Egalitarianism in a Global Economy

Andrew Glyn

"Globalization" has become the great conversation-stopper in contemporary policy debate. According to a now-conventional story, an increasingly competitive international economy sharply restricts national governments' room to maneuver in policy-making. Constrained by the stringent demands of global markets, governments can no longer do much to improve citizens' economic welfare--particularly when those citizens are less wealthy, less skilled, less well-connected to the labor market. Economic globalization has, in short, killed egalitarianism.

The terrain for national policy-makers has certainly shifted since the 1960s; whether globalization is responsible for that shift is another matter. Back in palmier times, access to jobs, income distribution, employment rights, and welfare provisions were all on seemingly inexorable progressive trends. For egalitarians, the great policy problem was to sustain those trends and extend them to excluded groups. In the current setting, an egalitarian economic program faces the vastly more demanding task of reversing or avoiding disastrous increases in inequality.

Within Europe--and I will be concentrating on Europe here--growing economic inequality has highlighted three economic problems:

1. High rates of unemployment and labor market withdrawal, especially among the least qualified sections of the labor force;
2. Pressures to deregulate the labor market and allow a widening dispersion of pay and increasing numbers of insecure jobs with very low wages;
3. Demands to reduce budget deficits and finance tax cuts by reducing the provision of public welfare services and levels of income support.

These labor market and fiscal trends have been especially acute within the United Kingdom, where, between the late 1970s and the early 1990s, the nonemployment rate for the least-educated working-age men tripled to reach 33 percent, the pay of the worst-paid tenth of wage earners fell by 28 percent relative to the best paid, and the level of state pensions fell by more than 25 percent compared to average earnings. Such trends generated an astounding rise in inequality. By the end of the 1980s, the United Kingdom was challenging the United States for the title of most unequal of the advanced countries.1 Even Sweden, the widely admired model for egalitarians, is suffering from high unemployment, a widening of the previously compressed pay distribution, and pressure on its generous welfare programs.2

The details of an egalitarian economic program suited to correcting such trends will of course vary across countries. But the core package, of general applicability in Europe, seems straightforward: create new jobs by increasing aggregate demand; reverse the deteriorating employment opportunities at the low-pay end of the labor market; and maintain standards of welfare services.

Can such a package succeed politically in today's global economy? Would it achieve the desired effects? To answer these larger questions we must first explore the costs and viability of the standard egalitarian package and examine the domestic hurdles to its implementation. We can then assess current trends toward internationalization and evaluate the additional constraints they impose on policy choices for European countries. Such analysis shows, in my view, that globalization is not the insurmountable obstacle to egalitarian policy that many commentators assume it to be: in fact, the large constraints on egalitarian policy are still the old-fashioned domestic ones.

Egalitarian Policies

In high-unemployment European econ-omies, a sensible egalitarian policy must start by creating jobs. The conventional job-generation strategy is to increase employment by expanding demand. Some economists argue that this strategy will no longer work: because technical progress now tends to be labor-saving, they argue, output growth in response to expanding demand no longer generates many additional jobs.

If this argument is right, a successful expansion of aggregate demand would have little impact on the labor market. But the evidence does not support this view. The slowdown in labor productivity growth since 1973 suggests that the worry about labor-saving bias is seriously exaggerated. More to the point, rapid expansions in output do still lead to faster job creation. For example, between 1985 and 1988, an increase in output of 4.7 percent per year in the United Kingdom translated into employment growth of 2 percent per year; from 1990 to 1992 employment declined by 3 percent per year in response to a 1.4 percent annual decline in output. In the western part of Germany employment increased 3 percent per year between 1989 and 1991, in response to 5 percent annual growth during the "unification boom." In the period 1982-93, for every 1 percent per year that a country increased its output growth (as compared to the previous cycle), employment grew by nearly 1 percent per year.3 If current trends in OECD countries' growth in labor productivity are any indication, a rate of 4 percent growth sustained over several years will provide jobs at an annual rate of around 1.5-2.5 percent.

According to a second and more subtle argument, expansions may create jobs, but they don't help low-wage workers. This claim has some plausibility: underlying economic trends, arising from the bias of technological progress and/or trade with low-wage countries, together with weakened capacity by unions to resist job cuts, are surely reducing demand for the least qualified section of the labor force. Still, changes in the general state of the labor market do not leave those at the bottom end unaffected. To the contrary, their position seems to be disproportionately dependent on the overall demand for labor. Thus when the unemployment rate of the most qualified fell by 0.7 percent in the United Kingdom between the mid-1980s and the end of the decade, the unemployment rate of the least qualified fell by 4.7 percent. The same broad pattern characterized other countries.4

Far from being irrelevant, then, rapid expansions in demand will have marked effects on the employment prospects of the least qualified.

Expanding demand will not, however, suffice to address the deteriorating position of the least qualified. With skill shortages and resulting inflationary problems, expansions may end prematurely. This suggests that expansions should be twisted towards sectors that use the types of workers whose unemployment is highest. Moreover, general expansions will have limited impact on the most severe concentrations of joblessness among the unskilled, which have resulted from rapid deindustrialization, cutting off many young people from realistic prospects of a job. Such trends could probably only be reversed over extended periods of time and with infusion of additional resources in geographically targeted programs. Thus a generalized expansion of labor demand needs a strong regional and industrial focus. And improving income distribution will also require policies focused directly on compensation and working conditions at the bottom end of the pay distribution--for example, increasing minimum wages.

Internal Constraints

Expanding demand, combined with proper targeting of policies and benefits, can, thus, have a major effect on the job market, and improve conditions at the low end. But for both domestic and international reasons, such expansions are not straightforward to achieve. Let's start with the constraints of domestic origin.

Generating Investment

In a model Keynesian economic expansion, increases in government spending and/or reductions in taxation fuel demand for private sector output, and growing private sector output generates further increases in income and consumption. Expanding demand, in turn, provokes new investment, especially if interest rates are kept from rising or even reduced. With a sufficiently strong investment response, income and tax revenue grow enough to eliminate the budget deficit produced by the initial rise in government spending. That spending can then be scaled back. In such an expansion the government deficit plays a temporary, "pump-priming" role, and full employment imposes no costs: previously unemployed workers gain because their take-home pay is greater than the level of state unemployment benefits; other workers gain because their taxes, no longer required to pay for unemployment benefits, are reduced ; employers gain because sales and aggregate profits rise.

Things are more difficult when private investment responds only sluggishly to an expansion. A government can then try to sustain demand at high employment rates by running a persistent budget deficit. Though no one immediately pays for the expansion, the state piles up debt, with the implied obligation to raise taxes in the future to cover interest payments. Alternatively, the surplus of private savings can be eliminated if the government undertakes a parallel expansion of government spending and taxation of a sufficient magnitude. An expansion of this kind relies on redistributing consumption from employed to unemployed workers. So part of society must pay the costs of full employment by reducing its private consumption, though it may also receive benefits from improved public services where the extra workers are employed.

Rising investment is the key to avoiding the costs of full employment. And if rising demand could be relied on to elicit the desired private investment response, then other policies specifically directed at stimulating investment would be unnecessary. But this is unfortunately not the case. While investment rose rapidly in Europe in the boom of the 1980s, it was typically concentrated in the service sector, and often highly speculative (office development, for example). The OECD noted that for continental Europe "business investment was generally weak during the early 1990s, and it has been slower to respond to improving conditions than in previous upswings."5 In the United Kingdom as well investment made virtually no contribution to recovery. If growing demand does not suffice to call forth new investment, the government can try incentives, but these are often expensive and unreliable. So it is chancy to rely on private investment to play a strongly expansionary role, especially given employers' likely evaluation of the difficulties in sustaining expansion as a result of other constraints considered below.

Conflicting Claims

By generating extra resources, expansionary programs have the potential to make all social groups better off. But even if this potential is realized, conflicts over income shares may still intensify as a tightening labor market enhances the bargaining position of workers. Moreover, unless the expansion leaves the unemployment rate above the pre-existing "natural rate of unemployment"--the rate below which inflationary pressures emerge--or is accompanied by measures to reduce the natural rate, accelerating inflation will develop. But rising inflation could well be associated, as in the 1960s and 1970s, with severe profit squeezes. Furthermore, prolonged full employment might well threaten business interests by generating a range of demands from labor for limiting managerial prerogatives over the organization of work and allocation of investment (restrictions on dismissals, industrial democracy, planning agreements, nationalizations, etc.).

After the social turbulence of the late 1960s through the mid 1970s, the subsequent period of mass unemployment delivered low inflation, a striking recovery of profitability, dramatically reduced industrial conflict, and the virtual disappearance of threats to business's freedom to maneuver. Does this mean that the natural rate of unemployment has fallen? The answer to this question is hotly debated. But whatever the right answer, and whatever the current location of the natural rate, beliefs about the impact of expansion on inflation can have important effects. If employers think that an expansion is not sustainable because it will generate social conflict, they are less likely to respond to higher demand with increased investment. Their hesitation in turn weakens productivity growth, which could otherwise have moderated the conflict over real incomes by increasing the size of the cake. Focusing expansion on low-wage workers helps to moderate inflationary pressure, but at some point those pressures will build up unless contained within the wage-bargaining system. The central problem is to devise a convincing institutional framework within which to resolve the social conflicts generated by high employment.

Government Deficits

In the simplest case considered above, an expansionary program produces a short-lived government deficit, which disappears when private investment expands sufficiently to drive up incomes and tax revenues. But if the private sector response is not strong enough, budget deficits become permanent fixtures. Why is this problematic?

The rise in the ratio of government debt to the gross domestic product (GDP), if the deficit exceeds that which would stabilize the debt burden, heightens the potential for distributional struggle, as an increasing share of output is used to cover interest payments. Workers attempt to offset taxes raised to pay the interest, thus increasing inflationary pressures. Governments are tempted to accommodate such inflation as a way to reduce the real value of their outstanding liabilities. Accordingly, at any given rate of unemployment, the higher the deficit (at least beyond certain roughly defined limits6) the greater the likelihood of faster inflation in the future. This is the rational basis for the financial markets' adverse response to deficits.

Financial Markets

The response of financial markets to impending difficulties is likely to hasten troubles. Anticipating faster inflation, investors tend to push up market interest rates to avoid future losses on their financial assets. Even if real interest rates are not affected, concerns about future inflation and capital losses inhibit spending and stifle the expansion. The crucial role of financial markets is to anticipate future developments and these very anticipations have real effects. Moreover, we should expect such effects in any developed financial market, whatever the degree of international financial integration.

Controls over the financial system might be devised to prevent market anticipations from dominating current activity. But in the circumstances just described such controls would suppress the warning symptoms of a problem. If an expansion is going to collapse in faster inflation and profit squeeze, then preventing financial markets from registering these prospects can only postpone the destabilizing pressures.

Taxation

Even weak private sector investment does not inevitably imply budget deficits at full employment. An alternative policy to maintaining demand by deficit financing is to increase government spending even further, while simultaneously increasing taxation to cover government costs. Citizens in general benefit from the improvements in public services or infrastructure, but there is no longer a "free lunch": someone must reduce consumption. Such an expansion, with all the extra demand emanating from the public sector, is an extreme case. But the public focus may be necessary to compensate for weak private sector spending. Moreover, such focus has desirable consequences given the generally egalitarian distribution of benefits from improvements in the public services. The expanded public role may create problems of public sector debt and financial instability, however, unless the financing of the net costs of the programs comes from higher tax rates.

Who will pay such higher tax rates? An egalitarian program would impose tax increases in a progressive manner, especially in countries like the United Kingdom, where the tax system has been shifting in a highly regressive direction. Whatever the possibilities for widely-supported increases in tax progressivity, political circumstances may dictate that tax increases extend beneath "top incomes" and impose an increased tax burden on middle incomes. But political support for such increases is hard to come by, even where tangible benefits accrue in the form of improvements in health and education services (not to mention the less tangible benefit of reducing social exclusion of groups at the margin of the labor market). And even if that support can be secured, institutions regulating wage bargaining have also to be strong enough to ensure that burdens accepted at the ballot-box are not then pushed aside through compensating wage increases gained at the bargaining table (not least by an expanded public sector).

To summarize: If the private sector is very dynamic, a mere nudge may push the economy towards high employment. Increasing demand will require neither sustained budget deficits nor a leading role for the public sector. In contrast, weak private investment requires an enhanced role for the public sector. If government is prepared to play that role, it may face opposition from financial markets concerned about the inflationary effects of deficits, or citizens resistant to higher taxes. If government overcomes such opposition, higher levels of demand and employment will inevitably intensify conflicts over wages and working conditions. And the resolution of those conflicts will require appropriate institutions to negotiate disagreements between workers, owners, and government. The fundamental lesson of the 1960s and 1970s was that domestic hurdles to maintaining high employment and an expanding welfare state have deep-seated economic and political roots.

Internationalization

Thus far, I have left the international economy entirely out of the picture. Before considering its impact on expansionary programs, we need some discussion of its scope. Because there is no agreement on what globalization is, or how to measure its importance, I will consider three elements of international economic interdependence: trade, foreign direct investment, and financial flows.

Trade

In a recent overview of historical trends, Angus Maddison measures the importance of trade in an economy--its degree of "trade integration"--by the ratio of exports to GDP at constant prices.7 The results for Europe are particularly dramatic: after declining from 16 percent in 1913 to 9 percent in 1950, the export ratio rose to 21 percent in 1973 and 30 percent in 1992. Trade thus appears to have tripled in importance since 1950, and is now far more important than it was during the classical free trade period.

"Constant price" comparisons are, however, misleading. Export prices rise systematically more slowly than do prices for output as a whole (because the average growth of labor productivity in the export sector is greater than for the economy as a whole). Indeed, over the period from 1960 to 1994, roughly two-thirds of the discrepancy between export and GDP growth was offset by the slower growth of export prices. To correct for these differential price changes, we should focus on export shares at current prices.8 We then find that the 1994 share of exports in GDP did not much exceed 1913 levels (see Table 1). This is true for the OECD countries in total and for the United States, Europe, and Japan taken separately. More immediately, the increase over the post-war period is relatively modest. Trade shares lurched up in the early 1970s as OECD countries exported more to pay their oil bills, and this was repeated in the early 1980s. Since then, however, the export share has continued to rise only in the United States. In Europe it stabilized; in Japan, it fell sharply. This is hardly a picture of rampant globalization

It might be argued, however, that the export share understates the importance of international competition: Imported manufactures represent an additional competitive pressure on substantial sections of the economy over and above that deriving from export markets. And the share of manufactures in total OECD imports has nearly doubled over the past 30 years to exceed 75 percent, with the importance of machinery (including cars) growing strongly. The impact of international competition within domestic economies is most clearly displayed in the degree of import penetration of the domestic market for manufactures (see Table 2).

Increasing import competition was noticeable even in the 1950s and, with the important exception of Japan, it has continued unabated, with import market shares typically doubling between 1971 and 1993. Most of this competition comes from other OECD countries. Although the European market share of so-called "emerging economies" has doubled since the late 1960s, they still take only 20 percent of the market for imported manufactures in Europe and the United States, and one third in Japan. This growing import-penetration of domestic manufacturing markets is surely responsible for the general impression of increasing international integration through trade. But manufacturing only constituted 18 percent of OECD employment in 1994 (ranging from 16 percent in the United States to 26 percent in Germany); for the OECD this represents a decline of one third as compared to 1974. Does internationalization amount, therefore, to increasingly fierce competition about a diminishing and relatively small sector?

Not quite. The significance of manufacturing is underplayed by its share of employment; other sectors contribute substantially to manufactured commodities. So part of the output of agriculture, mining, energy , construction, transport, and finance and business services is dependent on the success of domestic manufacturing. Pieces of these sectors are, at one remove, subject to the international competition within manufacturing markets. To indicate the magnitudes involved, in 1994 in the United Kingdom, other sectors contributed some $112 billion to manufacturing output (with finance and business services the biggest supplier) as compared to value added within manufacturing itself of $198 billion. Manufacturing is thus arguably at least 50 percent more important than its contribution to value added or employment suggests. If we extended the calculation to include the value of agricultural and mining output (which is extensively, if far from freely, traded internationally) it would seem that around 30 percent of the UK economy is directly or indirectly contributing to the production of internationally traded goods.

Of course, some services are traded directly as well. But they are concentrated in a narrow band (international transport, international finance, consulting, and so forth) and are irrelevant for the mass of domestic service producers. (Tourism is one exception, being in competition with a broad range of domestic services.) There is no obvious way to quantify what part of services is seriously internationalized; but any plausible estimate would leave a majority of employment in OECD countries--possibly a substantial and probably a growing majority--largely untouched by international competition.

So the impact of internationalization through trade is complicated. For manufacturers and their suppliers, and for some specialized enclaves in the service sector, internationalization has intensified considerably: It has increased strongly in the United States, but from a very low level, and in Europe as well, but much less in Japan. For the rest of the economy, covering probably a growing majority of those in employment, international competition is of little direct relevance--though it is not immune, of course, to the macroeconomic consequences of success and failure in the traded-goods sectors.

Foreign Direct Investment

The 1980s saw a substantial increase in foreign direct investment (FDI); according to one major source for data on FDI and multinationals, the inward stock of FDI in developed countries rose from 4.8 percent of GDP in 1980 to 8.6 percent in 1994. Although the increase seems impressive, the measure does not provide a good indicator of the importance of FDI in an economy. A less problematic measure is the ratio of the flow of FDI to the corresponding flow of domestic investment. This summarizes the relative importance of foreign and domestic sources of new investment in an economy.

As Table 3 indicates, FDI still represents a relatively small proportion of capital expenditure. Nor does all FDI represent the construction of new production facilities by overseas companies. Typically well over half of FDI inflows into OECD countries represents cross-border mergers and acquisitions. Some of this represents international intrusion into the competitive structure of the industry, but some acquisitions are closer to portfolio investments, involving a change in ownership but with relatively little impact on industry behavior.

Financial Capital

Indicators of the internationalization of financial capital are virtually unanimous in showing spectacular growth. The only difficulty is the now-familiar one of judging whether a particular aspect of internationalization has yet become "large" in relation to its domestic counterparts. As a ratio to world GDP, the stock of cross-border bank lending has grown from 6 percent in 1972 to 37 percent in 1991; cross-border transactions in bonds and equities have grown from less than 10 percent of GDP in 1980 to 80 percent of GDP for Japan, 160 percent for the United States and 200 percent for Germany.9 These statistics and numerous others like them do show an explosive growth of international transactions. However even in a highly open financial market such as the United Kingdom, less than one quarter of company securities are owned abroad and only one fifth of the assets of UK pension funds are in overseas assets. Thus the degree to which these rapidly growing and volatile flows has translated into a genuine internationalization in the ownership of financial assets is still quite limited . Even so, the size and volatility of these financial flows obviously enhances the power of financial markets to punish economic policies that are felt to threaten their interests.

Internationalization and Egalitarian Expansion

This sketch of trends in trade, FDI, and financial markets suggests a need for caution in discussing the pace and scope of internationalization. But what about the implications of these trends for egalitarian policy? How does internationalization interact with the domestic constraints on expansion summarized earlier?

Investment

Internationalization may have contributed to the weakness of European investment in a couple of ways. First, international financial integration has made it nearly impossible for individual countries to avoid recent worldwide increases in real interest rates. In the period from 1990 to 1993, real long-rates in European countries were in the range of 4-8 percent, compared with rates of around 2 percent in the 1960s and frequently negative rates in the inflationary 1970s.10 But this factor can be overemphasized. Interest costs have to be set against profitability, which in manufacturing recovered quite spectacularly in most European countries in the 1980s. Moreover, empirical studies almost universally find that interest costs are of secondary importance in investment decisions.

Second, part of any demand stimulus "leaks abroad"--it leads to new imports. So it can only be sustained if the initial balance of payments is strong enough, or exchange rate depreciation is feasible as a route for maintaining current account balance. Fears about a worsening balance of payments, then, add to other anxieties about whether expansionary programs can be maintained. And uncertainty over exchange rates, at least with countries outside the European Union (but also with fellow members after the 1992 collapse of the Exchange Rate Mechanism), increases the unpredictability of returns, particularly for manufacturing and related sectors.

Conflicting Claims

During an expansion, increased imports ease distributional struggles, as long as a current account deficit can be sustained. But such deficits eventually have to be reined in--and with integrated capital markets, sooner rather than later. Unless there are fixed exchange rates, deficits can be straightforwardly remedied by allowing the exchange rate to depreciate sufficiently to generate the extra exports required.

Depreciation, however, increases import costs, thus reducing the resources available for moderating distributional conflict; in particular, maintaining the gain in competitiveness requires workers' real wages to decline (relative to the trend in productivity): once more, additional employment must be paid for with lower real wages for those already in work. Even where additional employment would otherwise have been profitable on the existing capital stock without real wage cuts, the deterioration in the terms of trade may make wage reductions necessary.

So distributional struggles will be heightened by efforts to maintain a current account balance. But because such struggles raise troubles for economic expansion for purely domestic reasons, and because increases in trade are relatively modest, those increases can hardly provide much of a constraint on economic expansion. If European countries had no reason to hold back from Keynesian policies other than the effects on their payments balances, then it is hard to see why coordinated macroeconomic expansion at the European level for example would not happen. Domestic constraints to expansion make the failure to engineer coordinated expansion easily explicable: the big problem is fear of renewed distributional conflict.

The impact of internationalization on natural rates of unemployment in Europe is complex. Greater flows of FDI represent greater flexibility for employers to relocate production, and the threat of relocation may give employers greater bargaining strength over wages and other conditions of work. Fixed exchange rates within Europe could also lessen wage pressure since the competitiveness of industries within a national economy can no longer be restored by depreciation. But the much lower inflation rate in Europe, which is imported via a fixed exchange rate, can itself cause strains. Within complex bargaining systems, like that of Sweden, moderate inflation may be necessary to permit all levels of bargaining to deliver at least money wage increases.11 A fixed exchange rate requires not just convergence of inflation rates, but systematically lower wage increases to mimic depreciations for countries whose export shares are tending to decline. Moreover, to the extent that trade with less developed countries really is twisting the pattern of demand away from the less qualified, it tends to worsen skill shortages and increase the natural rate of unemployment.12 An overall balance sheet is hard to draw.

Internationalization, then, clearly has a mass of complex, even contradictory effects on struggles over income shares. It must complicate the task of devising institutional means to resolve them. But the source of these struggles still lies in the conflictual nature of capitalist employment relations.

Deficits

Where real interest rates exceed trend growth rates, a primary surplus on the budget is necessary to prevent government debt from rising faster than GDP. The extent to which real interest rates can be affected by monetary policy on a world scale is debatable; whatever the cause of the high real interest rates of the 1980s and 1990s, international financial integration makes it very difficult for individual governments to shake them off. Between 1979 and 1995 government net interest payments increased from 1.9 percent of Europe's GDP to 4.8 percent.13 This increase provided much of the impetus towards cutting deficits enshrined in the Maastricht targets. Only an unexpected burst of inflation can reduce the real burden of interest payments, and the near impossibility of keeping control of such a situation makes it unviable as a deliberate strategy.

Financial Markets

In general, the best support for an expansion would be interest rates as low as possible combined with a steady fall in the real exchange rate. But under conditions of free capital mobility, an expansionary program is liable to lead to a much larger fall in the exchange rate, along with very large depreciations. Moreover, those depreciations would not occur smoothly over the life of the program, as real depreciation became necessary or higher inflation occurred. Rather they would tend to happen in a rush, in anticipation of such developments. As discussed earlier in relation to domestic finance, these markets have no interest in giving expansionary policies the benefit of the doubt. On the contrary, they are geared to anticipating possible future problems; portfolios are shifted to avoid the consequences. Such anticipations will tend to "front load" the impact on real wages of an expansionary program by generating large initial depreciations. The task of holding the line on distributional conflict must be made more difficult by the likely response of foreign exchange markets.

The fact that financial markets can force the abandonment of a politically unrealistic deflationary policy, as in the case of the United Kingdom in the autumn of 1992, has not suddenly made them the workers' friend. Whatever policy stance is regarded as unsustainable and thus not credible can be tested out by the markets. There is thus every reason to explore all the possibilities for limiting the destructive power of the financial markets (including Tobin taxes, deposit requirements on foreign exchange transactions, and dual exchange rates).14 But while such restrictions might limit the power of the markets to destroy an egalitarian package that had a fighting chance of success, markets could never preserve a package that was unviable for the fundamental internal reasons discussed earlier.

Taxation

Egalitarian programs will tend to be slanted towards an expansion of public spending on welfare because of its redistributive effect. Moreover, where the private sector responds sluggishly to expansion, sustained growth of the public sector may be necessary to secure high employment. With substantial debt burdens in many European countries, high real interest rates, and sensitive financial markets, the possibility for more than modest and temporary increases in government deficits is remote. So egalitarian programs will require tax financing for public spending, at least in countries like the United Kingdom which have relatively low shares of taxation.15 But this raises the fundamental political problems noted earlier--gaining support for tax increases and maintaining discipline over wage bargaining.

What additional constraints does internationalization impose on the design of taxation packages? The most dramatic suggestion is that taxes on corporate profits and withholding taxes on interest income will tend to zero as countries compete to obtain direct investment and retain portfolio funds. This would leave only resident-based taxes on profit incomes, which could only function if all European, and probably other, governments provided the relevant information about cross-border interest and dividend payments. The European Union's Expert Committee16 found a general tendency across member countries to reduce the variability of tax systems in the 1980s. Corporate tax rates declined a little, but their decline was balanced by cuts in the value of investment incentives so that the importance of corporation tax revenue actually increased. Top rates of tax on dividends and interest income fell, but this applied to relatively less-mobile labor income as well, and was not obviously a response to tax competition. Taxation of capital has been rising only a bit slower than taxation of labor over the past 20 years.17 Thus fears of a headlong collapse of taxation of profits are overstated, though of course coordination at the European level is highly desirable.

There seems little convergence in average shares of GDP taken in public expenditure. The idea that Europe can no longer "afford" its welfare state can only mean that excessive levels of private and social consumption are squeezing the resources for investment in physical capital, R&D, and so forth which are necessary to maintain productivity growth and competitiveness on world markets. The claim is really that workers have gained too high a share of national output, and the most easily squeezed element (politically) is social spending. This argument is particularly prominent in Germany, where profitability is at an exceptionally low level, union strength is relatively intact, and the burdens of German unification have imposed additional strains on public financing.18 This problem is not fundamentally due to internationalization. But it can be expressed in terms of weakened competitiveness, which provides ammunition for business in its political campaign against public spending within countries, and against attempts to level up standards of social security across the European Union.

Conclusion

Trends to internationalization obviously do constrain domestic policy in a variety of ways. Above all, the internationalization of financial markets has accelerated the speed, drama, and costs of financial markets' retribution on governments whose policies are not deemed credible. But it is important not to identify this problem of credibility solely with narrow class prejudice. Sustainable egalitarian packages face very fundamental domestic obstacles. The costs of these packages have to be voted for; the conflict likely to be released by a sustained expansion of employment has somehow to be contained and channeled into directions consistent with maintaining the momentum of expansion. Internationalization increases some of the costs to be shouldered in an expansion, and can affect feasible distributions of those costs. In some cases, the impact of internationalization could well tip the balance against an otherwise viable program.

If we were really sure that the basic domestic constraints could successfully be negotiated, then attention could legitimately be focused on the costs of internationalization and how to circumvent them, particularly on the issue of international coordination. But given the experience of the past 30 years, we should be wary of blaming the narrow scope for egalitarian programs mainly, let alone exclusively, on internationalization.

---

1 Anthony B. Atkinson, Lee Rainwater, and Timothy M. Smeeding, Income Distribution in OECD Countries (Paris: OECD, 1995).

2 Trends in employment and pay are analysed in the OECD's OECD Jobs Study: Evidence and Explanations (Paris: OECD, 1994).

3 Andrea Boltho and Andrew Glyn, "Macroeconomic Policies, Public Spending and Employment," International Labour Review (1995): Table 3.

4 Stephen Nickell and B.Bell, "The Collapse in Demand for the Unskilled and Unemployment Across the OECD," Oxford Review Of Economic Policy 11, 1 (1995).

5 OECD, Economic Outlook (June 1995): 30.

6 Willem H. Buiter, Giancarlo Corsetti, and Nouriel Roubini, "Excessive Deficits: Sense and Nonsense in the Treaty of Maastricht," Economic Policy 16 (1993): 58-100.

7 Angus Maddison, Monitoring the World Economy (Paris: OECD, 1995), pp. 37-38.

8 This measure itself exaggerates trade dependence because it also includes the import content of exports.

9 Business for International Settlement, Annual Report (Basel: BIS, 1997).

10 OECD, Historical Statistics 1960-93 (Paris: OECD, 1993): Table 10.10.

11 Lars Calmfors, "Lessons from the Macroeconomic Experience of Sweden," European Journal of Political Economy 9 (1993): 25-72.

12 Adrian Wood, North-South Trade, Employment and Inequality (Oxford: Oxford University Press, 1994).

13 OECD, Economic Outlook (December 1995).

14 See Barry Eichengreen, James Tobin, and Charles Wyplosz, "Two Cases for Sand in the Wheels of International Finance," Economic Journal 105 (January 1995): 162-172.

15 Taxation and social security contributions as a percentage of GDP at market prices ranged in 1993 from 33.1 percent in the UK to 42-44 percent in France, Germany, and Italy, 52-53 percent in Denmark and Sweden. CSO Economic Trends (November 1995).

16 Commission of the European Communities, Report of the Committee of Independent Experts on Company Taxation (Luxemburg: Office for Official Publication of the European Communities, 1992).

17 Dani Rodrik, "Trade, Social Insurance, and the Limits to Globalization," National Bureau of Economic Research Working Paper (1997).

18 Wendy Carlin and David W. Soskice, "Shocks to the System: The German Political Economy Under Stress," National Institute Economic Review (February 1997).