Wage Dynamics in Europe Findings from the Wage Dynamics Network - - PowerPoint PPT Presentation

Wage Dynamics in Europe Findings from the Wage Dynamics Network (WDN)

Frank Smets European Central Bank

• Two main objectives:

– Identify the sources and features of wage and labour cost dynamics that are most relevant for monetary policy; – Clarify the relationship between wages, labour costs and prices, both at the firm and macro-economic level

• Inspired by the IPN finding that cross-sector

differences in the flexibility of prices was highly negatively correlated with the labour share, suggesting that stickiness in wages and labour costs may be a driving factor behind the slow adjustment of prices.

WDN Objectives

2

• Institutions and wage structure
• Micro facts and macro interpretations

– Frequency, timing, indexation and wages of new hires – Downward wage rigidities – Pass-through of wages into prices

• Labour markets and the current crisis
• Conclusions

Overview

3

Wage bargaining institutions: basics

Source: EIRO 2006, 2009

SK SE PT NL IT IE GR DE FR DK AT SI LU FI ES CY BE UK BG LT HU EE CZ PL

1 2

20 40 60 80 100

Collective bargaining coverage (% workers) Dominant level of bargaining

firm higher

4

• Union density has declined;
• But because of widespread extension procedures,

the coverage rates have basically remained constant;

• There is some trend towards more

decentralisation (from sector to firm level).

Wage bargaining institutions: trends

5

Wage structure across sectors

• 0.4
• 0.3
• 0.2
• 0.1

0.1 0.2 0.3 0.4 0.5 Sectors(2 digits-nace2)

Wage Differentials: Observed

Observed

6

Wage structure across sectors

• 0.4
• 0.3
• 0.2
• 0.1

0.1 0.2 0.3 0.4 0.5 Sectors(2 digits-nace2)

Wage Differentials: Observed

Observed

Petroleum and chemical, financial and insurance, research and development, utilities, are among the highest paying industries Clothing, leather and textiles, and retailing are among the lowest paying industries

Finance!

7

Wage structure across sectors

• 0.4
• 0.3
• 0.2
• 0.1

0.1 0.2 0.3 0.4 0.5 Sectors(2 digits-nace2)

Wage Differentials: Observed vs. Conditioned

Conditioned Observed

8

• Evidence of non-competitive and rent-sharing

explanations of wage setting.

Wage structure across sectors

TABLE 2.4 Rent sharing and institutions as explanations of wage differentials

(1) (2) (3) (4) (5) (6) Levels Change 0.049*** 0.038*** 0.074*** 0.045*** 0.026* Rents Real gross operating surplus per worker (GOS) (0.014) (0.011) (0.020) (0.016) (0.015) ‐0.347*** ‐0.295*** PM competition % of small firms in the industry (0.057) (0.076) 0.030* Bargaining structures % firms with firm‐level collective agreement *GOS (0.016) 0.062*** Collective agreement

coverage* GOS

(0.020) Observations 526 517 423 229 206 260 R2 0.18 0.24 0.37 0.51 0.60 0.08

Source: Du Caju, Kátay, Lamo, Nicolitsas and Poelhekke (2009). Dependent variable are estimated industry wage differentials after controlling for a large number of observed characteristics as deviations from a measure of aggregate. SES data

9

• Institutions and wage structure
• Micro facts and macro interpretations

– Frequency, timing, indexation and wages of new hires – Downward wage rigidities – Pass-through of wages into prices

• Labour markets and the current crisis
• Conclusions

Overview

10

• The frequency of wage changes is lower than that of

prices:

• Average duration of wages is about 15 months

Frequency of wage changes

Frequency of price and wage changes % of firms

10 20 30 40 50 60 70 more frequent than once a year

• nce a year

less frequent than once a year no pattern

prices wages

Source: WDN survey

11

Frequency of wage changes

Table 3.2 –Price and wage rigidity

(frequency of wage and price changes)

(ordered probit estimates)

PRICES WAGE Construction

• 0.369**
• 0.198**

Trade

• 0.77**

0.108** Market services

• 0.035

0.12** Financial intermediation

• 0.672**

0.21* 20-49

• 0.018
• 0.094*

50-199

• 0.124**
• 0.207**

>200

• 0.168**
• 0.331**

Labour cost share 0.508** 0.054 competitive pressures

• 0.301**

0.01 Export (% of sales)

• 0.139*
• 0.013

share of white collars 0.167** share of high skilled workers 0.087* workforce turnover

• 0.15**
• 0.144**

share of bonuses on total wage bill 0.01

• 0.172**

collective agreement outside the firm

• 0.067
• 0.055

collective agreement at the firm level

• 0.03
• 0.112*

coverage of collective agreement 0.055 0.089* EPL 0.104** Country dummies yes yes Observations 5340 8993 Notes: (*) and (**) denote statistical significance at 5% and 1%,

• respectively. Source: Druant, Fabiani, Kezdi, Lamo, Martins and

Sabbatini (2009). The dependent variable increases with the degree of rigidity.

• Wages change

– less often with high coverage and strong employment protection; – more often with firm level bargaining and indexation.

• Prices change

– less often with higher labour share; – And more often with higher degree of competition.

12

Timing and synchronisation

Timing of wage and price changes Percentage of firms that change wages and prices in (a) particular month(s)

Note: Germany, Bulgaria and Cyprus not included in the calculations. Source: Druant, Fabiani, Kezdi, Lamo, Martins and Sabbatini (2009).

13

14

Indexation of wages to inflation

Past Expected Past Expected

Total 13.6 3.9 12.9 7.0 36.3 Euro area 16.7 4.1 10.2 5.8 35.7 BE, LU, ES 54.8 11.8 7.9 3.6 78.1 Euro area, no aut. Ind. 6.3 1.9 10.8 6.3 23.9 Non euro area 5.5 3.2 19.8 10.2 38.1

Firm policy of adjusting base wages to inflation: overview

Note: BG, CY and NL not included in the aggregates. Country details in Table 3.3 of the WDN report.

Informal Firm-level policy of adjusting base wages to inflation Automatic Total

– High degree of formal indexation in Belgium, Luxembourg, Cyprus and Spain, consistent with the institutional setting. – In other countries, about one third of the firms have a “policy” that adapts changes in base wages to inflation, mostly to past inflation.

Wages of newly hired workers

• Most of the variation in hours worked over the

business cycle occurs at the extensive margin. Wages of new hires are a key determinant of employment.

• Micro-evidence for the United States and various

European countries suggests that wages of job movers are more pro-cyclical than wages of incumbents:

• Survey by Pissarides (2009) suggests an elasticity
• f about 3; Confirmed by WDN evidence for

Portugal.

• However, direct survey evidence suggests that firms

are reluctant to differentiate wages of newly hired workers and incumbents.

⇒

15

Wages of newly hired workers

• Around 80 % of firms report that internal factors are

the most important factors driving wages of new hires:

Percentage of firms reporting each factor

16

Wages of newly hired workers

• But external labour market conditions are more

important in CEE (36%) than in euro area (15%), mostly reflecting lower coverage.

Pseudo R2: 50%

17

Wages of newly hired workers

• Main reasons are:

– fairness (32/39%) – the fear that such a differentiation may have a negative impact on worker morale and effort (36/35%)

• In low-coverage countries, external labour market

conditions matter more for firms that

– face more competition; – employ more high-skilled workers; – and face a higher turnover.

18

Macro implications

• An estimated DSGE model for the euro area

(1990:Q2-2008:Q4) with sticky prices and wages, wage indexation and labour market frictions conforms quite well with the micro findings:

– The average duration of a wage contract of incumbent workers is estimated to be 4.4 quarters. – Wages of newly hired workers are estimated to be as sticky as those of incumbent workers. – The degree of indexation is estimated to be one third.

19

Macro implications

• Together, these features help replicate the relative

volatility and persistence in hours worked, real wage and inflation in the euro area quite well

Data source: AWM16 database and hours data by Joachim Schroth for the largest 5 economies of the EA. 0.491 0.395 0.274 0.125

• ser. corr.

0.441 0.276 0.192 0.175

• rel. std. dev.

Inflation 0.728 0.708 0.759 0.826

• ser. corr.

1.566 0.542 0.590 0.624

• rel. std. dev.

Real wage 0.862 0.812 0.837 0.785

• ser. corr.

0.637 0.787 0.742 0.721

• rel. std. dev.

Total hours worked Flexible wages for new hires Full wage indexation Benchmark model Euro area data

20

Macro implications

• The considerable degree of real wage rigidity leads to a small,

but persistent real wage response and a relatively stronger employment response to various shocks.

• Example: Tightening of monetary policy

2 4 6 8 10 12 14 16 18 20

• 1.5
• 1
• 0.5
• utput (blue), price deflator (red)

2 4 6 8 10 12 14 16 18 20

• 1.5
• 1
• 0.5

0.5 real hourly wage (blue) & total hours (red) 2 4 6 8 10 12 14 16 18 20 0.05 0.1 0.15 0.2 0.25

• nom. rate

21

Macro implications

• Confirmed by astructural methodologies: e.g. FAVAR

5 10 15 20

• 3
• 2.5
• 2
• 1.5
• 1
• 0.5

0.5 x 10

• 3

Real GDP and Consumption deflator

RGDP PCD 5 10 15 20

• 3
• 2.5
• 2
• 1.5
• 1
• 0.5

0.5 x 10

• 3

Nominal Compensation and Employment

CMP/EMP EMP 5 10 15 20

• 0.2
• 0.1

0.1 0.2 0.3

AWM short interest rate

22

Macro implications

• May explain why the elasticity of wages with respect to

changes in unemployment is smaller in the euro area compared to the US. United States Euro area

Source: Duarte and Marques

23

Policy implications

• Higher real wage rigidity complicates the conduct of

monetary policy in the pursuit of price stability, as it increases the cost of stabilising inflation in the face of cost-push shocks:

– Puts a premium on a firm anchoring of inflation expectations; – Suggests a medium-term orientation to avoid excessive volatility in interest rates and economic activity

• Differences in indexation may lead to persistent

differences in relative unit labour costs, inflation and associated changes in competitiveness:

– Automatic indexation to past inflation should be avoided;

24

• Institutions and wage structure
• Micro facts and macro interpretations

– Frequency, timing, indexation and wages of new hires – Downward wage rigidities – Pass-through of wages into prices

• Labour markets and the current crisis
• Conclusions

Overview

25

26

Downward wage rigidity: IWFP methodology

Figure 3.4 Downward nominal and real wage rigidity across countries.

IWFP Methodology. (Fraction of workers)

0.2 0.4 0.6 0.8 1 N e t h e r l a n d s G r e e c e U S S w i t z e r l a n d F r a n c e I t a l y H u n g a r y G e r m a n y I r e l a n d A u s t r i a U K N

• r

w a y D e n m a r k P

• r

t u g a l S p a i n B e l g i u m F i n l a n d S w e d e n L u x e m b

• u

r g DNWR DRWR

Source: the figures for Belgium, Denmark, Portugal and Spain are from (Messina et al 2009), figures for Hungary are from Katay (2008b), for Luxembourg are from Lünnemann and Wintr (2009b), the rest are IWFP figures from Dickens et al (2007).

27

Downward wage rigidity: Survey evidence

• Excluding Germany, nominal cuts in base wages are very rare:
• Labour market institutions matter: Higher level bargaining and

strict employment protection legislation increase DWR.

• Fairness and efficiency wage arguments are the most important

explanations, in addition to labour regulations.

2 4 6 8 10 12 14 16 Cuts Freezes Total Euro area Non-euro area

% of firms

28

Macro and policy implications (1)

• 60 percent of the firms surveyed reply they have used

alternative strategies to reduce costs: – Reduction of bonus payments is the most popular method in non-euro-area countries; – Euro area firms rely to a greater extent on the replacement of departing workers by cheaper hires.

• Nevertheless, DNWR does affect macro-economic

dynamics: – Evidence in sectoral wage bills: Holden and Wulfsberg (2008, 2009); – Wages respond more sluggishly to unemployment in a recession than in booms.

29

Macro and policy implications (2)

• In response to shocks that require a negative real wage adjustment,

DWR shifts the burden of adjustment towards employment:

• In response to cost-push shocks, cross country differences in the degree
• f nominal and real DWR in a monetary union imply that the regions

which exhibits real DWR will suffer from a loss of competitiveness relative to the region that has more flexible markets.

5 10 15 20

• 2
• 1

1

Wage inflation

5 10 15 20 0.2 0.4

Nominal interest rate

5 10 15 20

• 0.5

0.5

Inflation

5 10 15 20

• 3
• 2
• 1

Consumption

5 10 15 20

• 2
• 1

1

Real wage Effect of a -1.5σ technology shock under different wage setting specifications

5 10 15 20

• 2
• 1

1

Hours worked flexible nominalsymmetric nominalasymmetric realasymmetric

30

Macro and policy implications (3)

• Central bank may want to maintain a small positive

inflation buffer “to grease the wheels of the economy”

• WDN research suggests that a buffer between 0 and

2 percent is appropriate (lower than for the US):

2 - 5 1 - 2 Inflation buffer United States Portugal / Germany Belgium / Finland

31

• Institutions and wage structure
• Micro facts and macro interpretations

– Frequency, timing, indexation and wages of new hires – Downward wage rigidities – Pass-through of wages into prices

• Labour markets and the current crisis
• Conclusions

Overview

32

At the micro level, the strength of the link depends on the labour share

• Firms with a high labour cost

share report more frequently that there is a tight link between price and wage change

• The frequency of price

adjustment is lower in firms with a high labour cost share, confirming the IPN evidence

Link between wages and prices

Frequency of price changes vs labour cost share

5 10 15 20 25 30 35 40

0‐20 20‐40 40‐80 80‐100

Labour cost share % firms

more freq. than

• nce a year

less freq. than

• nce a year

Link between price and wage changes vs labour cost share

10 20 30 40 50 60 70 80 0‐20 20‐40 40‐80 80‐100

Labour cost share % firms

No link Link no pattern Link and pattern

33

Link between wages and prices

• In addition there is evidence of a statistically

significant relationship from the frequency of wage changes to that of prices, whereas the effect in the

• pposite direction is not significant.
• freq. of
• freq. of
• freq. of
• freq. of

wage change price change wage change price change Frequency of price change 0.069 0.065 Frequency of wage change 0.507** 0.135*

Country dummies No No Yes Yes Observations 5217 5217 5217 5217

Notes: (*) and (**) denote statistical significance at 5% and 1%, respectively. Source: Druant et al. (2009). Other regressors include sector and firm size (both equations), and labour cost share, competitive pressures, export, share of white collars, share of high skilled workers (price equation), share of bonuses on total wage bill, workforce turnover, collective agreement at the firm level, coverage of collective agreement, EPL (wage equation). details in Table 3.9 of the WDN report.

System estimates on price and wage change frequencies

(3 stage least squared)

34

Link between wages and prices

• Over 60 percent of firms surveyed declare that they would use a

strategy of increasing prices when faced with a permanent unexpected increase in wages.

Adjustment strategies to a permanent wage shocks

(firms answering "relevant" or "very relevant", percentages)

Wage shock

Reduce costs

59.0

Adjust prices

59.2

Reduce margins

49.8

Reduce output

22.5

35

Link between wages and prices

• This evidence of a substantial, but partial, pass-

through of wages into prices is more difficult to

• btain using micro data.

– Using a high-quality data set on both wages and prices, an elasticity of 0.3 is found for Sweden. – Other estimates for Italy and France are smaller, but there are substantial measurement issues.

• In contrast, prices of intermediate goods seem to

pass through much stronger.

36

• Institutions and wage structure
• Micro facts and macro interpretations

– Frequency, timing, indexation and wages of new hires – Downward wage rigidities

• Pass-through of wages into prices
• Labour markets and the current crisis
• Conclusions

Overview

37

Employment and wages in the crisis

Source: Eurostat and ECB calculations. Notes: Annual growth rates

Total employment and hours worked in the euro area Compensation per employee and hourly labour cost in the euro area

• 7.0
• 5.0
• 3.0
• 1.0

1.0 3.0 03Q1 04Q1 05Q1 06Q1 07Q1 08Q1 09Q1

Total Employment euro area Total hours euro area

• 1.0

0.0 1.0 2.0 3.0 4.0 5.0 03Q1 04Q1 05Q1 06Q1 07Q1 08Q1 09Q1

• 1.0

0.0 1.0 2.0 3.0 4.0 5.0

Compensation per employee Total hourly labour costs CPE, GDP deflated Hourly labour costs, GDP deflated

38

Diverse wage adjustments

Nominal compensation per employee

Source: Eurostat and ECB calculations. Notes: Annual growth rates.

• 5.0

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 06Q1 07Q1 08Q1 09Q1

• 5.0

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0

EA ES IE EE LV CZ LT

• 2.0

0.0 2.0 4.0 6.0 8.0 10.0 03Q1 04Q1 05Q1 06Q1 07Q1 08Q1 09Q1

• 2.0

0.0 2.0 4.0 6.0 8.0 10.0

EA ES IE DE IT FR NL

39

Labour markets and the current crisis

• How can the WDN findings be used to better

understand the labour market response in the current crisis?

• Difficult because

– differences in the depth of the recession, the nature of labour market policies and labour market institutions; – labour market adjustment is still ongoing

• Some hints from the follow-up survey held in 10

countries in the summer 2009: – AT, BE, CZ, EE, ES, FR, IT, LU, NL, PL

40

Evidence on the source of the crisis

• Most firms perceive the crisis as a fall in demand;
• Financial constraints are relatively less important.

5 10 15 20 25 30 35 40 45 50 Demand fall Difficulty being paid Financial constraints Marginal Moderate Strong Very strong

41

Adjustment to the current demand shock

• Cost reduction is the most relevant adjustment strategy to the

current fall in demand;

• Price reduction the least relevant.

Hypothetical Current crisis Hypothetical Current crisis reduce cost 84.7 73.4 86.1 85.9 reduce prices 56.3 37 47.2 40.6 reduce margins 63.1 43.2 61.9 49.1 reduce output 51.8 39.5 76.6 79.5

Adjustment strategies to demand shocks

(Percentage of firms replying relevant or very relevant)

Very strong crisis

Notes: All figures are employment-weighted. The sample includes AT, BE, CZ, EE, FR, IT, LU, NL and PL (ES not included). The construction sector is not covered by the survey in ES, FR and IT. The financial intermediation sector is not covered by the survey in CZ, EE, ES and FR. Details in Table 5.3 of the WDN report.

All

42

Adjustment to the current demand shock

• Reduction in labour costs occurs mostly through reducing the

quantity of labour, rather than its price; – Reduction of temporary, rather than permanent employees; – Reduction in flexible wage components, rather than cuts in base wages.

Reduce labour cost Adjust the amount of labour Reduce number of temporary employees Reduce number of permanent employees Reduce hours worked per employee Adjust wages Reduce flexible components Reduce base wages 27.4 25.6 17.1 19.1

Cost-cutting strategies during the current crisis (percentages)

61.7 64 Main Strategy Hypothetical Current crisis

Notes: All figures are employment-weighted. The sample includes AT, BE, CZ, EE, ES, FR, IT, NL and PL. The construction sector is not covered by the follow-up survey in ES, FR and IT. The financial intermediation sector is not covered by the follow- up survey in CZ, EE, ES and FR. details in Table 5.4 of the WDN report.

7.7 14.1 8.6 1.5 11.8 1.7

43

Adjustment to the current demand shock

• Preliminary analysis confirms that institutions matter:

– The presence of centralised collective wage agreements hinders the adjustment of wages, even the flexible components … – … and induces firms to reduce labour costs through the intensive margin. – Strict employment protection is associated with a higher recourse to layoffs of temporary employees and a lower reduction of hours worked.

44

• … with a number of exceptions: e.g. Estonia (44%);
• Number of firms that have frozen wages or intend to do so has

increased dramatically;

• Institutions play a role in explaining cross-country variation:

– Strict EPL in combination with higher-level collective bargaining reduces downward wage flexibility.

Downward wage rigidity still prevalent…

Incidence of wage cuts and freezes during the crisis: follow-up survey

% of firms cutting wages % of firms freezing wages Original survey Follow-up survey Original survey Follow-up survey did cut will cut did freeze will freeze Total 2.6 3.2 3.1 9.5 34.5 34.5 Euro area 1.3 2.1 3.3 7.6 37.1 43.1 Non-euro area 6.4 6.5 2.7 14.8 27.4 10.3

Notes: Source: Messina and Rõõm (2009). Figures for the original survey have been calculated including only the firms that are in the 2009 sample. Figures are employment-weighted and rescaled excluding “do not know” answers. The sample includes AT, BE, CZ, EE, ES, FR, IT, NL and PL. The construction sector is not covered by the follow-up survey in ES, FR and IT. The financial intermediation sector is not covered by the follow-up survey in CZ, EE, ES, and FR. Country details in Table 5.2 of the WDN Report.

45

• Institutions and wage structure
• Micro facts and macro interpretations

– Frequency, timing, indexation and wages of new hires – Downward wage rigidities

• Pass-through of wages into prices
• Labour markets and the current crisis
• Conclusions

Overview

46

A common theme: – Collective bargaining institutions, employment protection legislation and product market competition are all important factors shaping the response of wages, employment and prices to economic developments; – Higher-level bargaining, stringent EPL and lack

• f goods market competition lead to higher real

wage rigidity and a stronger (temporary) employment response to shocks. Those rigid wages in turn explain a relatively mute and persistent response of prices.

Conclusions (1)

47

• Wage rigidity complicates monetary policy in at

least three ways: – Increases the cost of stabilising inflation in the face of cost-push shocks; – Nominal DWR contributes to inflationary pressures and provides a rationale for maintaining a small positive inflation buffer (< 2 percent); – Wage rigidities may lead to persistent inflation differentials within the euro area and induce competitiveness losses vis-à-vis more flexible economies, reducing the cohesion of EMU

Conclusions (2)

48

• Structural reforms to improve the functioning of

the labour market should be comprehensive: – Increasing wage flexibility by moving to firm- level bargaining, protecting workers rather than inefficient jobs, avoiding a dual approach to employment adjustment, and improving competition in goods and services markets are mutually reinforcing mechanisms to ensure an efficient reallocation of resources.

• Balancing efficiency and social protection requires

a comprehensive assessment of the consequences

• f employment protection legislation, bargaining

procedures and unemployment benefits.

Conclusions (3)