Offshoring, not enough to beat Italy's productivity slowdown

Francesco Daveri , Cecilia Jona-Lasinio

29 November 2007

a

A

The offshoring of activities of manufacturing firms and industries often features at the centre-stage of the political arena for its allegedly negative effects on domestic employment. During the 2004 US presidential campaign, the concern that outsourcing had gone too far creating more hardships than necessary for American unskilled workers was one of the hot political issues. Not by chance academic research on this topic (Feenstra and Hanson, 1996 and 1999, being perhaps the most celebrated contributions in this area)1 has mostly focused on such effects.

Yet the fear of potentially adverse labour market outcomes of offshoring ended up obscuring in the public debate the very reason that pushes a company to delocalize its activities: the search for efficiency gains. Luckily, an array of McKinsey2 and other business consultancy studies have also found that offshoring has been a crucial ingredient to enable the American economy to take full advantage of the potential productivity gains brought about by the IT revolution. And consistent with this evidence, the statistical analysis for US firms and industries (see Amiti and Wei, 2006)3 has also indicated that the offshoring of services and, less strongly, of intermediates has been associated with productivity gains. The same correlation seems to hold in the French and German manufacturing sector and in the British and Irish business services.

In a recent paper,4 we provide empirical evidence on the relation between offshoring and productivity growth in the Italian manufacturing industries. In the last few years, Italy has been on a declining productivity path. Yet this occurred in parallel with an acceleration of the opening up of the economy, also implemented by delocalizing abroad the manufacturing activities previously carried out within the domestic borders. Hence our research question is whether manufacturing offshoring has counteracted or possibly added to the declining productivity trends lately experimented by the Italian economy.5

We plug our offshoring indicators in labour productivity growth regressions, where the growth rate of value added per full time equivalent employed (“labour productivity”) is explained by the growth of capital-labour ratios (so as to control for capital deepening), industry and period fixed effects controls, as well as intermediate and service offshoring indicators – our main variables of interest. Given that offshoring in all likelihood is also simultaneously affected by how well a given industry is doing as well as by a number of other determinants, we also provide instrumental variable estimates. These should control for the the reverse causality problems associated with the feedback effect of past capital accumulation and other industry and period specific influences on offshoring.

Overall, our statistical evidence shows a remarkably consistent pattern of correlation between offshoring and labour productivity growth with various statistical techniques. First of all, it appears that not all types of offshoring positively correlate with productivity growth. The type of good being outsourced indeed matters: the offshoring of intermediates is positively related to productivity growth, while the international outsourcing of services is either not related or - at times - even negatively related to productivity growth.

The scatter-plots reported below provide snapshots of the positive correlation between productivity growth and material offshoring and the absence of correlation between productivity growth and service offshoring.

Figure 1: The close correlation between labour productivity growth and the off-shoring of materials in the Italian manufacturing


Figure 2: The absence of correlation between labour productivity growth and service off-shoring in the Italian manufacturing


Our estimates (Table 1) indicate that the change in the imported intermediate shares in an average Italian manufacturing industry (some 0.7 percentage points per year in 1995-2003) may account for an increase of the yearly growth rate of labour productivity ranging between 0.1 and 0.2 percentage points, depending on the exact statistical specification. Thus intermediate offshoring accounted for some 8-16% of the yearly increase in labour productivity in manufacturing (equal to +1.2 percentage point per year) in this period of time.

Table 1: Narrow off-shoring direct index for intermediate products in the Italian manufacturing industries

 
1995
2003 (1995-03)
Food products and beverages
23.8
27.0
3.2
Tobacco
2.3
13.4
11.1
Textiles
24.9
26.8
1.9
Wearing and apparel
15.7
49.3
33.6
Leather
23.4
31.6
8.2
Wood and wood products
22.1
22.2
0.1
Pulp, paper and paper products
57.1
50.3
-6.8
Publishing and printing
7.8
6.3
-1.6
Chemicals and pharmaceuticals
62.4
73.9
11.5
Rubber and Plastics
17.6
18.2
0.5
Non-metallic mineral products
14.0
11.5
-2.5
Basic metals
72.0
83.3
11.3
Fabricated metal products
9.6
8.6
-1.1
Machinery and equipment n.e.c.
43.8
47.3
3.5
Office machinery and computers
70.8
99.3
28.5
Electrical machinery & apparatus nec
38.5
42.1
3.7
Radio, TV and TLC equipment
82.6
77.7
-4.8
Medical, precision and optical instrs
57.1
64.9
7.9
Motor vehicles, trailers and semi-trails
60.8
58.9
-1.9
Other transport equipment
32.9
47.7
14.8
Furniture; manufacturing n.e.c.
24.8
19.0
-5.9
Average manufacturing industry
36.4
41.9
5.5

Source:own calculation from ISTAT – National Accounts

Here the driving delocalizing industries were wearing and apparel (with a change in the imported share of intermediates of 4.2 percentage points per year in 1995-2003) and office machinery and computers (with a share increasing by 3.6 percentage points per year in the sample period) potentially accounting for an average increase of the yearly growth rate of sectoral labour productivity respectively of 0.7 and 0.6 percentage points.

Our results indicate that off-shoring has partly counteracted the declining forces at work in the Italian economy, but certainly not providing enough impetus to stop such forces.

Improving upon the standard measures of offshoring

Our analysis is based on an original data set inclusive of input-output tables recently released by the Italian statistical institute. Such tables, by splitting the imported and domestic content of the inter-industry transactions of goods and services in the economy, allow us to directly measure the extent of intermediate and services off-shoring on the part of manufacturing industries. This is different and – in our opinion – preferable to using the methodology previously employed by Feenstra and Hanson (and repeatedly used in the other studies in this area). Since their measure is unobserved, the extent of outsourcing has to be inferred from trade data assuming that any purchasing industry would import intermediates or services in the same proportion as any other industry in the economy.

Our calculations do not have to rely on such assumptions that are restrictive and which our data suggest are unwarranted. We provide a direct measure of offshoring not based on untested assumptions and then we econometrically test whether using our indicator or the Feenstra-Hanson indicator makes a big difference. It seems it does. We find a positive relation between intermediate offshoring and productivity growth when using our (direct) measure of offshoring. However, the correlation disappears altogether when the standard Feenstra-Hanson measures of external outsourcing are employed.

We find our results of general interest, over and above the discussion of the case of Italy and we intend to pursue this line of research further in the close future.


Footnotes

1 See Feenstra R.C. and G.H. Hanson (1996), “Globalization, Offshoring and Wage Inequality”, American Economic Review, 86(2) and Feenstra R.C. and G.H. Hanson (1999), “The Impact of Outsourcing and High-Technology Capital on Wages: Estimates for the United States, 1979-1990”, Quarterly Journal of Economics, 114, 3

2 See McKinsey Global Institute (2004), Can Germany Win from Offshoring? and McKinsey Global Institute (2005), How Offshoring of Services Could Benefit France, McKinsey and Company.

3 Amiti, M. and S.-J. Wei (2006), “Service offshoring and productivity: evidence from the United States”, NBER Working Paper w11926, January

4 Daveri, F. and C. Jona-Lasinio, “Off-shoring and productivity growth in the Italian manufacturing industries”. LUISS Lab of European Economics, Working Document n.53, October 2007 as well as University of Parma Discussion Papers, 2007-EP08, November 2007. This paper is available for downloading at either: http://www.luiss.it/ricerca/centri/llee/publications.html or: http://ideas.repec.org/p/par/dipeco/2007-ep08.html

5 A recent detailed study with industry data is in Daveri and Jona-Lasinio, (2005), “Italy’s decline: getting the facts right”, Giornale degli Economisti e Annali di Economia, December, available for downloading at: ideas.repec.org/p/par/dipeco/2006-ep01.html

a

A

Topics:  Labour markets

Tags:  Italy, manufacturing, offshoring, productivity growth

Professor of Economics, University of Parma

Senior researcher at the National Accounts’ Division of the Italian Statistical Institute (ISTAT) in Rome and Lecturer at LUISS University of Rome

Events