The process of German unification has proved more costly than originally anticipated. This cost includes not only the domestic tax bill and the government borrowing required to finance subsidies to the East, but also the rise in unemployment, fall in domestic output and productivity, and dislocation generated by the financial collapse of many enterprises. There is a widespread perception that the current policy for German integration is seriously inefficient: that its aims could probably be achieved at a fraction of its current cost. Many different employment policy packages have been proposed to address this problem. They have generally been made in isolation from one another, are often mutually exclusive and may be expected to have very different macroeconomic consequences.

These employment policy packages were the focus of a CEPR joint conference with the Ifo-Institut für Wirtschaftsforschung, München, on `Employment and Productivity Policy for German Integration', on 18 May, attended by academic economists and policy-makers. The conference was organized by Dennis J Snower, Professor of Economics at Birkbeck College, London, and Klaus F Zimmermann, Professor of Economics and Director of the Seminar for Labor and Population Economics at the Universität München, Co-Directors of CEPR's Human Resources programme. Financial support from the Anglo-German Foundation for the Study of Industrial Society is gratefully acknowledged. The morning session emphasized the theoretical evaluation of some possible instruments, while the afternoon session concentrated on the discussion of the strategies and policies actually pursued. Each session included three papers followed by a general discussion.

Theoretical Models

William Vickrey (Columbia University) in his paper, `Chock-full Employment Without Inflation', proposed a scheme of marketable rights to value added, or `gross mark-ups', to control the inflation that might otherwise inhibit the adoption of vigorous full employment policies. Each firm should be allotted a quantity of tradable rights to gross mark-ups determined by its investment, for a specified accounting period, at the end of which firms with insufficient rights to cover their operations should face a penalty tax. This should not be expected to raise significant revenues but merely to act as an enforcement device. The method of issuing such rights would control inflation: the sale of additional output at constant prices would require a corresponding additional issue of warrants, so any desired inflation rate could be achieved through a proportionate increase in warrant allotments. At the margin, such a scheme may resemble a subsidy to labour and investment, but the subsidy element in the mark-up scheme is self-financing and does not incur a budgetary drain for the government.

In his paper, `The Fallout from Employment Policies: How to Reduce the Cost of Raising East German Employment', Dennis Snower favoured a revenue- or profit-sharing subsidy, which may be expected to incur lower social and budgetary costs than a wage subsidy under East Germany's current wage bargaining structure. Snower evaluated the two types of subsidies using a model in which unions maximize the utility of a representative consumer according to a Nash bargaining process, incorporating the key features of the East German labour market. Wages are negotiated at sectoral and regional levels and hold uniformly for productive and less productive firms, so wage subsidies are also equal for both types of firms. Whereas wage subsidies compatible with full employment may be expected to generate real wages far in excess of the government's wage targets, profit-sharing subsidies can be shown to permit the achievement of both full employment and target wages.

Richard Portes (CEPR and Birkbeck College, London) stressed the perceived need to limit migration by reducing East-West wage differentials; this leads to strong trade union pressures, in the absence of countervailing opposition by management. In his joint paper with David Begg, `The Case for Wage Subsidies', he called for a simple uniform wage subsidy, declining over time according to a prespecified schedule, and the elimination of other subsidies which have created incentives for lobbying and rent- seeking. As the problem is in the labour market and not in the capital market, this wage subsidy would be a state-contingent, not a time-contingent, second-best policy. The current morass of distortions arising from different investment subsidies may reflect political weakness that the imposition of a uniform wage subsidy could overcome. Moreover, moral hazard will become irrelevant if national agreements lead to wage equality in the two regions by 1994.

Opening the discussion of the morning session, Gerd Illing (Universität München) stated that the design of policy for Eastern Germany's transformation may seem to be a task like that of the West after World War II, when there was also enormous unemployment. Western privatization rules currently prevent Eastern German workers from sharing the former state assets through distribution of vouchers or shares; they can participate only through high wages. He called for subsidies to the marginal firms that cannot survive if market wage conditions are violated, rather than the uniform subsidies suggested by Portes. Another means of attacking high wages is through profit-sharing to increase efficiency. Although this may not be attractive in the short run (when it could lead to loss-sharing), it may enable workers to accumulate necessary investment, transform current into future earnings, and share in firms' accumulated assets.

Heinz König (Zentrum für Europäische Wirtschaftsforschung, Mannheim) questioned Illing's comparison of the current situation with that after World War II, when the wage structure was not biased by trade unions and the Deutschmark was undervalued. Dennis Snower noted that marketable warrants may be an efficient instrument for East Germany, but inflation seems to be more important in the long run; it might be interesting to evaluate a combination of wage subsidies and marketable warrants. Hans- Werner Sinn (Universität München and CEPR) argued that wage subsidies may yield further benefits if the efficient level of migration to the West is considered; entitlements to investment do not reduce migration incentives, but the effects of wage subsidies depend on whether the wage rate is correct for East Germany. Klaus F Zimmermann maintained that there is a case for wage subsidies to slow down adjustment and therefore reduce some of the attendant pressures, although investment wages are more effective in the long run and also reduce mobility. Michael Burda (INSEAD and CEPR) objected to Snower's static approach, arguing that the distortion is not in the labour market but rather in the capital market so that a discussion of capital formation is required. With constant returns to scale, profit-sharing could induce a trade-off between the formation of physical and human capital.

Policies in Practice

Knut Emmerich (Institut für Arbeitsmarkt- und Berufsforschung, Nürnberg) opened the afternoon session with his paper, `Can We Overcome the Employment Crisis in East Germany by Instruments of the Employment Promotion Act?', written with Friedrich Buttler. East German data indicate that labour market policy is playing only a complementary but nevertheless important part in the transformation, where the central challenge is to increase the capital stock through private and public investment. Active labour market policies must take priority over the financing of unemployment. Until now, labour market policy has carried the main burden of restructuring. Short-time work is decreasing in importance, while job creation and further education are playing a greater role. Emmerich reported their study of the Federal Employment Agency's `Mega-Job Creation Measures', which indicated that the Agency is overseeing many projects that the Federal government, Länder and local authorities are obliged to finance collectively. This reflects a failure to combine the Employment Promotion Act with other measures to stimulate employment and economic activity, which a different combination of financial resources would enhance as well as further improving the `bridging function' of labour market policy.

In their paper, `Trade Unions, Wages and Structural Adjustment in the New German States', Michael Burda and Michael Funke (Freie Universität Berlin) showed that the East German `high-wage policy' is not wholly negative in its effects. They first developed a simple two-sector model stressing the role of union- driven collective bargaining, which is expected to slow down structural change by preventing the release of workers into the secondary or new private sectors. In their second, dynamic, `learning-by-doing' model, in which human capital external to the firm accumulates on the job, they considered the alternative possibility that industrial policy is required to guide the economy towards a superior, high-productivity equilibrium. They conclude that multiple equilibria are possible. A high-wage policy rules out a learning-by-doing equilibrium in which the economy specializes in the low-tech, low-human-capital sector. Unions often argue that high wages encourage investment in human capital, maintaining that East Germany's future requires a human capital constellation similar to that in the West, which can only be achieved in a decentralized market economy by offering private agents sufficient incentives to undertake the necessary human capital investment themselves. A high-wage policy is associated with high initial unemployment, which is accompanied by households' retraining in the hope of improving their earnings prospects.

In his paper, `The Role of Government in Restructuring East German Industry', Wolfgang Gerstenberger (Ifo-Institut für Wirtschaftsforschung, München) stated four main objections to a centralized industrial policy: serious information problems; the risk of overtaxing public budgets if existing capacities are maintained at a level far above that justified by market demand; the risk of massive investment errors by the Treuhand; and the need to avoid inconsistencies in adjustment policy. East German performance in terms of the size of capital inflows induced into the new Länder and the speed with which new or modernized production capacities are being created is far better than postulated by those arguing for a more active government role. Nevertheless, a regional industrial policy is necessary to influence industrial portfolios to establish the best possible foundation for regional economic welfare.
Opening the second general discussion, Roland Vaubel (Universität Mannheim) maintained that analyses of policy-makers' real wage problem requires a strict separation of allocational and distributional effects. On allocation, wage flexibility must be established in the growing sector even for unemployed union members, so wages should be below union wages. Unemployment benefits and payments by unions to their unemployed members may make work below union wages unattractive; Vaubel therefore proposed replacing the current system, in which `unemployment payments' are replaced after one year by `unemployment assistance', with prolonged unemployment assistance. On distribution, those with the lowest incomes should receive the largest transfers and work at very low market-clearing wages; this could be implemented through a negative income tax scheme and may even outperform wage subsidies. Models of human capital formation coming from the endogenous growth literature assume full employment; with unemployment at hand, incentives to accumulate human capital are reduced.

Knut Emmerich responded that not all firms in Eastern Germany pay union wages; a negative wage drift can be observed, and some medium-sized firms have even opted out of employers' associations in order to avoid paying union wages, so there is a problem of incentives to move to the right job. Dennis Snower agreed that Vaubel's policy would need to be supplemented by incentives to engage in non-union work and reported that debate is starting in the UK over whether unemployment benefits should be conditional on work on government projects, such as cleaning up the environment.


The papers presented at this conference will be published early next year, in a volume to be edited by Dennis J Snower and Klaus F Zimmermann.