Wage Policy in East Germany - (SSRN) Papers

Wage Policy in East Germany Matthew Creagan Courtnage University of Oregon and

Libby Rittenberg Colorado College

This paper assesses the impact of East Germany moving toward wage parity with West Germany within a few years on the goal of achieving quick convergence. Results of a simple econometric model of historical growth in West Germany for 1969-1985 suggest that the high wage-to-productivity policy has lowered overall economic growth. We apply these results to East Germany to estimate the amount of investment that would be required to achieve high growth targets in light of the growth-depressing effect of current wage policy. We find that the amount ofinvestment required is far in excess of what Germany has historically maintained. Finally, our analysis shows that high growth becomes a more feasible outcome if wages are raised by less than productivity improvements.

With the collapse of the Berlin Wall in November 1989 and the unification of Germany in October 1990, many predicted that convergence of the two Germanies would be a relatively smooth and easy process. Clearly, some part of the optimism was politically motivated and another part may have been the result of overestimating the strength of the East German economy. But given such a well-endowed "big brother," was it not reasonable to assume that the transformation of East Germany would become the envied model for the rest of the former Soviet-bloc countries? As documented in Lipschitz and McDonald (1990), the initial conditions were nearly ideal. In West Germany, inflation was low, growth was high, the government budget balanced, and financial resources for rebuilding

Comparative Economic Studies

19

Vol. 36, No. 3, Fall 1994

20

COURTNAGE AND RITTENBERG

seemingly abundant. Moreover, East Germany is in fact relatively small compared to West Germany, with about 30% of total German land and 20% of population. Thus, the reasons for the grim statistics on the post-unification economic performance of East Germany have been the subject of considerable concern and speculation. Although the East German economy is growing, the unemployment rate is in excess of 16% (rising to 23% if involuntary parttime employment is also counted), while inflation is running above 8%. Explanations have pointed to external factors such as the sudden collapse of COMECON, and to internal factors such as privatization policy, including both restitution and sales by Treuhandanstalt (Sinn 1992). Also prominent in the discussions has been the impact of wage policy. With wages in East Germany moving rapidly to West German levels and rising far in excess of productivity gains, the argument goes, East German firms have become increasingly unprofitable and thus East Germany's path to recovery increasingly prolonged. The purpose of this paper is to consider the extent to which wage policy is hampering the process of economic convergence. Based on a simple model of growth, we show the decreased likelihood of achieving quick convergence because of the massive investment levels required to compensate for the high and increasing wages under current policy. We conclude that wage levels in East Germany are not economically sensible, especially in relation to the low levels of productivity in the east and the extremely low wageto-productivity ratios in other former eastern-bloc countries. This high-cost policy is diverting resources away from job creation and a much needed capital inflow. In the context of this conclusion, we attempt to estimate the extent to which an alternative, less-aggressive wage policy might enhance growth and thus speed convergence. Section 1 briefly reviews wage policy in West Germany and wage policy that has evolved in East Germany since unification and looks at the theoretical implications of the policy. A simple econometric model of growth is presented in section 2 which allows us to isolate the impact that the West German wage-to-productivity ratio has had on overall West German growth before unification. These results are then used in section 3 to assess, albeit indirectly, the implications of an East German wage structure which is set to converge within West German levels within one year. We then suggest the possible effects of an altered policy which better reflects productivity levels in East Germany. Conclusions are drawn in section 4.

WAGE POLICY IN EAST GERMANY

21

Wages in East and West Germany The labor market in West Germany is characterized by both high wages and high unemployment. Policies leading to these outcomes are germane to the present study because they have been adopted in large measure by East Germany. Figure 1 depicts the unemployment rate in West Germany for the twenty years prior to unification. Most notable is the ratcheting upward of the unemployment rate in the early 1970s and again in the early 1980s. The increase in unemployment is not believed to be the result of structural changes in the labor force that led to a substantially higher natural rate of unemployment. Coe and Krueger (1990) estimate that the natural rate of unemployment in West Germany peaked at 4.25% in 1976 and averaged about 3.25% for the 1980s. Thus, the gap between the actual unemployment rate and the natural rate of unemployment was nearly five percentage points for most of the 1980s. Variations of the insider-outsider model of wage determination have been offered to explain the chronic discrepancy between the actual and natural rates of unemployment. The basic argument is that the wage bargaining process in West Germany has created a situation where insiders (that is, employees via their trade unions), negotiate salaries without regard to the outsiders (that is, the unemployed). They are able to do so because the cost of replacing insiders with outsiders may be high due to employment security provisions, severance pay, the cost of work disruptions, and the like. Excess unemployment persists so long as the negotiated wage does not exceed the competitive market wage plus this turnover cost. This wage bargaining process reduces employment and output, and thus also growth, from what it would otherwise be.' The drop in the West German unemployment rate to 7% at the end of the decade can be attributed to the increase in demand resulting from unification rather than to a decrease in the power of wage bargaining institutions. On the other hand, the double digit unemployment resulting from high wage levels in East Germany does attest to the power of West German wage bargaining institutions. Upon unification, wage rates in ostmarks were exchanged into deutchemarks at par. Thus, in 1990 East German wages were already about 45 % of the West German average. Many East German firms became unprofitable overnight and one would have expected the resulting unemployment to have led to wage decreases (Akerloff 1991, p. 2-3). Instead, the opposite


22

FIGURE

1

FRG Unemployment Rate

1967-1989

Unemployment Rate

8

7

6

s

4 3

2

1967

n

619

711

73

75

77

79

811

n83n85n

817

89

Year

Source: David T. Coe and Thomas Kreuger. 1990. "Wage Determination, the Natural Rate of Unemployment, and Potential Output," in Occasional Paper 75 German Unification: Economic Issues. Washington, D.C.: The International Monetary Fund, p. 115.

has occurred. Wages in East Germany have continued to climb. As of March 1993 they were at about 65% of the West German level (Cohen 1993) and are expected to reach parity toward the end of 1994 or early 1995, though some delay is possible. The perverse impact of these wage developments is illustrated in Figure 2, which represents the theoretical labor market and output scenarios for East and West Germany The bottom portion 00' gives the total labor supply of East and West Germany. Initially, the West German labor supply amounts to OA and East Germany has 0' A. The MRPL curves for East Germany and West Germany are drawn to represent the higher productivity level in West Germany. Had wages in both Germanys been at competitive


23

FIGURE 2

Output and Employment Effects of Wage Policy in East and West Germany

Western

Eastern Germany

Germany

Wage=

Wage=

MRP MR?L

B' Wb'

A*

A

E

B'

Total labor supply for both regions of Germany

levels prior to unification, the wage level in West Germany would have been OW and in East Germany would have been OW. Prior to unification the West German practice of wage bargaining already distorted the labor market picture there. Firms in West Germany faced the higher wage rate of Wb and thus employed only OB workers, instead of OA workers at a wage of W. Involuntary unemployment amounted to BA. Total output in West Germany was thus already below what it would have been if wages had been determined competitively. Specifically West German output was OFB*B instead of OFA*A. Upon unification labor became free to migrate. If the labor market were competitive and undistorted throughout Germany, the wage rate in both parts of Germany would eventually equalize at We following migration of AE workers from East to West. Output in Germany would be at the theoretical maximum of OFE*E plus 0 ' GE*E. The practice of wage bargaining and the security it provides, however, to those already employed would discriminate against migration tendencies. Workers moving to the West in search of employment would

24


effectively lower the market wage rate; however, as long as the competitive wage level plus the turnover cost remained greater than the prevailing wage rate, there would be no incentive to hire these workers. The high level of social security payments provided by the German government also reduces the profitability of migrating to the West. When workers are faced with the minimal likelihood of gaining a relaltively high wage job in the West and the high level of compensation offered by unemployment benefits, the risk factor will induce people to remain in the East. Migration has occurred to some extent in the German economy, but only slowly and not to the extent that free-market, competitive wage theory would predict. As of March 1992, approximately 380,000 workers had chosen to migrate to West Germany, with another 480,000 choosing to commute (Dornbusch 1992, p. 241). What has occurred instead is a maintenance of the wage levels in West Germany and a fairly rapid transfer of these levels to the eastern market. In terms of Figure 2, as the wage rate in East Germany rises above Wn, unemployment there increases and output is less than it would otherwise have been. As wages rise toward the West German level of Wb, output for all of Germany becomes OFB*B plus 0 ' GB '*B ' , output foregone is BB* E*E plus B' B' *E*E, and unemployment in all of Germany reaches BB ' . The intention of reducing wage differentials and overall inequality between East and West Germany, and hence the attraction to migrants, could actually promote the very problem it is trying to solve. Rapid increases in German wages, in advance of underlying productivity gains, are likely to increase unemployment in East Germany, which could be a stronger "push" than the wage differential "pull" (OECD 1992, p. 73). To gain a better perspective on the level of wages in West Germany and East Germany, it is helpful to make a comparison of wage and productivity levels by international standards. Table 1 provides information for hourly compensation and GDP per worker in relation to productivity for selected countries based on 1991 data. Using an index with U.S. wage-productivity ratio set at 100, it is clear that substantial differences exist across countries. Japan and West Germany, which tend to be very capital intensive, have relatively high ratios of wages-to-productivity. The West German ratio of two-to-one stands out among the other countries, reflecting to a large extent the influence of wage bargaining policies. Countries such as Mexico, on the other hand, have large labor surpluses and low wage levels. As of the end of the 1991, the wage to productivity ratio for East Germany was nearly 3:1. This is already extremely high and it certainly does not reflect a capital-scarce, labor-abundant nation. This disregard for standard market principles, which would have wage rates reflecting both productivity


TABLE

25

1

Wages and Productivity in Selected Countries (Index: United States= 100) Country Germany: Western Eastern United States Japan Korea Portugal Mexico Poland Hungary

Manufacturing Wages

GDP Per Worker

146 66

70 23

100 86 26 25 12

100 65 33

7

21

7

30

31

39

Source: Rudiger Dornbusch and Holger Wolf. 1992. "Economic Transition in Eastern Germany," Brookings Papers on Economic Activity 1: 235-272.

and labor scarcity, will constrain output and hamper investment for the foreseeable future. Lastly, it is important to look at the ratios which exist in Hungary and Poland, East Germany's neighbors. For Poland, the ratio of wages-toproductivity equals 1:3 and for Hungary an even lower 1:4. With the fastpaced economic change and development taking place in most of eastern and central Europe, the competition for the world's supply of available capital and investment is great. With wage rates of ten times those existing in nearby countries and productivity at roughly the same level, the incentive for global corporations to invest in East Germany cannot be very strong. This section has documented the extreme misalignment of East German wages. In the next two sections we estimate its impact on the ability of East Germany to achieve convergence with West Germany.

2. Estimating the Impact of Wage Policy on Growth

We estimate the impact that the distorted wage policy of West Germany has had historically on economic growth, using this information to predict the possible repercussions for East Germany as it attempts to achieve convergence. The model is based on a production function which uses investment share and the ratio of wages-to-productivity to predict growth. The production function is estimated using West German data for the period 1969 to 1985.

26


The simplest neoclassical growth model suggests that overall GDP growth depends on growth of factors of production and on technological change, that is, AY Y

=

AA A

+a

AN

N

+13

AIC

K

(1)

where Y=GDP, N= level of employment, K= capital stock, A=level of technology, and a > 0,13 > 0 In order to focus on the effect of the wage-productivity ratio on employment and hence on overall growth, we expect that employment growth MT

is inversely related to the wage-productivity ratio, that is,

(2)

where W= wage rate, P= productivity , and 7> 0 This equation allows for the fact that wages in a country where the insider/outsider model is applicable may not be at the competitive wage equilibrium and, more precisely, that an above-equilibrium wage is expected to discourage employment and hence lower overall growth. In addition, it is important to account for the interdependence of the world's economies. Thus, an additional variable was added to reflect the influence of world growth on German growth. Substituting equation (2) into equation (1) and adding the world growth variable yields the following estimating equation: AY

W AK AYR =a+b+c +d P K YR

(3)

where YR= world GDP. All coefficients except b are expected to be positive. Because East Germany adopted the West German model, the regression is based on West German aggregate data for 1969-1985. While the application of the results to the East German future may not be ideal, it is appropriate because of the wholesale unification that is now underway. The


27

use of data on pre-unification East Germany seems even less appropriate because of the dramatic paradigm shift that occurred after 1989, even if the data on East Germany were reliable. Moreover, since East Germany has essentially adopted West German economic policies, in particular those related to the labor market, the estimates obtained seem to offer a likely future scenario for East Germany. The results are shown in Table 2. The data are from the World Bank's World Tables (1990). Growth is measured by the growth rate of real GDP. The capital growth variable is measured as the share of investment in GDP, I/Y, expressed as a percent. The data for wages and productivity are based upon an index with 1980 equal to 100. The first regression contains only the aforementioned variables with no correction for serial correlation. The second regression includes a dummy variable for 1976, which was a significant outlier in the model. This may be due to the fact that 1975 was a recession year for West Germany. The actual bounce-back of the economy was much greater than that predicted by the model for 1976. The third regression corrects for first-order serial correlation, which brought the Durbin-Watson statistic very close to the upper bound. This regression will be used in order to estimate the impact of wage convergence on East Germany in terms of growth and investment levels required to achieve specific levels of growth. All of the coefficients have the expected signs and are statistically significant. They suggest that the higher wage-productivity ratio, the lower the overall growth rate, while a higher investment percentage and a higher world growth rate encourage country growth. The estimated impact of the wage-to-productivity ratio on the overall growth rate is substantial. For example a 0.5 increase in that variable would, ceteris paribus, lower the overall growth rate by about 2.25 % . Coe and Krueger (1990) similarly concluded that high wage rates led to unemployment above the natural rate and reduced Germany's growth in the 1970s and 1980s. These results will be used in the next section to analyze the impact of the current high wage policy for East German growth and to assess the benefits to Germany of slowing down the movement toward wage parity. 3. Implications

for East Germany

As shown in Table 1, current productivity levels in the eastern part of the country are only about 30% of those in the western part. Based upon


28

TABLE 2

Estimating Growth in Western Germany, 1969-1985 (Dependent Variable: GDP Growth AY/Y) Independent Variables Constant

(1)

(2)

(3)

5.72 (2.90)

6.18 (4.29)

4.98

-4.05

W/P

-6.33

(2.50)

-4.49

(1.61)

(3.27)

(2.08)

I/Y

0.44 (3.19)

0.58 (5.38)

0.43 (2.90)

AYR/YR

0.85 (5.93)

0.70 (6.19)

0.71 (6.77)

2.50 (3.55

2.38 (4.74)

1976 Dummy

0.36 (1.31)

Autocorrelation Coefficient Adj.

R2

D-W

.87 1.35

.95 1.04

.97 1.65

T-statistics in parentheses.

the present estimates of wage levels in East Germany, they averaged about 70% of western levels for 1993. This will increase to 85% in 1994 and finally achieve parity in 1995. In the regression, West Germany wages/productivity averaged out to around one over the sample period, using the indexed scale. For East Germany, this ratio will depend upon productivity growth in the coming years. Productivity will only begin to rise as the economy adapts to the free-market economy and new investment replaces the region's outdated capital stock. The regression equation estimated for West Germany can be usefully applied to East Germany because it comes out of the system adopted by East Germany and because it can account for differences between the two regions. Specifically, these differences are a much lower productivity level in East Germany than in West Germany and a different wage structure, although this is set to soon converge. Given that we know the projected pattern of wage convergence along with an estimate for world growth, we can use hypothetical levels of GDP growth and productivity gains to estimate the


29

various levels of investment required to achieve specific rates of growth. This will show that the situation in East Germany requires very large investment rates under any scenario. However, the difference between a wage policy which reflects productivity to some extent and the current policy of wage convergence is substantial. The following calculations truly suggest a need for an alternative approach which is more along the lines of economic sensibility. The scenarios shown are for the period during which wage convergence will take place, 1993-1995. Beyond this time frame, wages will be at parity, providing a greater burden to both regions of Germany and making any renegotiation of wage levels extremely difficult. Table 3 summarizes the various levels of investment which would be required for specific levels of East German GDP growth during the period 1993-1995. This table is based on the third regression equation in Table 2, World Bank projections for OECD growth of 2.1%, and hypothetical levels of GDP and productivity growth for East Germany. The results are based upon the trend toward wage convergence expected in 1995 (that is, that wages are 70% of the West German level in 1993, rise to 85% of the West German level in 1994, and achieve parity in 1995). For example, suppose over the three-year period that GDP were to grow at 6% annually and East German productivity closed the east-west gap at an annual rate of 5%; that is, suppose East German productivity rises from 30% to 33% of the West German level between 1993 and 1995. This would require an average investment level equal to 57.6% of East German GDP. With the bulk of West German transfer payments being used for welfare payments, such a figure is highly unlikely or impossible. In comparison, during the period from 1969 to 1985, West German investment relative to GDP averaged out to 23.3% annually. Supposing a more modest GDP growth of 5% and ahighly optimistic 7% annual reduction in the east-west gap, investment would still have to amount to 45% to GDP. Realistically, the East German wage/productivity ratio should be below the western level of one-to-one. This is because the eastern economy faces a large labor surplus and lack of capital. A low ratio would create an incentive for invesment and thus for an increased inflow of capital. Over time this would raise productivity and the labor surplus would begin to diminish. As shown earlier in Table 1, the wage/productivity ratio for former eastern bloc economies is considerably below one. The East German ratio is likely to remain between 2.4 and 2.65 over the next three years, depending upon productivity growth. These figures are enormous when compared


30

TABLE 3

Projected Investment Requirements for 1993-1995 Under Alternate East German GDP and Productivity Growth Rates* Required Yearly Investment (as % of GDP)

East German Annual GDP Growth

Productivity Growth (% reduction in

Assumptions

East-West gap)

47.13 46.00 45.02 58.72 57.60

5%

3.0 5.0 7.0 3.0 5.0 7.0

56.61

*Assuming current wage projections. Based on regression (3), Table 2: AY/Y expected growth)

50/0

5% 10% 10% 10%

-4.98

-4.49 (MP)

+ 0.43 (IN)

4.71 (OECD

Notes: Expected growth rate for OECD countries is assumed to be 2.1%, as projected in World Development Report 1992 (New York: Oxford University Press, Inc., 1992), 32. Wages are assumed to be 70% of WG level in 1993, 85% of WG level in 1994, and parity in 1995. Productivity in EG is assumed to start at 30% of the WG level. The required annual investment rate is the average required for 1993-1995 under the various assumptions concerning EG GDP growth and EG productivity catch-up.

to any country, but especially in comparison to its Central European neighbors. The problem is that, while in theory wage reduction is economically sensible, social and political constraints will not allow East German wages to fall to a level comparable to its eastern neighbors. Table 4 uses the same information, but assumes that wages are allowed to better reflect productivity. Two different scenarios are used in this analysis. The first would set the wage-to-productivity ratio at one, the same ratio that exists in West Germany. The second scenario places the ratio at .5, which is only half the West German level but still higher than the ratios in Poland and Hungary. The results demonstrate the large difference between the existing wage policy and one that would reflect productivity. The level of investment that would be required to attain similar rates of GDP growth drops substantially. For instance, with a wage to productivity ratio of one and GDP growth of 5 % , the level of investment required falls from around 46% to only 30% of GDP. Similarly, for 10% GDP growth, investment falls from 58% to 42%. In either scenario, the required level of investment remains high, but


31

TABLE 4

Projected Investment Requirements for 1993-1995 Under Alternative East German GDP Growth Rates and Wage/Productivity Ratios* Required Yearly Investment (as % of GDP)

East German Annual GDP Growth

Assumptions

East German W/P Ratio

30.09 24.89 41.68 36.48

5%

1.0

5% 10% 10%

0.5

*Assuming a policy which allows wages to better reflect productivity. Based on regression (3), Table 2: AY/Y = -4.98 -4.49 (W/P) + 0.43 (IN)

1.0

0.5 + .71 (OECD

expected growth)

these levels are much more realistic than those which are required under current wage policy to achieve similar growth targets. 4. Conclusions A policy which significantly reduced current wages in East Germany would have four important effects. This regression analysis has focused on the first of these effectsthat the level of investment required to achieve specific levels of GDP growth falls substantially. The second implication centers around the demand for labor. Wages which were allowed to reflect the value of labor would have the effect of increasing the demand for labor. Firms would simply become more competitive and be in a better position to increase their work forces. The result is a lower unemployment rate and a reduced percentage of the population dependent upon western welfare transfers. In effect, a large portion of West German transfer payments is freed up for investment purposes, which is what the eastern economy truly needs. Third, as firms become more competitive because of the reduced wages, the incentive for foreign investment is greatly enhanced. Again, this would reduce the West German burden of revitalizing the eastern economy. Finally, from a negative standpoint, a discrepancy between eastern and western wages could lead to massive migration. This is the main argument used by opponents of a lower wage policy. The evidence, however, does not support such a conclusion. Current unemployment in East Germany is already quite high. Most of the funds that are transferred to East Germany are not being used for investment purposes, due to the massive welfare

32


burden. If this trend continues, migration could become a major force as the East Germany population is faced with minimal employment prospects. A wage policy which substantially cuts wages could induce some migration. However, if unemployment rates were to fall into single digits with the wage cuts, at least citizens would have a job in their native region to weigh against the decision to migrate. The current scenario involves long-term depressed growth, high unemployment, and wages at parity. The alternative scenario would involve higher rates of growth, lower unemployment, and wages which would start from a low base and gradually work upward.

References Akerlof, George A., Helga Hessenius, Andrew K. Rose, and Janet L. Yellen. 1991. "East Germany in from the Cold: The Economic Aftermath of Currency Union," Brookings Papers on Economic Activity 1: 1-105. Coe, David T., and Thomas Krueger. 1990. "Wage Determination, the Natural Rate of Unemployment, and Potential Output," in Occasional Paper 75 German Unification: Economic Issues. Washington, D.C.: The International Monetary Fund. Cohen, Roger. 1993. "The Growth Weight of Germany's Unification," New York Times, 8 March: Cl. Dornbusch, Rudiger, and Holger Wolf. 1992. "Economic Transition in Eastern Germany," Brookings Papers on Economic Activity 1: 235-72. Lipschitz, Leslie, and Donogh McDonald, eds. 1990. German Unification: Economic Issues. Washington, D.C.: The International Monetary Fund. Mayer, Thomas. 1990. "Immigration into West Germany: Historical Perspectives and Policy Implications," in Occasional Paper 75 German Unification: Economic Issues. Washington, D.C.: The International Monetary Fund. , and Gunther Thumann. 1990. "German Democratic Republic: Background and Plans for Reform," in Occasional Paper 75 German Unification: Economic Issues. Washington, D.C.: The International Monetary Fund. Organisation for Economic Co-operation and Development. 1992. OECD Economic Surveys: Germany 1991-1992. Paris: OECD. Sinn, Hans-Werner. 1991. "Macroeconomic Aspects of German Unification," National Bureau of Economic Research, Inc., Working Paper Series No. 3596, January: 1-57. The World Bank. 1992. World Development Report 1992: Development and the Environment. New York: Oxford University Press. . 1990. World Tables 1989-90 Edition. Baltimore, MD: The Johns Hopkins University Press.