Information about Europe Europa
Journal Issue
Share
Article

Switzerland

Author(s):
International Monetary Fund
Published Date:
June 1998
Share
  • ShareShare
Information about Europe Europa
Show Summary Details

Introduction

1. This Selected Issues paper contains three chapters. The first chapter evaluates Switzerland’s long-run growth and productivity performance. The second chapter analyzes the behavior of Swiss fiscal policy over the business cycle. And the third chapter takes a fresh look at the nature of the tradeoff between inflation and economic activity in Switzerland. A statistical data appendix is also included.

2. Since the mid-1970s, there has been a significant reduction in the rate of economic growth in most industrial countries. This growth slowdown has been particularly marked in Switzerland. During 1960-74, real value added in the business sector grew by about 3½ percent, or close to the growth rate in major European industrial countries. During 1976-96, however, real growth in Switzerland fell to about 1¼ percent compared with averaged real growth of about 2½ percent for other major European industrial countries. Chapter I seeks to shed light on the sources of Switzerland’s disappointing growth performance during the last two decades.

3. Using a standard growth accounting framework, the chapter finds that the growth slowdown in Switzerland since the mid-1970s is largely accounted for by a drastic decline in total factor productivity gains. At the same time, the average rate of return on capital also declined since the mid-1970s. Reflecting a high and relatively stable national savings rate, the economy adjusted by increasing its foreign investment and decreasing its domestic investment. Consequently, the current account moved into a surplus that averaged close to 5 percent of GDP during 1975-95 compared with near balance during 1962-74. At the sectoral level, it is found that the growth slowdown since the mid-1970s was most pronounced in the “sheltered sectors,” i.e. services, construction, and agriculture. The chapter concludes that Switzerland’s marked relative long-run growth slowdown largely reflects long-standing structural rigidities, which have stifled competition and adjustment in the “sheltered sectors” and have increasingly become a brake on growth in a global environment of rapid economic and technological changes. Although Switzerland achieved considerable progress in confronting structural rigidities in recent years, a more comprehensive and accelerated approach to structural reforms could pay substantial dividends over time.

4. Switzerland’s long tradition of sound public finances has been reflected in relatively low fiscal deficits and public debt, notwithstanding the recent marked deterioration of fiscal positions as a consequence of a seven-year stagnation during the 1990s. At the same time, it has often been argued that fiscal policies have contributed little to macroeconomic stability over the business cycle, due to the weakness of automatic fiscal stabilizers and the, at times, procyclical stance of discretionary fiscal policy. Chapter II reports estimates of the automatic and discretionary responses of general government finances to cyclical output movements during 1970-96 and examines the main options for improving the stabilization role of Switzerland’s fiscal policy over the business cycle.

5. The analysis of the cyclical behavior of fiscal balances during 1970-96 suggests that the long lags between income accrual and income tax collections (2-4 years) considerably weakened automatic fiscal stabilizers. Moreover, notwithstanding Switzerland’s moderate inflation rates, the extraordinary length of the income tax collection lags also reduced the real value of tax collections. Discretionary fiscal policy is found to have often operated in a procyclical manner, undermining the operation of the already weak automatic fiscal stabilizers, in particular in years of large excesses or shortfalls of aggregate demand. The chapter argues that three adaptations of present fiscal policy rules would prove beneficial to enhancing macroeconomic stability: one, a significant shortening of the long lags in the collection of income taxes; two, the adoption of fiscal rules that avoid pronounced procyclical discretionary fiscal policies, in particular during periods of significant excesses or shortfalls of aggregate demand; and three, an enhanced coordination of fiscal policy stances among the different levels of governments. Several cantons including the large canton of Zürich are considering shortening the long collection lags for income taxes. At the Confederation level, a constitutional amendment has been proposed that would aim, after the year 2001, at a balanced budget over the business cycle at the Confederation level. At the same time, the new policy rule would provide room for countercyclical discretionary fiscal policy in the face of large aggregate demand disturbances. The chapter concludes that the proposed new fiscal policy rule would likely improve on the present fiscal policy behavior at the Confederation level. At the same time, prospects remain uncertain for enhancing fiscal policy coordination among the different tiers of government, reflecting in part voters’ preferences for a highly decentralized fiscal decision making process.

6. The nature of the tradeoff between inflation and cyclical measures of real activity has important implications for the conduct of macroeconomic stabilization policies. Recent research has argued that a non-linear relationship (“Phillips curve”) between inflation and cyclical measures of real activity places a high premium on successful stabilization policies, in particular if the economy is subject to significant aggregate demand shocks. More specifically, given a non-linear Phillips curve, large swings in cyclical activity are costly because they reduce the economy’s average level of activity. Chapter III reports the results of an empirical investigation of the tradeoff between inflation and economic activity in Switzerland.

7. The empirical analysis in chapter III mainly focuses on the short-run tradeoff between inflation and real activity, although the possibility of a long-run tradeoff is also considered in view of recent research stressing the long-run welfare cost of even moderate inflation rates. The chapter’s analysis supports the hypothesis that a non-linear Phillips curve provides a marginally better fit compared with the common linear specification. Statistical inference is, however, hampered by significant structural breaks in the data series in the mid-1970s and the early 1990s. In view of the comparatively high cyclical variability of economic activity in Switzerland, the chapter concludes that the benefits of effective stabilization policies could nevertheless be substantial. At the same time, the chapter’s results do not support the hypothesis of a long-run tradeoff between inflation and economic activity in Switzerland.

I. Switzerland’s Long-run Growth Slowdown1

8. Since 1976, the average annual rate of growth in real GDP for Switzerland has been substantially less than that recorded during the period 1960-75. Moreover, the real growth rate for Switzerland has been only roughly half the growth rate experienced by other major industrial countries during the past two decades (Figure I-1). How can Switzerland’s disappointing growth performance be explained? Is it due to less rapid expansion in factor inputs or slower productivity growth? Does it reflect other countries “catching up” with Switzerland’s high income level as Switzerland’s technological advantage is reduced? Or has the Swiss economy “run out of steam,” suffering increasingly from structural rigidities and lack of entrepreneurial innovation?

Figure I-1.Switzerland: Business Output, Employment and Capital Stock, 1962-95

(1962=100)

Sources: Analytical Database and National Accounts, OECD.

1/ Figures refer to Western Germany.

9. This chapter attempts to shed some light on these questions. Initially, the standard neo-classical growth accounting framework is used to decompose real GDP growth into the contributions from inputs of productive factors—capital and labor—and from total factor productivity (TFP), the “unexplained” residual. This decomposition shows that more than two thirds of the growth slowdown in Switzerland between 1960-75 and 1976-96 reflects less rapid TFP growth, with most of the remaining slowdown stemming from less rapid capital accumulation. In section B, possible reasons for the slower TFP growth are examined. In the final section, possible routes the economy could take in adapting to lower TFP growth are explored. In this connection, lessons are drawn concerning the beneficial effects on real growth obtained from structural reforms in New Zealand.

A. Accounting for the Long-Run Growth Slowdown

10. During the period 1960-75, average annual real GDP growth in Switzerland (at about 3½ percent) was not out of line with average annual real growth rates of major European industrialized countries, and the United States (Table I-1). From 1976 to 1996 however, Switzerland’s average annual real growth rate dropped by more than 60 percent to just below 1½ percent. Average annual growth rates also fell in other major industrial countries in this period compared with 1960-75, but the growth slowdown in these countries was less pronounced than that experienced by Switzerland. Average annual real growth for the G-7 countries during 1976-96 was nearly 3 percent or roughly twice the average annual real growth rate for Switzerland during this period.

Table I-1.Real GDP Growth, 1961-96
1960-751976-96
Switzerland3.51.4
United States3.62.8
Japan8.23.4
Germany3.72.8
France4.92.2
Italy4.72.4
United Kingdom2.62.1
Canada5.22.7
New Zealand3.51.6
G-74.72.8
EU-44.02.4
Sources: Analytical Databank, OECD.
Sources: Analytical Databank, OECD.

11. To understand the factors that have contributed to this growth slowdown, the standard neoclassical growth accounting framework was utilized. Output growth was decomposed into the contributions from; (i) labor (L)—employed workers or hours worked; (ii) capital stock (K)—the value of buildings, machinery and utility vehicles; and (iii) an “unexplained” residual (s), called TFP growth. The contributions of other inputs to the production process that are not explicitly included—for example, human capital, land or natural resources—would be captured in the residual along with technological progress. Using a Cobb-Douglas function with constant returns to scale, the growth accounting framework weights the production inputs according to their respective income shares:

where α is the capital income share and the notation ΔX/X denotes percentage change. Total factor productivity growth is equal to the weighted (by income shares) sum of capital productivity growth (output growth minus percentage change in the capital stock) and labor productivity growth (output growth minus percentage change in employment or hours worked).

12. The results from the growth accounting exercise for Switzerland and selected other advanced economies are summarized in Table I-2. The sample period—1961-1996—was split in 1975 because that year has been identified as a structural break point for Switzerland.2 The two sub-periods—1961-75 and 1975-1996—contain 15 and 20 annual observations, respectively. In the case of Switzerland, output growth for the business sector (real GDP less the government sector) fell from 3½ percent per annum (pre-1975 period) to 1¼ percent per annum (post-1975 period). This slowdown in average annual output growth was almost 2¼ percentage points or a decline of 60 percent.3 The major industrial countries also experienced slower growth of private sector output but their slowdowns were less sharp than the one exhibited by Switzerland; for the 4 largest European economies, the growth slowdown averaged about 1 percentage point and the slowdown in the United States was around ½ percentage point. By contrast, the Japanese economy experienced a growth slowdown more in line with Switzerland. In Japan, annual average output growth dropped from more than 8¼ percent during 1960-75 to 3½ per cent during 1975-96—a growth slowdown of almost 5 percentage points, or a decline of around 60 percent.

Table I-2.Contributions to Real Value Added Growth in the Business Sector for Selected Countries, 1961-96(Annual percent changes)
(A)(B)(C)(D)(E)
1961-1975Real value added Business Sector 1/CapitalEmploymentTFP Unadj.Avg. Hours WorkedTFP
Switzerland3.51.30.41.8-0.22.0
United States3.60.71.02.00.02.0
Japan8.34.10.63.5-1.14.6
Germany 2/3.71.7-0.42.4-0.73.1
France3.61.60.21.8-0.62.3
Italy5.11.7-0.33.7....
United Kingdom2.61.0-0.31.9....
Canada4.71.91.41.4....
New Zealand3.51.11.01.3....
EU-43.71.5-0.22.4
G-74.71.80.52.4
Sources: Analytical Databank, OECD Economic Outlook, and Employment Outlook 1997 and 1998, OECD.

In the cases where hours worked data are available, the annual growth rate of real value added in the business sector is composed of the contribution from capital (A) plus the contribution from employment (B) plus an adjustment from average hours worked (D) (average hours worked times the labor income share) and growth in TFP (E).

Includes only western Germany.

In the case of average hours worked, Italy is excluded.

(A)(B)(C)(D)(E)
1976-1996Real value added Business SectorCapitalEmploymentTFP Unadj.Avg. Hours WorkedTFP
Switzerland1.30.90.5-0.1-0.50.4
United States3.00.91.50.6-0.10.7
Japan3.52.00.70.9-0.61.5
Germany 2/2.70.90.31.6-0.62.1
France2.40.9-0.11.5-0.62.2
Italy2.70.90.21.6....
United Kingdom2.60.60.31.6-0.42.0
Canada2.91.91.1-0.1-0.30.2
New Zealand1.70.60.60.5-0.61.2
EU-4 3/2.60.80.21.6-0.52.1
G-7 3/3.01.20.81.0-0.31.1
Sources: Analytical Databank, OECD Economic Outlook, and Employment Outlook 1997 and 1998, OECD.

In the cases where hours worked data are available, the annual growth rate of real value added in the business sector is composed of the contribution from capital (A) plus the contribution from employment (B) plus an adjustment from average hours worked (D) (average hours worked times the labor income share) and growth in TFP (E).

Includes only western Germany.

In the case of average hours worked, Italy is excluded.

Sources: Analytical Databank, OECD Economic Outlook, and Employment Outlook 1997 and 1998, OECD.

In the cases where hours worked data are available, the annual growth rate of real value added in the business sector is composed of the contribution from capital (A) plus the contribution from employment (B) plus an adjustment from average hours worked (D) (average hours worked times the labor income share) and growth in TFP (E).

Includes only western Germany.

In the case of average hours worked, Italy is excluded.

13. Estimates of the contributions to output growth from the capital stock and employment are provided in Table I-2. The same capital income share—the average for 1960-96—was used for both periods to avoid biasing the calculations; this also accords reasonably well with actual developments in income shares. Data on capital stock data was calculated by the OECD4, except for the case of Switzerland. The Swiss capital stock was constructed from national sources and includes the large revision to the investment series released in 1997 (see Annex I for details). Employment data for the total number of employees in the nongovernment sectors were taken from the OECD5, while most of the data on hours worked were obtained from national sources.6

14. As shown in Table I-2, more than two thirds of the Swiss growth slowdown can be attributed to slower TFP growth (columns C and E). At the same time, the contribution to output growth from capital accumulation fell by nearly ½ percentage point between the pre-1975 period and the post-1975 period. This lower contribution reflected lower average capital productivity across the two periods as the capital-output ratio continued to rise rapidly (Figure I-2). Indeed, capital productivity growth fell throughout the period 1962-95, and in 1995 capital productivity growth was roughly half its 1962 level. Total factor input contributions were also affected by lower average hours worked (column C), which dropped the labor contribution by about ¼ percentage point. However, Switzerland’s relatively slow output growth in the post-1976 period is only partly explained by slower growth in capital and labor. Switzerland’s annual growth rate for TFP (column E) during 1976-1996 was less half that for G-7 countries and only about one fifth of the TFP growth rate for the largest four European economies (EU-4).

Figure I-2.Switzerland: Labor, Capital and Total Factor Productivity, 1962-96

(1962=100)

Sources: Analytical Database and National Accounts, OECD.

1/ Total Factor Productivity (TFP) growth is equal to a weighted average of the growth In labor and capital productivity. The sample period averages for capital and labor shares are used as weights.

2/ Figures refer to Western Germany.

15. No consistent picture emerges from a comparison with selected industrial countries, although it does appear that TFP growth declined in most industrial countries. The growth contribution from capital formation declined generally among the selected industrial countries with the notable exceptions of the United States and Canada. Among the major industrial countries, the contribution from capital increases fell by around ½ percentage point, while the contribution of capital formation in the four largest EU economies slipped on average by close to ¾ percentage point.7 With the exception of France and Canada, the contribution to growth from employment in the G-7 countries was the same, or larger, during the second sub-period compared with the first sub-period. Average hours worked fell by around ½ percentage point in most countries, with the notable exception of the United States.

16. Hours worked data—which adjusts for changes in statutory work week, overtime, vacation periods, and the share of part-time work—were not readily available for all countries. In the absence of hours worked data, TFP growth was calculated as a residual after substituting the contributions from capital and employment. The results are very similar to those obtained using hours worked data: unadjusted Swiss TFP growth (column C) is estimated to have averaged approximately zero in the post-1976 period, dropping from 1¾ percent per annum during 1960-75. The average decline in unadjusted TFP growth for the EU-4 countries from the pre-1975 to the post-1976 period was around ¾ percentage points. Thus, the Swiss slowdown in TFP growth was more than twice that experienced by the EU-4 economies. By either measure, Switzerland recorded a pronounced decline in TFP growth.

B. Explaining Slower Productivity Growth

17. The quest for explanations for Switzerland’s slower growth during 1975-96 requires an understanding of the growth process. In traditional neoclassical growth models, the steady state rate of growth is determined by exogenous factors, such as growth in the labor force and technological change. In these “older” models, economic policies affect the level of output by removing economic distortions or by changing the capital/labor ratio, but economic policies would not affect long-term growth rates. A broader set of theories, such as endogenous growth models and institutionally based theories, have introduced channels (e.g. via human capital formation, technological change, infrastructure externalities) by which economic policies can change long-term growth rates. In particular, such growth rates could be raised by: (i) a stable macroeconomic environment; (ii) high investment rates in physical capital, including infrastructure, and in human capital; (iii) reduced distortions in both international and domestic trade; and (iv) quick adaptations to a rapidly changing economic environment.

18. Against this background, Switzerland’s relative poor growth performance is all the more puzzling since Switzerland would not seem to lack the preconditions to grow as fast as other industrial countries. Clearly, macroeconomic policies have been “stability oriented” and should be conducive to long-term economic growth. Average annual inflation in Switzerland has been one of the lowest among industrial countries during the postwar period. Public finances have been sound: general government deficits in Switzerland have been smaller on average than those in other industrial countries and the gross public debt/GDP ratio in Switzerland has changed little from 1975 to 1996 (in spite of a rise during the 1990s). Recent evidence points to a negative impact on real growth rates from high marginal tax rates on both human and physical capital.8 However, this consideration does not explain the relative greater growth slowdown in Switzerland: Swiss tax rates have generally been lower than in the most EU countries and the relative positions have not changed significantly. Public spending on education and health care has been found to have a statistically significant long-run effect on real GDP growth9 in Switzerland. However, public spending on these expenditures categories has increased relative to GDP and therefore can not explain the growth slowdown.

19. Insufficient fixed investment has also not been a problem in Switzerland. The gross investment-GDP ratio for Switzerland averaged around 25 percent of GDP during 1976-96—one of the highest ratios in the world. Owing to high domestic investment, the capital-output ratio in Switzerland has trended upwards over the period 1962-96 (Figure I-3). Consistent with a rising capital-output ratio, capital productivity in Switzerland declined. The average rate of return on capital in Switzerland has drifted steadily downwards since the mid-1960s.10 The decline in the rate of return on investment in Switzerland is also reflected in a swelling current account surplus as investors sought the higher rates of return available abroad. While the national saving-GDP ratio remained fairly constant during the 1960-1996 period, the domestic investment-GDP ratio dropped by about 5 percentage points in the mid-1970s. Consequently, the current account position increased from an average of near balance during 1962-74 to an average surplus of about 5 percent of GDP during 1975-95 (Figure I-4).

Figure I-3.Switzerland: Capital-Output Ratio and Average Rate of Return, 1962-96

Sources: Analytical Database and National Accounts, OECD.

1/ Business sector capital (Total economy less government sector).

2/ Figures refer to Western Germany.

Figure I-4.Switzerland: Saving, Investment and the Current Account, 1962-95

Sources: Analytical Database and National Accounts, OECD.

1/ Figures refer to Western Germany.

20. Investment in research and development (R&D) has also been high in Switzerland and has been on an upward trend since the 1970s. Spending on R&D averaged around 2¼ percent of GDP per annum during the 1970s and rose to an annual average of 2¾ percent during the 1990s—which placed Switzerland among the leading R&D spenders in the OECD. According to the number of patents awarded, technical innovation is high. However, the technological content of exports and patent data seems to indicate that Switzerland may have slipped in the knowledge-intensive and high technology areas (see OECD Economic Survey of Switzerland 1996).11 The Swiss workforce is also highly educated with human capital underpinning the country’s high labor productivity (see OECD Economic Survey of Switzerland 1997). The proportion of the adult working-age population that has completed at least upper secondary education is the third highest in the OECD. The apprenticeship system also enhances the quality and effectiveness of the human capital formation and facilitates the transition from education to work. Except for international protection of agriculture and food products, Switzerland scores highly on openness to international trade. Yet, in spite of all these favorable elements, Swiss economic growth seems to have “run out of steam.”

21. In standard neoclassical growth theories, real GDP per worker levels in different economies tend to converge over time in the absence of major impediments. This convergence or “catch up” hypothesis has also found empirical support. Thus, a high-income country like Switzerland would naturally grow less rapidly than lower-income countries. Looking at per capita real GDP (at purchasing-power-parity exchange rates), Switzerland has a higher income level than nearly all industrial countries, which is consistent with the hypothesis of catch up growth. However, the relatively faster growth rate of the United States and its higher relative per capita real GDP level are inconsistent with this theory. In addition, based on real GDP per hour worked (again at PPP exchange rates), Switzerland is not a labor productivity “leader” (Table I-3). The average labor productivity level of the four largest EU countries in 1992 was about 5 percent higher than Switzerland’s labor productivity, while productivity in the United States was around 13 percent higher than Swiss productivity. The largest EU countries and the United States grew faster than Switzerland since 1975 by about 1 percentage point and 1½ percentage point (per annum), respectively. In sum, economic developments in Switzerland compared with those in other economies suggest that income divergence, rather than, income convergence has been taking place.12

Table I-3.Real GDP per Capita and Labor Productivity Levels
GDP per Capita 1/

1994
GDP per hour worked 1/

1992
Switzerland92.387.2
United States100.0100.0
Japan86.468.8
Germany84.694.7
France79.6101.8
Italy72.784.5
United Kingdom72.582.4
Canada81.387.0
Australia75.877.5
New Zealand66.8
G-788.789.8
EU-478.692.1
Source: A. Maddison, “Monitoring the World Economy: 1820-1992,” (Paris: Development Centre of the OECD), 1994.

Adjusted by relative PPPs.

Source: A. Maddison, “Monitoring the World Economy: 1820-1992,” (Paris: Development Centre of the OECD), 1994.

Adjusted by relative PPPs.

22. A clue to Switzerland’s relatively poorer growth performance may be found in structural rigidities. Switzerland has been frequently characterized as a dual economy with a highly competitive open sector engaged in external trade and a sheltered sector oriented toward the domestic market. With the exception of agriculture, which accounted for less than 3½ percent of domestic value added in 1996, the open sector in Switzerland has relative low tariff protection. On the other hand, the sheltered sector has been extensively protected.13 The distribution sector is highly concentrated and, at least until recently, anti-competitive practices, such as price-fixing, market-sharing arrangements, pooling of sales or outright cartel arrangements, have been common.14 In addition, the domestic market was segmented until recently by canton, owing to differences in technical standards, licencing and public procurement practices. Public monopolies in transport, postal services, telecommunications, and electricity resulted in relatively high prices, inefficient operations, and less innovation. In addition, the domestic market led to higher input costs for other sectors and may have hampered the diffusion of the benefits from new technologies.

23. The opportunity costs of the reported rigidities in sheltered product and service sectors have probably been large. While during the 1950s and 1960s, innovations occurred largely in manufacturing industries, in the 1980s and 1990s, rapid technical change has centered in telecommunications, information technologies, and biotechnology, which are part of the sheltered sector in Switzerland. Deregulation can open up new possibilities for productivity gains and growth. In the United States, for example, deregulation of telecommunications, airlines, and trucking reduced unit costs by 25 to 50 percent (Council of Economic Advisors, February 1997). In addition, deregulation helped fuel an impressive process of product innovation. The economic benefits were widespread: input costs to manufacturing industries fell, the distribution system was improved and new marketing techniques were introduced; enhanced telecommunication and information technology have facilitated globalization of production and finance. Some econometric evidence also suggests that product market rigidities have lowered long-run growth in some EU countries.15 A statistically significant negative correlation was found between real growth rates and indices of product and labor market rigidities across different EU countries.16 The regression showed that product market rigidities were about twice as damaging to real growth as labor market rigidities—slower TFP growth was the major transmission mechanism.

24. The level of Swiss consumer prices relative to other European industrial countries can serve as an indicator of the relative degree of domestic competition. An international price comparison based on a standard basket of consumer goods and services was conducted by the OECD. It found that consumer prices in Switzerland in 1990 were about 37 percent higher than prices in EU countries (EU-12). In 1993, Swiss consumer prices were still 35 percent higher than in the same EU countries (Table I-4). Although changes in definitions, productivity effects or VAT differences make relative price comparisons suspect, these factors cannot fully explain the large observed differences.17 It seems reasonable, therefore, to conclude that persistent price premiums in Switzerland reflect in that market.

Table I-4.Switzerland: Price Level Comparison between Switzerland and the EU (12)(Percent difference between Swiss prices and EU prices)
19901993
Food52.051.1
Clothing and footwear12.420.4
Car fuel (rent and power)48.960.2
Household equipment17.024.3
Health care60.443.0
Transportation13.018.3
Education, recreation21.129.8
Other44.621.2
Total private consumption36.634.7
Source: Purchasing Power Parities and Real Expenditure, OECD, 1992 and 1995.
Source: Purchasing Power Parities and Real Expenditure, OECD, 1992 and 1995.

25. The dual character of the Swiss economy would suggest that the sheltered sectors are the principal factor explaining the growth slowdown. However, as sectoral data on the capital stock are not available, the growth accounting framework cannot be utilized. Instead, the annual average growth rates for labor productivity in the manufacturing (a proxy for the “open” sector) sector manufacturing private (or “sheltered”) sector.18 The average annual growth rate for labor productivity in the Swiss manufacturing sector (which accounted for about 30 percent of value added), was 2¾ percent during 1965-75 or substantially below the average annual growth rates in labor productivity in other industrial countries (Table I-5). However, the average annual growth rate for Swiss labor productivity in the manufacturing sector slowed by only ¼ percentage point from 1965-75 to 1976-96. This slowdown in labor productivity growth in manufacturing in Switzerland was much less than the average growth slowdown experienced in the other industrial countries. Thus, Swiss labor productivity growth in manufacturing during 1976-96 was roughly comparable to its growth rate in the pre-1975 period and Switzerland improved its relative position with the average for selected industrial countries. Clearly then, the manufacturing sector has not been the major source of the marked slowdown in Switzerland’s real GDP growth performance. On the other hand, the implied growth slowdown for the sheltered sectors—the non-manufacturing private sector—has been considerable. The average annual growth rate of labor productivity in the sheltered sector fell by 2 percentage points from about 2½ percent during 1965-75 to ½ percent during 1976-96. This contrasts with an implied productivity growth of nearly 2 percent per annum (1976-1996) in the sheltered sector of the EU-4 countries.

Table I-5.Labor Productivity in Manufacturing(Percent annual changes)
1965-751976-96
Switzerland2.72.4
United States3.33.0
Japan8.22.9
Germany 1/4.83.5
France4.13.4
Italy4.11.7
United Kingdom5.53.7
Canada3.41.8
New Zealand6.412.7
G-74.83.0
EU-44.63.2
Source: WEO database.

From 1968-75.

Source: WEO database.

From 1968-75.

26. Preliminary sectoral data from the service sector for the period 1990-1995 provides some more recent evidence on the restructuring process. These data indicate that large shifts—both of production and employment—have taken place among various sub-sectors within the service sector (Figure I-5). While overall real output of the service sector contracted by close to 5 percent, some sub-sectors—such as telecommunications, air transport, insurance and, to a lesser extent, health services—expanded (and in some cases, significantly) in the same period. Wholesale and retail trade, and tourism, on the other hand, experienced a sharp contraction. On average, employment fell in the service sector by close to 4 percent, but some sectors—such as telecommunications, and health services—were net “creators” of jobs. Only two sub-sectors—telecommunications and real estate—posted productivity gains together with employment expansion, while financial services and air transport achieved improved labor productivity through labor shedding.

Figure I-5.Switzerland: Sectoral Change in Real Value Added and Employment, 1990-95 Main Service Sectors

Source: Staff estimates based on country submission.

1/ Sectors to the left (right) of the dotted line experienced an increase (decline) in productivity.

27. The sectoral patterns in the service sector contrast sharply with the sectoral data on manufacturing industries (Figure I-6). While the total manufacturing sector expanded by some 5 percent, some sub-sectors—e.g., machines and vehicles, electronics, and textiles and clothing—contracted. In all sub-sectors of manufacturing, employment fell, contracting, on average, by as much as 16½ percent. Thus, the increased labor productivity in manufacturing is explained by massive labor shedding.

Figure I-6.Switzerland: Sectoral Changes in Real Value Added and Employment, 1990-95 Manufacturing Industries

Source: Staff estimates based on country submission.

1/ Sectors to the left (right) of the dotted line experienced an increase (decline) in productivity.

28. In spite of the preliminary nature of the sectoral data and measurement problems (in particular the split between real output and price mark-ups in the service sector) the sectoral data does suggest that, under the surface of lackluster productivity growth, a process of structural change is taking place in the whole economy. Reflecting its exposure to foreign competition, it is not surprising that the process is much faster, and the productivity gains larger, in the manufacturing sector. But the process of structural change is also taking place in the service sector, partly reflecting the anticipation of increased competition (air transport, financial services, and telecommunications) or structural shifts in foreign demand (tourism).

C. Redressing the Structural Growth Problem

29. Is relatively slow real GDP growth in Switzerland a problem? The answer is not straightforward because real GDP may not be the appropriate measure of social welfare. The quality of life is also influenced by environmental issues and pursuit of leisure—neither are captured in standard national income accounts. Even presuming that per capita real consumption is the appropriate measure, then national income, and not real domestic production is the appropriate income variable to maximize.

30. With these caveats in mind, it is nevertheless useful to ask how the economy might adapt to slower productivity growth and a lower total rate of return on investment. Possibilities include (i) shifting more savings abroad and expanding further the current account surplus, (ii) reducing savings, and (iii) raising productivity growth. To the extent that rates of return remain lower in Switzerland than abroad, holding a greater share of real capital abroad would increase national income.19 A higher steady state ratio of net foreign assets to GDP could be achieved by running a larger current account surplus. While the increase in the current account surplus relative to GDP during the 1990s has undoubtedly an important cyclical component, it may also reflect efforts by the private sector to invest more wealth abroad to take advantage of the higher rates of return. Alternatively, domestic savings could decline in the long run as private households face a lower opportunity cost for current consumption, that is a lower rate of return on domestic investment.20

31. The best solution would be to increase the growth rate of TFP and the rate of return on investment to international levels. Structural reforms aimed mainly at the product market could contribute to faster TFP growth and higher rates of return on investment. With this in mind, the Government’s “revitalization” program has made important progress towards redressing structural rigidities in the sheltered sector. This revitalization program was the Government’s response to the rejection of Switzerland’s participation in the European Economic Area in a referendum in December 1992.

32. One major element of this program is the new Cartel Law. This law was approved by Parliament in October 1995 and came into force on July 1, 1996, although a phase in period continued until January 1, 1997. Under this law, the Competition Commission is given clearer guidelines and an improved instrument to combat anti-competitive commercial practices. The law does not prohibit cartels outright as that would have required a time consuming amendment to the constitution and possibly a referendum. Rather, the new law prohibits the “abuse of dominant” market position and is, in effect, considered to be close to the EU law.21 The law contains a presumption that horizontal agreements that set prices, production volumes or territorial distribution, act to eliminate effective competition and hence are illegal. Limitations of competition—e.g., vertical integration—are not prohibited, if they can be shown to increase economic efficiency, in some way advance a positive economic goal (e.g., consumer welfare), or meet political/social needs. Mergers are to be reported where they exceed certain thresholds—annual turnover of Sw F 2 billion worldwide or Sw F 500 million in Switzerland. These notification thresholds have been criticized for being too high.22 It is, however, too early to assess the economic impact of the law since much depends on how strictly the law is implemented.

33. Another important plank in the revitalization program was the “domestic market law”, which was also approved by Parliament in October 1995. This law came into effect in early 1996. It establishes the fundamental principles for market access throughout the Confederation. Under the new law, all persons established or headquartered in Switzerland have the right to offer goods and services, including personal services, throughout Switzerland. In doing so, it should reduce the market segmentation across cantons. This law is based on the principle of “mutual recognition” of qualifications and technical standards. (This is the same principle adopted in the EU rules on the internal market.) Thus, access to cantonal markets will no longer be regulated by the canton and will no longer be restricted through specific cantonal qualifications. The law on the internal market also opens up public procurement at the federal, cantonal and communal levels, providing for non-discriminatory access for all suppliers and public offers to enhance transparency. Non-discriminatory access to public procurement is however subject to the threshold levels agreed in the Uruguay Round. These relatively high thresholds—given the small scale of some cantons—may effectively leave incumbent producers shielded from competition.

34. The third leg of the revitalization program was the removal of the legal basis for technical barriers to trade in order to open Swiss markets to greater foreign competition. This provision was approved in October 1995 and came into effect in July 1996. It established the guiding principles for the preparation, adoption, and periodic examination of regulations on technical standards. The new technical regulations must be compatible with those in Switzerland’s main trading partners and should effectively reduce the protection stemming from specific Swiss technical standards.

35. Deregulation of key network services is also scheduled, following closely developments in the EU. In line with the opening of competition in the EU, free entry (subject to an assessment of the technical capabilities of the applicant) into the telecommunication market was established from January 1998, and Swiss Telecom, the previous monopoly, is scheduled to be partly privatized (49 percent) in the second half of 1998. The key issue—the level of interconnection prices—is still to be decided, however. Too high a level could limit the degree of effective competition in the industry, while too low a level would put Swiss Telecom at a cost disadvantage. Electricity production and distribution are subject to a regime of concessions, with utilities having a geographical monopoly. Moreover, three-quarters of the electricity sector is owned by local governments. This situation has contributed to Swiss electricity prices being among the highest in Europe and to large price differences within Switzerland. To open up the electricity market, a reform based on the principle of “third party access” has been proposed. This reform would align Switzerland with reforms introduced in the United Kingdom, Sweden, Norway, and New Zealand, and make it compatible with the EU’s internal electricity market. It is intended that the Swiss reform be in place by 1999, the date envisaged for the opening of the EU electricity market.23

36. These comprehensive structural reforms will create more competition in the sheltered sectors, lowering prices to consumers and producers in the traded goods sector. Since the various laws have been in force for a year or less, it is too early to assess their impact on the Swiss economy. In particular, will these reforms pay off in terms of faster TFP growth? In order to answer this question, it might be instructive to review the main lessons from New Zealand’s experience. As in the case of Switzerland, structural reforms in New Zealand came after a long period of slow growth: per capita real GNP in New Zealand dropped from nearly equal that of the United States prior to World War II to half that of the United States in the early 1980s. Prompted by crisis in 1984, a new Government embarked on an impressive agenda of economic reforms, including privatization and deregulation of industry and reduced agricultural protection.24 The bulk of the reforms were implemented between 1984 and 1987, although labor market liberalization was delayed until the early 1990s.

37. New Zealand’s structural reform program consisted inter alia of: (i) an industrial and competition policy focused on the development of a competitive environment in which no sector was singled out by policy interventions; (ii) privatization and commercialization of network services such as telecommunications and electricity; (iii) a comprehensive tax reform; (iv) a sharp reduction in the support and regulation of agrculture; and, finally, (v) comprehensive labor market reforms, including a replacement of centralized wage bargaining structures by decentralized enterprise (or individual) bargaining. In particular, the deregulation of telecommunications led to a rapid fall in prices and improved services. The implementation of the reforms were greatly helped by reforming several sectors at once; in this way, the cost to one sector were compensated by gains to other sectors. The delayed introduction of labor market reforms did probably, however, increase the transition costs by slowing the needed shift of labor from contracting to expanding sectors. Switzerland, however, with its relatively flexible labor market is less likely to encounter such sequencing problems.

38. There is now some evidence that the reforms have redressed the pace of productivity growth in New Zealand. The average rate of return on capital in the business sector increased from about 12 percent in the late 1980s to around 16½ percent during 1991-1996. The annual rate of per capita real growth rose to over 2½ percent during 1991-96 or nearly almost triple its per capita growth rate during 1974-91. While average annual growth rates for New Zealand and Switzerland were similar during 1960-75 and 1976-96, current estimates of the growth rate for potential output in New Zealand are close to 3 percent per annum or at least twice the growth rate for potential output in Switzerland.

39. In short, the main lessons from the New Zealand experience may be summarized as follows. First, comprehensive structural reform can yield more rapid growth. Second, the reform process cannot be expected to be painless; structural reform may result in a period of lower growth, in particular if the reform is incomplete or if there are problems with the sequencing of reforms. Restructuring means moving “resources” to more efficient uses, often by closing businesses and laying off workers. Retraining is therefore important to facilitate the transition of labor from declining to expanding sectors. As evidenced by the sectoral data presented in the previous section of this paper, significant economic restructuring has already been taking place in Switzerland in the early 1990s. Given the relatively flexible labor market in Switzerland and high level of education, the conditions should therefore be congenial for more rapid growth in the future.

Appendix I Swiss Capital Stock Data

40. The Swiss data were constructed directly from published national investment series, using the same methodology as B, Lüscher, and E. Ross (1996). The resulting capital stock series produce a significantly higher capital-output ratio than the OECD data for Switzerland; for 1960, the capital-output ratio of the capital stock—based on national data—was around 2¼ compared to 1½ using the OECD capital stock data. The capital-output ratio capital stock based on the national data is more in line with the level observed in other countries: in the same year, German and the United States capital-output ratios were 2 and 2½, respectively. Abstracting for the level differences, the two series are, however, very similar.

41. The capital stock data were constructed on the basis of three sources:

  • i. “Séries longues des comptes nationaux de la Suisse”, (Bern: Federal Statistical Office), 1992;
  • ii. “Le système de comptabilité nationale”, Séries longues de 1980 a 1995, Méthodes et résultats (Bern: Federal Statistical Office), 1997; and
  • iii. Lüscher, and E. Ross, “Entwicklung der potentiellen Production in der Schweiz,” Geld, Wahrung und Konjuktur, Quarterly Bulletin No. 14 (1), (Zurich: Swiss National Bank), 1996.

42. The capital stock series was constructed in two steps. First, the new investment series from (ii)—covering only 1980-1996—was converted into 1980 prices—the price-base used in the old investment series in—and extended back to 1960 by linking it with the old series from (i). This calculation was undertaken separately for investment in machinery and equipment and commercial construction (total construction minus residential construction)25. A level adjustment was applied to the old series based on the ratio of the new investment to old investment in 1980 for the two categories of investment (Table I-A1, Column A-C) Second, the real investment series (in 1980-prices) was combined with the observed capital stock in 1966 taken from (iii), to construct a capital stocks series from 1960 to 1996 in 1980 prices (Table A1, Column D). A depreciation rate of 20 percent was assumed for machinery and equipment and 4 percent for commercial buildings. This is the same depreciation rates that were used in B. Lüscher, and E. Ross (1996). The original capital stock data from (iii) was adjusted by the same level adjustment applied to the old investment series.

Table I-A1.Switzerland: Investment and Capital Stock for the Business Sector(In million Sw F; 1980 prices)
Investment in Machinery and EquipmentCommercial ConstructionTotal Business Sector InvestmentCapital Stock
196010,3837,52817,911177,259
196112,2898,60320,892184,169
196214,2989,18423,482192,804
196315,1389,93225,070201,890
196415,45711,11426,571211,346
196515,19310,79225,985219,173
196615,19310,66025,853226,064
196715,39510,59525,990232,425
196815,81510,94726,762238,958
196916,50011,69528,195246,318
197018,76512,41831,183255,988
197120,35313,75534,108267,525
197220,80414,66235,467279,169
197320,71915,37136,089290,265
197420,65614,38635,042299,321
197516,99012,78029,770302,313
197614,62411,67026,295301,952
197715,24711,70026,947302,831
197817,25511,97929,234306,320
197918,17312,56830,741311,143
198020,23613,19133,427318,283
198120,00613,94233,948325,190
198219,04113,95532,996330,529
198319,44214,28633,728336,270
198420,15214,62134,772342,668
198521,22314,84836,071349,869
198623,50915,45338,962359,310
198724,56516,11740,682369,438
198826,76217,62144,383382,167
198927,54619,09846,644395,742
199029,49719,99749,494410,772
199128,83019,91448,744423,459
199224,97219,84344,815431,021
199323,72118,81742,538436,072
199425,10418,55043,653442,290
199527,63517,26944,904449,491
199628,09116,20244,294455,375
Source: Staff estimates.
Source: Staff estimates.
1Prepared by Ketil Hviding
2For details on the statistical analysis related to identifying 1975 as a series break for Switzerland, see the chapter “Estimates of Potential Growth and Cyclical Output Gap” in the IMF Staff Country Report No. 97/18 for Switzerland. This series break coincides with the negative supply shock generated by the first oil price shock and the introduction of floating exchange rates.
3This slowdown is even more striking if 1975 is excluded from the sample period (between 1974 and 1976, nongovernment output fell by more than 9 percent). From 1960 to 1974, nongovernment output increased by close to 4 per cent per annum compared with 1½ percent per year during 1976-96, or an annual growth slowdown of 2½ percentage points.
4In the OECD estimates, depreciation rates follow the so-called “sudden death” assumption—the value of capital equipment drops to zero after a specified number of years which differ by type of capital. “Scrapping rates” were set equal for all countries and only productive capital of the business sector was included.
5Data on capital stock, employment and output for western Germany after 1991 where taken from national sources.
6In the case of the United Kingdom, Canada, and New Zealand, the hours worked data were taken from OECD Employment Outlook, 1997.
7For Germany, the data excluded eastern Germany in order to reduce the bias from the reunification, particularly the introduction of a large number of relatively unskilled (or “inappropriately” skilled) labor from the eastern Länder. Using unified Germany for the period after 1990, does not change the results significantly, except for the contribution from reduced working hours in 1976-1996. Using unified Germany data, the contribution from working hours was -0.4 percent compared with -1.1 percent for western Germany.
8See W. Leibfritz, J. Thornton, and A. Bibbee, “Taxation and Economic Performance”, OECD Economics Department Working Paper, (Paris: OECD), 1997).
9For an empirical study of endogenous growth and public spending in Switzerland, see R. Jan Singh and K Weber, “The Composition of Public Expenditure and Economic Growth: Can Anything be Learned from Swiss Data?”, Swiss Journal of Economic and Statistics, Vol.133, 1997, pp. 617-634.
10A large degree of uncertainty is associated with these data. The rate of return on capital is not only dependent on the level of the real capital stock, but also its income share and the unit price of capital used to calculate the nominal value of capital. Observations based on the rate of return on comparable financial assets in Switzerland from 1979 to 1994 suggest that the rate of return on Swiss financial assets is 2-4 percent lower than the rate of return on similar assets denominated in deutsche mark, French francs, lira, yen, or US dollars (see “Switzerland—Selected Background Issues”, SM/95/2).
11These activities tend to be centered in the sheltered sectors which face less competitive pressures.
12Improved environmental standards may explain some of the slowdown in growth. Conventional national accounts do not measure environmental quality. Thus, international growth comparisons based on national accounts could bias the outcome against countries like Switzerland with relatively tough environmental standards (for an extensive discussion of Swiss environmental policies, see OECD Economic Survey of Switzerland, 1996).
13OECD Economic Surveys of Switzerland of 1993, 1996 and 1997.
14OECD Economic Survey of Switzerland, 1993.
15K. Kremers and J. Koedijk, Economic Policy, October 1996, pp. 445-467.
16An index of product market regulation was created as an unweighted average of regulations on business establishment; competition policy; extent of public ownership; restrictions on shop hours; and extent of implementation of the single market.
17Indeed, correcting for VAT differences is likely to reinforce the basic message because indirect taxes are lower in Switzerland than in most EU countries.
18This categorization does not correspond perfectly to the distinction between the open and sheltered sectors, but it is the best that can be done given the data limitations.
19Portfolio theory also requires an analysis of the risks—the variance and covariance matrix of returns. Data on the variability of rates of return on capital are not readily available; however, calculations on the variability of real GDP in Switzerland, western Germany and the United States—a proxy for variance—indicate that real GDP variability in Switzerland is greater than in Germany and the United States. Also, foreign business cycles are likely not to be perfectly correlated with the Swiss business cycle.
20Such action would only be welfare enhancing in the presence of capital account restrictions that effectively foreclosed the option of investing abroad to obtain the higher rates of return available there.
21See D. Neven and T. von Ungern-Sternberg, “Competition Policy in Switzerland”, Center for Economic Policy Research Discussion Paper, No. 1416, June 1996.
22See OECD Economic Survey of Switzerland 1996-1997.
23“Third party access” is the general model for the planned EU electricity market, although France and Austria opted for the “single buyer” model.
24For an accessible account of New Zealand’s reforms see L. Evans, A. Grimes, B. Wilkinson, and D. Teece, “Economic Reform in New Zealand 1984-95: The Pursuit of Efficiency,” Journal of Economic Literature, December 1996.
25Since for the period 1960-1975 no residential construction data were available, commercial construction was assumed to be a constant fraction of total construction. The fraction was set equal to the average of 1980 to 1996.

Other Resources Citing This Publication