Can industrial Europe be saved?

cadot layout 18/10/02 5:06 PM Page iii

Can industrial Europe be saved?

Olivier Cadot & Pierre Blime

cadot layout 18/10/02 5:06 PM Page iv

iv

cadot layout 18/10/02 5:06 PM Page v

v

cadot layout 18/10/02 5:06 PM Page vi

vi

ABOUT THE AUTHORS Olivier Cadot is Assistant Professor of Economics at INSEAD, the business school in Fontainebleau near Paris. Having trained at Princeton and McGill universities, he has carried out research for both the European Commission and the French government. His publications have focused on international trade, technology policy and environmental policy. Pierre Blime is an independent consultant who has written on trade policy in developing countries. He has worked on international relations at Harvard University’s Kennedy School, and on industrial policy at INSEAD.

ACKNOWLEDGEMENTS Several officials at the European Commission kindly provided background material; special thanks are due to Jerôme Vignon.

★ The views expressed in this publication are those of the authors themselves and do not necessarily reflect the opinions of the CER or of its trustees.

cadot layout 18/10/02 5:06 PM Page vii

vii

Contents

About the authors

vi

Acknowledgements

vi

1 Introduction

1

2 Industrial competitiveness: legitimate concern or dangerous obsession?

3

3 European industry’s performance in global markets: a basic diagnosis

7

4 Europe’s dwindling surpluses: a consequence of macroeconomic disequilibria?

15

5 The failure of Europe’s high-tech industries

25

6 Globalisation, jobs and the welfare state

33

7 Conclusion

43

cadot layout 18/10/02 5:06 PM Page viii

viii

cadot layout 18/10/02 5:06 PM Page 1

1

1 Introduction

I remain a Euro-optimist. Europe retains its strategic position. It has the world’s largest market; it will 1 Lester write the trading rules for the twenty-first century; it Thurow, has a unique set of complementary talents not found “Head to in either North America or the Pacific Rim. It may Head: The not win the race for the 21st century, but it certainly coming starts from the strongest position.1 economic battle among Japan,

Over the last 20 years, the European economy’s Europe & potential rate of growth has shrunk; unemployment America”, has been steadily rising from cycle to cycle; the Nicholas investment ratio has fallen by five percentage points; Brealey, 1994 our competitive position in relation to the USA and Japan has worsened as regards employment, our shares of export markets, R&D and innovation .... European Community industry improved its position on markets experiencing slow growth (railway equipment, cotton, textile and sewing machinery, miscellaneous textiles, 2 “Growth, tanning and dressing, animal slaughter and meat Competitiveness, preparation, grain processing and ethyl alcohol & Employment”, distillation) while its performance deteriorated on European markets with high value-added such as office Commission, automation, information technology, electronics, 1994 and medical and surgical equipment.2 These two opposing views on Europe were expressed in the same year, 1994, respectively by Lester Thurow, Dean of MIT’s Sloan School, and by the European Commission in its White Paper. Who is right? One would like to think Dean Thurow. Unfortunately, the trade data seems to support the commission’s view. Industrial Europe is facing declining market shares in most sectors, and our performance in crucial highgrowth industries such as electronics is particularly worrying. The questions are: why, and what can be done about it? This essay argues that

cadot layout 18/10/02 5:06 PM Page 2

2

Can industrial Europe be saved?

one of the most popular explanations for our industrial difficulties, high labour costs, is far from being the whole story. Europe has been suffering, over the last quarter-century or so, from a wide array of ills. Many continental European countries have been handicapped by overvalued exchange rates, which, together with interest rates that were until recently much too high, have stifled export growth and investment. And Europe’s failure in the electronics industry seems to have had more to do with wrong-headed industrial policies and poor management than high labour costs. Is the problem of Europe’s lacklustre industrial performance related to that of its persistently high unemployment rates? If one looks at Japan, the answer seems to be Yes: market share gains enabled Japan to keep large numbers of blue-collar jobs that were lost in Europe. But if one looks at America, the answer appears to be No: service-sector job creation has more than offset manufacturing job destruction. As Europe is, at this point, unlikely to regain the blue-collar jobs it has lost, its challenge is to remove the obstacles to taking on employees in services, especially for the young. Some critics, especially in America, argue that this implies a major dismantling of the welfare state, on the grounds that it is costly and sets the wrong incentives. This critique is far from warranted. Europe does need some labour-market deregulation, but not necessarily across the board, and even less a frontal attack on its welfare state. Recent British experience suggests that however aggressively labour-market deregulation is pursued, it will not bring back full employment unless part of a sustainable growth package. If full employment is to return—and any statement in this regard should be made with humility—we may need both structural reforms and sustained expansionary macroeconomic policies at the pan-European level. First of all, we shall examine Europe’s competitive position, starting with a brief review of the conceptual issues.

cadot layout 18/10/02 5:06 PM Page 3

3

2 Industrial competitiveness: legitimate concern or dangerous obsession?

According to the Ricardian theory of comparative advantage, countries with abundant and cheap labour tend to specialise in labour-intensive industries such as garments, whereas countries with, say, a highlyeducated labour force tend to specialise in high-tech industries. The resulting pattern of international specialisation is economically efficient, and every single country benefits from trading, irrespective of 3 Paul Krugman, what it ends up producing. Thus, in the words of a former “Competitiveness: adviser to President George Bush, “it does not matter whether a dangerous a country produces computer chips or potato chips”.3 All that obsession”, matters is domestic productivity growth, which determines the Foreign Affairs evolution of living standards, and one country’s optimal 73, 1994 economic policy—necessarily a variant of laissez-faire—is largely independent from what other countries are doing. So why bother about ‘competitiveness’? In fact it is crucial for several reasons. When a country loses the ability to export successfully, economists would often argue that the simplest cure would be for the country to devalue its currency. But a devaluation is—rightly or wrongly—perceived by voters and the financial community as a failure; moreover, it inflicts a shock on the domestic economy, raising inflation and possibly unemployment as well. Understandably, few governments are willing to go down that route, except as a last resort. The only alternative is to finance the country’s imports by borrowing abroad. What’s wrong with that, one might ask? Well, borrowing is never a sustainable way of financing consumption. For a household, it may be necessary in order to spread the costs of a large one-off purchase. For a country, it makes sense only in two types of circumstances: as a way of smoothing out a temporary disturbance—say, the effects of an earthquake; or as a way of financing productive investment that will pay for itself in the form of increased exports. But it makes no sense as a way

cadot layout 18/10/02 5:06 PM Page 4

4

Can industrial Europe be saved?

of paying for a steady flow of imports, which are not matched by exports. Furthermore, a country running persistent trade deficits will soon lose the confidence of financial markets, inducing capital flight, which the central bank will have to fight with high interest rates, losing its ability to run the country’s macroeconomic policy independently. Worse, this kind of fight is invariably lost, leading to brutal currency crises and their painful consequences. In sum, no country can ignore ‘external constraints’. It may also matter, after all, whether a country specialises in computer chips or potato chips and other snacks. In his recent book, Lester Thurow wrote: Ask Japan, Germany and the United States to name those industries that they think are necessary to give their citizens a world-class standard of living in the first half of the twenty-first century, and they will return remarkably similar lists—microelectronics, biotechnology, the new materialsscience industries, telecommunications, civilian aviation, robotics plus machine tools, and computers plus software.What was an era of niche competition in the last half of the twentieth century will become an era of head-tohead competition in the first half of the twenty-first century. Niche competition is win-win. Everyone has a place 4 Lester where they can excel; no one is going to be driven out of business. Head-to-head competition is winThurow, op. cit. lose. Not everyone will get those seven key industries. Some will win; some will lose.4 Thus the somewhat idyllic view of countries smoothly specialising in what they are good at may not be quite appropriate in a world where some types of specialisation are more desirable than others. There are two basic arguments here. First, if productivity is growing faster in some industries than others, countries which specialise in those industries will see their standards of living rise faster than others. However attached the French are to their agriculture, few politicians in Paris would be thrilled by the idea of France becoming solely an exporter of wine and cheese. Second, the demand for some types of products—the high-tech ones, it turns out—grows faster than that for other products. As a result, a

cadot layout 18/10/02 5:06 PM Page 5

Industrial competitiveness: Legitimate concern or dangerous obsession?

5

country locked into low-growth sectors would see its exports and GDP grow slower than others, and vice versa. In other words, there are “good specialisations” and “bad specialisations”. Recognising this has 5 See for instance formidable implications for economic policy, according to some Laura D’Andrea American economists.5 One country’s predatory tactics may Tyson, “Who’s permanently alter not only its own pattern of industrial Bashing Whom?” specialisation, but also that of its trading partners. As the world Institute for divides into powerful blocs, each aggressively promoting its International national champions in strategic sectors—a policy epitomised, Economics, 1992 for many Americans, by Europe’s Airbus—the United States, they argue, rather than clinging to an outdated model of 6 See Alexis unilateral free trade, should simply do the same. Although few Jacquemin, “Les European economists would recognise this aggressive vision of enjeux de la international trade as theirs,6 some of its implications cannot be compétitivité européenne...”, entirely shrugged off. Revue du Marché Commun et de l’Union Européenne , Mars 1996

For one thing, contrary to fashionable belief, manufacturing still matters: no country can live on its citizens laundering each others’ shirts. In order to pay for necessary imports, some hard products must be made and exported. Those hard products can be made in so-called “lights-out factories” (factories with no one inside, so they don’t need to be lit), which means that manufacturing employment can decline. However, manufacturing output cannot go down indefinitely. Furthermore, even when one does not subscribe to the grim view of international trade to be found in Thurow’s Head 7 Clyde V. to Head or Prestowitz’s Trading Places, 7 comparing the Prestowitz, Jr, performance of Europe’s corporations against those of Japan Trading places: and America remains a useful exercise. In a world of globalised “How we are capital markets, Europe’s ability to attract the capital necessary giving our future to finance its growth—be it in manufacturing or in services— to Japan and how depends on the quality of its infrastructure, labour force, R&D to reclaim it”, capabilities, and so on. A simple way of assessing the quality of Basic Books, these factors of production is to look at how firms located in 1988 Europe, which draw on those factors, perform on global markets. One last caveat is called for. When speaking of the competitiveness of European firms, two perspectives can be taken, depending on which definition of “European” one chooses. The first consists of defining a firm as European if it is European-owned. According to this definition,

cadot layout 18/10/02 5:06 PM Page 6

6

Can industrial Europe be saved?

the relocation of production, say to a developing country, so long as it is undertaken by a European firm, does not harm European competitiveness, even though it means a reduction in the number of European jobs. The second definition of competitiveness focuses on Europe’s attractiveness as a site of production, irrespective of the ownership of the firms operating on its soil. In this perspective, outward job relocation is a loss of competitiveness. As long as one is concerned about jobs, clearly the latter approach is relevant; but if one is concerned about the survival of European firms, or by shareholder welfare, it is the former. The difference between the two approaches has very important implications for industrial policy, which can be illustrated with an example. At the end of the 1970s, the British and French automobile industries suffered from similar ailments—low productivity, hostile labour relations, and an unimpressive quality record. Both urgently needed some kind of medicine; but the British and French governments chose differently. The British government’s gamble was to transform Britain into an attractive place for producing automobiles; in order to do this, it would welcome whoever was ready to pick up the challenge, which meant, at that time, the Japanese. The French government, by contrast, chose to bet on turning round its own carmakers, with the help of quota protection and subsidies. Given their respective objectives, it can be said that both governments succeeded: Britain is back among automobile-making countries, with modern factories, young, non-union labour, and a record of quality and high productivity. France’s newly privatised Renault makes profits producing cars that sell in Germany and win reliability contests, having— under state ownership—introduced Japanese-style work practices in factories that used to be outposts of the Soviet Union. But clearly the British and French approaches had starkly different implications for both shareholders and workers; in fact, it is an irony that a British Conservative government chose the job-focused approach, while a French Socialist one chose the shareholder-focused one. Throughout the analysis below, we will take the ‘job-focused’ approach and consider the exports of goods produced in Europe, irrespective of the producer’s ownership, as European.

cadot layout 18/10/02 5:06 PM Page 7

7

3 European industry’s performance in global markets: a basic diagnosis

Although opinions vary on the severity of Europe’s competitiveness problem, concern is fuelled by figure 1, which shows the EU’s importcoverage ratio for manufactured products over the period 1967-93. (A country’s import-coverage ratio is that country’s exports as a percentage of its imports; a value above one hundred means a trade surplus, whereas a value below one hundred means a trade deficit.) FIGURE 1: EU’S IMPORT COVERAGE RATIO MANUFACTURED GOODS, 1967-93 140

I MPORT- COVERAGE RATIO IN %

135

130

125

120

115

110

105

100 67

69

71

73

75

77

79

81

83

85

87

89

91

93

SOURCE: CEPII- CHELEM NOTES: ALL TRADE FIGURES REFER TO EXTRA-EU TRADE THE EU IS DEFINED AS THE EU-15 FOR ALL YEARS IN THE SAMPLE

cadot layout 18/10/02 5:06 PM Page 8

8

Can industrial Europe be saved?

It is apparent from figure 1 that the EU’s competitive position in manufactured goods has been deteriorating throughout the 1980s, up to the beginning of the 1990s, when it recovered sharply. The 1993 recovery, unfortunately, cannot been interpreted as really good news, since a deep recession like the one Europe had just been through reduces demand for imports, inducing a quasi-mechanical trade-balance improvement. This kind of phenomenon is likely to vanish as soon as the recovery takes hold—if it ever does. Since manufactured products represent over 90 per cent of the EU’s trade with the rest of the world, and manufactured exports are used to pay for imports of fuels and raw materials, this deterioration of the EU’s market position is clearly worrying. Furthermore, OECD data on trade in services does not suggest that service-trade surpluses have the potential to replace Europe’s dwindling manufactured-good surpluses in the foreseeable future, at least for continental Europe (Britain might be the only exception in the EU). Thus, unless the improvement of 1992-93 proves long-lived, the trend apparent during the 1980s will not be sustainable: the EU would soon have to borrow either against future export earnings to pay for imports, as the United States has done throughout the 1980s; or let its currencies (or currency, if unique) depreciate, impoverishing its residents in the process (the United States is in fact doing both). Has this declining trade performance been associated with higher growth in the EU? If it had, there would be nothing to worry about, as a longrun growth slowdown (as distinct from a short-lived recession like that of 1993) would be enough to bring Europe’s foreign trade back into equilibrium. The answer is ambiguous. Over 1980-93, real per capita income rose by 25.8 per cent (an average of 1.65 per cent a year) in the EU 15, against 19.8 per cent (1.3 per cent a year) in the United Status and 52.4 per cent (3.05 per cent a year) in Japan. Thus, Europe did worse than Japan, but nevertheless slightly better than the United States (the EUUnited Status growth differential at the end of the 13-year period is roughly two years’ worth of growth). Compared to the whole OECD, the EU’s growth performance is almost exactly average. The conclusion is that something more than just a growth differential is required to explain Europe’s trade performance. The geographic pattern of EU manufacturing trade shows that Europe, like the other two blocs, lost considerable ground in emerging markets,

cadot layout 18/10/02 5:06 PM Page 9

European industry’s performance in global markets

9

primarily to non-Japanese Asian producers (Japan has been losing market share at the same rate as Europe since the mid-1980s). As Europe used to be the Third World’s main supplier of manufactured products, accounting at the beginning of the 1970s for 40 per cent of its imports, it stood to lose most from the recent development of South-South trade. The 15 per cent market share it lost during the last two decades went primarily to the four Asian Tigers (South Korea, Hong Kong, Singapore and Taiwan). But the EU has also been losing ground steadily in the North American market, from roughly a third of American manufactured imports at the beginning of the 1970s to little more than 20 per cent today, while Latin America and—even more—non-Japanese Asia were posting steady gains. The sectoral pattern of EU trade revealed by figure 2 is also striking. The sectors which have had the highest growth rates in world trade are also the sectors where the EU is weakest, as measured by its trade balance. Productive resources seem to have been channeled into the wrong sectors, suggesting either distorted incentives or, worse, a structural weakness in FIGURE 2: E UROPE’S SPECIALISATION IN INTERNATIONAL TRADE, 1993 200 Specialised machinery

EU’S-15 IMPORT- COVERAGE RATIO,

1993

Shipbuilding

Machine tools Const. equip

150 Agri. equip

Mechanical Engines

Pharm. prod

Iron/steel Vehicles Chemicals Aeronautics Precis. instr

Arms

Electrical Telecom equip.

Wood/paper

100 Food & agri

Textile

Electronic comp Computer equip

Consumer electronics

50 0

5

10

15

20

ANNUAL RATE OF GROWTH OF WORLD IMPORTS (DOLLARS), 1983-93

SOURCE: CEPII- CHELEM

cadot layout 18/10/02 5:06 PM Page 10

10

Can industrial Europe be saved?

the factors on which high-growth industries depend, such as skills, R&D capabilities and risk capital. We will return to the specific problem of high-tech industries in section 4, since Europe’s failure to establish solid positions in those industries is by far the most disturbing symptom of its decline. But our difficulties are more widespread. Varying degrees of deterioration in the EU’s market positions are also apparent in industries which traditionally formed the core of its manufacturing exports: automobiles, machinery and chemicals. Let us consider two key sectors, automobiles and machine tools, which illustrate the differing fortunes of EU exporters in the last twenty years. In automobiles, the EU has been losing market share ever since the mid1960s, steadily retreating from what used to be one of its main outlets: the North American market. Figure 3, showing the EU’s bilateral trade balance in the automobile FIGURE 3: EU’S BILATERAL AUTOMOBILE TRADE BALANCE WITH SELECTED PARTNERS, 1967-1993

BILATERAL TRADE BALANCE AS % OF EU GDP

4

3 United States

2

1 Latin Am. Asia (excluding Japan) 0

Japan -1

-2 67

69

71

73

75

77

79

81

83

85

SOURCE: CEPII- CHELEM

87

89

91

93

cadot layout 18/10/02 5:06 PM Page 11

European industry’s performance in global markets

11

sector (expressed as a percentage of its overall GDP in order to exclude inflation), illustrates a troubling phenomenon. The confrontation with Japanese carmakers in the North American market, which could be considered neutral ground, ended badly. European cars, which had up to the mid-1980s successfully occupied several niches left open by Detroit, such as luxury cars, super-minis and minis, proved, when compared to what the Japanese could offer, fairly expensive for their level of reliability and finish. While their sophistication was thought to be a strong selling argument, they often ended up being “over-engineered” for North American use and driving conditions. The EU’s automobile trade balance with Japan, which deteriorated right through the 1970s, stabilised in the 1980s. Were EU manufacturers waking up? Unfortunately, the answer probably has more to do with the “voluntary export restraint agreements” that EU member-states negotiated with Japan. Had protectionist measures not been adopted, the result of the confrontation with Japan might have been even worse, as suggested by the results of an international study of the automobile industry published by MIT in 1989.8 Although the core of the study was a comparison of production processes in America and Japan, 8 F. Jones & the European data suggested that European carmakers were lagging significantly behind their Japanese competitors in terms F. Wollack, “The Machine of production process improvements. Thus the European that Changed Commission’s 1991 agreement with Japan (promising full the World”, liberalisation of the EU automobile market by 1999) might MIT Press, appear, in the light of the American experience, a rather risky 1989 move, given the number of jobs at stake and the potential overcapacity created by massive Japanese investment in Britain. On the other hand, experience suggests that firms do not shape up under permanent protection. In fact, it may be argued that protected carmakers such as Renault and Peugeot finally grew leaner only when the threat of liberalisation became credible. The picture is more complex in machine tools, a key export item, particularly for Germany. Figure 4 shows the evolution of the EU’s bilateral trade balance with some of its partners in machine tools. The turning point for machine tools, as with several other sectors, seems to have been the mid-1970s, when the EU started losing ground to the Japanese in Asian markets. At that time, Japanese machine tool producers,

cadot layout 18/10/02 5:06 PM Page 12

12

Can industrial Europe be saved?

FIGURE 4: EU’S BILATERAL MACHINE TOOLS TRADE BALANCE WITH SELECTED PARTNERS, 1967-1993

BILATERAL TRADE BALANCE AS % OF EU GDP

4 Asia (excluding Japan) 3 Latin Am. 2

1

0

United States

-1

Japan -2 67

69

71

73

75

77

79

81

83

85

87

89

91

93

SOURCE: CEPII- CHELEM

having acquired the industry’s basic know-how through licensing agreements with American firms, pioneered the use of computer-numerical controls, which allowed them to sell cheaper, better and more flexible machines than their European or American counterparts. During the first half of the 1980s, when the dollar was significantly overvalued, the retreat of North American producers gave extra pickings to EU manufacturers, helping them to limit the damage they had suffered from Japanese competition, and hiding some of their weaknesses. But as the dollar started its steady descent in the latter half of the 1980s, those weaknesses could not be hidden. Fortunately, the EU was able to maintain its positions in Latin America. Whereas the 1980s had been a disastrous decade for Latin America because of the debt crisis—which forced growth-stifling austerity policies on the whole continent—this traditional market for European exporters started picking up significantly in the early 1990s. Thus, the Latin American recovery, combined with the ongoing investment boom in Asia, helped Europe’s machine tool makers to limit the global erosion of their

cadot layout 18/10/02 5:06 PM Page 13

European industry’s performance in global markets

13

market positions right up to the early 1990s. By then, having significantly shaped up and maintained very strong positions in specialised machinery, they were able to reverse the downward trend with Japan. Thus, the situation in machine-tools is more complex and certainly not as alarming as that for automobiles. In fact, these two sectors illustrate the fairly broad range of interpretations that one can give to trade data. Nevertheless, it remains clear that Europe’s strongholds in traditional manufacturing sectors are unlikely to provide a viable alternative to the missed opportunities in high-tech industries. In sum, the EU is witnessing a trend that should be a cause for concern. First, it seems to be slowly locking itself into the wrong type of specialisation; namely, one in which low and medium-tech industries dominate. Of course, many observers would argue that European products in the mechanical sector are in fact high-tech products. But that is not the point. Rather, it is that low- and medium-tech industries do not grow as fast as high-tech ones, or at least have not in the recent past, as figure 2 shows. Under such conditions, competition in those industries increasingly takes the form of a zero-sum game where savage cost-cutting and downsizing are the key to corporate success, as they simply do not generate the expanding profits that hightech industries do. Second, Europe’s manufacturing trade surpluses with the rest of the world are dwindling overall. As Europe is still a net creditor to the rest of the world, it can afford to run trade deficits and pay for them with income from its foreign assets. But this process ultimately has to come to an end. As the United States is discovering, no country can live forever beyond its means; likewise, sooner or later, the EU will have to stop the slide of its foreign trade into the red. But how did this slide come about in the first place?

cadot layout 18/10/02 5:06 PM Page 14

14

Can industrial Europe be saved?

cadot layout 18/10/02 5:06 PM Page 15

15

4 Europe’s dwindling surpluses: a consequence of macroeconomic disequilibria?

When the United States ran large trade deficits in the mid-1980s it was fashionable in Washington to accuse Japan and Germany of unfair trade practices. The expression “Japan, Inc.” was coined to subsume a galaxy of industrial conglomerates, banks and MITI bureaucrats all colluding to deny foreigners access to overpriced Japanese markets. Protectionist politicians like Richard Gephardt came to the fore, demanding protection for American workers and unrestricted access to foreign markets. But economists had another explanation for the deficits. In general, they pointed out, trade imbalances reflect macroeconomic imbalances, and the best way of understanding the direction of flows of goods and services is to look at capital movements. The economists’ argument went on as follows. Because of President Reagan’s 1981 tax cuts and the simultaneous boost to defence spending, the federal government started in the early 1980s to borrow heavily on domestic and foreign financial markets. Since American households were reducing their savings rate, foreign savings had to come to the rescue; German and Japanese savers ended up, directly or indirectly, financing the American budget deficit. Money flowed into America to be invested in Treasury bills, attracted by sky-high interest rates, and the dollar soared. The strength of the dollar in 1985, far from representing—as the business press was claiming at the time—the renewed dynamism of the American economy, reflected the magnitude of the macroeconomic imbalances forced upon it. As a result, American products became grossly uncompetitive in world markets and the American trade deficit ballooned. The process reversed itself when foreign investors started worrying about possible capital losses on their dollar-denominated assets. As Japanese financial institutions became increasingly reluctant to buy American Treasury bills, the dollar started to lose altitude. This—among other

cadot layout 18/10/02 5:06 PM Page 16

16

Can industrial Europe be saved?

factors—forced Congress to face reality and cut the deficit, which, as a proportion of GDP, had anyway shrunk because of the American economy’s rapid growth over the period 1983-91. But in the process the United States switched from the world’s first creditor nation to its largest debtor. Since debts have to be serviced, in the long run, America must now generate trade surpluses to prevent its foreign debt from growing indefinitely. Some progress has been made, thanks to the depreciation of the dollar and appreciation of the yen: the Japanese goods that American consumers have become fond of are now significantly more expensive than they were ten years ago. In other words, since the bill has arrived, but only a small part been settled, the Americans will be busy making payments on it for some time to come. Are we seeing in Europe a replay of the same story? Can the dwindling EU trade surpluses be traced to similar—if, perhaps, less violent— macroeconomic imbalances? Figure 5 shows the evolution over the last 20 years of the real effective exchange rate of three currencies: the dollar, the D-mark and the yen. FIGURE 5: REAL EFFECTIVE EXCHANGE RATES 1975-1994 160

I NDEX 1975 = 100

140

Yen +61% over 1984-94

120 D-mark + 41% over 1984-94 100

US$ -45% over 1984-94

80

60 75

77

79

81

83

85

87

89

91

93

SOURCE: IMF, INTERNATIONAL FINANCIAL STATISTICS

cadot layout 18/10/02 5:06 PM Page 17

Europe’s dwindling surpluses

17

The dollar line clearly shows the overvaluation of the mid-1980s, the source of America’s trade deficit, and the subsequent depreciation, the source of the renewed dynamism of American exporters. But the evolution of the D-mark is equally spectacular: whereas the dollar lost 45 per cent of its value in real effective terms over 1984-94, the D-mark gained 41 per cent. Thus a German machine tool is 41 per cent more expensive on the German exporter’s average market, which goes a long way towards explaining the difficulties of German exporters in all sectors. It is not difficult to imagine that however fast the productivity growth, however savage the cost-cutting, and however thorough the quality upgrading, a cost disadvantage of that magnitude cannot be easily overcome. As the EU’s external trade data is to a large extent dominated by German extraEU trade, the D-mark effect on EU trade is very strong, even though several currencies of the European Monetry System have been devalued against the D-mark. Like the dollar appreciation of the 1980s, the D-mark appreciation of the 1990s has been hailed by the business press as an unambiguous sign of FIGURE 6: I NTEREST RATES AND CAPITAL FLOWS, G ERMANY 1974-1994 100 80

12

Long-term interest rate (right-hand scale) 10

60

US$ BILLION

8

20 6 0 -20

4

-40 net capital flows (left-hand scale)

2

-60 -80

0 75

77

79

81

83

85

87

89

91

93

SOURCE: IMF, INTERNATIONAL FINANCIAL STATISTICS

P ERCENTAGE POINTS

40

cadot layout 18/10/02 5:06 PM Page 18

18

Can industrial Europe be saved?

Germany’s new power and assertiveness. Americans blamed Japan’s unfair trade practices for their inability to export successfully; today the Germans blame their welfare state. In both cases, observers fail to relate imbalances on product markets to imbalances on capital markets. (Figure 6 shows net capital flows in and out of Germany.) The years 1990-92 featured a spectacular reversal in the direction of capital flows. Whereas Germany was until 1990 a net exporter of capital—the magnitude of the annual outflow reaching 70 billion dollars by 1988—within two years, under the shock of reunification, the country had become a massive capital importer. This sudden imbalance on world financial markets sent the D-mark skyward (while further depressing the dollar) as investors bought billions of D-marks on foreign-exchange markets to invest in Germany. Three factors contributed to making the process in Germany distinct from that in the United States ten years earlier. First, the reason for the capital inflow was different; in the German case, it was not just the federal government borrowing to finance its deficit, but also private agents investing in East Germany. Second, the resulting upward pressure on the D-mark was mitigated by the fixedparity system of the EMS—which proved nevertheless unable, in the end, to withstand the pressure. On the other hand, and this is the third factor, the upward pressure on the D-mark was aggravated by a long-run trend towards the yen and the D-mark replacing the dollar as international reserve currencies. But, by and large, the process and its impact on foreign exchange markets were much the same. Unsurprisingly, the consequences for Germany’s foreign trade were also similar. Indeed, EU trade data, when broken down for each member, reveals that Germany has been hardest hit by the erosion of market-share. Recognising the role of currency appreciation in the EU’s trade performance has fundamental implications for both the diagnosis and the remedies that follow. Whereas Germany still runs surpluses, albeit shrinking ones, with non-EU trading partners, other countries such as France, with currencies on the same trajectory as the D-mark, clearly suffer from currency overvaluation. If this trend were to persist, the EU as a whole would end up suffering from the same syndrome as the United States in the 1980s, namely living beyond its means. But will the trend persist? In the United States, Congress’s inability to tackle the budget deficit reinforced it. In Europe, fortunately, the Maastricht Treaty’s

cadot layout 18/10/02 5:06 PM Page 19

Europe’s dwindling surpluses

19

convergence criteria impose drastic reductions on the budget deficits of member states. Thus those much-decried convergence criteria do make economic sense, if one is concerned by the prospect of currency overvaluation and growing trade deficits. On the other hand, they force austerity measures on European countries at a time when they are suffering from slow growth and high unemployment. Is this a sensible choice? In other words, would a domestic consumption boom, fuelled by American-style budget deficits, be a way out of our current economic problems? There is no final answer to this question, but such an approach would likely be, in the end, extremely costly. In the long run, growth in per capita incomes is determined by investment and R&D; neither of these picked up significantly in America during the Reagan-Bush years. On the other hand, that country’s foreign debt is increasing exponentially: the United States is borrowing against the future. Keeping macroeconomic policy under control, as the Maastricht Treaty requires, provides a sort of guarantee against such excesses. Whereas the role of currency appreciation has undoubtedly been crucial to the EU’s trade performance, other macroeconomic factors have also been at work. In 1973 and 1979, the world economy was hit by oil shocks. These shocks, which increased the cost of energy to manufacturing firms, had the effect of reducing the proportion of sales revenue—value added—available to remunerate factors of production such as capital and labour. The question then was, how to allocate the burden of adjustment between capital and labour? In several European countries, particularly France, government policy—under pressure from pugnacious unions—set as its primary objective the defence of the working class’s purchasing power. Thus profit margins were squeezed. Business profits were further reduced by a brutal transition, in the early 1980s, from an inflationary environment in which negative real interest rates induced firms to borrow heavily, to a deflationary environment characterised by painful real interest rates. The result was a wave of bankruptcies and layoffs in the most vulnerable sectors. Firms which were able to withstand the shocks then spent most of the decade scrambling to clean up their balance sheets and to reduce indebtedness, and were hardly in a position to pursue market share aggressively. Japanese conglomerates, by contrast, largely sheltered from the widespread profit squeeze by tacit collusion in their home market, were able to

cadot layout 18/10/02 5:06 PM Page 20

20

Can industrial Europe be saved?

conquer markets worldwide. Access to long-term capital within the Keiretsu structure also allowed them to invest counter-cyclically, giving them a strong advantage at the beginning of each recovery, especially that of 1983. The Japanese strategy of investing in capacity during recessions was particularly conspicuous in the memory sector of the semiconductor industry, where it paid off in a spectacular way. The opposing strategic choices of Japanese and French businesses, in terms of profitability and market share during the 1980s, are illustrated in figure 7. FIGURE 7: PROFITABILITY AND MARKET SHARE IN MANUFACTURING (1980-92)

CHANGE IN WORLD MARKET SHARE (%)

3.5

Japan

3.0 2.5 2.0 1.5 1.0 0.5 Australia

0.0

Spain

Austria UK

Switz

-0.5

Belgium

Netherlands

Germany

Italy

France US

-1.0 -2

0

2

4

6

8

10

12

CHANGE IN THE PROFIT RATE (%)

SOURCE: OECD

Figure 7 is a useful reminder of the fact that market share alone is by no means the only reliable indicator of business performance. It is a matter of business strategy whether a firm goes for market share or profits; and which objective should be given priority in the short run depends, among other things, on the particular macroeconomic environment that the firm is facing. In fact, in the long run, the ultimate objective of the firm is, and

cadot layout 18/10/02 5:06 PM Page 21

Europe’s dwindling surpluses

21

should always be, to maximize shareholder value, i.e. expected profits; market share expansion is only an intermediate step towards the fulfillment of that objective. Viewed from this perspective, the behaviour of EU FIGURE 8: EXPORT MARKETS BY RATE OF GROWTH, 1980-1990 A: E UROPE more than 6% per year (11%)

less than 2% (19%)

between 3 and 6% per year (23%) between 2 and 3% per year (47%)

B: JAPAN

less than 2% (7%)

more than 6% per year (30%)

between 2 and 3% per year (52%)

between 3 and 6% per year (11%)

SOURCE: OECD, WORLD BANK & AUTHORS’ CALCULATIONS

cadot layout 18/10/02 5:06 PM Page 22

22

Can industrial Europe be saved?

market shares reflected the adjustment to the adverse business environment that firms had to suffer during the decade 1974-84. So, did the sharply improved environment that they enjoyed during 1984-91 enable them to take a less defensive position? This is precisely what the trend reversal of the early 1990s (before the severe 1993 depression) suggested. A further part of the explanation for European industry’s lacklustre performance during the last decade is provided by figure 8. The pie charts in figure 8 rank Japan’s and the EU’s export markets by their average growth rate over the decade 1980-90; thus, 11 per cent of the EU’s exports went to countries having recorded average GDP growth rates above 6 per cent per year during 1980-90, and so forth. Japan happens to be located in the middle of the world’s most dynamic zone, so that its exports are naturally pulled up by its neighbours’ GDP growth. Of course, there is a feedback effect, as Japanese investment has been instrumental in lifting the region out of poverty. But, on the other hand, Japan has never been a really open market, so that it is the United States which provided the large, dynamic, open market on which the Asian Tigers could sell their products; in that sense, Japan has been freeriding on the United States. Europe has not been able to benefit from similar growth synergies with its periphery, which consisted, before the fall of the Berlin Wall, of lowgrowth economic zones in Africa and the Middle East. The good news, in this regard, comes from the opening up of Central and Eastern Europe, a zone with high economic potential, particularly in terms of human capital. If the experience of post-war Western Europe is any guide, human capital plays a key role in the economic take-off of nations. Provided that Central and Eastern Europe can attract the capital necessary to build its infrastructure, its growth potential is probably similar to that of Western Europe in the early 1950s—in other words, enormous. For this it needs— in addition to foreign capital—open access for its products to EU markets, and encouragement in the difficult process of economic reform. Whereas the 1991 Europe Agreements with the so-called Visegrad three (Poland, Hungary and what was then Czechoslovakia) did go some way towards opening up EU markets, a lot remains to be done in the sensitive sectors of chemicals, steel, agriculture and textiles.

cadot layout 18/10/02 5:06 PM Page 23

Europe’s dwindling surpluses

23

Substantial currency appreciation, adverse macroeconomic shocks, and a low-performance hinterland all contributed to the EU’s mediocre export performance throughout the 1980s. In fact, these three factors taken together go a long way towards explaining the trends apparent in trade data. But something else is also at work. How can we explain Europe’s particularly poor performance high-tech industries? If an adverse macroeconomic environment was the only source of difficulty for EU exporters, the damage should have been least marked—according to the theory of comparative advantage—in high-tech industries where the EU has a strong resource base (human capital, R&D capabilities, and so on). Yet it is in those sectors that the damage has been most severe, suggesting specific problems to which we now turn.

cadot layout 18/10/02 5:06 PM Page 24

24

Can industrial Europe be saved?

cadot layout 18/10/02 5:06 PM Page 25

25

5 The failure of Europe’s high-tech industries

It is commonplace to stress the importance of high-tech industries in modern society. By 1990 the worldwide turnover of information 9 European technology (IT) industries—defined as computer hardware and Commission, software, integrated circuits, industrial and office automation “European systems and consumer electronics—was close to one trillion industrial dollars. About a quarter of that output was in the EU, which policy in the meant that IT industries represented roughly 5 per cent of the 1990s”, 1991 Union’s GDP.9 On current trends, turnover in these industries is will grow by 15 per cent a year; it is expected to account for 10 per cent of the EU’s GDP by the turn of the century. However, the EU makes only three quarters of the IT it consumes. At the beginning of the 1990s it had a deficit of $40 billion—half in computers, $13 billion in consumer electronics and $7 billion in components. The EU’s weakness in semiconductors is especially worrying, since they are an important component of many other industrial goods—such as factory automation systems, aeroplanes and cars—and increasingly determine the performance of those downstream products. Dependence on external sources for such a key technology may be dangerous in industries that are characterised by rapid technical change and in which a technological lead is crucial to competitiveness. This is especially problematic when European firms compete with integrated conglomerates which produce both semiconductors and their downstream applications. It is commonly argued that Europe has, by and large, missed the boat on high-technology industries, declining drastically in some of them and never having got anywhere in others. Figure 9 compares indices of the “net market position” of the EU, the United States, Japan and the Asian Tigers (South Korea, Taiwan, Hong Kong and Singapore) in hightechnology industries, defined as pharmaceutical products, computers and electronic components, telecommunications equipment, consumer

cadot layout 18/10/02 5:06 PM Page 26

26

Can industrial Europe be saved?

electronics, precision instruments and aerospace. Whereas the United States has been coming down from a very strong position, the EU has roughly maintained its overall position over the sample period. It lost heavily in computers and telecoms equipment and never made it into semiconductors, but kept reasonably strong positions in pharmaceuticals and aerospace and even recovered, to some extent, in software. Thus Europe’s overall market position in high-tech industries, although certainly not outstanding, is not quite as desperate as is sometimes argued. FIGURE 9: EU’S NET MARKET POSITION, HI -TECH INDUSTRIES, 1967-1993 SECTORAL TRADE BALANCE AS % OR WORLD EXPORTS

0.25

0.20 United States

0.15 Japan 0.10

0.05

0.00 Asian Nics EU -0.05 67

69

71

73

75

77

79

81

83

85

87

89

91

93

SOURCE: CEPII- CHELEM

As already mentioned, Europe’s difficulties in high-tech industries would seem to contradict the theory of comparative advantage: Europe’s wealth of scientists, research labs, highly qualified workers and infrastructure should surely make it a natural location for such industries. Numerous hypotheses based on case studies, industry-wide surveys, and statistical data have been proposed to explain Europe’s failure to establish a core of high-technology industries that can compete worldwide. We will consider four in turn: the problem of R&D and business-university links; managerial issues; industrial policies; and the environment for high-tech start-ups.

cadot layout 18/10/02 5:06 PM Page 27

The failure of Europe’s high-tech industries

27

At an aggregate level, Europe’s R&D effort is weaker than that of its competitors, as it spends only 2 per cent of its GDP on R&D, against 2.7 per cent in the US and 2.9 per cent in Japan. Between 1987 and 1993, public funding of R&D amounted to $90 billion in the EU, against more than $220 in America. Worse, of this smaller European pool of publiclyfunded R&D, only 14 per cent was carried out by private firms, against 27 per cent in the United States.10 Although in terms of basic science (as measured by the quantity of scientific articles) Europe is very 10 Alexis active, its production of commercially useful knowledge seems to Jacquemin, op. be significantly lower than that of the United States. For instance, cit. patent counts (which are one measure of R&D output) suggest that Europe’s share in the production of knowledge is steadily decreasing, whereas Japan’s share is growing. The raw numbers hide differences which are perhaps even more important in determining the impact of research on commercial technology development. European universities are, to a large extent, funded directly by the public sector, whereas many—and particularly the best—American universities are funded by business donations. In the United States such donations benefit from generous tax treatment, so a large part of the money effectively comes from the taxpayer in both systems. However, the American system favours the establishment of strong links between business and academia, ties which are much more tenuous and distant in Europe. In the United States the effectiveness of those links is reinforced by a strong tradition—going back to the beginning of this century—of empirical and business-oriented research. The importance of ties between business and academia is illustrated by the story of the development of high-definition television standards in America and Europe. After successfully blocking the adoption of the Japanese high-definition transmission standard, Muse, as a world standard in 1986, the Europeans scrambled to develop their own, the D2Mac/HD Mac family of standards. A consortium led by Phillips and Thomson chose an analogue standard, at that time the only feasible option. But in 1991 a joint team of researchers from MIT and General Instruments, an American defence contractor with expertise in encoding and decoding, showed that the bulky high-definition signal could be compressed enough for it to be transmitted digitally over conventional media. Suddenly, digital transmission—offering vastly improved

cadot layout 18/10/02 5:06 PM Page 28

28

Can industrial Europe be saved?

performance and multimedia compatibility—was possible. 11 The Europeans had gone down the wrong route. Fortunately, the breakthrough came early enough to stop the development of D2 11 See Olivier Mac, in spite of pressure from the Phillips-Thomson lobby to Cadot & H continue at all costs. But the episode illustrated the power of Landis Gabel, high-level university-business co-operation in setting the “High-definition direction of technological change. TV in Europe”, INSEAD, 1994

While the efficiency of higher education systems varies considerably across Europe, the weakness of business-university links in France is aggravated by the fact that most of the business and government elite is trained through a separate system of so-called Grandes Ecoles. The men and women who pass through this system have no loyalty to the universities upon which they often look down. Furthermore, having never been exposed to university research, they tend to not really understand what it is about (this is particularly true of the Ecole Nationale d’Administration). Worse, the Grandes Ecoles receive close to one third of French higher education funding, yet turn out only 4 per cent of degrees, thus depriving the universities of their fair share of resources. A different line of argument, mainly managerial, was stressed in a recent McKinsey study of the global electronics industry.12 McKinsey 12 Hubert Joly, researchers found that European electronics firms carried cost Jürgen Kluge disadvantages of up to 50 per cent compared to their Japanese and Lothar competitors. Of course, international cost differences often reflect Stein, “Europe’s exchange-rate misalignment and must therefore be treated with structural caution. But the interesting finding was that, typically, no more weakness”, than one fifth of that cost differential was attributable to higher McKinsey Quarterly, 1994 labour, capital, or material costs. The rest, four fifths, was attributable to operational inefficiencies, of which McKinsey analysts found ample evidence. Compared to their Japanese and American competitors, European firms generally failed to delegate responsibility for quality to line employees, and thus suffered high rates of defective products (for instance, twice as bad as the Japanese and three times worse than the Americans for printed circuit boards). Often, European firms rotated employees between sales and manufacturing, but seldom between manufacturing and R&D, as did the leading firms in this sector. Thus the Europeans often had designs that were difficult to manufacture and products that were unnecessarily complex. In the McKinsey sample, the

cadot layout 18/10/02 5:06 PM Page 29

The failure of Europe’s high-tech industries

29

number of parts per dollar of value in a European product was two to ten times the number for comparable American and Japanese products. Rigid and hierarchical organisations led to designs that could not be altered at later stages to suit shifts in technology or consumer preference. Apparently, as a result of inefficiencies in the development process, European firms did not innovate enough: whereas 60 per cent of the sales revenue of Japanese firms was derived from products less than one year old, the comparable figure for European firms was only 24 per cent. European firms have also been investing less than their competitors. Repeated product-development failures, coupled with the adverse macroeconomic environment already described, burdened their balance sheets and weakened their capacity to invest further. This created a vicious circle: products developed by firms which lacked the financial muscle to flood the market and reduce prices carried little credibility. In industries like consumer electronics, where the number of users is the key to determining which standard will ultimately dominate, such a lack of credibility was often the kiss of death. But the statistics hide the most important investment problem: European firms seem to have made the wrong type of investment, focusing on automation rather than capacity expansion. McKinsey analysts found that the most successful companies in the electronics industry invested less than average in automation. European management seems to have taken a simplistic view of productivity growth, based on substitution of capital for labour, on downsizing, and on outward job relocation; chasing down labour costs was easier than upgrading processes. But the impact on employee morale and, consequently, on the firms’ ability to innovate, was devastating. Well-intentioned industrial policies have also contributed to Europe’s difficulties in high technologies, as is illustrated by the history of the computer industry. The American computer industry emerged in the immediate aftermath of the second world war, with strong support from the federal government in the form of R&D funding—to both business and universities—and military procurement. When, in the mid-1950s, a commercial market started to emerge, the dominance of American producers (IBM, General Electric, National Cash Register, Sperry Rand, and RCA) was overwhelming. As the strategic importance of computers for both industrial development and national security became obvious, Germany, France and Britain all engaged in industrial policies designed

cadot layout 18/10/02 5:06 PM Page 30

30

Can industrial Europe be saved?

to catch up with the United States’ technological lead. These policies were generally based on the promotion of a single national champion: Britain formed ICL in 1967 by merging Elliott Automation and English Electric; France’s Plans Calcul engineered the merger of various computer makers into CII, later merged with Bull. Germany, whose first electronic data processing plan attempted unsuccessfully to foster the merger of AEG-Telefunken with Siemens, was perhaps the only exception. This policy of “engineered mergers”—still pursued in France—caused much harm. It prevented the exit of weak firms, clouded the balance-sheet of strong ones and diverted managerial attention to the complex task of digesting mergers. Thus chronically subsidised national champions developed a pattern of external growth by acquisitions, followed by waves of downsizing. The resulting absence of internal cohesion and low morale prevented the emergence of strong corporate cultures, and made the integration of technology development and corporate strategy difficult if not impossible. Sometimes, it even led to incompatible product lines being marketed by the same firm. As the inefficiency of national champion policies became apparent, a shift of emphasis in the early 1980s led to a wave of pan-European cooperative research projects, starting with ESPRIT. While these projects are generally considered to be useful, they did not lead, any more than the previous national policies, to the emergence of a world-class computer industry in the EU. Just as earlier industrial policies had failed to create credible challengers to IBM in mainframes, the later ones failed to help Europe join the bandwagon of personal computer makers which did put an end to IBM’s dominance. Why such a dramatic failure? Michael Kende argues that both sets of policies addressed the wrong problem.13 13 Michael For a long time, European governments held the view that the Kende, American superiority was derived primarily from economies of “Government scale; hence their over-emphasis on mergers rather than on support in the product development. Later, they failed to see the industry’s European IT transformation from a defence-contractor model to a massindustry”, consumption model, which meant that the challenge to IBM INSEAD, would come from new products and small-scale entrants rather 1996 than from government-sponsored supercomputer makers. What Europe needed—and still needs—is a support infrastructure for small and medium-sized enterprises (SMEs).

cadot layout 18/10/02 5:06 PM Page 31

The failure of Europe’s high-tech industries

31

Europe’s inability to generate a large number of high-technology start-ups, as the United States has done, is a subject of intense debate. Several factors are usually invoked. First, seed capital is scarce: venture-capital markets are, except in Britain, relatively under-developed, and less than 10 per cent of venture capital goes into start-ups. Second, Europe does not have a large enough pool of the researchers who can quickly analyse new products and firms. Furthermore, risk capital suffers from insufficient liquidity, because of the absence—until now—of a European NASDAQ where shares of start-ups can be traded. As investors have discovered that money invested in such ventures tends to get locked in, the flow of fresh money has dried up—leading to what some analysts call ‘financial constipation’. For all these reasons, mobilising capital for start-ups in Europe has proved significantly more difficult than it has in America. This has prompted the most daring European companies to cross the Atlantic and raise funds on NASDAQ—like Paris-based Business Objects, which raised $25 million in 1994, or Belgium’s Lernout & Hauspie, which raised $40 million in 1995. But the number of firms willing and able to make such a transatlantic move is very small: over 1994-95, only 34 European companies went to NASDAQ for financing. The imminent launch of EASDAQ, a European NASDAQ, should solve many of these problems, provided that the flurry of alternative markets set up by national stock exchanges—like London’s AIM or Paris’s Nouveau Marché—do not undermine it by excessive segmentation. Second, the already-mentioned gap separating Europe’s science from its business hinders the exchange of people and ideas. Few scientists in Europe, even among those with commercially viable ideas, are ready to take the jump and create their own business—a difficulty undoubtedly related to cultural backgrounds, but also to the insufficient spread of management training in Europe, especially among scientists. However, there are encouraging signs of change. Although American producers still account for more than three quarters of Europe’s software sales, firms such as Germany’s SAP (the world’s fifth largest software company, with projected sales of over $2 billion for 1996) or the Netherlands’ Baan (with projected sales of over $300 million for 1996) are carving out significant niches on global markets. The peculiarity of these start-ups is their strong American-style entrepreneurial culture and their international strategy. They typically go global right away, challenging the leaders in particular niches, rather than focusing on their domestic market. Such

cadot layout 18/10/02 5:06 PM Page 32

32

Can industrial Europe be saved?

strategies make their short-term growth potential considerable. In fact, several American analysts believe that, if Europe missed the computer revolution, it may yet have a head start in a wide range of informationhighway products. To sum up, what Europe seems to have lacked in the past, compared with America, is that intangible asset, entrepreneurial spirit, which is so difficult to nurture and even to measure. Nor has Europe had the capacity to support, promote and develop new ventures. Compared with Japanese firms, what European companies—especially the large ones—seem to have lacked is a clear strategic vision of internally-generated growth, based on market-driven innovation and, in particular, on the ability to link technology development with long-term business strategy. Many analysts believe that linkage is vital to corporate success. And this lack of strategic vision in many European high-tech firms can often be traced to a business history of engineered mergers and acquisitions, rather than of internal growth.

cadot layout 18/10/02 5:06 PM Page 33

33

6 Globalisation, jobs, and the welfare state

As the EU struggles to limit the erosion of its shares of global product markets, the term ‘globalisation’ has taken on a threatening overtone for many Europeans, becoming synonymous with lower job security and living standards. Such fears have prompted calls for very different solutions—drastic deregulation of labour markets on the one hand, and trade protection against developing countries on the other. These feelings of anxiety are based on two notions, one wrong, the other correct. The wrong notion is that, if some countries have higher labour costs than others, firms will steadily relocate, moving jobs to low-cost countries, cutting them in high-cost ones, and relentlessly tearing apart the latters’ industrial fabric. This dramatic view was popularised in 14 Jean Arthuis, France by a Senate report on job relocation which, although rapporteur, based on apallingly poor economics and facts, had a formidable impact on public opinion.14 Actually, if some countries have “Rapport sur l’incidence lower wages than others, it is precisely because they offer a less économique et attractive location for productive capital. The reasons may vary, fiscale des ranging from political instability to the unavailability of factors délocalisations”, of production that are complementary to capital, like skills and Sénat, infrastructure. But the notion that capital locates simply where June 1993 labour is cheaper is naive and is, as a matter of fact, contradicted by the most readily available figures on international capital flows. The correct notion is what is called in economic jargon “factor-price equalisation”, which means that if barriers to international trade 15 R Freeman, are reduced, the price of factors of production (like labour) tends “Are your wages to converge, even if those factors do not themselves move across set in Beijing?”, borders. The threat is then, in the words of R. Freeman, that Journal of “your wage will be set in Beijing”15 , or that, if wages are Economic inflexible, unemployment will prevail in rich countries—as indeed Perspectives 9, it does in Europe. While this reasoning is theoretically correct, Summer 1995

cadot layout 18/10/02 5:06 PM Page 34

34

Can industrial Europe be saved?

one has to ask why this process should develop now? The view seems to be that barriers to factor-price equalisation have been coming down significantly in the last decade or so, presumably because of what is called “globalisation”. Evidence for the extent of globalisation is mixed. On the one hand, financial globalisation is undoubtedly spectacular: the annual volume of international security transactions went up from less than 5 per cent of GDP to as high as 200 per cent for most OECD countries between 1975 and 1995. Increased capital mobility, fostered by both deregulation and technological change, implies that differentials in the rate of return on capital are now limited by arbitrage. This reduces the ability of national governments to set macroeconomic and tax policy independently. For instance, the international mobility of capital makes it more difficult to tax capital than labour, which is to a large extent immobile. On the other hand, although the flow of investment to developing countries has accelerated markedly during the 1990s (from 17 per cent of total foreign direct investment in 1990 to 44 per cent in 1994), the extent of capital mobility, especially between industrial and developing countries, should not be overstated. According to an OECD study, in 1992 only 21 per cent of the total stock of capital invested outside its country of origin went into developing countries, against 30.6 per cent in 1967. Similarly, foreign assets still represent less than 10 per cent of the portfolio of American pension funds, and claims on 16 Jean Pisanideveloping countries less than 0.5 per cent.16

Ferry, “Face à la mondialisation”, Témoin, forthcoming 1996

Real (non-financial) globalisation is even more elusive. The ratio of world trade to world GDP recovered to its 1913 level only in 1973, and has since crept further up, but at an unspectacular pace. The share of external trade in America’s GDP has been increasing steadily, but it is still only 20 per cent (combining exports and imports), a level comparable to the ratio of the EU’s external trade to its total GDP. As for trade-induced job losses, although estimates differ, the numbers are invariably small. In France, the most pessimistic estimates of cumulative job losses due to competition from low-wage countries are below 500,000, compared to a total of more than 3 million unemployed. Evidence at the microeconomic level seems to confirm this basic diagnosis. The McKinsey report on the electronics industry stressed that labour

cadot layout 18/10/02 5:06 PM Page 35

Globalisation, jobs, and the welfare state

35

costs were simply not a crucial issue. In fact, average labour costs in the EU are now lower than in Japan, and one of the countries worst hit by unemployment and market-share losses, France, has relatively low labour costs (even after taking into account indirect benefits and taxes on labour). Thus while it is undeniable that the EU has been losing market share in several sectors of manufacturing, there is little evidence linking those losses to high labour costs. By implication, the link between “globalisation”—itself a phenomenon of debatable extent—and unemployment is far from clear. What, then, accounts for Europe’s poor employment record over the last decade? Comparative evidence on job destruction and creation in the EU, America and Japan provides further insight into the issue. During the 1980s, in a sample of industries surveyed by McKinsey,17 the United States created 55.6 net new jobs per thousand of the working- 17 McKinsey age population, and Japan, 15.8; meanwhile, Germany destroyed Global Institute, 13.3, and France, with by far the worst record of the sample, Employment 36.5. Is the cross-industry pattern of job losses identified by the Performance, OECD, McKinsey, and other studies related to the pattern of November 1994. industrial strengths and weaknesses analysed in previous sections See also the of this paper? We have seen that the EU’s main weakness is in OECD Jobs high-tech industries, which are dominated by American and Study: Evidence Japanese firms. Indeed, relatively little net job creation took & Explanations, OECD, 1994 place in high-tech industries in Europe. But the same applies to the United States, which in fact destroyed jobs in computer hardware, and even to Japan (see table 1). It can almost be said that manufacturing computers is history as a job-creating industry in all developed countries. By contrast, the computer software industry has been a significant contributor to job creation, and its performance in this regard is, according to the McKinsey study, stronger in France than in the US. Thus, there is little evidence that Europe’s failure to develop strong hightech industries has much to do with its employment problem, and, by implication, that policies designed to boost the competitiveness of its high-tech industries would create significant numbers of jobs.

cadot layout 18/10/02 5:06 PM Page 36

36

Can industrial Europe be saved?

TABLE 1: N ET JOB CREATION /1000 WORKING -AGE POPULATION, 1980-90 US Computer hardware Computer software Automobile Construction Retail Manufacturing Services

Japan

Germany

France

-0.4 2.7 -0.6 3.0 4.9

1.3 4.4 0.7 -1.3 -2.3

0.3 n.a. 0.5 -8.1 -2.1

0.1 3.5 -5.3 -10.7 -3.6

-17.2 51.1

1.1 29.5

-19.8 16.9

-35.6 17.7

SOURCE: MCKINSEY (1994)

If one breaks down the evidence on job creation and destruction by level of skill, a striking pattern emerges. As a result of the past ten years’ globalisation and technological change, the demand for labour in OECD countries has been affected by two structural shifts: from manufacturing to services, and from low-skilled jobs to high-skilled ones. Everywhere, large numbers of high-skilled jobs have been created in services. In manufacturing, it turns out that Europe’s performance in terms of highskill job creation is surprisingly good, being roughly comparable to that of Japan and better than that of America. What sets Europe apart from Japan and the United States is the story of unskilled jobs. Throughout the 1980s Japan was, by and large, able to retain large numbers of blue-collar low-skilled jobs in all sectors of manufacturing. Why? First, growth of output matched rising productivity. Second, factory automation was not pursued as systematically as in Europe; although investment and productivity growth were high throughout the decade, the share of investment devoted to capacity expansion as opposed to labour-force reduction was higher in Japan than in Europe. To simplify the argument, Japanese management—in its quest for total quality and low costs—relied on worker involvement, whereas European management relied on automation. As a result of these and other factors (such as adverse macroeconomic conditions, already discussed) large numbers of blue-collar jobs were destroyed in Europe. A similar process of blue-collar job destruction occurred in the United States, yet America’s

cadot layout 18/10/02 5:06 PM Page 37

Globalisation, jobs, and the welfare state

37

unemployment rate is about half that of Europe. What accounts for the difference? OECD statistics suggest that America, unlike the EU, was able to create large numbers of unskilled jobs in services, particularly retailing. Those jobs enabled the American economy to integrate new entrants into the labour force, particularly the young and immigrants, thereby increasing their long-term employability. So Europe seems to have been losing on both counts: unlike Japan, it was not able to keep its traditional manufacturing jobs; but it could not replace them with new employment opportunities, as the United States did. Some observers—especially in America—argue that Europe’s problem is an unwillingness to accept that low-skill jobs should command low wages. The issue, in their view, is not so much one of international competition, but rather that high labour costs, unless matched by high skill levels, induce employers to replace man by machine—even in sectors that do not face foreign competition. In the words of Gary Becker, The rapid growth of labour costs throughout Europe appears to have been a principal cause of the explosion of unemployment. About half of Germany’s and France’s average labour costs result from social security, health, unemployment compensation, disability, and other taxes; other European nations have similar shares. Regulations that restrict layoffs and mandate numerous vacation days and other paid leaves also raise the cost of labour markedly.... When labour is expensive and firing employees is difficult, companies are reluctant either to replace departing workers or to expand employment when the economy picks up. This is why it now takes much longer than it did a decade ago to find a job in Europe if you are a young job-seeker, a mother returning to work after childbirth, or an immigrant.... Taxes and death may be inevitable, but not the degree of taxation and regulation prevailing in Europe. A large dent will not be made in their disgracefully high unemployment rates until European policies are recognised for what they are: a serious disease that badly infects their labour markets.18

18

Gary Becker, “Why Europe is drowning in joblessness”, Business Week, 8 April 1996

cadot layout 18/10/02 5:06 PM Page 38

38

Can industrial Europe be saved?

In more diplomatic language, the OECD “Jobs Study” of 1994 also argued that labour-market rigidity had something to do with Europe’s high unemployment rates, as did the European Commission’s 19 European White Paper19. Whereas Chicago School economists think that Commission, any kind of social protection is wasteful and sets the wrong “Growth, incentives, the OECD and the commission attempted to strike a Competitiveness, balance between the need to enhance labour-market flexibility & Employment: and the desire of many Europeans to preserve their societal The Challenges model. What is this model? Historically, it has been based on two and Ways Forward into the pillars: first, a comprehensive welfare state, involving unemployment insurance, health insurance, pensions, guaranteed 21st Century”, minimum incomes, and extensive public services; and second, a 1993 system of collective bargaining and tripartite (employers, unions and state) agreements on labour laws, including minimum wages, standards and training, and on the management of the welfare state. This system, inherited in large part from the social-democratic tradition of Northern Europe, has been—together with Keynesian demandmanagement policies—the bedrock of Europe’s prosperity and consensus since the end of the second world war. It put an end to the social convulsions of the first half of this century. But this economic and social order is in crisis. When the oil shocks of the 1970s sent inflationary pressures that had been building up for years spinning out of control, Keynesian policies became the first casualty. Now the whole model is being questioned and held responsible for Europe’s high levels of unemployment. Is there strong evidence that labour-market deregulation would significantly improve the EU’s job performance? Unsurprisingly, opinions vary. In Europe, Britain is the country which has pushed labour-market deregulation farthest. Wage contracts are increasingly decentralised, and there is less collective bargaining. Similarly, terms and conditions of employment (working times, firing rules, and so forth) have been left, increasingly, to the market. As a result, in the words of the 20 OECD OECD, “Greater flexibility has been reflected in the wide Economic distribution of working hours, low strike activity, ease of hiring Surveys: United and firing, greater decentralisation of pay fixing and working Kingdom, conditions, wider wage differentials.... and greater wage 1994-95 variation across regions” 20 . To some, these are worthy achievements of economic efficiency; to others, the OECD’s description

cadot layout 18/10/02 5:06 PM Page 39

Globalisation, jobs, and the welfare state

39

might sound like a catalogue of horrors. The question is whether “greater flexibility” means improved efficiency in terms of overall societal objectives, such as lower unemployment rates, shorter spells of unemployment, and so on, or whether it is simply the reflection of a shifting balance of power between employers and employees. Raw exploitation of employees by employers during periods of high unemployment is a historical reality which—in the current climate of free-market ideology—is often forgotten. How successful has been British-style labour-market deregulation in reducing unemployment? On one hand, by 1995, the British unemployment rate had converged down to the OECD average for the first time in 15 years, making Britain one of Europe’s best performers. The British economy’s job-creation performance was particularly strong between 1983 and 1990, with more than 3 million net new jobs. As in America, all net job creation was in the business sector, in contrast with continental Europe, where the public sector’s share of total employment kept expanding. Moreover, the average length of unemployment went down significantly during the same period: those out of work for more than a year shrank from more than 50 per cent in 1985 to 27 per cent in 1991. And although Britain’s youth unemployment deteriorated between 1990 and 1993, it has a much better record than Southern Europe (Britain’s youth unemployment rate in 1993 was slightly above 16 per cent, against an EU average of 22 per cent). Regional variations in unemployment rates have also been largely eliminated in Britain; again, this compares favourably to the failure of regional policies in many continental countries. However, some of the British economy’s successes during the 1983-90 recovery proved more short-lived than the depth of the structural reforms would have suggested. By the mid-1990s, massive job destruction during the 1990-93 downturn had pushed back overall employment to its 1979 level. Over the whole period 1985-95, ‘trend’ unemployment (the unemployment rate smoothed out to eliminate short-term fluctuations) merely stabilised at around 8-9 per cent. Although the actual (unsmoothed) rate of unemployment was in 1995 lower in Britain (8.5 per cent) than the EU average (10.7 per cent), the British economy’s performance in terms of job creation was, in that year, very similar to that of the EU as a whole (0.8 per cent against 0.7 per cent, respectively).

cadot layout 18/10/02 5:06 PM Page 40

40

Can industrial Europe be saved?

Meanwhile Britain’s long-term unemployment, often said to be the consequence of labour-market rigidities, had in 1994 returned to 45 per cent of total unemployment, very close to the 47 per cent average for the EU 15. Reflecting on these disappointing developments, the 1994-95 OECD study of the British economy noted that “greater microeconomic flexibility has yet to be significantly reflected in enhanced macroeconomic flexibility [meaning downward wage flexibility during recessions] or in sustained low unemployment”. Whereas the successes of labour-market deregulation proved somewhat short-lived, its costs appear substantial. Data on income inequality must be considered with caution, as it is less recent than employment data; the picture presented below may therefore have more to do with large levels of unemployment in the early 1980s than with the effect of structural reforms. Nevertheless, it is part of the picture, and a spectacular part of it. In 1986 Britain’s income inequality, while still less than that in America (see table 2), was rising sharply. TABLE 2: CUMULATIVE INCOME SHARES (%) FOR DECILE INCOME GROUPS Decile income group 1 (poorest-richest) US (1986) UK (1986) France (1984) Germany W (1984)

1.9 2.5 3.0 4.0

2

3

4

5

6

7

8

9

5.7 7.5 8.3 9.8

11.2 13.5 14.6 16.6

18.0 20.5 21.8 24.2

26.2 28.7 29.9 32.9

35.7 38.2 39.1 42.5

46.9 49.1 49.5 53.2

60.2 61.8 61.6 65.3

76.3 77.1 76.3 79.4

SOURCE: AB ATKINSON, “I NCOME DISTRIBUTION IN EUROPE AND THE UNITED STATES”; OXFORD REVIEW OF ECONOMIC P OLICY 12, 1996.

Thus, poorer households have participated less than richer ones in the general rise in British standards of living. The poorest 5 per cent have seen their incomes go down even in absolute terms. Moreover, this widening inequality does not seem to be related to increases in the number of single-parent households, but rather to changes in wages and employment status. One may wonder whether such a social evolution is not a high price to pay for the employment gains observed. Although it may be too early to judge, from a continental viewpoint it is not clear that the

cadot layout 18/10/02 5:06 PM Page 41

Globalisation, jobs, and the welfare state

41

achievements of the British model of labour-market deregulation warrant, at this stage, its replication elsewhere in Europe. There are several arguments in favour of caution. First, unlike in the United States, income inequality in Europe is compensated neither by high upward mobility nor by lesser wealth inequality; its social costs are therefore likely to be larger. Second, firing employees entails social and human costs which are not borne by the party making the decision; some sluggishness in workforce adjustment may well be socially desirable, if not privately optimal. Finally, the idea of deregulation (as applied to the labour or other markets) is based on the premise that the alternative to regulation is, in general, the free operation of the market. But this is a somewhat misleading view. Regulation codifies behaviour ex ante to deal with potential conflicts; the alternative is private contracts and litigation, which solves conflicts ex post. As the American experience suggests, litigation is far from being costless; in fact, it is both costly and unpredictable. In general, “transaction costs” have often been argued, at a theoretical level, to reduce the efficiency of the libertarian model of society promoted by economists like Gary Becker. The recent American experience suggests that such costs do matter in reality. Thus, not only is the libertarian model somewhat difficult for many Europeans to accept; its ability to improve economic efficiency is also largely unproven. Where does this all leave us? Libertarian critics of the welfare state are right in pointing out that low-skill jobs cannot command high wages, and that too much welfare destroys incentives to work. Conservative British governments have shown the way in making the welfare state more “employment-friendly”, whereas in many continental countries benefits are designed in such a way as to effectively discourage any desire to seek employment. In spite of such efforts, the British system still suffers from high effective marginal tax rates; Europe has yet to find a solution to the conflict between, on one hand, the desire for insurance against poverty and, on the other hand, the need to preserve incentives for people to seek employment and “work out” of poverty. Britain has also shown the way in tackling bravely the rigidities that plagued its labour market in the 1970s. But so far, to the continental eye, the reforms sometimes look more like the employers’ revenge against trade unions than a truly new solution to Europe’s unemployment problem. If some degree of labour-market deregulation is clearly needed throughout Europe, it is less clear that it has to be across all types of jobs and skills. The temporary nature of the

cadot layout 18/10/02 5:06 PM Page 42

42

Can industrial Europe be saved?

low-skill, low-pay retail-sector jobs created in America and taken up by the young, immigrants, and so on provides some justification for their low level of protection. But precisely because they fulfill a particular role (integrating newcomers and easing the transition of displaced individuals) it is logical that they should be treated differently from other jobs in terms of regulation and taxation. In some European countries, minimum wages combined with heavy social-security taxes have taken the cost of youth labour to absurd heights. Governments need to muster the courage to question ritual increases in minimum wages, and to pursue more vigorously programmes for exempting the young from social charges. So far, those programmes, pursued in isolation and in a somewhat incoherent and haphazard way, and invariably launched before elections, do not seem to be working. They might work better if designed in a more transparent way and used in conjunction with other instruments. However, the real constraints on low-skill job creation in services such as retailing may not be the labour market itself. The McKinsey employment study pointed to product-market regulations (like shopping hours) rather than labour-market regulations as the primary hindrance to hiring in retailing and other services. Although more information is needed, this finding, if confirmed, could be crucial for devising the right employment policies. Clearly, as Europe is unlikely to regain the blue-collar manufacturing jobs it has lost, it will have to create American-style service jobs. Understanding why it does not create them is more important than understanding why it missed the personal computer revolution. In the long run, if Europe wants to reconcile its view of income equality with the realities of its workers’ productivity, it must undertake a large-scale effort in favour of education and training. The greater part of that effort still lies ahead of us.

cadot layout 18/10/02 5:06 PM Page 43

43

7 Conclusion

Industrial Europe does have a problem, although it may not be as severe as some, especially in the United States, often argue—at least if the improvement in Europe’s trade position that was apparent in the early 1990s is more than short-lived. Europe’s problem is in large part due to adverse macroeconomic conditions, and should therefore correct itself over time, provided that some action is taken to reduce the extent of budget deficits and currency overvaluation. But the problem is also structural, since Europe’s worst market-share losses have been registered in high-tech industries in which, according to the theory of comparative advantage, Europe should specialise. Although a lot more research is needed on the high-tech problem, it seems to have more to do with specific managerial and industrial-policy failures than with high labour costs per se. However, market-share losses in high-tech industries can hardly be held responsible for mass unemployment, since these industries are not major job creators in the United States, nor even in Japan. In fact, Europe’s record at creating high-skill jobs is as good as that of America or Japan. Europe’s unemployment problem seems to be confined to lowskill jobs, which it has destroyed massively in manufacturing (unlike Japan) without creating alternative employment opportunities in services (as the United States has done). Therefore short-term relief from Europe’s unemployment problem is unlikely to come from market-share gains in high-tech industries. Instead, it might come, at least in part, from regulatory changes in labour markets: European firms need more incentives to hire low-skilled individuals in services. Such regulatory changes might involve a slower growth of minimum wages, which harm youth employment, and a rejigging of social-security charges. The objective would be to reduce charges on unskilled jobs, which are most vulnerable to labour-saving investment. The burden of financing social spending could also be shifted from taxation on income, and especially employers’ national insurance contributions, to taxation on capital—though that would require co-ordination at European, and possibly global level. Finally, rules on entitlement to benefits should be

cadot layout 18/10/02 5:06 PM Page 44

44

Can industrial Europe be saved?

designed in ways that minimize the adverse impact on incentives to work. This is likely to require simplification and harmonisation of the various benefit schemes simultaneously in force in most European countries, and a more gradual reduction of benefits for individuals who creep up the income ladder. Such changes do not imply the elimination of the welfare state as we know it, but rather the recognition that all jobs should not be regulated and taxed in the same way. In the long run, Europe needs to raise the productivity of its workers, not by downsizing yet again, but by investing massively in skills. However needed such structural policies are, the British experience suggests that they are unlikely, by themselves, to bring back full employment, so long as stop-go macroeconomic policies prevent the emergence of a strong, long-term recovery. Many economists argue that Europe also needs an urgent effort to boost growth, and that can only be done through macroeconomic policy co-ordination at the pan-European level. Recent social unrest in continental countries, in reaction to mild attempts at reforming the welfare state, suggests that any government attempting to impose painful structural reforms, after years of recession and high unemployment, will only be able to move extremely slowly. Reforms need to be sold, for instance as part of a convincing growth package; but endless austerity makes it hard for governments to sugar the pill. One can only hope that when the transition to a common currency is finally complete, the European Central Bank will be in a position to set Europe on a trajectory for sustainable growth; only then would Europeans be likely to accept painful but necessary structural reforms. ★