Kingdom of the Netherlands - Netherlands: Selected Issues

International Monetary Fund

doi:10.5089/9781451829303.002.A003

Jump to Content Jump to Main Navigation

Kingdom of the Netherlands - Netherlands: Selected Issues

Author: International Monetary Fund

Publication Date:: 26 Aug 1996

eISBN:: 9781451829303

Language:: English

Keywords:: ISCR; CR; wage; pension; employment growth; real wage; wage negotiator; wage restraint; growth in the Netherlands; wage indexation; wage excess; Employment; Real wages; Wage moderation; Wages

Download PDF (986.1 KB)

This Selected Issues paper examines the long-term issues with pension expenditures in the Netherlands. The paper highlights that the public pension for a single person is equal to 70 percent of the (statutory) minimum wage. The minimum wage and public pensions thus move in lock-step; they are both by law indexed to contract wages in the private sector. This paper examines the structural policies of the Netherlands. Real wages and employment growth are also analyzed.

Abstract

III. Real Wages and Employment Growth ^1/

1. Introduction

Since the early eighties, employment growth in the Netherlands has been strong, and well in advance of other European countries (Chart 1, top panel). Employment creation was strong also in comparison with historical experience in the Netherlands (Chart 1, bottom panel); when measured in full-time equivalents, employment grew at a rate not witnessed since the first half of the 1960s.

A key factor of the recent period was that the pattern of growth became more labor-intensive (Chart 2, top panel). Although output growth was higher than in the late seventies and early eighties, it was lower than in the late sixties and early seventies—a period marked by significantly slower employment growth. One possible reason for the more labor-intensive pattern of growth could be a change in wage behavior following a period of excessive real wage growth. As illustrated in Chart 2 (bottom panel), in the second half of the seventies, the Netherlands switched from rather rapid real wage growth to very moderate wage growth. This may have contributed to rapid employment growth in two ways: first, by slowing the process of labor shedding in industry; and second, by increasing relative demand for services and thereby—given the labor-intensiveness of that sector—fostering a pattern of growth that was particularly friendly to employment.

The paper is organized as follows. Section 2 seeks to shed some light on the question why wage growth in the Netherlands was so restrained in the eighties. It is argued that the disastrous economic situation in the early eighties convinced people of the necessity of wage moderation and that the consequent high rate of employment growth helped create a durable consensus on the desirability of sustained wage moderation. Section 3 discusses the impact of this moderation in real wage growth in the Netherlands on employment growth and the sectoral composition of output growth. It will argue that the sheltered sector profited even more from wage restraint than the open sector. ^2/ Section 4 concludes.

2. The reasons for low real wage growth

To understand the switch from fast real wage growth to very moderate growth that took place in the late seventies and early eighties, an historical detour is needed. Following very tight labor market conditions in the first half of the sixties, real wage growth in the second half of the sixties in the Netherlands was very strong. As wage growth outpaced labor productivity increases, the competitive position of the Dutch economy—as measured by the real exchange rate—gradually deteriorated (Chart 3, top panel), and profitability started to decline. ^1/ Profitability declined further in the seventies, when as a result of automatic indexation of wages to prices, the terms of trade deterioration associated with the oil price shocks was fully shifted to profits (Bakker and Sterks, 1989). As a result, the share of labor income in value added rose to record levels (Chart 3, bottom panel), ^2/ and investment, as a percentage of GDP, plummeted (Chart 4, top panel). Firms tried to soften the impact of wage increases on profitability by shedding of labor. As a result, labor productivity growth was strong (Chart 4, bottom panel) and employment growth weak.

It might have been expected that the decline in employment growth would result in lower wage growth. That this did not take place immediately may have been due to several factors. An important factor might have been that the increased resources from natural gas revenues were in large part used to increase government expenditures; the impact of the increase in government expenditures to some extent offset the impact of the loss in competitiveness. Apart from this, the functioning of the Dutch labor market may have prevented a swift adjustment. Wages are set in negotiations—at a branch level—between unions and employers organizations. The resulting collective labor agreements are declared binding for the whole branch by government. Negotiators did not adjust wage demands for three reasons. First, the stagnation in employment growth did initially not result in a large rise in unemployment, since employers and employees used the disability benefits scheme as an alternative unemployment benefit scheme; ^3/ early retirement became popular as an instrument to reduce unemployment; and discouraged workers withdrew from the labor force. Second, there was initially some fear that a moderation of wage growth—through its impact on demand—might have negative effects on employment. Third, wage negotiators had not sufficiently internalized the vicious circle whereby automatic wage indexation coupled with terms of trade deterioration shifted adjustment to profits and then to employment.

Gradually, however, the harmful effects of excessive wage growth on employment growth started to be seen. The combination in the early 1970s of substantial GDP growth and no employment growth had cast doubt on the view that unemployment was a Keynesian problem, and increasingly the idea gained ground that persistent unemployment was not caused by insufficient effective demand but by a lack of profitable productive capacity. This idea was underscored by the introduction at the Central Planning Bureau of a new generation of models, that contained supply-side as well as demand-side elements. In the new models, conventional Keynesian demand equations determining total expenditure were combined with a clay-clay vintage production function with embodied labor augmenting technical progress (Den Hartog and Tjan, 1976). According to these models, the excess growth of real wage costs over the rate of technical progress induces affected employment in two ways: first, it led to scrapping of old labor-intensive vintages of capital; and second, it implied lower profitability, which resulted in a lower level of investment, a lower growth of the capital stock, and fewer new jobs.

In the second half of the seventies wage growth became more moderate. However, the combination of the very high real exchange rate, the low level of profitability, and the recession that followed the second oil crisis, led to a massive shedding of labor and a strong decline in employment. Nonetheless, the combination of more moderate growth of wages and massive labor shedding led to an improvement in the Dutch competitive position, and the real exchange rate depreciated strongly. The gradually emerging consensus on the necessity of wage moderation was formalized in 1982, when unions and employers agreed to moderate wage growth in order to stimulate employment creation. The subsequent employment growth was in large part seen as the result of wage restraint, and by the late 1980s wage moderation had become a virtually unchallenged policy recipe among unions, employers, and political parties in the Netherlands.

It is notable that this shift to wage moderation took place against the background of a substantial growth in the labor supply: this reflected both an increase in the working age population, and an increase in labor force participation, mainly as a result of increased participation of women ^1/ (Chart 5 ^2/). In this connection, causality may have flowed in both directions: the increase in labor supply may in part have been encouraged by the rapid rate of employment creation. It seems reasonable to conclude that, in the 1980s, a virtuous circle was at work that increased labor supply and employment, and kept wages low.

In sum, the disastrous situation in the early eighties convinced people of the necessity of wage moderation: it undermined the “deficient demand” argument, and directed attention instead to the distortions caused by excess real wages. The subsequent high rate of employment growth sustained this consensus. The question arises, however, as to which channels or linkages resulted in the success of the policy of wage moderation.

3. The impact of wage restraint on employment growth

The positive impact of wage restraint on employment growth ^1/ in the Netherlands in recent years is generally considered to have occurred in two ways. First, it increased the competitiveness of the Dutch economy and profitability of new investment, thus encouraging the expansion of productive capacity. Second, it reduced the incentive for firms to acquire new capital equipment that is particularly economical in requirements for labor. On the face of it, these points appear valid: indeed profitability increased strongly in the eighties, and investment rebounded.

However, there are some observations that cast some doubt on whether this story can account fully for the strong Dutch employment performance. If the impact of wage restraint on employment growth would have been through its impact on competitiveness, we would expect that wage moderation was associated with a decline in the real exchange rate measured in terms of unit labor costs. However, as shown in Chart 6, in the last ten years the Dutch real exchange rate hardly changed. Moreover, one would also expect that production in the exposed sector ^2/ would have been stimulated more strongly than production in the sheltered sector—a sector that by definition does not benefit directly from an increase in competitiveness. However, the sectoral components of output growth portray a different picture. Chart 7 shows that whereas in the 1970-1982 period output growth in the sheltered and exposed sector were virtually equal, since the early eighties output growth in the sheltered sector has been substantially faster than in the exposed sector. ^1/

Several factors may have accounted for the strong output (and employment) performance of the sheltered sector. First, the improved performance of the exposed sector and the associated increase in incomes may have increased the demand for services; to the extent that income elasticity of demand for services is high, this could explain why the performance of the sheltered sector improved as well. However, this argument does not readily explain why the performance of the sheltered sector relative to the exposed sector in the eighties was better than in the seventies. A second reason may have been that restraint of real wage growth affected the relative price of services. As services are more labor intensive, a change in wages will have a larger impact on prices of services than on prices of manufacturing (De Gregorio, Giovannini and Krueger, 1993). ^2/ Thus, if real wages decline, the relative price of services will decline. The overall effect of a decline of real wages therefore becomes as follows. If real wages decline, output in the tradeable sector will increase. The associated increase in income raises consumption of non-tradeables. Consumption of non-tradeables also benefits from the decline in the relative price of services. Indeed—as is illustrated more formally in the Appendix—under certain assumptions the combination of the two effects might cause output in the non-tradeable sector to increase faster than output in the tradeable sector. Thus, as real wage growth in the eighties was lower than in the seventies, and, as a result, the relative price of services increased less, this would explain why the relative performance of the services sector was better in the eighties. ^1/

Combining both arguments would lead to the following explanation of the observed facts. With income elasticity of demand for services higher than that for goods, we would, with growing income and constant relative prices, expect an increase of the share of services in total demand. ^2/ This increase will partly be offset as a result of the relative price increase of services. With slow wage growth, the relative price increase will be relatively low, the offset would be low, and we would expect output in services to grow faster than output of goods; with high wage growth, the relative price increase will be higher, the offset will be higher, and the growth rates of goods and services would be much closer.

The impact of wage moderation on employment creation in the sheltered sector adds to its impact in the exposed sector, and a summary of the overall effects of the impact of wage moderation on employment growth would run as follows. For the exposed sector, wage moderation might have improved profitability and investment, and reduced the incentive of firms to increase labor productivity by substituting capital for labor. ^3/ The expansion of the exposed sector might have contributed to the expansion of production and employment in the sheltered sector; moreover, the wage moderation might also have mitigated the relative price increase of services, thereby stimulating demand—and hence output—for services. All in all, employment in manufacturing no longer declined, and employment in services expanded strongly (Chart 7); the overall effect was a strong increase in employment.

4. Conclusion

Since the early eighties, employment growth in the Netherlands has been strong; it was well in advance of other European countries, and also strong in comparison with historical experience in the Netherlands. The strong performance in employment creation was not so much due to high output growth, but rather to growth being labor intensive. A possible candidate for Dutch growth having become more labor intensive may have been the development of real wage growth. In the second half of the seventies, the Netherlands switched from rather rapid real wage growth to very moderate wage growth. The moderation of wage growth—which was in part a reaction to excessive wage growth in the seventies—contributed to rapid employment growth in two ways: first, it slowed the process of labor shedding in industry; and second, it increased relative demand for services, thereby—given the labor-intensiveness of that sector—fostering a pattern of growth that was particularly friendly to employment.

What policy lessons might be drawn from this experience? First, to the extent that real wage increases are excessive—i.e., not warranted by underlying productivity increases—they will have a negative impact on employment. Second, if there have been wage excesses in the past, and wage growth is subsequently moderated, employment growth may be expected to pick up; moreover, the pick-up may be most pronounced in the services sector.

APPENDIX: The Impact of Real Wages on the Composition of Output

The purpose of this appendix is to present a model that illustrates how in a situation of excess labor supply, a decrease of real wages might lead to an increase of the share of non-tradables in production. The model does not pretend to be a description of what actually happened; rather it formalizes the mechanisms that were outlined verbally in the text. The model is partial-equilibrium; it does not include a capital stock, and, hence, investment.

The model

We use the model of De Gregorio, Giovannini, and Krueger (1993). ^1/ In the original model of De Gregorio et al., two periods are taken into account, which allows for current account dynamics. As those dynamics are beyond the scope of this paper, we have changed the model to include only one period. The model is a two-sector model of an open economy characterized by non-competitive goods and labor markets. In the goods market, it is assumed that while the tradable goods sector takes international prices as given, the nontradable goods sector is monopolistically competitive similar to Dixit and Stiglitz (1977); in the labor market, wages are determined in a centralized bargaining arrangement between employers in the two sectors and trade unions.

Demand

The representative consumer solves the following maximization problem:

\begin{matrix} max_{c_{T, c_{N}}} log (c_{T}^{1 - φ} c_{N, 1}^{φ}) & (1) \end{matrix}

subject to:

\begin{matrix} P_{T} c_{T} + P_{N} c_{N} = P_{T} y_{T} + P_{N} Y_{N} & (2) \end{matrix}

The consumer derives utility from the consumption of a basket of nontraded goods (c_N) and a single traded good (c_N). He faces a budget constraint that limits the value of his consumption to the value of income derived from the production in he two sectors. Applying first-order conditions yields the following expressions for consumption in terms of tradable production:

\begin{matrix} c_{T} = y_{T} & (3) \end{matrix}

\begin{matrix} c_{N} = [\frac{φ}{1 - φ} \frac{P T}{P N}] y T & (4) \end{matrix}

Output

In the tradeable goods sector, there is a single firm with a decreasing-returns technology. The production function is:

\begin{matrix} Y_{T, t} = a_{t} L_{T, t}^{α} α < 1 & (5) \end{matrix}

The price of the tradable goods is given exogenously. Setting the marginal productivity of labor equal to the product wage, gives the equilibrium output supply for tradable goods:

\begin{matrix} y T = a_{T}^{\frac{1}{1 - α}} {[\frac{α P T}{W}]}^{\frac{α}{1 - α}} & (6) \end{matrix}

In the nontradable goods sector, the production function for a single firm that produces Y_N(i) (where i=1..n) units of the nontraded goods i employing N_N(i) units of labor is: ^1/

\begin{matrix} Y_{N} (i) = A_{n} (L^{N} (i) - F) L_{N} \geq F & (7) \end{matrix}

Each firm behaves as a monopolist in its own sector. It takes wages as given, and does not incorporate the spillover effects on prices charged by other companies. With these assumptions, prices are set as a markup over wages. Moreover, since the problem is identical for all firms, all goods prices in the nontraded sector are identical and the aggregate nontraded goods price level equals the price of the individual goods:

\begin{matrix} P_{N} = \frac{θ}{1 - θ} \frac{W}{a_{N}} & (8) \end{matrix}

Production in the non-tradable sector is equal to consumption of non-tradables. Substituting equation (8) in (3) yields the following expression for non-tradable goods output:

\begin{matrix} y N = yT [\frac{\frac{φ}{1 - φ} \frac{1 - θ}{θ} a N}{\frac{W}{P T}}] & (9) \end{matrix}

Labor market

Wages are determined by a centralized labor union. Given wages, the level of employment is decided by firms according to their labor demand schedules. ^1/ The union minimizes a quadratic loss function of the deviations of employment and real consumption from their targets: ^2/

\begin{matrix} min_{w} {(L - L)}^{2} + σ {(w - \bar{w})}^{2} & (10) \end{matrix}

Consumption wages have a one-to-one relationship with real product wages:

\begin{matrix} w = \frac{W}{P_{T}^{φ} P_{N}^{1 - φ}} = \frac{{[\frac{W}{P_{T}}]}^{1 - φ}}{{(\frac{θ}{θ - 1})}^{φ}} & (11) \end{matrix}

Thus, we can write demand for labor as a function of either real production or consumption wages. Writing it as a function of consumption wages, and assuming that the nonlinear demand for labor function can be approximated by a linear specification, yields:

\begin{matrix} L^{d} (w) = L_{0} - ϵ w & (12) \end{matrix}

The equilibrium real consumption wage thus equals:

\begin{matrix} w = \frac{σ}{σ + ϵ^{2}} \bar{w} + \frac{ϵ}{σ + ϵ^{2}} (L_{0} - \bar{L}) & (13) \end{matrix}

The impact of real wages on sectoral output

Using the above outlined model of De Gregorio et al, we will now assess the impact of real wages on sectoral output. From the above discussion we will use the following equations:

\begin{matrix} y T = a_{T}^{\frac{1}{1 - α}} {[\frac{α P T}{W}]}^{\frac{α}{1 - α}} & (14) \end{matrix}

\begin{matrix} \frac{P_{T}}{P_{N}} = \frac{\frac{θ}{(1 - θ) a_{N}}}{\frac{W}{P_{T}}} & (15) \end{matrix}

\begin{matrix} \frac{c_{N}}{c_{T}} = \frac{y N}{y T} = \frac{\frac{φ}{1 - φ} \frac{1 - θ}{θ} a N}{\frac{W}{P T}} & (16) \end{matrix}

What happens if real wages decline, for instance, because unions change their preferences, and attach greater importance to employment? First, as follows from equation (14), output in the tradeable sector will go up. Second, as follows from equation (15), the relative price of tradeables will increase. ^1/ As a result, and as demonstrated in equation (16), the share of non-tradeables in consumption and output will go up. The intuition behind this result is as follows. A decline in real wages leads to an increase in output of tradeable goods. The associated increase in income raises consumption of non-tradeables. Consumption of non-tradeables also benefits from the decline in the relative price of services, and the combination of the two effects causes output in the non-tradeable sector to increase faster than output in the tradeable sector.

References

Artus, Jacques R., “The Disequilibrium Real Wage Rate Hypothesis, An Empirical Evaluation,” IMF Staff Papers (1984), pp. 249–302.
- Search Google Scholar
- Export Citation
Bakker, B.B. and C.G.M Sterks (1989), “De onderschatte invloed van ruilvoetverslechteringen” (the underestimated impact of terms-of-trade deteriorations), Economische Statische Berichten, 74 (1989), pp. 1017–1018.
- Search Google Scholar
- Export Citation
Bruno, Michael, and Jeffrey D. Sachs, Economics of Worldwide Stagflation, Basil Blackwell, Oxford, 1985.
- Search Google Scholar
- Export Citation
De Gregorio, José, Alberto Giovannini, and Thomas H. Krueger, “The Behavior of Nontradable Goods Prices in Europe: Evidence and Interpretation,” IMF, WP/9¾5 (1993).
- Search Google Scholar
- Export Citation
De Gregorio, José, Alberto Giovannini and Holger C. Wolff, “International Evidence on Tradables and Nontradables Inflation,” European Economic Review, 38, (1994), pp. 1225–1244.
- Search Google Scholar
- Export Citation
Den Hartog, H. and H.S. Tjan, “Investment, Wages, Prices, and the Demand for Labor, a Clay-Clay Vintage Model for the Netherlands,” De Economist, Vol. 124, 1976, pp. 32–55.
- Search Google Scholar
- Export Citation
Dixit, A. and J. Stiglitz, “Monopolistic Competition and Optimum Product Diversity,” American Economic Review, Vol. 67 (June), (1977).
- Search Google Scholar
- Export Citation
Enders, Klaus, and Horst Herberg, “The Dutch Disease, Consequences, Cures and Calmatives,” Weltwirtschaftliches Archiv, (1983), pp. 473–497.
- Search Google Scholar
- Export Citation
OECD, Revenue Statistics, 1995.
- Search Google Scholar
- Export Citation
Wetenschappelijke Raad voor het Regeringsbeleid, “Een Werkend Perspectief” (a working perspective), Rapporten aan de Regering, 38, Den Haag, (1990).
- Search Google Scholar
- Export Citation

Prepared by Bas B. Bakker.

An appendix provides a model to explain why the sheltered sector might have profited more from wage restraint than the open sector.

The increase in revenues from natural gas may also have played a role in the real appreciation of the Dutch guilder (Dutch disease). In this mechanism, the higher revenues from gas would have increased the balance on current account, thereby exerting upward pressure on the exchange rate and damaging the competitiveness of the non-gas sector (Enders and Herberg (1983)).

As noted by Artus (1984), the increase in the share of labor costs in value added does not necessarily mean that the real rate was too high, as the increase could be warranted by, for instance, long-run changes in production techniques or in the relative availability of capital and labor.

The disability scheme was introduced in 1967. At that time it was expected that there would be no more than one hundred thousand disability benefits recipients. However, until 1993, the number of recipients increased each year, to a peak of 920 thousand in 1993.

The participation rate of women used to be rather low in the Netherlands (Wetenschappelijke Raad voor het Regeringsbeleid, 1990), but has increased strongly, and by 1992, the participation rate for women was above the EU-average (see SM/95/76, Supplement 1). Most women work part-time; survey results suggest that this reflects a preference among women.

The decline in labor supply in 1984 and 1985 is due to the fact that from 1984 onwards, the unemployed aged 57 ½ and older were no longer required to look for a job, and thus were withdrawn from the labor force.

Although moderate wage growth is seen as a very important factor, there might also have been other factors that contributed to Dutch employment growth. For instance, apart from the moderation of general wage increases, there was also a change in behavior of relative wages. Whereas, in the seventies, the minimum wage grew much faster than the average wage, in the eighties this was the other way around. An increase in the relative level of the minimum wage might have negatively affected employment growth in the seventies; the subsequent decline might have stimulated employment. Also, the wedge decreased somewhat in the eighties—although its level is still very high compared to other OECD countries (OECD, 1995).

The Dutch Central Planning Bureau has disaggregated the economy in a sheltered sector and an exposed sector: the sheltered sector roughly corresponds to services, and the exposed sector to manufacturing. Specifically, the open sector consists of agriculture, manufacturing, utilities, and transport, storage and communication; the sheltered sector consists of wholesale and retail trade, restaurants and hotels, finance, insurance, real estate and business services, and community, social and personal services.

In the 1970-1994 period, wage growth in the sheltered and exposed sector and sheltered sector was very similar. Thus, differences in output growth in both sectors are not due to differences in wage developments between both sectors.

It is well known that, over time, as real wages in the various sectors of the economy increase at roughly the same rate, the relative price of sectors with low productivity growth—such as services—go up. (Baumol effect). Thus, even in an economy where wage growth is not excessive (for instance, in the sense that labor markets clear, and average wage increases are in line with average productivity increases) we would observe a correlation between the growth of real wages and the increase in the relative price of services.

To the extent that differences in real wage increases between the seventies and eighties were the result of underlying productivity differences, we would not necessarily associate them with differences in the relative performance of the services sector. However, we argued previously that the low wage growth in the eighties was not the result of lower underlying productivity increases, but was in large part a correction on the excesses of the seventies.

This raise refers to a situation in which relative prices are constant. As relative prices of services increase over time (due to both real wage increases and lower total factor productivity increases in services), we would, with a same development in real output, expect an increase in the share of services in nominal output.

The negative impact excessive real wages might have on employment has been amply discussed in the literature (see, for instance, Bruno and Sachs (1985)).

De Gregorio et. al. used their model to explain the evolution of the relative price between tradable and nontradable goods in a group of European countries.

The presence of a fixed costs insures that each variety is produced by a single firm, and that the total number of varieties and firms is finite.

In the tradeable sector, real wages determine production, which in turn determine employment; in the non-tradeable sector real wages determine prices, which determine demand which in turn determines output and employment.

L could, for instance, be interpreted as labor supply. This interpretation would suggest that if labor supply rises, wage demands will decrease.

The underlying mechanism here is that non-tradeable prices decrease when wages decline, and tradeable prices are given, and hence do not change.

Other Publishers

Save
Cite
Email this content

Share Link

Copy this link, or click below to email it to a friend
Email this content
or copy the link directly:

https://www.elibrary.imf.org/view/journals/002/1996/080/article-A003-en.xml
The link was not copied. Your current browser may not support copying via this button.

Link copied successfully