Employment Dynamics on Business Estates

Abstract

The central question of this paper is: “What is the magnitude of different forms of employment dynamics on business estates in the Netherlands between 2000 and 2005 and to what extent are the employment dynamics on business estates different from employment dynamics in general?” In order to answer this question, we examine three kinds of local employment dynamics: (1) entries and exits, (2) inter-municipal relocation and (3) the growth/decline of existing firms. We find that new firm formation and inter-municipal relocation do not lead to employment growth on business estates. However, the in situ employment growth on business estates is higher than the general employment growth in the Netherlands. Therefore, it can be concluded that business estates play an important role in local employment dynamics by facilitating the growth of incumbent firms.

Notes

The competition between municipalities for relocating firms is often referred to as a zero-sum game, because no net employment is created when a firm relocates from one municipality to the other (Mariotti, 2005). It is interesting to note that, even from a national perspective, a redistribution of employment through (subsidized) firm relocation can lead to welfare gains, because firms might be located at suboptimal locations they are not able to leave due to large relocation costs (Owens & Sarte, 2002). Therefore, any costs made by the municipalities in the forms of subsidies or supply of cheap land do not necessarily result in a negative-sum game.

The official definition of a business estate in the Netherlands is “an estate of at least 1 ha which is intended and suitable for use by manufacturing, logistics, and commercial and non-commercial service firms”.

Seaport estates are a special kind of business estates that are only found in a few regions, predominantly in the Rotterdam area, Moerdijk, IJmond and in the area around Delfzijl in the North of the Netherlands. For the sake of these analyses, they are categorized as business estates.

It should be noted that we performed all analyses for several sub-regions of the Netherlands as well. The results were strikingly similar for all different regions studied. Therefore, we decided to disaggregate the results by sector only. The results of the analyses for different regions can be obtained from the authors on request.

These numbers refer to the employment in the four sectors considered in this research only. When taking into account all employment (including the government sector), the share of business estates is smaller (around 32%) but on the increase as well.

For example, the employment growth from new firm formation in the period of 2000–2001 has been divided by the total employment in the year 2000.

Nor as relocations in the case of business estates.

I should be noted that an even larger share of employment relocates between business estates (see note of Table 1). When taking this flow into account as well, the share of employment on business estates that is involved in a relocation in any given year is roughly comparable to the employment resulting from new firm formation. However, from the aggregated perspective business estates, relocation between business estates does not result in employment growth. It is, however, an interesting potential source of employment from a local perspective. Therefore, this stream of employment will be analysed in more detail in Section 5.

When taking inter-municipal relocations into account as well, 56.4% of the relocations to business estates originated from another business estate.

The main difference with the growth of existing firms discussed in Section 4 is that Section 4 only looked at the aggregated growth, whereas this section focuses on the growth rates of individual firms.

This implies that firms that were coded as an exit in either of these 2 years have been dropped from these calculations. Higher rates of exits in a certain group of firms might bias the results of these calculations. However, there is no evidence of significant differences in exit-rates between any of the groups (also see Section 4).

The years 2004 and 2005 have not been used as base years because the growth rate in the subsequent years could not be calculated. However, data with respect to these years have been used to calculate the growth rate of firms for the base years 2002 and 2003.

Significance levels of the differences between various groups are reported here because only firms that survived for at least two subsequent years have been included in the calculations. In contrast to the other calculations reported in the previous tables, Table 3 is therefore not based on population data. Because all other tables are based on population data, all differences automatically pertain to the population as a whole (i.e. they are automatically significant).

Additional information

Notes on contributors

Joris Knoben

Both have contributed equally to the manuscript.

Anet Weterings

Both have contributed equally to the manuscript.