Estimation of foreign MNEs spillovers in Spain

ABSTRACT

Using Spanish firm-level data, we estimate productivity effects of spillovers from foreign multinationals to domestic firms in both manufacturing and service sectors. We find evidence of a positive productivity effect from multinationals on domestic firms operating in the same industry. Analyzing inter-industry linkages, we find evidence consistent with positive productivity spillovers from forward linkages (i.e. from suppliers to buyers) and negative productivity spillovers from backward linkages (i.e. from buyers to suppliers). Our main results hold when analyzing differences between multinational and domestic firms, and for periods of economic growth and recession, although some differences arise. Interestingly, we find evidence supporting a positive role of spillovers during the last recession period.

KEYWORDS:

• Multinational firms
• FDI
• spillovers
• economic recession

JEL CLASSIFICATION:

• F23
• L50
• O12

Acknowledgements

Ramón Núñez-Sánchez conducted part of this study while visiting University College of Cork (Ireland) and is grateful to Eoin O’Leary and Eleanor Doyle for their hospitality during his stay. The authors also wish to acknowledge the useful comments received at the UCC School of Economics Seminar, the XLI Simposio de Análisis Económico at the UPV, and the XXXI Jornadas de Economía Industrial at the UIB. The authors thank an anonymous referee for useful comments.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ For a more detailed explanation, Girma, Görg, and Pisu (2008) summarize the theoretical arguments for both positive and negative effects of foreign MNEs spillovers on domestic firms’ productivity.

² This survey is being carried out by the INE (The National Statistics Institute) and it is placed at the disposal of researchers on the FECYT web site https://icono.fecyt.es/PITEC/Paginas/por_que.aspx.

³ We also drop from our sample those firms reporting a number of contingencies (including mergers and acquisitions). Table A1 in Appendix shows the composition in terms of time observations of the unbalanced panel sample used.

⁴ This variable takes a value of 0 for firms with 0 per cent of foreign ownership; a value of 0.05 for firms with a foreign ownership greater than 0 per cent and lower than 10 per cent; a value of 0.35 for firms with a foreign ownership greater than or equal to 10 per cent and lower than 50 per cent; and a value of 0.75 for firms with a foreign ownership greater than or equal to 50 per cent.

⁵ As pointed out by Javorcik (2004), this is done to consider only the production sold in the domestic market.

⁶ Output and physical investments are deflated to express values in real terms. The deflators are based on the industrial price index and the service sector price index provided by the INE. We use the GDP deflator when industry-level prices are not available.

⁷ The industry breakdown provided by the INE is: Food products, beverages and tobacco products; Textiles and clothing; Leather and footwear; Wood and products of wood and cork; Paper, publishing, printing and reproduction; Coke, refined petroleum products; Chemicals and chemical products; Rubber and plastic products; Other non-metallic mineral products; Metal products; Machinery and equipment; Electrical machinery, apparatus and electronic components; Transport equipment; Other manufacturing products; Wholesale, retail trade and repair of motor vehicles and motorcycles; Hotels and restaurants; Transport and communications; Financial intermediation; Real estate activities and professional, scientific and technical activities; Other services activities.

⁸ Following Huergo and Jaumandreu (2004), when age is older than 40 years, we change it to a unique category of 40 or more years. We group firms by industry into six categories: high-tech manufacturing firms; medium-high tech manufacturing firms; medium-low tech manufacturing firms; low-tech manufacturing firms; knowledge-intensive services; and non-knowledge-intensive services. Finally, we consider four Spanish regions: Madrid; Cataluña; Andaluca; and the rest of Spain.

⁹ We use the Olley-Pakes estimation package for Stata; see Yasar, Raciborski, and Poi (2008).

¹⁰ Results are robust to excluding firms with only one observation. These results are not reported here for space reasons.

¹¹ Again, results are robust to excluding firms with only one observation.

Additional information

Funding

This work was supported by the Fundación Ramón Areces, Spain (Convocatoria 2015).