Is the public sector holding back Italy’s productivity?

Raffaela Giordano, Sergi Lanau, Pietro Tommasino, Petia Topalova

04 August 2015



Since the late 1990s, Italy’s productivity has been stagnant. Real GDP per hour worked increased a meagre 3.5%, while total factor productivity fell by a cumulative 7.5% over 1998-2013. As a result, a productivity gap has emerged between Italy and several OECD economies (Figure 1). With GDP 9% below its pre-Crisis levels and public debt at 132% of GDP, addressing the productivity problem is crucial for Italy.

Figure 1. Italy: Evolution of aggregate productivity

Note: Labour productivity is measured as GDP per hour worked.
Source: Total Economy Database (TED).

Many possible hypotheses for the marked slowdown in productivity have been put forth: the sectoral specialisation of Italian manufacturing, a business model which relies predominantly on micro and small firms (Ciriaci and Palma 2008); cumbersome labour regulations (Daveri and Parisi 2010), judicial inefficiency (Giacomelli and Menon 2013, Esposito et al. 2013); and lack of human capital and/or managerial knowhow (e.g. Bandiera et al. 2008, Brasili and Federico 2008, Bloom et al. 2008). Recent Vox columns have highlighted the lack of reforms in credit, product and labour markets (Manasse 2013) and capital and labour misallocation (Hassan and Ottaviano 2013).

We argue that Italy’s inefficient public sector — one of the lowest ranked among OECD economies (Figure 2) — could be contributing to the low level of productivity of Italian firms. While the idea is not new (e.g. Gross 2011, Pellegrino and Zingales 2014), in a recent paper we provide new evidence on the importance of the public sector in shaping firms’ labour productivity (Giordano et al. 2015). Using data from more than 400,000 Italian firms, we exploit both differences in the quality of government service provision within Italy and differences across sectors in the need for efficient public services. Our findings suggest that public sector inefficiency hinders firm productivity. This effect is not only statistically but also economically significant. For example, for a firm in a sector with above median dependence on government, being in a province with above median public efficiency increases output per worker by 13%.

Figure 2. Italy: Quality of governance indicators

Sources: World Bank Doing Business Indicators, World Economic Forum, Transparency International, and IMF staff estimates.

Empirical approach

To identify the causal effect of public sector efficiency, our empirical strategy exploits the large disparities in government efficiency across provinces in Italy.1 Across a wide range of indicators of quality of governance, there is a significant gap between the relatively efficient centre-north and the lagging south.2 Figure 3 depicts our preferred measure of public sector efficiency: the province-level indicators computed by Giordano and Tommasino (2013) of the efficiency with which government units transform inputs into output relative to the most efficient province in the areas of health care, education, civil justice, child care, and waste collection.3

Figure 3. Italy: Public sector efficiency and firm labour productivity

Public sector efficiency

Firm labour productivity

Notes: Province-level public sector efficiency is from Giordano and Tommasino (2013). Firm labour productivity is measured as real output per employee cost. The map on the right panel plots the median for each province based on 2007 Orbis data.

Similarly, there are substantial regional disparities in firm productivity. In our sample, the median firm in the north produces 9.5% more per euro spent on employees than the median firm in the south, and the median return on assets is 180 basis points higher. Even a casual visual inspection reveals that provinces that have higher public sector efficiency also tend to have higher firm-level productivity. However, this simple correlation does not necessarily imply a causal link. Provinces with low public sector efficiency may differ in their industrial structure, size composition of firms, and a host of other ways that affect labour productivity, independently of government efficiency.

In order to establish a causal link between government efficiency and firm productivity, we employ a simple version of the Rajan and Zingales (1998) framework. In particular, our identifying assumption is that productivity of firms in sectors that are more reliant on the government would be more affected by government inefficiency. In other words, the causal effect of government efficiency is captured by the difference, across provinces with different government efficiency, in productivity gaps between firms operating in sectors more or less dependent on the government. Government dependence is proxied by the frequency of news about a certain sector mentioning the government (Pellegrino and Zingales 2014).

Public sector efficiency and firm productivity

Our main finding is that public sector inefficiency significantly reduces firm-level productivity. Firms in industries more dependent on the government are more productive in provinces where the public sector provides key public services more efficiently (Figure 4). We measure labour productivity as the ratio of operating revenue (or gross value-added) to costs of employees (or number of workers). Return on assets and overall output are used as alternative proxies of the productivity of the firm. Across all measures, we find a similar pattern of higher productivity of firms in government-dependent industries in provinces with more efficient governments.

Figure 4. Effect of public sector efficiency on firm productivity

Note: Point estimate and 95% confidence interval, percent)
Sources: Orbis et al. (2014); Giordano and Tommasino (2011); IMF staff estimates.

This effect is economically significant. To give a sense of the magnitudes involved, for a firm operating in a sector characterised by above-median dependence on government, being in a province with above median public efficiency increases output per euro spent on salaries by 11.3% and gross value added per euro spent on employees by 4.3%.4

We also find significant heterogeneity in the impact of government efficiency on different types of firms. Government inefficiency affects the younger and the largest firms most in our sample (establishments with more than 250 workers). Furthermore, there is evidence that the efficient provision of services, which are responsible for the local branches of the central government is more important than those provided by local governments themselves.5

Italy could realise significant productivity gains

Our analysis suggests that Italy could realise significant macroeconomic productivity gains if average public sector efficiency improved from currently low levels – if public sector efficiency rose to the frontier in all provinces, firm productivity, measured as output per euro spent on salaries, could increase by up to 22% in the sectors that depend the most on the public sector, while gross value added per employee costs could rise from 2% to 10%. For the average firm, output would expand by 3%.

The impact of increasing public sector efficiency could be potentially much more sizable than that of other interventions suggested by the existing empirical literature. For example, several studies have documented the importance of local financial development for growth and productivity in Italy (see Guiso 2004, d’Alfonso 2004, and Barra et al. 2013). In our working paper, we compute the increase in firm labour productivity if financial development in all provinces were to rise to the level of the most financially developed province. Figure 5 presents our findings for both public sector efficiency and financial development. The dividends from raising public sector efficiency appear to be substantially larger.

Figure 5. Italy: Gains from raising public sector efficiency and financial development


Bandiera, O, L Guiso, A Prat, and R Sadun (2010), “Italian Managers: Fidelity or Performance?”, in T Boeri, A Prat, and A Merlo (eds.) The Ruling Class: Management and Politics in Modern Italy, Oxford: Oxford University Press.

Barra, C, S Destefanis, and G Lavadera (2013), “Financial Development and Economic Growth: Evidence from Highly Disaggregated Italian Data”, CSEF Working Paper 346.

Bloom, N, R Sadun, and J Van Reenen (2008), "Measuring and Explaining Management Practices in Italy", Rivista di Politica Economica 98(2): 15-56.

Brasili, A, and L Federico (2008), “Recent Developments in Productivity and the Role of Entrepreneurship in Italy: An Industry View", Rivista di Politica Economica 98(2): 179-214.

Cannari L, M Magnani, and G Pellegrini (2010), Critica della ragione meridionale. Il Sud e le politiche pubbliche, Laterza Editori, Bari.

Ciriaci, D, and D Palma (2008), “The role of knowledge-based supply specialisation for competitiveness: A spatial econometric approach”, Papers in Regional Science 87(3): 453-475.

D’Alfonso, E (2010), “The Italian Financial Development and the Regional Impact on Growth,” Working Paper 3, UniCredit and Universities.

Daveri, F, and M L Parisi (2010), "Experience, Innovation and Productivity - Empirical Evidence from Italy's Slowdown," CESifo Working Paper Series 3123, CESifo Group Munich.

Esposito, G, S Lanau, and S Pompe (2013), “Judicial System Reform in Italy — A Key to Growth,” IMF Working Paper 14/32.

Hassan, F, and G Ottaviano (2013), "Productivity in Italy: The great unlearning", 30 November.

Giacomelli, S, and C Menon (2013), “Firm Size and Judicial Efficiency: Evidence from the Neighbour’s Court,” Bank of Italy Working Paper 898.

Giordano, R, and P Tommasino (2013), “Public Sector Efficiency and Political Culture”, FinanzArchiv 69(3): 256–288.

Giordano, R, S Lanau, P Tommasino, and P Topalova (2015), “Does Public Sector Inefficiency Constrain Firm Productivity: Evidence from Italian Provinces”, IMF Working Paper 15/168.

Gross, D (2011), "What is holding Italy back?",, 9 November.

Guiso, L, P Sapienza, and L Zingales (2004), “Does Local Financial Development Matter?”, The Quarterly Journal of Economics, 119: 929–69.

Manasse, P (2013), "The roots of the Italian stagnation", 19 June.

Pellegrino, B, and L Zingales (2014), “Diagnosing the Italian Disease”, Chicago Booth Working Paper.


1. In Italy, a province is an administrative unit between municipalities and regions. Italy is divided into roughly 20 regions, 100 provinces, and 8,100 municipalities.

2. For example, while in Lombardy, it takes about 150 days to obtain a construction permit, in Sicily, firms have to wait more than 300 days for a similar permit (World Bank Doing Business Indicators 2013).

3. Using province level data on public spending and outputs of five key public services, Giordano and Tommasino (2013) estimate an efficient production frontier using a nonparametric data envelopment analysis. The government efficiency of each of the 103 Italian provinces is assessed based on its distance to the production frontier.

4. The results are robust to several modifications of the baseline empirical approach, such as using data from different years, using alternative measures of dependence on government or alternative proxies for government quality.

5. This confirms earlier studies; see Cannari et al. (2010).



Topics:  Europe's nations and regions

Tags:  Italy, productivity, public sector inefficiency

Economist, Bank of Italy

Economist, IMF

Economist, Bank of Italy

Senior Economist, IMF