Thomson and Schneider, withdrew). The Giscard administration was less enamoured of national champions than its predecessors, and although the government agreed to support the new company over a four-year period the expectation was that it would then be able to stand on its own feet.