A. Introduction

106. Spain’s high GDP growth since the mid-1990s has been accompanied by relatively low growth in labor productivity. Between 1995 and 2005, Spain’s real GDP rose by an average of 3.7 percent a year, underpinned by an impressive increase in labor utilization. Meanwhile, labor productivity grew by a mere 0.3 percent a year (Figure 1) and average total factor productivity (TFP) even declined by 0.8 percent, well below the European Union (EU) and U.S. averages.

Spain: Output and Productivity Growth

(percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Source: EU KLEMS, March 2008.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Spain: Output and Productivity Growth

(percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Source: EU KLEMS, March 2008.

• Download Figure
• Download figure as PowerPoint slide

Spain: Output and Productivity Growth

(percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Source: EU KLEMS, March 2008.

• Download Figure
• Download figure as PowerPoint slide

107. With the growth model of the last decade all but exhausted, an economic resurgence will likely require substantial productivity improvements. Residential construction and private consumption are adjusting rapidly as the factors behind the recent boom—such as low real interest rates, ample credit availability, rising female participation, and immigration—are fading. These developments underline the need for robust productivity gains if the economy is to return to strong growth over the coming years. Implementing an ambitious program of structural reforms, including those delineated in the National Reform Program (Presidencia del Gobierno Español, 2005 and 2008), will be critical to achieve that goal (OECD, 2008).

108. The purpose of this paper is to identify the sources of Spain’s scant productivity performance and explore what policies can do about it. Section B examines Spain’s productivity facts in an international context. Section C presents scenarios of the impact that changes in sectoral trends can have on productivity growth. Section D estimates a productivity model to determine to what extent structural reforms can boost productivity growth, and Section E concludes.

B. What Are the Facts?

109. Despite improving somewhat in recent years, Spanish labor productivity growth has lagged behind the EU and the U.S. (Figure 2).⁷⁸ Accordingly, Spain’s productivity level relative to that of the EU dropped from 93 percent in 1995 to 84 percent in 2005. The corresponding values relative to the U.S. were 78 and 63 percent. To understand the sources of aggregate growth differentials with the EU and the U.S., we analyze the contributions of various inputs to productivity growth, and then focus on a sectoral perspective.

Labor Productivity Growth

(Percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Labor Productivity Growth

(Percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.

• Download Figure
• Download figure as PowerPoint slide

Labor Productivity Growth

(Percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.

• Download Figure
• Download figure as PowerPoint slide

A growth-accounting exercise

110. Under neoclassical assumptions and constant returns to scale, labor productivity growth can be decomposed into:

$\Delta \text{ln}\left(\frac{y_t}{h_t}\right)=\alpha \Delta \text{ln}\left(l{c}_t\right)+\beta \Delta \text{ln}\left(\frac{ki{t}_t}{h_t}\right)+\gamma \Delta \text{ln}\left(\frac{kni{t}_t}{h_t}\right)+\Delta \text{ln}\left(tf{p}_t\right),\left(1\right)$

where $\frac{y_t}{h_t}$ is gross value added per hour worked; lc_t is the labor input (accounting for differences in skill levels); $\frac{ki{t}_t}{h_t}$ is the information and communications technology (ICT) capital per hour; $\frac{kni{t}_t}{h_t}$ is the non-ICT capital per hour; tfp_t represents total factor productivity (TFP), a measure of the efficiency in combining a given amount of capital and labor to produce output; and the parameters α, β, and γ reflect the output elasticity of each input, adding to one.⁷⁹

111. According to this decomposition, TFP explains most of Spain’s sluggish productivity growth (Table 1). In particular, between 1995 and 2005, TFP contributed negatively to labor productivity growth by about 0.8 percentage point per year. By contrast, TFP made a small positive contribution in the EU and more than 1 percentage point in the U.S. Compounding this effect, ICT capital, although positive, made slightly smaller contributions to Spanish labor productivity than in the EU or the U.S. However, this was more than offset by changes in the labor composition that contributed about 0.4 percentage point to labor productivity growth, more than double the impact in the EU and the U.S. This suggests that during the recent past newcomers had more years of schooling than the existing labor force, thereby raising Spain’s overall skill level (Figure 3).

Table 1.

Contributions to Labor Productivity Growth ^1/

Sources: EU KLEMS, March 2008; and staff calculations.

^1/Contributions to total economy labor productivity growth during 1995-2005.

Table 1.

Contributions to Labor Productivity Growth ^1/

	Spain	EU	U.S.
Labor composition	0.4	0.2	0.2
ICT capital per hour	0.3	0.4	0.5
Non-Ict capital per hour	0.4	0.4	0.4
TFP	-0.8	0.2	1.2

Sources: EU KLEMS, March 2008; and staff calculations.

^1/Contributions to total economy labor productivity growth during 1995-2005.

View Table

Table 1.

Contributions to Labor Productivity Growth ^1/

	Spain	EU	U.S.
Labor composition	0.4	0.2	0.2
ICT capital per hour	0.3	0.4	0.5
Non-Ict capital per hour	0.4	0.4	0.4
TFP	-0.8	0.2	1.2

Sources: EU KLEMS, March 2008; and staff calculations.

^1/Contributions to total economy labor productivity growth during 1995-2005.

Hours Worked by High-Skilled Labor

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Hours Worked by High-Skilled Labor

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.

• Download Figure
• Download figure as PowerPoint slide

Hours Worked by High-Skilled Labor

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.

• Download Figure
• Download figure as PowerPoint slide

A sectoral perspective

112. The poor productivity performance of the Spanish economy has its roots in both the non-ICT and ICT sectors (Figure 4). In particular, the non-ICT sectors contributed -0.1 percent per year to productivity growth during 1995–2005, compared with sizable contributions in both the EU and the U.S. In addition, the ICT sectors made very small (although positive) contributions to productivity, particularly the ICT-producing industries, highlighting the small share of the high-tech sector in the Spanish economy. This is in contrast with the EU and more so the U.S., where a large proportion of the productivity increase since the mid-1990s originated in the ICT sectors.

Contributions to Labor Productivity Growth by Sectors

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Contributions to Labor Productivity Growth by Sectors

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.

• Download Figure
• Download figure as PowerPoint slide

Contributions to Labor Productivity Growth by Sectors

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.

• Download Figure
• Download figure as PowerPoint slide

113. From an industry perspective, construction and services appear to have been the main drag on productivity (Table 2). Construction and real estate made the largest negative contribution to productivity growth, -0.4 percent per year, followed by personal and social services. The rest of the industries made positive but almost negligible contributions. This is particularly notable in the case of manufacturing, distribution, and electrical, post, and telecommunications, which together accounted for 0.7 and 1.6 percent of the productivity growth differential with the EU and the U.S., respectively. The only sector where Spain appeared ahead was agriculture, partly reflecting the employment decline in that sector. This evidence raises the question of whether Spain’s weak productivity is explained by insufficient reallocation of resources between sectors (from low to high productivity), or is the result of low dynamism within sectors.

Table 2.

Contributions to Labor Productivity Growth, 1995-2005 ^1/

(Percent)

Sources: EU KLEMS, March 2008; and staff calculations.

^1/Numbers may not sum exactly due to approximation.

^2/Personal and social services include hotels and restaurants, other community, social and personal services, and private households.

Table 2.

Contributions to Labor Productivity Growth, 1995-2005 ^1/

(Percent)

			Spain			European Union			U.S.
			Total	Within	Reallocation	Total	Within	Reallocation	Total	Within	Reallocation
Primary			0.2	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1
	Agriculture, hunting, forestry, and fishing		0.2	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1
	Mining		0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Secondary			-0.1	0.1	-0.2	0.7	0.8	-0.1	1.2	1.5	-0.3
	Manufacturing, excl. electrical		0.0	0.1	0.0	0.3	0.3	0.0	0.5	0.6	-0.1
	Electrical, post, and communications		0.1	0.1	0.0	0.3	0.3	0.0	0.7	0.9	-0.1
	Utilities		0.1	0.1	-0.1	0.1	0.1	-0.1	0.0	0.1	-0.1
	Construction		-0.3	-0.2	-0.1	0.0	0.0	0.0	-0.1	-0.1	0.0
Services			0.1	-0.4	0.5	0.4	0.4	0.1	1.3	1.2	0.0
	Distribution		0.1	0.1	0.0	0.3	0.2	0.0	0.5	0.5	0.0
	Financial and business services		0.1	0.2	-0.1	0.1	0.1	0.0	0.4	0.4	0.0
	Personal and social services 2/		-0.1	-0.1	0.0	-0.1	0.0	-0.1	0.0	0.0	-0.1
	Nonmarket services		0.0	-0.6	0.6	0.2	0.1	0.2	0.3	0.3	0.0
		Of which: real estate 3/	-0.1	-0.8	0.6	0.1	0.0	0.2	0.3	0.2	0.1
Total			0.3	-0.2	0.5	1.2	1.2	0.0	2.3	2.8	-0.2

Sources: EU KLEMS, March 2008; and staff calculations.

^1/Numbers may not sum exactly due to approximation.

^2/Personal and social services include hotels and restaurants, other community, social and personal services, and private households.

View Table

Table 2.

Contributions to Labor Productivity Growth, 1995-2005 ^1/

(Percent)

			Spain			European Union			U.S.
			Total	Within	Reallocation	Total	Within	Reallocation	Total	Within	Reallocation
Primary			0.2	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1
	Agriculture, hunting, forestry, and fishing		0.2	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1
	Mining		0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Secondary			-0.1	0.1	-0.2	0.7	0.8	-0.1	1.2	1.5	-0.3
	Manufacturing, excl. electrical		0.0	0.1	0.0	0.3	0.3	0.0	0.5	0.6	-0.1
	Electrical, post, and communications		0.1	0.1	0.0	0.3	0.3	0.0	0.7	0.9	-0.1
	Utilities		0.1	0.1	-0.1	0.1	0.1	-0.1	0.0	0.1	-0.1
	Construction		-0.3	-0.2	-0.1	0.0	0.0	0.0	-0.1	-0.1	0.0
Services			0.1	-0.4	0.5	0.4	0.4	0.1	1.3	1.2	0.0
	Distribution		0.1	0.1	0.0	0.3	0.2	0.0	0.5	0.5	0.0
	Financial and business services		0.1	0.2	-0.1	0.1	0.1	0.0	0.4	0.4	0.0
	Personal and social services 2/		-0.1	-0.1	0.0	-0.1	0.0	-0.1	0.0	0.0	-0.1
	Nonmarket services		0.0	-0.6	0.6	0.2	0.1	0.2	0.3	0.3	0.0
		Of which: real estate 3/	-0.1	-0.8	0.6	0.1	0.0	0.2	0.3	0.2	0.1
Total			0.3	-0.2	0.5	1.2	1.2	0.0	2.3	2.8	-0.2

Sources: EU KLEMS, March 2008; and staff calculations.

^1/Numbers may not sum exactly due to approximation.

^2/Personal and social services include hotels and restaurants, other community, social and personal services, and private households.

114. To assess the contributions of reallocation between, and restructuring of, industries to productivity growth, we follow Baily, Bartelsman, and Haltiwanger (1996). Productivity growth is decomposed into three industry-specific components, namely “within” effect, “between” effect, and “cross” effect, as follows:

$\frac{L{P}_t-L{P}_{t-1}}{L{P}_{t-1}}=\frac{{\displaystyle \sum _j{s}_{jt-1}(L{P}_{jt}-L{P}_{jt-1})}}{\underset{Within}{{\displaystyle L{P}_{t-1}}}}+\frac{{\displaystyle \sum _j(s_{jt}-s_{jt-1})(L{P}_{jt-1}-L{P}_{t-1})}}{\underset{Between}{{\displaystyle L{P}_{t-1}}}}+\frac{{\displaystyle \sum _j(s_{jt}-s_{jt-1})[(L{P}_{jt}-L{P}_{jt-1})-(L{P}_t-L{P}_{t-1})]}}{\underset{Cross}{{\displaystyle L{P}_{t-1}}}},\left(2\right)$

where LP_jt is the level of labor productivity in industry j (total economy if industry index is missing) and time t; and s_jt is the employment share in industry j at time t. The first term in this decomposition represents the within effect, that is, the productivity growth within an industry keeping the employment shares fixed. These productivity gains could be the result of human or physical capital deepening or the introduction of new technological or organizational methods. The second term is the between effect, which reflects productivity growth due to changing employment shares. This term will be positive when employment shares increase for industries with higher-than-average productivity levels in the previous year. The third term represents a cross (i.e., covariance-type) term. This term will be positive when employment shares increase for industries with productivity growth above the average for the economy. The last two terms show the reallocation effect.

115. Overall, Spain’s lackluster productivity is due more to a lack of dynamism within industries, than to a reallocation of resources between industries (Table 2). The negative within effect is mainly driven by services (in particular, real estate) and construction, setting Spain apart from the EU and the U.S.⁸⁰ A notable exception is financial intermediation, which seems to have been performing well, particularly relative to the EU. On the other hand, the reallocation of employment across sectors had a positive impact on productivity except for construction, utilities, and the financial and business sector.

116. A key aspect of this finding is the negative contribution to TFP growth in almost all sectors. As with labor productivity, both non-ICT and ICT sectors are to be blamed (Figure 5). Foremost among the laggards is construction, in turn followed by all services, with the exception of financial intermediation, in which Spain is close to the U.S. and well above the EU (Figure 6). Surprisingly, the high-tech sector (electrical, post, and communications) also made negative contributions to TFP growth, in sharp contrast to the EU and the U.S. This could reflect the smaller size and technology intensity of the Spanish ICT sector (as argued by Mas and Quesada, 2007).

Contributions to TFP Growth by Sectors, 1995-2005

(percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.1/ Hotels and restaurants, personal services, and private households.2/ Manufacturing, excluding electrical equipment.3/ Electrical equipment, post, and communications.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Contributions to TFP Growth by Sectors, 1995-2005

(percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.1/ Hotels and restaurants, personal services, and private households.2/ Manufacturing, excluding electrical equipment.3/ Electrical equipment, post, and communications.

• Download Figure
• Download figure as PowerPoint slide

Contributions to TFP Growth by Sectors, 1995-2005

(percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: EU KLEMS, March 2008; and staff calculations.1/ Hotels and restaurants, personal services, and private households.2/ Manufacturing, excluding electrical equipment.3/ Electrical equipment, post, and communications.

• Download Figure
• Download figure as PowerPoint slide

Contributions to TFP Growth, 1995-2005

(percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Contributions to TFP Growth, 1995-2005

(percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

• Download Figure
• Download figure as PowerPoint slide

Contributions to TFP Growth, 1995-2005

(percent)

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

• Download Figure
• Download figure as PowerPoint slide

C. Productivity Scenarios

117. What do these findings suggest about TFP growth prospects? The previous section has identified construction, ICT, and services as the weakest in Spanish productivity. In this section, we consider three alternative scenarios to analyze the impact that a change in these sectors could have on overall TFP growth. The first scenario assumes a reallocation whereby the share of the construction sector in total value added declines by 30 percent (to about 8 percent of total value added), and that the most dynamic sectors within manufacturing take the place of construction. The second scenario considers that each industry within the ICT sector catches up to the productivity growth of the leader for that industry. The leader is defined as the country (among the EU and U.S.) with the highest productivity level (adjusted for purchasing power parity (PPP)) in the sector. Finally, the third scenario similarly analyzes the impact of each market service sector catching up to the productivity growth of the leader, where the leader is defined as the EU country with the highest productivity level in that sector. In the last two scenarios, we assume that each sector share in total value added is the same as in 2005.

118. The scenarios suggest that to close the productivity gap with the EU and U.S., TFP within each sector needs to be improved. The first scenario shows that by only shifting resources from construction to other sectors, without a change in sectoral TFP itself, TFP growth for the economy as a whole would increase by only 0.1 percent (Table 3). Thus, there is limited room for productivity gains should there only be a change in sectoral allocation of resources. By contrast, if Spain were to achieve the same TFP growth rate in the ICT sectors as the leaders, overall TFP growth would improve by 1.2 percent. Further, improving the performance in the services sector to that of the best performers in Europe would also yield a 1.2 percent increase in TFP growth. Thus, the key is to make sectoral TFP growth rates to catch up with the leaders. We try to address this question in the next section.

Table 3.

Spain: TFP Growth Scenarios ^1/

(In percent)

Source: Staff’s calculations.

^1/TFP growth rates for each sector are assumed to be the same as the 2000-05 average unless otherwise indicated. The share of each sector in value added is assumed at the 2005 level unless otherwise indicated.

^2/This scenario assumes that the construction share of value added declines to 8 percent and the most dynamic manufacturing sectors take the place of construction.

^3/This scenario assumes for each ICT sector the same growth rate as the leader (among the U.S. and EU countries) during 2000-05.

^4/This scenario assumes for each market service sector, the same growth rate as the leader among EU countries during 2000-05.

Table 3.

Spain: TFP Growth Scenarios ^1/

(In percent)

Scenario	TFP
Average 1995-2005	-0.8
Scenario 1: Construction decline in total value added 2/	-0.7
Scenario 2: ICT sectors catching up with the leaders 3/	0.4
Scenario 3: Services catching with the EU 4/	0.4

Source: Staff’s calculations.

^1/TFP growth rates for each sector are assumed to be the same as the 2000-05 average unless otherwise indicated. The share of each sector in value added is assumed at the 2005 level unless otherwise indicated.

^2/This scenario assumes that the construction share of value added declines to 8 percent and the most dynamic manufacturing sectors take the place of construction.

^3/This scenario assumes for each ICT sector the same growth rate as the leader (among the U.S. and EU countries) during 2000-05.

^4/This scenario assumes for each market service sector, the same growth rate as the leader among EU countries during 2000-05.

View Table

Table 3.

Spain: TFP Growth Scenarios ^1/

(In percent)

Scenario	TFP
Average 1995-2005	-0.8
Scenario 1: Construction decline in total value added 2/	-0.7
Scenario 2: ICT sectors catching up with the leaders 3/	0.4
Scenario 3: Services catching with the EU 4/	0.4

Source: Staff’s calculations.

^1/TFP growth rates for each sector are assumed to be the same as the 2000-05 average unless otherwise indicated. The share of each sector in value added is assumed at the 2005 level unless otherwise indicated.

^2/This scenario assumes that the construction share of value added declines to 8 percent and the most dynamic manufacturing sectors take the place of construction.

^3/This scenario assumes for each ICT sector the same growth rate as the leader (among the U.S. and EU countries) during 2000-05.

^4/This scenario assumes for each market service sector, the same growth rate as the leader among EU countries during 2000-05.

D. Can Policies Help?

119. Policymakers and academics have long recommended strengthening product market competition to boost productivity growth. Product market rigidities can impair productivity.⁸¹ First, they dampen productivity growth by reducing incentives to invest, adopt latest technologies, or innovate (Crafts, 2006). This effect could be stronger for industries closer to the technology frontier as they rely on innovation rather than imitation. Second, rigidities raise entry costs and curb competition by hindering resource reallocation. Limited competition among suppliers increases the cost of inputs and makes products supplied less innovative and of poorer quality, thereby lowering productivity in downstream sectors. Following Aghion and Griffith (2005), empirical studies have indeed found that regulatory rigidities curb productivity growth (see, for example, Nicoletti and Scarpetta, 2003; Conway and others, 2006; and Arnold, Nicoletti, and Scarpetta, 2008). However, others find more limited evidence of the impact of regulatory barriers on TFP growth (Inklaar, Timmer and van Ark, 2008) or even a positive effect (Dew-Becker and Gordon, 2008).

120. Easing labor market rigidities could also enhance productivity. One of the premises behind this argument is that labor market policies could distort incentives to investment in education, thereby lowering productivity by reducing the stock of human capital. They can also depress productivity by preventing firms from adjusting to changes in technology or product demand that require labor reallocation or downsizing (see, for example, Hopenhayn and Rogerson, 1993; and Saint-Paul, 2002; and Haltiwanger and others, 2008). The empirical evidence on the impact of labor market policies is mixed, however. In particular, while some papers find that employment protection legislation (EPL) can dampen productivity levels and growth (see, for example, Besley and Burgess, 2004; Scarpetta and Tressel, 2004; and Micco and Pagés, 2007), others suggest that some aspects of labor market policy (e.g., minimum wages) can even have a positive effect on productivity (e.g., Bassanini and Venn, 2007).

121. To assess the effect of product and labor market rigidities on productivity, we estimate a model based on Aghion and Howitt (2005). In this model, TFP growth in a given country and sector depends on its ability to keep pace with the growth in the country with the highest productivity level in that sector (the productivity leader). This is, in turn, affected by the policy environment that the sector confronts in the country of operation. The estimated equation is

where the indices i,j, t denote countries, industries and years, respectively; TFP denotes total factor productivity; TFP gap is the productivity gap (measured as the log ratio of the level of productivity in each sector relative to that of the productivity leader); Policy is an indicator of product market regulation or labor market rigidity; X denotes other control variables; C represents country dummies; J, industry dummies; and T, time dummies.⁸² In this model, TFP shocks in the leader can have a direct impact on the followers’ productivity growth. In addition, the model allows differences in productivity levels between the leader and the follower to have an impact on TFP growth as captured by the coefficient σ. A negative coefficient indicates that the farther a sector is from the technology frontier the greater the scope for productivity improvements arising from technological catch-up. We estimate this model over the period 1976 to 2003 for a sample of 25 industries in 10 EU countries (including Spain) and the U.S.

122. Barriers to competition are still significant in Spain. Despite progress in several areas—such as some network industries—and reform of the competition law, improvements are needed in the energy and transport sectors, telecommunications, professional services, and retail trade, where sectoral regulations remain strict and are seen as impeding competition (Figure 7). Increasing competition in these sectors could have spillover effects throughout the economy as they provide intermediate inputs for other sectors. To capture this effect, we use the regulation impact indicators developed by Conway and Nicoletti (2006). For each sector in a particular country, the regulation impact indicator is calculated as a weighted average of indicators in nonmanufacturing sectors covering transport, energy, communications, retail distribution, banking, and business services. The weights used in the calculation are total input coefficients derived from input-output tables that measure the degree to which intermediate inputs from each of the nonmanufacturing sectors are used in the final output of each sector in the economy. In our estimation, we allow product market regulation to influence TFP growth both directly and indirectly by affecting the speed of productivity catch-up (coefficient α). A positive value of a implies that product market regulation hinders technology diffusion.

Product Market Regulation

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Source: OECD, Going for Growth 2008.1/ Index scale of 0-6 from least to most restrictive.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Product Market Regulation

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Source: OECD, Going for Growth 2008.1/ Index scale of 0-6 from least to most restrictive.

• Download Figure
• Download figure as PowerPoint slide

Product Market Regulation

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Source: OECD, Going for Growth 2008.1/ Index scale of 0-6 from least to most restrictive.

• Download Figure
• Download figure as PowerPoint slide

123. Labor market rigidities are also an important factor in Spain. One key aspect is the duality of the labor market between permanent and temporary (i.e., fixed-term) contracts. The latter accounts for about one-third of wage earners in Spain, the highest in the EU. The incidence of temporary contracts is particularly high among the young. The extensive use of these contracts (with low redundancy costs) is partly due to the uncertainty⁸³ and rigidity of the EPL for permanent jobs (Figure 8).⁸⁴ By inducing the extensive use of temporary contracts among the young and making it costly to fire older (less productive) workers, EPL may contribute to the underutilization of young (skilled) workers. Also, the widespread use of temporary contracts could dampen workers’ effort, decreasing TFP.⁸⁵ Finally, EPL may slow the reallocation of employment to more innovative, high-productivity sectors since it discourages worker mobility, as those who change jobs voluntarily lose their protection. To capture this effect, we include the OECD index of EPL for permanent workers. To test whether the use of temporary workers may hamper productivity growth, we introduce an interaction term between the share of temporary workers in the sector and EPL. This coefficient would be negative if the use of temporary workers in the presence of strict EPL reduces productivity growth.

Temporary Employment and Employment Protection Legislation

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: OECD; and Eurostat.1/ Temporary employment as percentage of total employees of the group as of the first quarter of 2008.2/ Employment protection legislation for permanent workers: index scale of 0-6, moving from least to most restrictive, 2006.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Temporary Employment and Employment Protection Legislation

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: OECD; and Eurostat.1/ Temporary employment as percentage of total employees of the group as of the first quarter of 2008.2/ Employment protection legislation for permanent workers: index scale of 0-6, moving from least to most restrictive, 2006.

• Download Figure
• Download figure as PowerPoint slide

Temporary Employment and Employment Protection Legislation

Citation: IMF Staff Country Reports 2009, 129; 10.5089/9781451812299.002.A004

Sources: OECD; and Eurostat.1/ Temporary employment as percentage of total employees of the group as of the first quarter of 2008.2/ Employment protection legislation for permanent workers: index scale of 0-6, moving from least to most restrictive, 2006.

• Download Figure
• Download figure as PowerPoint slide

124. Overall, we find evidence that product market regulation slows TFP growth, but results for labor market rigidities are less conclusive (Table 4). TFP growth in the leading sector has positive spillover effects on TFP growth in less productive sectors. In addition, the coefficient of the productivity gap is negative, indicating the importance of international technology diffusion. Regarding product market regulation, we find a negative direct effect of the anticompetitive indicator on TFP growth (column 2). This effect appears to be particularly important in the ICT sectors (column 4). The strictness of regulation also affects the speed at which sectors catch up to the leader (column 1), but this result is not robust to the inclusion of the labor market indicators. EPL does not seem to have a direct effect on TFP growth.⁸⁶ Nor does it seem to dampen the speed of productivity convergence with the leader (column 3). However, it appears that productivity growth is boosted by resorting to temporary workers when EPL is strict (column 2). This suggests that temporary contracts give firms a flexibility not present in permanent ones.

Table 4.

Productivity Growth Model

Sources: EU KLEMS, March 2008; Eurostat; OECD; Conway and Nicoletti (2006); and staff calculations.Notes: *** p<00.01, ** p<0.05, * p<0.1. Robust standard errors in parentheses. All equations include country, industry and time dummies.

^1/Interaction of product market regulation and productivity gap.

^2/Interaction of EPL and productivity gap.

Table 4.

Productivity Growth Model

	(1)	(2)	(3)	(4)
ICT capital share in total capital	0.0*	0.0***	0.0	0.0***
	(0.0)	(0.0)	(0.0)	(0.0)
Change in the productivity in the technology leader	0.14***	0.05	0.05	0.05
	(0.05)	(0.09)	(0.09)	(0.09)
Gap in productivity levels	-0.08***	-0.13***	-0.05	-0.14***
	(0.02)	(0.04)	(0.05)	(0.04)
Product market regulation	0.03	-0.05*	-0.05*	…
	(0.02)	(0.03)	(0.03)	…
Effect of regulation catch-up 1/	0.07**	-0.06	-0.03	-0.05
	(0.03)	(0.08)	(0.08)	(0.07)
EPL	…	0.00	0.00	…
	…	(0.01)	(0.01)	…
EPL* Temporary employment	…	0.03***	0.04***	0.03**
	…	(0.01)	(0.01)	(0.01)
Effect of ELP on catch-up 2/	…	…	-0.04	…
	…	…	(0.02)	…
Product market regulation ICT-intensive sectors	…	…	…	-0.09***
	…	…	…	(0.03)
Product market regulation non-ICT intensive sectors	…	…	…	0.00
	…	…	…	(0.03)
EPL ICT-intensive sectors	…	…	…	0.00
	…	…	…	(0.01)
EPL non-ICT intensive sectors	…	…	…	0.01
	…	…	…	(0.01)
Constant	-0.01	-0.07**	-0.06**	-0.09***
	(0.01)	(0.03)	(0.03)	(0.03)
Observations	6863	2746	2746	2746
R-squared	0.11	0.14	0.14	0.14

Sources: EU KLEMS, March 2008; Eurostat; OECD; Conway and Nicoletti (2006); and staff calculations.Notes: *** p<00.01, ** p<0.05, * p<0.1. Robust standard errors in parentheses. All equations include country, industry and time dummies.

^1/Interaction of product market regulation and productivity gap.

^2/Interaction of EPL and productivity gap.

View Table

Table 4.

Productivity Growth Model

	(1)	(2)	(3)	(4)
ICT capital share in total capital	0.0*	0.0***	0.0	0.0***
	(0.0)	(0.0)	(0.0)	(0.0)
Change in the productivity in the technology leader	0.14***	0.05	0.05	0.05
	(0.05)	(0.09)	(0.09)	(0.09)
Gap in productivity levels	-0.08***	-0.13***	-0.05	-0.14***
	(0.02)	(0.04)	(0.05)	(0.04)
Product market regulation	0.03	-0.05*	-0.05*	…
	(0.02)	(0.03)	(0.03)	…
Effect of regulation catch-up 1/	0.07**	-0.06	-0.03	-0.05
	(0.03)	(0.08)	(0.08)	(0.07)
EPL	…	0.00	0.00	…
	…	(0.01)	(0.01)	…
EPL* Temporary employment	…	0.03***	0.04***	0.03**
	…	(0.01)	(0.01)	(0.01)
Effect of ELP on catch-up 2/	…	…	-0.04	…
	…	…	(0.02)	…
Product market regulation ICT-intensive sectors	…	…	…	-0.09***
	…	…	…	(0.03)
Product market regulation non-ICT intensive sectors	…	…	…	0.00
	…	…	…	(0.03)
EPL ICT-intensive sectors	…	…	…	0.00
	…	…	…	(0.01)
EPL non-ICT intensive sectors	…	…	…	0.01
	…	…	…	(0.01)
Constant	-0.01	-0.07**	-0.06**	-0.09***
	(0.01)	(0.03)	(0.03)	(0.03)
Observations	6863	2746	2746	2746
R-squared	0.11	0.14	0.14	0.14

Sources: EU KLEMS, March 2008; Eurostat; OECD; Conway and Nicoletti (2006); and staff calculations.Notes: *** p<00.01, ** p<0.05, * p<0.1. Robust standard errors in parentheses. All equations include country, industry and time dummies.

^1/Interaction of product market regulation and productivity gap.

^2/Interaction of EPL and productivity gap.

125. A scenario analysis based on these results suggests that adopting the least restrictive practices could yield productivity gains. In particular, assuming Spain implements the least restrictive product market regulations within the sample, TFP growth could improve by 0.3 percent. This result is just indicative since we do not take into account policy changes that may affect the underlying relationships. Moreover, we may underestimate the overall effect since we do not consider the potential reallocation of resources toward high-productivity sectors that a change in regulation could spur.

126. These findings also have some caveats. First, the OECD indicators of regulation do not capture well all the complexities of product market regulation and interactions with labor market rigidities. Second, the EPL indicator is country based and, therefore, does not reflect differences across industries. In fact, recent studies have argued that EPL should affect more those industries where, absent regulation, firms would rely more on layoffs (i.e., labor shedding) for organizational purposes than those that rely more on voluntary turnover (see Micco and Pagés, 2007; and Bassanini and Venn, 2007). Third, we do not take into account the skills mismatch that EPL legislation may engender. This effect could be particularly important in Spain since temporary workers are usually young and better-qualified workers than older cohorts. Finally, our productivity data may suffer from measurement problems (for example, in the services sector or input deflators), which could bias the calculation of productivity gaps (see, for example, Inklaar, Timmer, and van Ark, 2008).

E. Conclusions

127. Spain’s relatively poor productivity performance during the last decade has resulted in a widening gap with the EU and the U.S. Underlying this weakness are declining TFP, the relatively small share of ICT-producing sectors, and the paucity of productivity growth in construction and services (with the notable exception of the financial sector). Reallocation effects between sectors can explain only a small part of the productivity growth gap. Rather, the burden appears to lie more within sectors. The scenarios presented in this paper suggest that catching up with the leaders in ICT and services could deliver substantial productivity gains.

128. Empirical analysis indicates that reforms could help improve TFP growth. These reforms are particularly important for the ICT sectors, where Spain stands to get the largest gains from productivity catch-up. Moreover, there is some evidence that product market reforms can increase the speed at which Spain converges to the productivity leaders. Evidence that EPL directly limits productivity growth is more difficult to find, at least in our models, which do not take into account differences across sectors, the interaction with product market policies, or the potential skill mismatches that EPL induces. Finally, there is evidence that fixed-term contracts assist productivity by providing marginal flexibility to firms to adjust to shocks.

129. Implementing an ambitious reform program will be important if the economy is to resume strong growth after the current downturn. Reforms to sharpen competition in transport, postal, and professional services should be priorities given the positive spillover effects on other sectors that use their services as inputs. In particular, consideration should be given to (1) putting the operation of regional passenger rail services out to tender on a compulsory and regular basis; (2) removing restrictions to entry and consolidation in the road transport sector; (3) ensuring appropriate access of competitors to the public postal network; and (4) limiting the range of professional services for which Spanish regulations require specific qualifications requirements and reducing potential regulatory differences across regions. Also, entry barriers in retail should be removed to bolster productivity in this sector. Product market reform is in fact already under way with the transposition of the European Services Directive, reductions in administrative burdens and reforms in network industries, and the more ambitious these efforts are, the more likely their beneficial impact on productivity and confidence will be.⁸⁷ In the electricity industry, retail prices should recover costs in contested and unbundled markets to eliminate distortions. Finally, equalizing dismissal costs of temporary and permanent contracts at low levels should help reduce labor market segmentation and broaden the flexibility benefits beyond those provided by temporary contracts. This would also facilitate higher penetration by new cohorts into better jobs and improve the return on human capital formation, where Spain scores below the OECD average.