This Selected Issues paper and Statistical Appendix presents estimates of potential output and the output gap for Austria to identify the scope for sustainable noninflation growth and allow an assessment of the current stance of macroeconomic policies. The estimates of the cyclical fluctuations in Austria are compared with those of the other countries of the European Union to provide the basis for an assessment of the relative economic benefits and constraints for Austria in the context of its participation in European Monetary Union, both in the short and longer term.

Abstract

This Selected Issues paper and Statistical Appendix presents estimates of potential output and the output gap for Austria to identify the scope for sustainable noninflation growth and allow an assessment of the current stance of macroeconomic policies. The estimates of the cyclical fluctuations in Austria are compared with those of the other countries of the European Union to provide the basis for an assessment of the relative economic benefits and constraints for Austria in the context of its participation in European Monetary Union, both in the short and longer term.

I. Potential Output And The Output Gap in Austria: A European Perspective1

A. Introduction

1. This chapter presents estimates of potential output and the output gap for Austria, in order to identify the scope for sustainable noninflationary growth and allow an assessment of the current stance of macroeconomic policies.2 The estimates of the cyclical fluctuations in Austria are then compared with those of the other countries of the European Union (EU), in order to provide the basis for an assessment of the relative economic benefits and constraints for Austria in the context of its participation in EMU, both in the short and longer term.

2. The chapter is organized as follows. Section B briefly describes alternative methods for estimating potential output and the output gap. Section C presents the estimates obtained for Austria and discusses recent developments in the determinants of potential output—including the NAIRU—and the medium-term outlook. Section D compares the main features of the Austrian business cycle, particularly its amplitude and degree of synchronization, with those of other EU countries, and completes the investigation by an econometric analysis of the principal characteristics and correlations of demand and supply shocks in these countries. Section E summarizes the main findings.

B. Approaches to Estimating Potential Output and the Output Gap

3. The definition and estimation of the trend and cyclical components of output raise a number of theoretical and empirical questions, which reflect the ongoing controversy over the origins of economic fluctuations. As potential output is an unobserved variable, a number of statistical and economic approaches have been developed to estimate it and the corresponding output gap. Since such measures are known to be fairly uncertain, this paper presents estimates derived from four different techniques, highlighting the sensitivity of the results to alternative methodologies. These methods avoid some of the shortcomings of more traditional approaches, such as the split time-trend.

The Hodrick-Prescott filter

4. The Hodrick-Prescott (HP) filter is a simple smoothing procedure that has become increasingly popular because of its flexibility in tracking the characteristics of the fluctuations in trend output. Trend output (denoted by y*) derived using the HP filter is obtained by minimizing a combination of the gap between actual output (y) and trend output and the rate of change in trend output for the whole sample of observations (t):

(1)MinΣt=0T(ytyt*)2+λΣt=2T1[(yt+1*yt*)(yt*yt1*)]2,

where λ determines the degree of smoothness of the trend.

5. The properties and shortcomings of the HP filter have been well documented (Harvey and Jaeger, 1993). A major drawback comes from its inability to extract properly the trend and cyclical components for most macroeconomic variables and the difficulty in identifying the appropriate detrending parameter λ—which is generally overlooked by using arbitrary values popularized by the real business cycle literature. In particular, mechanical detrending based on the HP filter can lead to spurious cyclically with integrated or nearly integrated time series and an excessive smoothing of structural breaks. A second important flaw of the HP filter arises from its high end-sample biases, which reflect the symmetric trending objective of the method across the whole sample and the different constraints that apply within the sample and at its edges. This flaw is particularly severe when the focus of attention is directed at the most recent observations in the sample in an effort to draw conclusions for policy implementation and make projections for the immediate future.

The nonparametric approach by Coe and McDermott (1997)

6. This method identifies trend output from a nonparametric regression whose functional form does not need to be specified. The approach assumes that the trend has an adequate number of derivatives so that it is smooth enough compared with the cyclical component and allows quite flexible functional forms. The originality of this method is that the degree of smoothing—based on the size of the data window used for each observation—is determined by minimizing a fairly general global error criterion. Contrary to the HP filter, there is thus no need to specify an arbitrary smoothing parameter and the method can be implemented uniformly for different countries.

7. However, this nonparametric method is also subject to some fundamental shortcomings that it shares with other detrending techniques. The general error criterion used—and to some extent the shape of the data window (the so-called kernel)—may not be the most appropriate for extracting the trend and cyclical components of output in each country, and estimates appear also to be affected by strong end-sample biases. Insofar, this method has the same operational drawbacks for policy oriented work as the HP filter method.

The structural VAR approach by Blanchard and Quah (1989)

8. This method stems from the traditional Keynesian and neoclassical synthesis, which identifies potential output with the aggregate supply capacity of the economy and cyclical fluctuations with changes in aggregate demand. Based on a vector autoregression (VAR) for output growth and unemployment, the original method identifies structural supply and demand disturbances by assuming that the former have a permanent impact on output, while the latter can have only temporary effects on it.3

9. Formally, a structural model for growth and unemployment, yt = (ΔlnYt, Ut)’

(3)yt=μ+Γ(L)etwhereΓ(L)=Σi=1ΓiLi,Γ0=I2,

with et=(etd,ets)’ the vector of demand and supply shocks, E(et)=0 and E(etet’)=I2, can be derived from the assumed VAR representation of its reduced form:

(4)Φ(L)yt=c+ϵtwhereΦ(L)=Σi=1pΦiLi,Φ0=I2,

with E(ϵt)=0 and E(ϵϵet’=Ω,

and by identifying the transformation ϵt = Aet, A = [aij]1,2, with the standard constraint AA’=Ω and and the long-run restriction: [Φ(1)1]1111a+[Φ(1)1]1212a=0, imposed b the condition that demand disturbances cannot have a permanent effect on output.

10. Compared with other multivariate detrending techniques, this method relies on clear theoretical foundations and does not impose undue restrictions on the short-run dynamics of the permanent component of output. In particular, the estimated potential output is allowed to differ from a strict—and most often unrealistic—random walk (Dupasquier et al., 1997). In addition, the output gap estimates derived by this method are not subject to any end-sample biases. One obvious drawback of this approach is that the identification chosen may not be appropriate in all circumstances. This is true when changes in the unemployment rate do not provide good indications of cyclical developments in output. Standard deviations of the output gap estimates also suggest that these measures are particularly uncertain.4

The production function approach

11. In its simplest form, this approach postulates a simple two-factor Cobb-Douglas production function for the business sector (Giorno et al., 1995):

(5)Ln(Y)=c+αLn(L)+(1α)Ln(K)+tfp+e,

where Y, L, and K are the value added, employment, and capital stock of the business sector, respectively; tfp, the trend total factor productivity (in log form); c, a constant; and e, the residual.

12. With parameter α approximated by labor’s share in value added, the contributions of labor and capital to output can be computed and subtracted from the value added of the business sector (in log form). The trend total factor productivity is then derived by smoothing the residuals of the equation.

13. Potential output for the business sector is then computed as:

(6)Ln(Y*)=c+αLn(L*)+(1α)Ln(K)+tfp,

where L* is the trend labor input of the business sector calculated as:

(7)L*=PwaPart*(1NAIRU)EG,

with Pwa the working age population; Part* the trend participation rate; NAIRU an estimate of the nonaccelerating-inflation rate of unemployment; and EG employment in the government sector.

14. Potential output for the whole economy is then computed by assuming that output of the government sector—measured by the government wage bill—is always at its potential.

15. Compared with other methods, the production function approach can provide useful information on the determinants of potential growth. This approach relies, however, on an overly simplistic representation of the economy, and the estimates of potential output and the output gap are crucially dependent on the NAIRU estimates and sensitive to the detrending techniques used for smoothing the components of the factor inputs. In particular, the estimates from the production function approach also share the end-sample biases that affect the underlying detrending techniques that are used. These estimates may also be affected by measurement errors in factor inputs, particularly in the capital stock.

C. Empirical Estimates of Potential Output and the Output Gap

Estimations

16. The HP filter and the non-parametric filter were implemented on a sample of annual observations for 1960–97, enlarged by medium-term staff projections for real GDP growth in order to reduce the end-point biases for the most recent years. These projections assume growth of 2¾ percent in 1998 and 3 percent in 1999, in line with the latest projections by the Austrian Institute of Economic Research (WIFO, April 1998), followed by 2½ percent in 2000 and 2½ percent each year in 2001–03. As expected, historical estimates of the output gap from the HP filter depend significantly on the choice of the detrending parameter λ. However, for values of λ in a range of 25 to 400, these differences appear quite marginal for recent years. This parameter was thus fixed at 100.

17. The Blanchard and Quah decomposition of output was obtained from a bivariate VAR model for changes in real GDP (in log form) and deviations of the unemployment rate from its trend (derived in turn from the HP filter, including medium-term staff projections in line with WIFO) estimated on annual data for 1963–97.5 Potential output was then computed by a full-sample dynamic simulation with actual supply disturbances while demand disturbances were set at zero.

18. In the production function approach, the parameter α was assumed to be ⅔, close to—albeit somewhat above—simple measures of labor’s share in GDP from national accounts estimates. The NAIRU was assessed on the basis of staff estimates (see below) and other production inputs were smoothed with the HP filter, using staff medium-term projections to reduce end-sample biases. Data for the capital stock of the business sector are from the OECD’s Economic Outlook database, entailing a depreciation rate of about 3 percent a year for 1960–97.

Measures of potential output and the output gap

19. The measures of potential output and the output gap obtained from the four different methods are shown in Table I-1 and Figure I-1. Interestingly, measures derived from the HP filter, the nonparametric, and the production function methods are very similar, while estimates using the decomposition of Blanchard and Quah share their main characteristics.6 All the measures indicate that potential output growth experienced a sharp slowdown in the 1970s, from 4½-5 percent a year at the beginning of the decade to 2–2½ percent in the early 1980s, and that it has fluctuated in this range thereafter. These developments are broadly in line with the experience of most other industrial countries, particularly in the EU, although some of them achieved higher potential growth during the period (Giorno et al., 1995).

Table I-1.

Austria: Alternative Estimates of Potential Growth and Output Gap Estimates

article image
Sources: WIFO; OECD, Economic Outlook; and staff caclulations.

Hodrick-Prescott filter with λ equal to 100 and real GDP growth projections for 1998-2003 identical to those by WIFO for 1998-99 (April 1998).

Non-parametric method of Coe and McDermott (1996) and growth projections for 1998-2003 identical to those by WIFO for 1998-99 (April 1998).

V.A.R. structural approach from Blanchard and Quah (1989) based on a VAR. model of order 2 on annual changes in real GDP (in log) and deviations of the unemployment rate from its trend (estimated with the HP filter). Period of estimation: 1975-97.

Production function approach with projections for 1998-2003 for real GDP and factor inputs identical to those by WIFO for 1998-99 (April 1998).

Figure I-1.
Figure I-1.

Austria: Output Gap Measures from Different Methods

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Sources: WIFO; OECD, Economic Outlook; and staff estimates.1/ Applied on 1960-2003 with λ=100 and real GDP growth projections for 1998-2003 in line with those of WIFO for 1998-99 (April 1998).2/ Method by Coe and McDermott (1997) on 1960-2003 with growth projections for 1998-2003 in line with WIFO for 1998-99 (April 1998).3/ Based on Blanchard and Quah (1989) for 1963-97 with changes in real GDP (in log form) and deviations of the unemployment rate from its trend.4/ Production function approach (Giorno et al., 1995) with medium-term projections for growth and factor inputs in line with WIFO for 1998-99.

20. Fluctuating between 2¼ percent and 2½ percent in the 1990s, potential growth appears to have been limited to the lower part of that range in the most recent years. The different estimates also suggest that the amount of slack in the economy is relatively modest at present. The HP filter, the nonparametric and the production function approach indicate that the output gap was between ¾ percent and 1 percent of GDP in 1997—after 1 percent in 1996—while the VAR structural approach points to a somewhat smaller output gap. With real GDP growth envisaged at 23¾ percent in the latest projections by WIFO (April 1998), this output gap would be reduced to ½ percent of GDP in 1998 and all but closed by 1999. These results are broadly consistent with the findings of existing empirical studies (Hahn and Rünstler, 1996; Fritzer and Glück, 1997; and OECD, 1998). The absence of significant inflationary pressures could, however, reflect a higher amount of slack in the economy.

The NAIRU

21. In order to assess the amount of slack in the labor market, this section presents estimates of the nonaccelerating-inflation rate of unemployment (NAIRU), which were used as a component of the production function approach. Such estimates are known to be highly uncertain,7 and therefore this section will be confined to simple analysis.

22. The Austrian labor market appears more flexible than those of most other EU counties (see chapter II). However, several features suggest that the increase in the overall unemployment rate since the early 1980s—from 1½ percent close to 6½ percent of the total labor force in 1997—reflects a notable rise in its structural component. The top left part of Figure I-2 does indeed point to an adverse shift in the short-run Phillips curve and a marked increase in the NAIRU during the period. Apparent shifts in the Beveridge curve and in the standard Okun relation also support this view.

Figure I-2.
Figure I-2.

Austria: Structural Unemployment Indicators1/

(In percent unless otherwise noted)

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Sources: WIFO; OECD, Economic Outlook; and staff estimates.1/ Unemployment to total labor force (including self employment) ratio.2/ Based on compensation rate in business sector.3/ Vacancies to total employment ratio (in percent).4/ Trend GDP by Hodrick-Prescott filter.5/ See Elmeskov (1993).

23. Simple estimates of the NAWRU (nonaccelerating-wage rate of unemployment, which is assumed identical with the NAIRU in this section) were derived from the simple approach proposed by Elmeskov (1993). This method essentially assumes that changes in wage inflation are proportional to the gaps between actual unemployment and the NAWRU:

(8)Δ2Ln(W)=a(UNAWRU),

where W is an index of nominal wages.

24. With the additional assumption that the NAWRU does not change significantly from one year to another, the NAWRU can thus be simply approximated by:

(9)NAWRU=U(ΔU/Δ3Ln(W))Δ2Ln(W),

and the resulting variable is then smoothed with the HP filter after outliers are eliminated by interpolations.

25. As expected, the results display a marked increase in the estimated NAWRU and indicate that its present level could stand at close to 6 percent of the total labor force (Figure I-2). Indicators of structural unemployment based on a flexible Okun curve and unfilled vacancies and a flexible Beveridge curve—obtained as in Elmeskov (1993)—point to a similar increase in structural unemployment, although of a somewhat smaller magnitude for the latter.8 These estimates of the NAIRU appear consistent with, although somewhat higher on average than, those of existing empirical studies, which would tend to point to an unemployment gap of ¼ to 1 percent of the labor force in the most recent years. These studies also suggest that the rise in the NAIRU since the early 1980s would be the result of the increase in labor taxation, particularly social security contributions, and of the increase in the share of long-term unemployment (Pichelmann, 1997) and the slower integration of young workers into the labor force (Fritzer and Glück, 1997).

Determinants of potential output and medium-term outlook

26. The production function approach helps to identify the main determinants of potential growth, assessing the respective contributions of capital, labor, and total factor productivity (Table I-2; and Figures I-3, I-4, and I-5). It appears that the sharp slowdown in potential growth experienced in the 1970s was mainly (and almost equally) due to (i) a pronounced reduction in business investment—inducing a decline in capital accumulation from 7–8 percent a year in the early 1970s to about 3 percent a year in the first half of the 1980s—and (ii) a decline of trend factor productivity growth from 2½ percent at the beginning of the decade to 1 percent a year in the early 1980s, a level that is relatively low by international standards. Despite modest growth in the working age population, the contribution of labor to potential growth remained slightly negative, owing to a more than offsetting decline in the participation rate—reflecting in part the increased recourse to early retirement and disability pension schemes—and the rise in the NAIRU.

Table I-2.

Austria: Contributions to Potential GDP Growth and Output Gap in Production Function Approach 1/

article image
Source: Staff calculations.

Production function approach with projections for 1998-2003 for real GDP growth and factor inputs identical to those by WIFO for 1998-99 (April 1998).

Estimated trend total factor productivity.

Figure I-3.
Figure I-3.

Austria: Potential GDP and Output Gap Estimated by Production Function Method 1/

Actual and Potential GDP in Business Sector

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Sources: WIFO; OECD? Economic Outlook; and staff estimates.1/ Computations based on the standard production function approach (see Giorno et al., 1995) and staff medium-term assumptions for real GDP growth and factor inputs in line with WIFO projections for 1998-99 (April 1998).
Figure I-4.
Figure I-4.

Austria: Labor Input for Production Function Method

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Sources: WIFO; OECD, Economic Outlook; and staff calculations.1/ Population of age 15-64 (source OECD).2/ Ratio of total labor force (including self-employment) to working age population.3/ Unemployment in percent ot total labor force (including self-employment).4 / Staff calculations.
Figure I-5.
Figure I-5.

Austria: Capital Input and Total Factor Productivity for Production Function Method

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Sources: WIFO; OECD, Economic Outlook; and staff calculations.1/ Volume of capital stock in business sector excluding residential (Source OECD).2/ Saff calculations.

27. For the most recent years, low growth in the working age population and a further rise in the NAIRU, together with a stabilization of the labor force participation rate and trend factor productivity growth, seem to have limited potential growth to 2¼ percent a year, despite a slight pick-up in capital accumulation. The recent and projected strong expansion of business investment—particularly for machinery and equipment—and a likely stabilization of the NAIRU could, however, allow a modest acceleration of potential growth to, say, 2½ percent per annum in the next few years. A gradual decline in the NAIRU, allowed by recent and prospective labor market reforms, as well as strong productivity increases, such as those recently achieved in the manufacturing sector, might even push it somewhat beyond that level in the medium term. This underscores the importance of proceeding with further structural reforms in the labor and product markets (see chapter II for a review of these reforms).

28. The present estimates of the NAIRU and the output gap point to a small amount of slack and its virtual elimination by 1999, while most business surveys do not envisage any significant production bottlenecks or labor shortages, and wage and price inflation are projected to remain subdued in the near term. The estimates are undoubtedly subject to a high level of uncertainty, as illustrated by the difficulty of extracting the trend component of labor input and factor productivity growth (Figures I-4 and I-5), and may overestimate the NAIRU and underestimate the remaining output gap. Nevertheless the growth assumptions (based on the latest WIFO projections) do not envisage any significant overheating in the medium term, as output would remain rather close to, albeit somewhat above, its potential in the period ahead. Besides, the recent disinflation is partly explained by increased competition resulting from Austria’s membership in the EU and the opening of the Central and East European countries (CEECs). These factors should continue to prevail—particularly with the implementation of EMU—contributing to moderate inflationary pressures in the years ahead.

D. A European Perspective

29. As Austria is set to participate in EMU from the outset, monetary policy will be determined by the European Central Bank (ECB) from January 1999, on the basis of EMU-wide economic conditions, and fiscal policy will be constrained by the Stability and Growth Pact. With these prospects, interest rates in the countries selected to participate in EMU should further converge this year. Looking ahead, Austria’s balance of economic benefits and costs resulting from EMU will depend on how closely it can form an optimal currency area with the other countries of the union. This section attempts to shed some light on these aspects by assessing the relative amplitude and correlation of Austrian cyclical fluctuations with those of other EU countries.

The output gaps in Austria and the EU

30. Cyclical fluctuations in Austria appear to be of a smaller amplitude than in most other EU countries (Table I-3). The volatility of growth over 1970–97 was indeed similar to that of other core EMU countries, such as Belgium, France, Germany, and the Netherlands, but somewhat lower than in most other EU countries. Meanwhile, employment and particularly the unemployment rate, have been more stable than in practically all other EU countries. Owing to the high credibility of the exchange rate peg to the deutsche mark, inflation also was nearly as stable as in Germany over the period. These cyclical features were also associated with smaller fluctuations in real wages.

Table I-3.

Austria: Amplitude of Cyclical Fluctuations in Austria and the European Union, 1970-97

article image
Sources: WIFO; OECD, Economic Outlook; IMF, World Economic Outlook; and staff calculations.

Compensation rate of business sector deflated by consumer price index.

HP filter applied with λ=100 on 1960-2003 with real GDP growth projections for 1998-2003 by staff for Austria and from the WEO (April 1998) for other countries.

Non-parametric method by Coe and McDermott (1996) applied on 1960-2003 with real GDP growth projections for 1998-2003 by staff for Austria and from the WEO (April 1998) for other countries.

Estimates from the production function for Austria (with medium-term projections for growth and factor inputs in line with WIFO for 1998-99) and from the WEO (April 1998) for other countries.

Based on West Germany only, for fluctuations in growth, employment and the unemployment rate, as well as for output gaps from the HP filter and the non-parametric method.

31. With a standard deviation of 1½ percent of GDP over 1970–97, measures of the output gap from the HP filter, the nonparametric method, as well as from the WEO (based on the production function approach for Austria) converge to indicate that the output gap in Austria has been less volatile than in other EU countries, with the possible exception of France and the Netherlands.9 With a maximum output gap of 2¾ percent of GDP obtained from the HP filter and the production function approach, and 2 percent of GDP obtained from the nonparametric method, the troughs of recessions in Austria appear to have been significantly less deep than in other EU countries. Conversely, phases of overheating were also less severe, with output exceeding its potential by no more than 2½–3 percent during 1970–97.

32. The lower amplitude of the Austrian business cycle could be attributed to higher aggregate real wage flexibility (see chapter II)—resulting from the highly centralized wage negotiations under the social partnership system—and pronounced consumption-smoothing patterns—possibly associated with greater job security under the social partnership system and a large stock of household savings—while macroeconomic policies may have also been more successful, with monetary policy subject to the deutsche mark peg since the early 1980s and a significant countercyclical role for fiscal policy, particularly during recessions. It is likely that some of these factors influenced others, but ascertaining any such interdependence is beyond the scope of this paper.

33. Table I-4 shows that the different measures of the output gap for Austria appear closely correlated with their counterparts in other core EMU countries, as well as with those in Portugal and Spain over 1970–97. Correlation coefficients with output gaps in the EU as a whole and the group of 11 countries selected to participate in EMU from the outset (EU11) reaches as high as 0.7–0.8 over the period. This relatively high synchronization with the core countries of the euro area suggests that monetary policy by the ECB should generally be adapted to Austrian economic conditions and that the net economic benefits of EMU should be higher for Austria than for some other EU countries.

Table I-4

Austria: Correlations of Cyclical Fluctuations in Austria and the European Union, 1970-97

article image
Sources: WIFO; OECD, Economic Outlook; IMF, World Economic Outlook; and staff calculations.

EU countries, except Denmark, Greece, Sweden and the UK. Based on the west part of Germany only.

See footnote 2/ in Table 3.

See footnote 3/ in Table 3.

See footnote 4/ in Table 3.

See footnote 5/ in Table 3.

34. At present, WEO estimates indicate that the amount of slack in the Austrian economy is smaller than in major EU countries such as Germany, France, and Italy and the future euro area as a whole (Figure I-6). This would support the view that a moderate monetary policy tightening in the run-up to, or just after the beginning of, EMU would be unlikely to derail the present recovery in Austria and might be helpful to maintain low inflation. Finally, it is intriguing that estimates of the output gap using the HP filter method point to much closer cyclical positions for Austria and other EMU countries.

Figure I-6.
Figure I-6.

Austria: Output Gap in Austria and the EU

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

1/ Applied with a detrending parameter λ=100 on 1960-2003, with projections for real GDP growth for 1998-2003 by staff for Austria and the WEO (April 1998) for other EU countries.2/ Estimates from the production function for Austria and WEO (April 1998) for other EU countries.3/ Based on the west part of Germany alone.

Demand and supply disturbances in Austria and the EU

35. The present investigation was extended to compare the respective size, correlations, and economic impact of demand and supply disturbances in Austria and in other EU countries. Such disturbances were derived from a Blanchard and Quah decomposition (see sections B and C) for all EU countries.

36. The identification chosen, based on changes in real GDP (in log form) and deviations in the unemployment rate from its trend, may not provide sufficient information on supply and demand shocks for all EU countries, given in particular the likely importance of the unemployment hysteresis effect. The results obtained share some important characteristics with those derived by Bayoumi and Eichengreen (1992) from another identification based on growth and inflation, but there are also some significant differences between the two sets of results.

37. Figure I-7 to I-9 show the correlations of demand and supply shocks for all EU countries with Germany and the entire prospective euro area (EU11), on the basis of estimates for 1963–97, as well as for 1963–89 in order to abstract from the impact of German unification in 1990. From an Austrian perspective, demand shocks appear to have been highly correlated in the past with those of Germany and the EU11 as a whole, a characteristic shared with Belgium and the Netherlands, and—maybe to a smaller extent—France and Spain. While the correlations of supply shocks within the core group of EMU countries appear somewhat smaller than in Bayoumi and Eichengreen (1992), the present estimates indicate that those for Austria are also significantly correlated with those in Germany and the EU11.

Figure I-7.
Figure I-7.

Austria: Correlations of Demand and Supply Disturbances for EU Countries, 1963-97 1/

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Source: Staff calculations.1/ Demand and supply shocks identified by the procedure of Blanchard and Quah (1989).2/ Based on the west part of Germany only.3/ EU11 gathers all EU countries, except Denmark, Greece, Sweden and the United Kingdom.
Figure I-8.
Figure I-8.

Austria: Correlations of Demand and Supply Disturbances for EU Countries, 1963-1989 1/

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Source: Staff calculations.1/ Demand and supply shocks identified by the procedure of Blanchard and Quah (1989).2/ Based on the west part of Germany only.3/ EU11 gathers all EU countries, except Denmark, Greece, Sweden and the United Kingdom.
Figure I-9.
Figure I-9.

Austria: Correlations of Demand and Supply Disturbances for EU Countries, 1963-97 1/

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Source: Staff calculations.1/ Demand and supply shocks identified by the procedure of Blanchard and Quah (1989).2/ Based on the west part of Germany only.3/ EU11 gathers all EU countries, except Denmark, Greece, Sweden and the United Kingdom.4/ Standard deviation of residuals of V.A.R. reduced form due to demand and supply shocks respectively.

38. According to Figure I-9, demand disturbances in Austria would have been of a somewhat smaller amplitude than in most other EU countries, while supply disturbances might have been somewhat higher than the average. However, the difficulty of interpreting the experience of some other countries—including France, Italy, and the United Kingdom—suggests that these results be interpreted with great caution.

39. Figures I-10 and I-11 show how demand and supply disturbances affect output over time in the different EU countries. One striking result for Austria is that the impact of demand disturbances recedes much more quickly than in other EU countries. The typical impact of a demand shock declines steadily and disappears completely after four years, while most other countries seem to experience an overadjustment and some—including Belgium, France, Greece, and Spain—show considerable persistence. Supply disturbances also reach their long-term impact relatively quickly in Austria, in about 4 years, but a number of other EU countries seem to adjust as quickly.

Figure I-10.
Figure I-10.

Austria: Impulse Response Functions of Output to Demand Disturbances 1/

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Source: Staff calculations.1/ Based on an identification of demand and supply shocks from Blanchard and Quah (1989).2/ EU except Denmark, Greece, Sweden and the UK.
Figure I-11.
Figure I-11.

Austria: Impulse Response Functions of Output to Supply Disturbances 1/

Citation: IMF Staff Country Reports 1998, 107; 10.5089/9781451802313.002.A001

Source: Staff calculations.1/ Based on an identification of demand and supply shocks from Blanchard and Quah (1989).2/ EU except Denmark, Greece, Sweden and the UK.

40. These results show that one of the key characteristics of the Austrian economy is its capacity to adjust quickly to demand and—to a lesser extent—to supply shocks. This would tend to suggest that real-wage flexibility and consumption-smoothing patterns are the primary factors—rather than macroeconomic policies—explaining the relatively low amplitude of cyclical fluctuations in Austria. This point would, however, deserve further empirical work.

E. Conclusion

41. This chapter has presented new estimates of trend output and the output gap for Austria according to four different methods. These estimates suggest that potential growth slowed sharply during the 1970s and remained at a modest 2–2½ percent a year thereafter. This slowdown reflected lower capital accumulation and a marked decline in total factor productivity growth during the 1970s, while modest growth in the working age population, lower participation rates for older workers, and a notable rise in the NAIRU restrained labor input growth during the 1980s. These trends underscore the need for further structural reforms in the labor and product markets, in order to stimulate business investment and enhance labor utilization and efficiency.

42. Based on the present calculations, potential growth was close to 2¼ percent per annum in the last few years, but it could accelerate to 2½ percent a year or slightly higher in the medium term, in light of the projected expansion of business investment and the stabilization of the NAIRU. These estimates—though particularly uncertain—point to a relatively small output gap at present, which would likely be all but eliminated by 1999. This argues that a modest monetary tightening in Europe would not be inconsistent with Austria’s cyclical position. But, despite output projected to remain near, or even somewhat above, its potential, the present estimates also indicate that inflationary pressures should remain modest in the coming years, inasmuch as greater competition will continue to exert downward pressure on prices and continued piecemeal reforms improve the functioning of the labor and product markets.

43. Finally, the analysis in this chapter has shown that the Austrian business cycle is of smaller amplitude than those in most other EU countries and is highly correlated with those of countries that have been selected to participate in EMU from the outset, suggesting that Austria could reap higher net benefits from EMU than some of the other participants. These conclusions are supported further by a study of the main characteristics of demand and supply disturbances in Austria and in other EU countries. In particular, it appears that the impact of demand shocks on output tends to recede more quickly in Austria than in other EU countries, indicating that Austria’s high aggregate real-wage flexibility and pronounced consumption-smoothing patterns have contributed to dampening Austria’s cyclical fluctuations in the past.

References

  • Bayoumi T., and B. Eichengreen, 1992, “Is There a Conflict Between EC Enlargement and European Monetary Unification?”, NBER Working Paper No 3950.

    • Search Google Scholar
    • Export Citation
  • Blanchard O.J., and D. Quah, 1989, “The Dynamic Effects of Aggregate Demand and Supply Disturbances”, The American Economic Review, Vol 79.

    • Search Google Scholar
    • Export Citation
  • Coe D.T., and CJ. McDermott, 1997, “Does the Gap Model Work in Asia”, IMF Staff Papers, Vol 44, No 1.

  • Dupasquier C, A. Guay, and P. St-Amant, 1997, “A Comparison of Alternative Methodologies for Estimating Potential Output and the Output Gap”, Bank of Canada Working Paper No 975.

    • Search Google Scholar
    • Export Citation
  • Elmeskov J., 1993, “High and Persistent Unemployment: Assessment of the Problem and its Causes”, OECD Economics Department Working Paper No 132.

    • Search Google Scholar
    • Export Citation
  • Fritzer F., and Glück H., 1997, “A System Approach to the Determination of the NAIRU, Inflation and Potential Output in Austria”, in Monetary Policy and the Inflation Process, BIS-Conference Papers, Vol 4.

    • Search Google Scholar
    • Export Citation
  • Giorno C, P. Richardson, D. Roseveare and P. van der Noord, 1995, “Estimating Potential Output, Output Gaps and Structural Budget Balances”, OECD Economic Department Working Paper No. 152.

    • Search Google Scholar
    • Export Citation
  • Hahn F.R., and G. Runstler, 1996, “Potential-Ouput-Messung fur Österreich”, WIFO Monatsberichte, No 31996.

  • Harvey A.C., and A. Jaeger, 1993, “Detrending, Stylized Facts and the Business Cycle”, Journal of Applied Econometrics, Vol. 8.

  • King R.G., Plosser C. I, Stock J.H., and M.W. Watson, 1991, “Stochastic Trends and Economic Fluctuations”, The American Economic Review, Vol 81.

    • Search Google Scholar
    • Export Citation
  • OECD, 1998, Economic Surveys - Austria.

  • Pichelmann K., and A.U. Schuh, 1997, “The NAIRU Concept”, OECD Working Paper No 178.

  • Staiger D., Stock J.H., and M.W. Watson, 1996, “How Precise Are Estimates of the Natural Rate of Unemployment”, NBER Working Paper No 5477.

    • Search Google Scholar
    • Export Citation
1

Prepared by Antoine Magnier.

2

ln this chapter, the terms “trend” and “potential” output are used interchangeably.

3

Extensions of the method and other types of identification can be found in King et al. (1991) and Bayoumi and Eichengreen (1992).

4

See Staiger et al. (1997) for a conclusions on estimates of the NAIRU.

5

Deviations of the unemployment rate from its trend were used rather than the unemployment rate itself as the latter appears to be a series integrated of order 1, reflecting a steady rise since the beginning of the 1980s (see Blanchard and Quah, 1989, for a discussion of this issue).

6

This feature is not overly surprising, however, as Coe and McDermott have shown that their approach delivers, under certain circumstances, results similar to those of the HP filter method and since the production function approach relies on smoothing some components of factor input with the HP filter.

7

See, for instance, Staiger, Stock and Watson (1996) for a recent study on the precision of conventional and unconventional econometric estimates of the NAIRU in the United States.

8

The indicator based on the Beveridge curve appears less reliable, however, since it is subject to the strong end-of-sample bias associated with the HP filter.

9

The cyclical fluctuations for the EU and the euro area as a whole are of a similar amplitude as in Austria. This low amplitude however reflects the aggregation of national cycles that are not perfectly synchronized.

Austria: Selected Issues and Statistical Appendix
Author: International Monetary Fund
  • View in gallery

    Austria: Output Gap Measures from Different Methods

  • View in gallery

    Austria: Structural Unemployment Indicators1/

    (In percent unless otherwise noted)

  • View in gallery

    Austria: Potential GDP and Output Gap Estimated by Production Function Method 1/

    Actual and Potential GDP in Business Sector

  • View in gallery

    Austria: Labor Input for Production Function Method

  • View in gallery

    Austria: Capital Input and Total Factor Productivity for Production Function Method

  • View in gallery

    Austria: Output Gap in Austria and the EU

  • View in gallery

    Austria: Correlations of Demand and Supply Disturbances for EU Countries, 1963-97 1/

  • View in gallery

    Austria: Correlations of Demand and Supply Disturbances for EU Countries, 1963-1989 1/

  • View in gallery

    Austria: Correlations of Demand and Supply Disturbances for EU Countries, 1963-97 1/

  • View in gallery

    Austria: Impulse Response Functions of Output to Demand Disturbances 1/

  • View in gallery

    Austria: Impulse Response Functions of Output to Supply Disturbances 1/