Labor Productivity Dynamics in Spain: A Firm-Level Perspective: Spain

Nicolas Arregui; Yu Shi

doi:10.5089/9798400235634.018.A001

Jump to Content Jump to Main Navigation

Labor Productivity Dynamics in Spain: A Firm-Level Perspective: Spain

Author:

Mr. Nicolas Arregui

Mr. Nicolas Arregui
Search for other papers by Mr. Nicolas Arregui in
Current site
Google Scholar
PubMed

and

Ms. Yu Shi

Ms. Yu Shi
Search for other papers by Ms. Yu Shi in
Current site
Google Scholar
PubMed

Publication Date:: 28 Feb 2023

eISBN:: 9798400235634

ISBN:: 9798400235634

Language:: English

Keywords:: log TFP; investment rate; firm size; firm Age; firm investment

Download PDF (658.7 KB)

This paper uses firm-level data that covers over 1.2 million Spanish firms during the period of 2003-2019 to provide an updated assessment of the drivers of labor productivity in the Spanish economy, focusing on both TFP and firm investment. The empirical analysis shows significant differences in production constraints in both the capital market and the labor market, across firm size and age. This paper also includes a review of Spain’s ambitious reform commitments under the Recovery, Transformation and Resilience Plan and concludes with recommendations for further action.

Abstract

Labor Productivity Dynamics in Spain: A Firm-level Perspective^¹

A. Introduction

1. Labor productivity has been a long-standing structural challenge in Spain. Productivity performance has been weak across several dimensions: labor productivity levels are significantly lower than in some peer economies, its growth rate has been low and not favored convergence, and differences in output per hour worked across Spanish regions is considerable (IMF, 2018). Moreover, weak labor productivity performance relative to peers holds across sectors, suggesting that cross-cutting drivers are likely to play a more significant role than the productive structure of the economy.

2. In the aftermath of the Global Financial Crisis (GFC), labor productivity in Spain exhibited a counter-cyclical pattern driven by the large reduction in employment. The relatively strong productivity growth from 2009 to 2013 was mainly driven by developments in the labor market. The rapid decline in working hours resulted in a faster increase in the capital-labor ratio. Additionally, employment destruction had a bias towards less productive occupation, implying an increase in the average productivity for those who remained occupied. Once labor market conditions normalized, capital deepening^² settled at a lower contribution level (negative in 2014–19) to labor productivity, reflecting the lower rate of capital accumulation since the GFC. The adoption of measures during the pandemic, particularly with the role played by the strengthened furlough scheme (ERTE), resulted in better labor market dynamics than in past crises. On the flip side, labor productivity performed worse in Spain than in other countries.

3. Total factor productivity (TFP) has been consistently low and lagging peers for decades. A number of factors have been identified as contributing to the sluggish TFP growth in Spain, including the structural weaknesses in the labor market, such as the high share of temporary workers and the wide use of sector-level collective bargaining agreements (Hospido and Moreno-Galbis, 2015), capital misallocation linking to financial frictions from size-dependent borrowing constraints (Gopinath et al. 2017; García-Santana et al. 2020), product market regulation (Andrews and Cingano, 2014; Bank of Spain, 2015), size-related regulations and policies (Almunia and Lopez Rodriguez, 2018; IMF Country Report, 2018), and weak business innovation (OECD, 2021). Spain, nonetheless, has recently put in place reforms that could have a positive impact on TFP. For example, substantial changes in the regulation of temporary contracts were introduced in December 2021 and early results are showing a decline in the share of temporary workers. Reforms also established incentives for the provision of training to workers and revamped the vocational and training system. Furthermore, Spain’s Recovery, Transformation and Resilience Plan (RTRP) contemplates improvements to active labor market policies.

4. Sustained policy focus on raising productivity will be important to increase living standards, help rebuild fiscal buffers and make growth more inclusive. A previous study by the IMF (IMF, 2015) discussed the determinants of TFP in the aftermath of the GFC and emphasized the importance of product market regulation as well as the size-related tax incentives. Building on this earlier work, this chapter provides an updated assessment of the drivers of labor productivity in recent years, focusing on both TFP and firm investment. The empirical analysis benefits from the rich information in firm financial statements to provide a deep-dive study on differences across firms based on their size and age. Given that firm-level data is available with a significant lag, the analysis of productivity developments during the pandemic is beyond the scope of this chapter. The chapter also includes a review of Spain’s ambitious reform commitments under the recovery plan and concludes with recommendations for further action.

B. Data and Empirical Strategy

Data

5. This chapter uses firm-level data that covers over 1.2 million Spanish firms^³ during the period of 2003–2019. The microdata is from the Orbis Bureau Van Dijk (BvD) database, complied by the IMF’s Research Department (Díez et al., 2018). The Orbis BvD database includes all companies that report to the business registry, and thus is considered as well-representing the business dynamics of the Spanish economy (Gopinath et al., 2017). Nevertheless, there could be an underrepresentation of the smaller firms as their self-reported information can be less accurate compared to large firms. These firms are also subject to lighter reporting standards which could result in more missing information. To ensure that business dynamics in our data sample are consistent with census, we select 10 economic sectors for which the database matches well both the sectoral employment and value-added growth from the National Statistical Office since early 2000s. These sectors include manufacturing, construction, wholesale and retail trade, transport, accommodation and food, ICT, professional and technical activities, education, health, entertainment, and other services. They cover about 80 percent of the total value added and 70 percent of total employment in the Spanish economy (see Annex I for details).

Empirical Strategy

6. To understand the factors that are relevant to firm investment, we include standard firm-level variables and other macroeconomic variables of our interest. The empirical model follows the standard specification for firm investment behaviors, which regresses firm-level investment on the above-mentioned variables (Gebauer et al., 2018; Dejuán and Ghirelli, 2018; etc.):

I n v_{i t} = β_{0} + β_{1} L e v e r a g e_{i, t - 1} + β_{2} \log (F i x e d a s s e t_{i t}) + β_{3} D e b t R a t i o_{i t} + β_{4} R O A_{i t} + β_{5} S a l e s G r o w t h_{i, t + 1} + γ X_{i t} + κ_{s r t}

Firm net investment, Inv_it, is the change in fixed assets normalized by the one-year lagged value of fixed assets. The normalization allows the investment variable to be comparable between large and small firms. The normalized investment rate can also be interpreted as the percent change of firm capital stock. Leverage_it and DebtRatio_it represent the debt-to-asset ratio and the debt service ratio, respectively, which serve as proxies for firm balance sheet health. ROA_it and SalesGrowth_it characterize the profits and sales growth, which indicate the performance of each firm. Conceptually, firms with higher leverage ratios and debt service ratios have more stressed balance sheets and find it more difficult to finance their investments. Firms with higher profits or future sales growth, on the contrary, should invest more. Finally, X_it includes other explanatory variables of interests, such as firm size, age, and local product market concentration; K_srt is sector-region-year fixed effects.^⁴

7. The construction of misallocation and TFP measures also follows the standard approaches in the literature.

a. Misallocation. A conventional measure of misallocation is the dispersion of firm-level return to capital, measured as the log marginal revenue product of capital (MRPK) and the return to labor, as measured by the log marginal revenue product of labor (MRPL). The two variables are derived from a growth accounting framework that has been widely used in the literature^⁵ (Hsieh and Klenow, 2009; Hsieh and Song, 2015; Gopinath et al. 2017), which solves each firm’s profit maximization problem^⁶:
$\begin{matrix} \max_{p_{i s t}, k_{i s t}, l_{i s t}} π_{i s t} = p (y_{i s t}) y_{i s t} - (1 + τ_{i s t}^{K}) (r_{t} + δ_{s t}) k_{i s t} - (1 + τ_{i s t}^{L}) w_{s t} l_{i s t}, & (1) \end{matrix}$

where $y_{i s t} = A_{i s t} k_{i s t}^{α} l_{i s t}^{1 - α}$ is the production function of firm i. Standard first-order conditions equalize the marginal revenue product of capital (labor) to the cost of the production factor:
$\begin{array}{l} M R P L_{i t} = ((1 - α) / μ) (p (y_{i s t}) y_{i s t} / l_{i s t}) = (1 + τ_{i s t}^{L}) w_{s t} \\ M R P K_{i t} = (α / μ) (p (y_{i s t}) y_{i s t} / k_{i s t}) = (1 + τ_{i s t}^{K}) (r_{t} + δ_{s t}), \end{array}$

where $τ_{i s t}^{L}, τ_{i s t}^{K}$ are reduced-form measures of labor and capital market frictions, respectively. A higher τ implies that a firm pays a higher cost for the production factor, and therefore it faces more frictions in the factor market. Note that here we assume sector-specific factor prices, therefore the exercise only analyzes misallocation within each sector, but not across sectors.
b. Firm-Level TFP. The estimation of firm-level TFP is conducted by the IMF’s Research Department. Since the Orbis BvD database provides information on material costs for Spanish firms, TFP is then estimated using the method proposed by De Loecker and Warzynski (2012), and also by Ackerberg, Caves, and Frazer (2015), and uses value-added on the left-hand-side.

8. Common explanatory variables for investment, misallocation, and TFP analyses are:

a. Local Market Concentration. We define a unique local product market for each pair of NACE 2-digit sector and autonomous communities. For example, all firms conducting manufacturing of machinery and equipment in Madrid are considered as in the same product market. A similar firm operating in Valencia, however, is considered as in a different market. There is a total of [209] local product markets based on this definition. We measure the concentration of each product market using the Herfindahl–Hirschman index (HHI), calculated by squaring the market share of each firm competing in the market and then summing the resulting numbers.
b. Firm Size. This chapter adopts two different approaches to identify the size of each firm. The first approach follows the standard definition by relying on the number of employees: firms with 0–9 employees, 10–49 employees, 50–249 employees, and 250 employees or more are considered as micro, small, medium-sized, and large enterprises, respectively. The second approach defines the size of a firm based on its book value of fixed assets. We divide the sample into 4 quartiles and define firms from the lowest to the highest quartiles as micro, small, medium, and large firms.
c. Firm Age. The age of a firm is calculated as the number of years since establishment. At establishment, the firm is considered as one year old.

C. Results

Firm-Level Investment

9. While firms generally invested less after the GFC amid a deleveraging process and tighter financial conditions, large firms were disproportionately more affected. Before the GFC, firm net investment rate, which can be also interpreted as the annual percent change of firm capital stock, averaged about 20 to 30 percent. It dropped to negative territory after the crisis and has only recently started to increase again. Besides, firm investment rate is typically positively correlated with firm size (measured by the total value of fixed assets): before the GFC, a 10-percent increase in firm asset value could raise investment rate by about 0.6 percentage points. This difference between large and small firms declined rapidly after the GFC: by 2015, there was almost no difference between the investment rates of large and small firms. Such a change in the size-dependency of firm investment over time turned out to be mainly driven by three financial market and policy variables—real interest rate, total deposits growth, and economic policy uncertainty. Large firms tend to disproportionally lower investment when the real interest rate increases, total deposit growth declines, and economic policy uncertainty rises. This result is consistent with recent findings in the literature that large companies are more sensitive to economic cycles in advanced economies (Cravo, 2017) and contributing the most to economic fluctuations (Crouzet and Mehrotra, 2020). It suggests that supporting systemic firms during economic downturns could help better sustain the investment potential of the business sector.

10. Firm-level characteristics, such as balance sheet health and growth potentials, have also shown to be significant determinants for firm investment. Firms with healthier balance sheets, i.e., the ones with lower leverage ratios and debt service ratios, tend to invest more regardless of the business cycle. These two ratios are often used in the literature (IMF, 2015; Gebauer et al. 2018) as proxies for firm financing constraints—companies with higher leverage ratio and debt service ratio have more limited financing space to cope with any negative shocks. Our results validate the significance of firm financial constraints on investment decisions. Moreover, we find that the relevance of balance sheet stress increases when financial conditions are tighter,^⁷ and that it is the tangible capital investment that responds more strongly to changes in financial conditions. Finally, higher firm-level investment is also positively correlated with stronger future sales growth and return on asset.

Resource Misallocation

11. Resource misallocation has been a long-standing candidate for explaining the productivity differences across countries. Although misallocation can happen in various forms, for this chapter we focus on how the allocation of a given amount of capital and labor across heterogeneous producers can be distorted (Restuccia and Rogerson, 2017). Following the firm optimization problem described in ¶7, each firm will increase the amount of its inputs until the marginal products of inputs equal the marginal costs. In a frictionless market, an efficient allocation of inputs that maximizes aggregate output will thus equalize the marginal products of labor and capital across all producers with the same costs. By examining the dispersion of firm marginal revenue products of capital (MRPK) and labor (MRPL), and the statistical correlation between these marginal products and firm characteristics, we could identify possible sources^⁸ of misallocation and discuss policy solutions to improve the allocative efficiency of production inputs. We also find significant and positive correlations between the model-based measure of cost of capital, MRPK, and observed indicators for financial constraints. For example, firms that have a higher MRPK also pay a higher level of effective interest rate, and hold on average a larger share of intangible assets. For the rest of this chapter, we use the marginal revenue product of capital (labor) and the marginal cost of capital (labor) interchangeably.

12. The within-sector allocative efficiency in Spain’s capital market has stabilized in recent years, consistent with rising TFP. Nevertheless, Spain remains more inefficient relative to early 2000s compared to more productive countries in the region such as France and Germany.^⁹ The stabilization of capital misallocation in recent years (Moral-Benito and Fu, 2015; García-Santana et al., 2020) could partly be associated with the “cleansing effect” of credit contractions (Osotimehin and Pappadà, 2017), as bank lending declines. Research has also pointed to the positive impact of accommodative monetary policy on improving allocative efficiency (Albrizio et al., 2021). The still-elevated level of misallocation implies a large variation in firms’ marginal productivity of capital. Firms showing a higher level^¹⁰ of marginal product of capital are facing a larger financing cost, and thus are more financially constrained. For the remainder of this section, we focus on understanding the allocative efficiency along two observable dimensions, firm size and age, and empirically test if there exist clear distortions in the capital market.

13. Firms with lower fixed assets are subject to higher financing costs. Small and medium-sized enterprises (SMEs) are usually considered as having more difficulties in accessing external financial sources. However, what is often neglected in the discussion is the relevant criteria in classifying firm size, which can include employment, revenue, asset value, etc. In official statistics, SMEs are usually defined based on employment and revenue (European Commission, 2003). We find that the relevant criteria for capital market distortions is the level of capital stock, measured as the total value of fixed assets. Micro, small, and medium-sized firms defined based on fixed asset value pay a higher cost of capital compared to large firms, both before the euro area crisis and in recent years. This finding is consistent with the literature emphasizing the importance of collateral constraint (the maximum amount of external financing is determined by the value of firms’ existing capital stock) for scaling up production. Firm size measured by employment, on the contrary, does not show a robust correlation with the cost of external financing. As a robustness check, we also find that firms with less capital stock on average pay higher effective interest rates, which can serve as a proxy for average borrowing cost.

14. Start-ups with an age of less than two years also face significantly higher frictions in the capital market. Compared to EU peers, Spain’s business sector tends to have lower entry rates and lower post-entry survival rates, suggesting that the young firms face significant barriers not only in entering the market but also in maintaining their businesses. Although the gaps in entry rates and the survival rate within the first two years have been declining before 2015, they became larger again in recent years. Another interesting finding is that Spanish firms seem to be facing a tougher environment only in the first two years—conditional on surviving after 2 years, the probability of surviving within 3 to 5 years is larger in Spain than in a median EU country. In addition, only firms that are 1 or 2 years old seem to be facing significantly higher cost of capital compared to established firms with 10 or more years of experience. Therefore, policies should focus on supporting startups to help them survive through the first two years.

15. In addition to capital market imperfections, misallocation could arise due to the large number of size-dependent regulations in Spain. The prevalence of small and micro firms is high in Spain compared to other European countries, and it is one of the factors that explains Spain’s low aggregate productivity, since SMEs have on average significantly lower TFP compared to large firms. Size related rules and regulations can potentially create a “small business trap” (IMF Country Report, 2018; CEPYME, 2021). More than 100 size related regulations are present in Spain, ranging from accounting, financial, insurance, labor, to tax areas. These have created advantages for small business but could have also reduced incentives to grow and created inefficiencies. For example, firms with less than 10 employees and with 10–49 employees have less responsibilities in forming a workplace representation, which could imply a lower cost of labor from the firms’ perspectives. Using the growth accounting framework in 117, we also find that firms with less than 10 employees and with 10–49 employees pay less for labor, which could imply size-related misallocations in the labor market

Examples of Size-Dependent Regulations

Source: CEPYME.

^{^1/}The accounting regulations would apply if companies meet both conditions related to assets and sales, regardless of its total number of employees.

Examples of Size-Dependent Regulations

	Labor	Accounting ^1/	Tax
>= 10 employees	Employees have the right to elect representatives with 15 hours per month paid	Not eligible for simplified accounting if the company has either 1) assets of more than EUR 1 mn or 2) sales of more than EUR 2 mn	Monthly VAT and income tax retention settlement if company has sales over EUR 6mn.
>= 50 employees	The employee representatives are elected as members of a works council (comité de empresa)	Not eligible for simplified balance sheets if the company has either 1) assets of more than EUR 4mn or 2) sales of more than EUR 8mn

Source: CEPYME.

^{^1/}The accounting regulations would apply if companies meet both conditions related to assets and sales, regardless of its total number of employees.

Examples of Size-Dependent Regulations

	Labor	Accounting ^1/	Tax
>= 10 employees	Employees have the right to elect representatives with 15 hours per month paid	Not eligible for simplified accounting if the company has either 1) assets of more than EUR 1 mn or 2) sales of more than EUR 2 mn	Monthly VAT and income tax retention settlement if company has sales over EUR 6mn.
>= 50 employees	The employee representatives are elected as members of a works council (comité de empresa)	Not eligible for simplified balance sheets if the company has either 1) assets of more than EUR 4mn or 2) sales of more than EUR 8mn

Source: CEPYME.

^{^1/}The accounting regulations would apply if companies meet both conditions related to assets and sales, regardless of its total number of employees.

16. The long-standing structural challenges in the labor market can be another source of inefficiency preventing more efficient allocation of the workforce, particularly for the young firms. Dolado et al. (2011) found that across different sectors, young firms in general use a larger share of temporary contracts, which also implies a higher risk of employment instability for these firms. This can be rationalized by the fact that newer firms are forced to make a more widespread use of flexible temporary contracts for precautionary reasons. In addition, young firms could face higher labor search costs compared to established franchises (Minkler, 1992). Finally, the rigidity in the labor market could also have prevented these young firms from expanding their production to the desirable scale. Using the growth accounting framework, we verify again that startups (with an age of less than 2 years) and young firms (with an age of 3–5 years) have a higher labor wedge, suggesting more distortions facing them also in the labor market. Therefore, supports to startups could consider including non-monetary measures such as promoting employee ownership and facilitating startup spin-offs from science and universities (Comprehensive startup strategy, Germany). Spain has recently modified its labor regulation to promote the use of permanent contracts rather than temporary ones. Since the implementation of the reform, there has been a significant increase in the share of permanent contracts, but it is still too early to make a full assessment. The new legal framework has also strengthened short-time work schemes to provide flexibility to firms facing shocks. These changes could have a positive impact on workforce allocation, including across young firms.

Other Drivers of TFP

17. Product market competition, which has been affected by the regional dispersion of regulatory frameworks, could also have an impact both firm-level TFP and investment. Spain approved the Market Unity Law in 2013, which aimed to align regulatory requirements by the central, regional, and local authorities. Nevertheless, the application of the Market Unity Law remains slow due to the null and void declaration of key principals by the Spanish Constitutional Court in 2017, and as a result, regulatory differences across autonomous communities and municipalities have not improved materially. A recent study by Llano-Verduras et al. (2021) found that fragmentation of the Spanish market has been increasing since the GFC, and that sub-national regulations have been significant contributors. An earlier study by the IMF (IMF, 2015) has stated the importance of product market competition on firm-level TFP by exploring cross-industry variations in ex-ante regulation stringency. Based on an ex-post measure of local market concentration (¶8), we find that a more competitive local product market implies both higher investment and higher TFP at the firm level. This finding is consistent with the literature finding (Gutiérrez and Philippon, 2017; IMF, 2019) that increases in the market power of already-powerful firms could weaken investment, deter innovation, and reduce labor income shares.

Importance of the Factors in Firm Growth: 2015 vs 2019

(in percent)

Note: Data drawn from the module devoted to opinion on the business environment, of the INE business confidence index. It represents the difference between the percentage of firms giving a high importance to those factors and the percentage of firms giving them a low importance.

Importance of the Factors in Firm Growth: 2015 vs 2019

(in percent)

	Total		Industry		Construction		Trade		Transport		Other Services
	2015	2019	2015	2019	2015	2019	2015	2019	2015	2019	2015	2019
Product demand	44.4	40.52	54.3	49.6	43.8	32.04	43.7	41.24	35.9	36.77	43.4	39.61
Macroeconomic environment	24.4	24.76	31.1	31.13	22.5	16.83	17.3	19.65	16.8	21.73	28.5	27.71
Economic regulation	19.7	23.82	18.3	21.76	20.1	19.25	14.9	18.35	11.1	20.62	26	30.03
Taxation	17.9	17.97	17.4	14.97	20.8	26.7	17	12.54	9.2	13.79	20.8	21.03
Default	13.9	5.14	13.4	1.89	22.8	19.74	5.3	-3.04	-6.3	-9.75	23.2	11.66
Availability of financing	13.5	4.08	14.7	5.19	29.5	16.67	7	-0.19	6.9	-0.97	14	3.79
Efficiency of the job market	7.6	14.1	7	10.58	7.6	14.24	3.5	7.19	3.4	15.19	11.6	18.9
Adaptation of human capital	1.1	11.3	5.1	10.28	-4.4	11.98	-4	1.85	-5	10.58	5.6	16.67
Input costs	-5.2	-4.55	13.1	13.47	-8.3	-1.46	-6	-6.18	-6.2	-7.8	-11.9	-12.08
Infrastructures	-11.6	-7.63	-4.8	-2.1	-7.7	-1.3	-12.5	-13.29	-6.6	4.04	-17.1	-13.17
Insufficient equipment	-17.1	-14.84	-4	-9.59	-18.9	-13.75	-21.8	-20.57	-15.8	-9.47	-20.6	-16.53
Legal costs	-33.3	-35.62	-38.8	-43.51	-31.5	-30.58	-38	-43.36	-41.9	-44.16	-25.8	-26.48

Importance of the Factors in Firm Growth: 2015 vs 2019

(in percent)

	Total		Industry		Construction		Trade		Transport		Other Services
	2015	2019	2015	2019	2015	2019	2015	2019	2015	2019	2015	2019
Product demand	44.4	40.52	54.3	49.6	43.8	32.04	43.7	41.24	35.9	36.77	43.4	39.61
Macroeconomic environment	24.4	24.76	31.1	31.13	22.5	16.83	17.3	19.65	16.8	21.73	28.5	27.71
Economic regulation	19.7	23.82	18.3	21.76	20.1	19.25	14.9	18.35	11.1	20.62	26	30.03
Taxation	17.9	17.97	17.4	14.97	20.8	26.7	17	12.54	9.2	13.79	20.8	21.03
Default	13.9	5.14	13.4	1.89	22.8	19.74	5.3	-3.04	-6.3	-9.75	23.2	11.66
Availability of financing	13.5	4.08	14.7	5.19	29.5	16.67	7	-0.19	6.9	-0.97	14	3.79
Efficiency of the job market	7.6	14.1	7	10.58	7.6	14.24	3.5	7.19	3.4	15.19	11.6	18.9
Adaptation of human capital	1.1	11.3	5.1	10.28	-4.4	11.98	-4	1.85	-5	10.58	5.6	16.67
Input costs	-5.2	-4.55	13.1	13.47	-8.3	-1.46	-6	-6.18	-6.2	-7.8	-11.9	-12.08
Infrastructures	-11.6	-7.63	-4.8	-2.1	-7.7	-1.3	-12.5	-13.29	-6.6	4.04	-17.1	-13.17
Insufficient equipment	-17.1	-14.84	-4	-9.59	-18.9	-13.75	-21.8	-20.57	-15.8	-9.47	-20.6	-16.53
Legal costs	-33.3	-35.62	-38.8	-43.51	-31.5	-30.58	-38	-43.36	-41.9	-44.16	-25.8	-26.48

18. Spain continues to rank relatively low in business innovation, which can be critical for productivity growth. Business research and development (R&D) spending in Spain as a share of GDP lags European and OECD peers and it has not increased materially since 2008. Despite the relatively high implied tax subsidy rate for R&D activities, the take-up rate of R&D tax incentives has been limited in the past. Overall government support of business R&D (both direct and indirect) fall short compared to the frontier of the region (OECD, 2021). The number of firms conducting R&D activities and the number receiving R&D tax reliefs have both been declining in recent years (Xifré, 2018; OECD, 2021). In addition to R&D, Spanish businesses also lag in terms of other innovative activities, including business and public-private collaborations, process innovations, and business non-R&D innovative expenditures.

19. Increasing the educational attainment level of workers and employers is crucial to boosting productivity. Despite a 13-percentage point decline over the last decade, the rate of early leavers from education and training (age 18–24) in Spain is still among the highest in Europe. In addition, enrollment in vocational education and training (VET) remains low compared to the region. The structural transformation of the economy, including digitalization and greening, is expected to raise demand on certain skills, such as communication & creation, and information skills (IMF, 2022 Selected Issues). An ageing population also requires emphasis on learning and re-skilling throughout employees’ work lives. Recent reforms to the vocational and training system and the introduction of incentives for the provision of training are aligned with this objective. The reforms to active labor market policies included in the RTRP also aim at promoting learning and re-skilling throughout employees’ work lives by creating personalized itineraries for low employability groups, with focus on those skills most demanded by the market.

D. Recent Policies to Promote Productivity Growth

20. Spain has committed to undertake ambitious reforms as part of its recovery plan. These initiatives cover areas that are key determinants of productivity, as discussed in the previous sections. While the plan identifies the right priorities, its success will ultimately depend on design and implementation details.

Human Capital. One of the main objectives of the labor reform enacted in 2022 was to combat the high prevalence of temporary employment, which hinders the accumulation of workers’ human capital. While it is still too early to assess the impact of the reform, initial data suggests that the proportion of permanent hires has increased significantly. The recovery plan also includes ambitious education reforms to help boost human capital, including investments in digital skills, expansion of vocational training, modernization of the education system and revamp active labor market policies (which is important to limit the loss of human capital during unemployment). It will be important to evaluate the effectiveness of the new programs once they have been in place for some time.
Innovation. The recently approved Law on Science, Technology and Innovation sets a goal for R&D public funding of 1.25 percent of GDP by 2030 (3 percent including private investment). The plan to increase budgetary support for R&D and innovation is welcome but should be complemented with a review of the existing incentives framework. Further efforts are needed to strengthen collaborations between the public and private sectors and identify impediments to business innovation. The adoption of large strategic projects (PERTEs) provides a good opportunity to work on the synergies between public and private investment and foster innovation.

Area	Key Initiatives
Human Capital	Labor reform to address duality, education law, digital skills PERTEs, Law on science, technology, and information, target 2 percent of GDP in R&D investment Law on business creation and growth, Startup law, sectoral conference on better regulation and business climate, insolvency reform
R&D
Business demographics and dynamics

Area	Key Initiatives
Human Capital	Labor reform to address duality, education law, digital skills PERTEs, Law on science, technology, and information, target 2 percent of GDP in R&D investment Law on business creation and growth, Startup law, sectoral conference on better regulation and business climate, insolvency reform
R&D
Business demographics and dynamics

21. There are several ongoing initiatives to improve business dynamics. The recently approved Law on Business Growth includes measures to eliminate financial and administrative barriers to firm growth and to reduce delinquency rates. A new Start-up Law is also being legislated to provide incentives and facilitate the creation of innovative companies. The new sectoral conference for Regulatory Improvement and the Business Climate, established in 2021, aims at enhancing cooperation and sharing of best practices among different levels of the government to foster business activity. Further efforts should focus on reviewing size-dependent regulatory thresholds in the labor and tax spheres (which may be discouraging business growth), and better aligning the regulatory framework across regional and local authorities to promote market integration and business competition.

E. Conclusion

22. This chapter studies the development and drivers of labor productivity in Spain using both macro and firm-level data. The low labor productivity growth in recent years was mostly driven by a lack of capital deepening because of weak firm investment, amid a continued deleveraging process since the GFC. The analysis finds that large firms lowered investment disproportionally more in response to the unfavorable macroeconomic shock, and that firm balance sheet health and growth potential are significant factors affecting investment decisions. Moreover, allocative inefficiency remains relatively high in Spain compared to neighboring countries. Misallocation across firms’ size and age groups is identified both in the capital market and in the labor market (based on different measures of firm size). This suggests that financial frictions due to size-dependent financial constraints remain a significant determinant of TFP. In addition, the size-dependent labor regulations (and possibly other regulations, including in the areas of tax and accounting) could have disincentivized a more efficient allocation of workers. Startups with less than 2 years of experience are the most vulnerable age group, which show a much lower survival rate compared to other European countries and significant disadvantages in both the capital and the labor market compared to older firms. Finally, the analysis finds that high market concentration is associated with both weak investment and weak TFP growth.

23. Safeguarding firm balance sheets and supporting large and systemic companies during downturns could help limit the impact of negative macroeconomic shocks on investment. Firm support measures taken during the pandemic, including the state-backed guarantee program and the solvency support to strategic companies and other businesses affected by COVID have successfully preserved the environment for resuming corporate investment. Reducing economic policy uncertainty and reforms in product market regulation would also help boost investment. Public investment in large strategic projects (PERTEs), and effective implementation of structural reforms, could crowd in private investment in tangible as well as intangible capital (Alloza et al., 2022).

24. Further efforts on improving the allocative efficiency of production factors could enhance firm TFP growth. The several initiatives planned in the context of Spain’s recovery plan go in the right direction. Going forward, continued policy effort is needed to alleviate distortions faced by startups and SMEs. These include: improving access to finance for disadvantaged firms (e.g. startup with less than two years of experience or firms with limited physical capital to use as collaterals), reviewing the large number of size-related rules and regulations, and expanding non-monetary support to startups (including facilitating the search for talent, providing easier access to data, and reducing regulatory costs). Continued work on aligning the regional regulatory framework and improving product market competition is also warranted.

References

Ackerberg, D., Caves, K., & Frazer, G. (2015). Indentification Properties of Recent Production Function Estimators. Econometrica.
- Search Google Scholar
- Export Citation
Albrizio, S., González, B., & Khametshin, D. (2021). A Tale of Two Margins: Monetary Policy and Capital Misallocation in Spain. Working paper.
- Search Google Scholar
- Export Citation
Alloza, M., Leiva-León, D., & Urtasun, A. (2022). The Response of Private Investment to an Increase in Public Investment. Bank of Spain, Economic Bulletin.
- Search Google Scholar
- Export Citation
Almunia, M., & Lopez-Rodriguez, D. (2018). Under the Radar: the Effects of Monitoring Firms on Tax Compliance. American Economic Journal: Economic Policy.
- Search Google Scholar
- Export Citation
Andrews, D., Cingano, F., & Conconi, P. (2014). Public policy and resource allocation: evidence from firms in OECD countries. Economic Policy.
- Search Google Scholar
- Export Citation
Bank of Spain. (2015). Annual Report: Business Dynamics in Spain. Bank of Spain.
- Search Google Scholar
- Export Citation
BMWK. (2022). Start-up-Strategie der Bundesregierung. Bundesministerium für Wirtschaft und Klimaschutz (BMWK).
- Search Google Scholar
- Export Citation
CEPYME. (2021). Improving Business Situation of SMEs in Spain. CEPYME.
- Search Google Scholar
- Export Citation
Cravo, T. (2017). Firm Size and Business Cycles. Institute for the Study of Labor (IZA).
- Search Google Scholar
- Export Citation
Crouzet, N., & Mehrotra, N. (2020). Small and Large Firms over the Business Cycle. American Economic Review.
- Search Google Scholar
- Export Citation
De Loecker, J., & Warzynski, F. (2012). Markups and Firm-level Export Status. American Economic Review.
- Search Google Scholar
- Export Citation
Dejuán, D., & Ghirelli, C. (2018). Policy Uncertainty and Investment in Spain. Bank of Spain.
- Search Google Scholar
- Export Citation
Díez, F., Fan, J., & Sanchez-Villegas, C. (2018). RES-Orbis Dataset: Cleaning Procedure and TFP Estimation. International Monetary Fund.
- Search Google Scholar
- Export Citation
European Commission. (2003). Commission Recomendation of 6 May 2003 Concerning the Definition of Micro, Small, and Medium-sized Enterprises. European Commission.
- Search Google Scholar
- Export Citation
Fu, C., & Moral-Benito, E. (2018). The Evolution of Spanish Total Factor Productivity since the Global Financial Crisis. Bank of Spain.
- Search Google Scholar
- Export Citation
García-Santana, M., Moral-Benito, E., Pijoan-Mas, J., & Ramos, R. (2020). Growing Like Spain: 1995–2007. International Economic Review.
- Search Google Scholar
- Export Citation
Gebauer, S., Setzer, R., & Westphal, A. (2018). Corporate Debt and Investment: A Firm-level Analysis for Stressed Euro Area Countries. Journal of International Money and Finance.
- Search Google Scholar
- Export Citation
Gopinath, G., Kalemli-Özcan, §., Karabarbounis, L, & Villegas-Sanchez, C. (2017). Capital Allocation and Productivity in South Europe. Quarterly Journal of Economics.
- Search Google Scholar
- Export Citation
Gutiérrez, G., & Philippon, T. (2017). Declining Competition and Investment in the US. NBER Working Paper No.23583.
- Search Google Scholar
- Export Citation
Hospido, L., & Moreno-Galbis, E. (2015). The Spanish Productivity Puzzle in the Great Recession. Madrid: Bank of Spain.
- Search Google Scholar
- Export Citation
Hsieh, C.-T., & Klenow, P. (2009). Misallocation and Manufacturing TFP in China and India. Quarterly Journal of Economics.
- Search Google Scholar
- Export Citation
Hsieh, C.-T., & Zheng, S. (2015). Grasp the Large, Let Go of the Small: The Transformation of the State Sector in China. Brookings Papers.
- Search Google Scholar
- Export Citation
IMF. (2015). Spain: Selected Issues. International Monetary Fund.
- Search Google Scholar
- Export Citation
IMF. (2018). IMF Country Report 18/330: Spain 2018 Artical IV Staff Report. International Monetary Fund.
- Search Google Scholar
- Export Citation
IMF. (2018). Spain: Selected Issues. International Monetary Fund.
- Search Google Scholar
- Export Citation
IMF. (2019). The Rise of Corporate Market Power and its Macroeconomic Effects. IMF World Economic Outlook.
- Search Google Scholar
- Export Citation
Llano-Verduras, C., Pérez-Balsalobre, S., & Rincón-Aznar, A. (2021). Market Fragmentation and the Rise of Sub-national Regulation. The Annals of Regional Science.
- Search Google Scholar
- Export Citation
Minkler, A. (1992). Why Firms Franchise: A Search Cost theory. Journal of Institutional and Theoretical Economics.
- Search Google Scholar
- Export Citation
OECD. (2021). R&D Tax Incentives: Spain, 2021. OECD.
- Search Google Scholar
- Export Citation
Osotimehin, S., & Pappadà, F. (2015). Credit Frictions and the Cleansing Effect of Recessions. The Economic Journal.
- Search Google Scholar
- Export Citation
Restuccia, D., & Rogerson, R. (2017). The Causes and Costs of Misallocation. Journal of Economic Perspectives.
- Search Google Scholar
- Export Citation
Xifré, R. (2018). Spanish Investment in R&D+I in the Wake of the Crisis: Public versus Private Sector. Spanish Economic and Financial Outlook.
- Search Google Scholar
- Export Citation

Annex I. Data Coverage and Sample Selection

The table below summarizes the comparison of sectoral value added and wage bills between 2019 and 2003 in both the Orbis BvD database and in Eurostat. The 10 sectors selected for our firm-level analysis also show similar dynamics in the developments of the two variables from the two data sources, but they are not shown in the annex due to space limits.

^{Annex Table I.1.}

Spain: Sectoral Value Added and Wage Bills

Note: The table shows the ratios of the nominal value of 2019 value added and wage bills over the relevant nominal value in 2003, from both the Orbis sample and macro statistics in Eurostat.

^{Annex Table I.1.}

Spain: Sectoral Value Added and Wage Bills

		Eurostat (2019 over 2003)		Orbis BvD (2019 over 2003)
NACE 1-digit Sector	No. of Observations	Value Added	Wage Bills	Value Added	Wage Bills
Manufacturing	1,240,793	1.12	1.01	0.94	0.93
Construction	1,654,263	0.81	0.74	0.58	0.72
Wholesale and retail trade; repair of motor vehicles and motorcycles	2,514,824	1.56	1.49	1.11	1.18
Transporting and storage	398,805	1.44	1.49	1.60	1.40
Accommodation and food service activities	614,057	1.37	1.65	1.53	1.51
Information and communication	278,940	1.22	1.53	1.09	1.30
Professional, scientific and technical activities	978,172	1.94	2.22	1.47	1.47
Education	137,630	1.51	1.56	1.90	1.91
Human health and social work activities	216,716	1.86	1.93	2.01	1.96
Arts, entertainment and recreation	164,213	1.82	1.66	1.78	1.97
Other services activities	156,871	2.06	1.87	1.77	1.49

Note: The table shows the ratios of the nominal value of 2019 value added and wage bills over the relevant nominal value in 2003, from both the Orbis sample and macro statistics in Eurostat.

^{Annex Table I.1.}

Spain: Sectoral Value Added and Wage Bills

		Eurostat (2019 over 2003)		Orbis BvD (2019 over 2003)
NACE 1-digit Sector	No. of Observations	Value Added	Wage Bills	Value Added	Wage Bills
Manufacturing	1,240,793	1.12	1.01	0.94	0.93
Construction	1,654,263	0.81	0.74	0.58	0.72
Wholesale and retail trade; repair of motor vehicles and motorcycles	2,514,824	1.56	1.49	1.11	1.18
Transporting and storage	398,805	1.44	1.49	1.60	1.40
Accommodation and food service activities	614,057	1.37	1.65	1.53	1.51
Information and communication	278,940	1.22	1.53	1.09	1.30
Professional, scientific and technical activities	978,172	1.94	2.22	1.47	1.47
Education	137,630	1.51	1.56	1.90	1.91
Human health and social work activities	216,716	1.86	1.93	2.01	1.96
Arts, entertainment and recreation	164,213	1.82	1.66	1.78	1.97
Other services activities	156,871	2.06	1.87	1.77	1.49

Note: The table shows the ratios of the nominal value of 2019 value added and wage bills over the relevant nominal value in 2003, from both the Orbis sample and macro statistics in Eurostat.

Annex II. Definition of Variables and Summary Statistics

Variable Definition (additional)

Net investment rate, Inv_it: the change in the value of fixed assets, divided by the total capital stock (1-year lagged value of fixed assets).

Leverage ratio, Leverage_it: total outstanding debt minus cash and its equivalents, divided by total assets.

Debt service ratio, DebtRatio_it: interest payments plus financial costs divided by gross operating revenue plus gross financial revenue.

Sales growth rate, SalesGrowth_it: the percent change of sales revenue relative to the previous year. The variable takes a value of zero if firm sales are zero in two subsequent years.

Return on asset, ROA_it: sum of gross operating revenue plus gross financial revenue minus financial costs and interest payments, divided by total assets.

Summary Statistics

^{Annex Table II.1.}

Spain: Descriptive Statistics, Full Sample

Note: All the variables are winsorized to eliminate the impacts from the bottom and top 1 percent. Observations with effective interest rate higher than 100 percent are also excluded.

^{Annex Table II.1.}

Spain: Descriptive Statistics, Full Sample

Variable	No. of Observations	Mean	Std. dev.	1st Percentile	99th Percentile
Net Investment Rate	6,542,441	0.32	1.75	-1.0	13.6
Leverage Ratio	5,177,887	0.3	0.3	0.0	1.7
Log(Fixed Assets)	7,719,755	11.5	2.1	6.0	16.6
Debt Service Ratio	5,846,711	0.3	1.1	-4.8	6.6
ROA	8,055,687	0.0	0.2	-1.4	0.7
Age	8,062,505	13.5	9.2	2	45
EMPL	6,694,141	9.6	18.1	1	131
Market Share (percent)	7,850,550	0.2	2.2	0	100
Effective Interest Rate (percent)	4,022,777	8.6	12.4	0.0	100
TFP Growth (percent)	4,066,854	-0.3	14.1	-51.6	51.8
Sales Growth (percent)	7,024,140	24.4	125.3	-90.7	935.5
HHI (0–1)	8,065,049	0.05	0.11	0	1

Note: All the variables are winsorized to eliminate the impacts from the bottom and top 1 percent. Observations with effective interest rate higher than 100 percent are also excluded.

^{Annex Table II.1.}

Spain: Descriptive Statistics, Full Sample

Variable	No. of Observations	Mean	Std. dev.	1st Percentile	99th Percentile
Net Investment Rate	6,542,441	0.32	1.75	-1.0	13.6
Leverage Ratio	5,177,887	0.3	0.3	0.0	1.7
Log(Fixed Assets)	7,719,755	11.5	2.1	6.0	16.6
Debt Service Ratio	5,846,711	0.3	1.1	-4.8	6.6
ROA	8,055,687	0.0	0.2	-1.4	0.7
Age	8,062,505	13.5	9.2	2	45
EMPL	6,694,141	9.6	18.1	1	131
Market Share (percent)	7,850,550	0.2	2.2	0	100
Effective Interest Rate (percent)	4,022,777	8.6	12.4	0.0	100
TFP Growth (percent)	4,066,854	-0.3	14.1	-51.6	51.8
Sales Growth (percent)	7,024,140	24.4	125.3	-90.7	935.5
HHI (0–1)	8,065,049	0.05	0.11	0	1

Note: All the variables are winsorized to eliminate the impacts from the bottom and top 1 percent. Observations with effective interest rate higher than 100 percent are also excluded.

Prepared by Nicolas Arregui and Yu Shi (both EUR).

Defined as the ratio of capital services (i.e. stock adjusted by quality) over total hours worked.

The data sample is an unbalanced panel with about 8 million total observations .

For a complete discussion of the construction of variables, see Annex II for details.

The specification of the wedge parameter, τ can be slightly different depending on whether the analyses focus more on the factors markets (capital, labor) or the product market. In this paper, we model the frictions in the capital market and the labor market following Hsieh and Song (2015).

π, p, y, r, δ, k, w, l represent profit, output price, output, market interest rate, sector-specific depreciation, capital stock, and employment, respectively; i, s, t refer to firm, sector, and year.

Tighter financial conditions imply that real interests are higher and that the total deposits growth in the financial system is lower.

Common drivers of misallocation that have been discussed in the literature include tax and labor regulations, discretionary governance or credit provision, imperfect competition in the capital, labor, and product markets, etc. (Restuccia and Rogerson, 2017).

This exercise fixes the sectoral share of gross value added at its initial level and normalizes MRPK dispersion in 2005 as 1 for all countries. A potential caveat of the exercise is that although the Orbis BvD database is relatively representative for Spain, it could be less representative for other countries. In the comparison chart, we show results starting 2005 to avoid any big jumps in sample coverage. In addition, the comparison for manufacturing firms is consistent with the one discussed in Gopinath et al. (2017).

After controlling for aggregate financial market conditions and the specifics of each industry.

Other Publishers

Save
Cite
Email this content

Share Link

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

https://www.elibrary.imf.org/view/journals/018/2023/012/article-A001-en.xml
The link was not copied. Your current browser may not support copying via this button.

Link copied successfully

Collapse
Expand

Labor Productivity Dynamics in Spain: A Firm-Level Perspective: Spain

Author:

Mr. Nicolas Arregui

and

Ms. Yu Shi

Volume/Issue:: Volume 2023: Issue 012

Publisher:: International Monetary Fund

ISBN:: 9798400235634

ISSN:: 2958-7875

Pages:: 24

DOI:: https://doi.org/10.5089/9798400235634.018

Keywords
Labor Productivity Dynamics in Spain: A Firm-level Perspective1
E. Conclusion
References
Annex I. Data Coverage and Sample Selection
Annex II. Definition of Variables and Summary Statistics
- Variable Definition (additional)
- Summary Statistics

View raw image

Labor Productivity

(GDP in const. 2015 prices divided by hours worked)

View raw image

Labor Productivity Growth by Sector 1996–2021

(annual average growth; in percent)

View raw image

Labor Productivity Growth

(annual growth rate, in percent)

View raw image

Spain Labor Productivity Growth Decomposition

(in percent)

View raw image

Capital Deepening Growth

(in percent)

View raw image

Firm Investment Rate: Average and Size Dependency

View raw image

Size-Dependent Firm Investment and Macro Conditions

(Adjusted R-squared, in percent)

View raw image

Macroeconomic Variables: Financial Market Indicators

(in percent)

View raw image

Macroeconomic Variable: Economic Policy Uncertainty

(Index, Mean = 100)

View raw image

Factors Limiting Manufacturing Production: Financial Access

(Percent of surveyed firms)

View raw image

Investment and Balance Sheet Health

(coefficients)

View raw image

Marginal Cost of Capital (MRPK) and Effecitve Interest Rate

View raw image

Marginal Cost of Capital (MRPK) and Intangible Assets

View raw image

Dispersion of Marginal Revenue Product of Capital (MRPK)

(whole sample, 2005 = 1)

View raw image

Dispersion of Marginal Revenue Product of Capital (MRPK)

(manufacturing sector, 2005 = 1)

View raw image

Firm Marginal Cost of Capital Across Different Sizes

(in log differences)

View raw image

Firm Effective Interest Rate Across Different Sizes

(in percentage points)

View raw image

New Entrants

(in percent of active enterprises)

View raw image

Survival Rates Within the First Two Years

(in percent of total entrants within the past 2 years)

View raw image

Conditional Survival Rate: 3 to 5 Years After Birth

(in percent of the number of enterprises survived after 2 years)

View raw image

Firm Marginal Cost of Capital and Age

(in log differences)

View raw image

Euro Area: Share of Firms with 1–9 Employees

(in percent of total enterprises, 2019)

View raw image

Total Factor Productivity (TFP) and Firm Size

(in log differences)

View raw image

Firm Marginal Cost of Labor Across Different Sizes

(in log differences)

View raw image

Firm Marginal Cost of Labor Across Age Groups

(in log differences)

View raw image

Regulatory Barriers: Regional Distribution

(in percent)

View raw image

Produce Market Concentration, Investment, and TFP

(in percentage points)

View raw image

OECD: R&D Intensity

(in percent of GDP)

View raw image

R&D and Other Innovations, 2021

View raw image

Early Leavers From Education and Training

(Percent of total 18–24 population; 2021)

View raw image

Share of Upper Secondary Students in VET Programs

(Percent of total students enrolled in upper secondary education; 2020)

Foresee

^{Annex Table II.1.}

Spain: Descriptive Statistics, Full Sample

Variable	No. of Observations	Mean	Std. dev.	1st Percentile	99th Percentile
Net Investment Rate	6,542,441	0.32	1.75	-1.0	13.6
Leverage Ratio	5,177,887	0.3	0.3	0.0	1.7
Log(Fixed Assets)	7,719,755	11.5	2.1	6.0	16.6
Debt Service Ratio	5,846,711	0.3	1.1	-4.8	6.6
ROA	8,055,687	0.0	0.2	-1.4	0.7
Age	8,062,505	13.5	9.2	2	45
EMPL	6,694,141	9.6	18.1	1	131
Market Share (percent)	7,850,550	0.2	2.2	0	100
Effective Interest Rate (percent)	4,022,777	8.6	12.4	0.0	100
TFP Growth (percent)	4,066,854	-0.3	14.1	-51.6	51.8
Sales Growth (percent)	7,024,140	24.4	125.3	-90.7	935.5
HHI (0–1)	8,065,049	0.05	0.11	0	1

Note: All the variables are winsorized to eliminate the impacts from the bottom and top 1 percent. Observations with effective interest rate higher than 100 percent are also excluded.

Topics

Countries

Series

About

Resources

Abstract

Labor Productivity Dynamics in Spain: A Firm-level Perspective1

A. Introduction

B. Data and Empirical Strategy

Data

Empirical Strategy

C. Results

Firm-Level Investment

Resource Misallocation

Other Drivers of TFP

D. Recent Policies to Promote Productivity Growth

E. Conclusion

References

Annex I. Data Coverage and Sample Selection

Annex II. Definition of Variables and Summary Statistics

Variable Definition (additional)

Summary Statistics

Other IMF Content

Other Publishers

Inter-American Development Bank

The World Bank

Share Link

Table of Contents

Headings

Figures

Metrics

Foresee

Labor Productivity Dynamics in Spain: A Firm-level Perspective^¹