The aim of this article is to provide an empirical evidence of the role of the price of the energy for industrial consumers in the foreign direct investment (FDI), testing whether or not it works as a determinant to attract long term fluxes of capital to the countries within the European Union. Panel Data of 23 European countries were analyzed among the several FDI triggers proposed by the UNCTAD in the United Nations Conference For Trade and Development report.
FDI, which stands for foreign direct investment, is an international investment made by a foreigner in an economy pursuing a lasting interest in an enterprise resident in such economy, i.e. not financial resources that goes in and out fast and are not due to productive purposes (also known as flying capitals). Each country registers the amount of FDI in the balance of payment accounts. A lasting interest indicates the presence of a long-term relationship between the foreign investor and the enterprise, and an investor's significant influence on the management of the enterprise.
The minimum standard to consider a direct investment as it, is when the direct investor owns at least the 10% of the shares of the company or when he has voting rights (for listed companies) or the equivalent (Eurostat, 2013). Some cases of FDI are the German car manufactures Opel building a factory to in Figueruelas (Spain), the American Microsoft buying the 40% of the shares of Nokia in Finland or Glencore buying allowances to explode a coal mine in Colombia for the next 40 years.
The FDI brings expertise to the countries in new technologies and therefore improves the technical processes, enhancing the competitiveness and therefore the underlying welfare of domestic consumers. These new technologies are in need of qualified human resources that must be supplied to guarantee the quality of the products being produced and the ancillary services attached as maintenance, supply chain and others.
It is not a secret that certain countries within Europe are in need of sources of employment. Their economies are stuck and despite of the painful efforts imposed by the politicians to the citizens in order to generate economic activity, the situation is turning worst as time goes by. The credit to SME companies is not flowing; the unemployment is growing very fast, the welfare of people is diminishing by the cut of social services, the factories are closing due the decrease of the consumption or are migrating to emerging markets with better prospective, cheaper hand jobs and greater availability of natural resources. A breakpoint to unravel all this trouble is to point out policies and efforts to attract FDI.
The UNCTAD has defined the determinants for the inward FDI, as a part of their Potential Index of Inward FDI that aid governments around the world in the improvement and succeed of policy making. Four key economic determinants are measured to determine the attractiveness of a country for the foreign direct investors and to rank the different countries. The next table summarizes the determinants:
Table 1. Determinants for inward FDI (UNCTAD, 2012).
Even more, in our models we will group the determinants in three main groups: market size, cost factors and infrastructure. In future chapters a brief explanation will be given to support our assumptions.
The econometric analysis plays a fundamental role in the explanation of phenomena in a truthfully way. By using a statistical approach, the economists are able to test dependencies and correlations among variables from real data. This is what we do to test whether the price of electricity and gas for industrial purposes are determinants for the FDI. Different ways to test this dependency may be applied: we chose a fixed effects model (FEM), useful for the panel data of the different determinants by country and year. Bearing in mind that the EU countries are below the same umbrella of regulations, currency, macroeconomic risk, among others, therefore a suitable model to describe the relations between inwards FDI and determinants may be a FEM.
The fixed effect model is a statistical model which treat as non-random determined the quantities of the observed explanatory variables. This means that there are some constant values among the different years that can be identified and rid of the model. In contrast, random effects model considers as random these changes in the variables from one year to another. Bearing in mind the scope of this paper, we will not go deeper in the theory of these models. Therefore we encourage curious readers to have a look of chapter 10 of the book of Wooldridge.
Eurostat is a powerful on-line database developed by the European Community to measure and publish important information regarding the countries within the union. We use this huge database to retrieve the data used in our models.
By looking the different determinants in Table 1 we have searched the exact correspondent or a near proxy. The next table summarizes the gathered data:
Table 2. Gathered variables from Eurostat.
However, as we will see in the next chapters, not all the variables were used to explain the FDI.
Another important issue is that for the sake of balance of the panel data, we got rid of data earlier than 2003 and later than 2010. This ensures a less biased model and better estimations. Notice that the real GDP growth rate is only since 2003 until 2012, whilst taxation variables go from 2001 to 2010.
We collected data for EU23 countries. Some countries were dropped: Switzerland, Luxemburg, Liechtenstein and others. This is due to the nature of their businesses that behaves as outliers for our data. Remember that they are focused on financial services and some rules on investment are not completely clear with them.
Another assumption in the model is about the information and telecommunications (IT) services: European countries are in the top of development in the world. Therefore IT services are transversal to all analyzed countries and won’t be included as a differential point to attract FDI when comparing the nations among themselves. The presence of natural resources is neither considered in our model. Recall that the core of the EC countries are manufactured goods and an economy of services. Therefore it is not a crucial point to increase the inflows of FDI.
The variables related to cash (GDP, Labour Cost and others) have been downloaded from Eurostat taking into account the Purchasing Power Standard (PPS). The PPS is an artificial currency unit, made to equalize the prices among nations. PPS comes from the division of any economic aggregate of some country in their national currency by the world famous Purchasing Power Parity (PPP) of the stated country.
To create our model, we will start declaring how we will face the structural differences among countries within the European Community. As we know, we have several speeds regarding infrastructure and we have to be careful with that: we cannot compare the roads of Spain with the small paths of Romania or Bulgaria. For that reason our study will have two different groups: EU15 countries and the rest. Within EU15 group there are: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg (not included), Netherlands, Portugal, Spain, Sweden and UK (not included). The rest of the countries are east European ones (see table 3 for further explanation).
Our model will be based in the next key parameters (as suggested by the Prof. Paulina Beato at the papers’ presentations at 20/02/2013):
- Inward FDI flow as a variable to be explained.
- Market size and attractiveness, represented by the explanatory variables GDP growth and GDP.
- Cost factors, given by the Labour Index Capital (LCI), electricity price and gas price.
- Infrastructure, not represented by any variable, but incorporated in the model when we split the information in EU15 and the rest of countries.
We want to test whether the investors make their investment decision taking into account the present variables (same year tuples). This is a simple fixed effect regression of the panel data with robust standard error. First the test EU15 countries are presented in the next table:
Table 6. FE model results for EU15.
As we can see in the Table 6, we just have the LCI being statistically significant when explaining FDI. Within a confidence level of 6.3% (P-value), an increase in in 1 unit of LCI, decreases in -2.05% the inward FDI flow for a EU15 countries. No other coefficient associated to any variable is significant. We reject that the electricity and gas prices influences the FDI during the same year. This is intuitive with the present situation of the world: competitiveness and investment are migrating to emerging markets, where the labour is very cheap. Inside EU15, the same effect appeared and is reflected by our model. However let test what is happening when comparing the rest of the countries among them (in development economies of East Europe). In the next table we have the results of the regression:
Table 7. FE model results for east European countries.
Particularly, in this case the price of the electricity is significant at 5.1% of confidence level. Also the GDP real growth rate is significant at any level of confidence. This result is powerful: If a foreign investor has to decide where to invest within the bunch of developing countries, will check these two variables to proceed. An increase of 1% in the real GDP growth rate will increase the inward FDI flows in 0.23% and a rise of 0.01 euros/kwh in the price of electricity for industrial purposes will reduce in the FDI in -0.48%.
If an investor has to do an investment decision in the bunch of countries of the EU15 (developed ones), he will assess the most the LCI for each country, searching for low prices of labour costs. If we agreed in that all EU15 countries offers high skilled and sufficient amount of manpower, therefore a country with the lowest salaries will be more attractive for the FDI. Also is important to mention that the labour resource is not a constraint in the European Community, because the possibilities of migrating resource without any restriction from a country to another where the work places are being created (for instance the migration of Spaniard, Italians, Portuguese and Greeks to Germany).
Reviewing the case of Eastern European countries, the investment in this case is not related to high technology manufacturing. Therefore the key determinant is the growth of the real GDP and the price of electricity, and not the labour costs. The kind of industry in those markets is associated with agricultural economy, manufacturing of goods with low level of technology, some economy of services and employment in big governmental structures that belongs to their communist heritage.
No hay comentarios:
Publicar un comentario