Big Guys Eat Big Cakes: Firm Size and Contracting in Urban and Rural Areas

ABSTRACT

A great deal of attention has been devoted to the analysis of different levels of privatization in urban and rural areas. However, until now no empirical study has been conducted on what types of firms are present in different geographical environments. We find that large firms that operate on a national basis dominate the contracts in the most populated and urban municipalities, and these firms seem to have closer relationships with nation-wide political parties. On the contrary, small firms that operate at a local level usually have the contracts in the less populated and isolated municipalities. This market structure may be harmful for competition in both types of municipalities, damaging the likelihood of obtaining cost savings from privatization.

ACKNOWLEDGEMENTS

Our research on local privatization has received financial support from the Spanish Ministry of Science and Innovation (ECO2009-06946/ECON), from the Autonomous Government of Catalonia (SGR2009-1066), and from the Catalonian Competition Commission. Germà Bel acknowledges as well support from ICREA-Academia. We have benefited from comments received when the paper was presented in seminars at Universitat de Barcelona and Universidad Pública de Navarra, as well as in the XI Encuentro de Economía Aplicada (Salamanca, Spain) and the 10th National Public Management Research Conference (Columbus, OH). Particularly interesting suggestion have been received from Pablo Arocena, Alejandro Bello, Trevor Brown, Peter Claeys, Ignacio Contín, Emilio Huerta, Matthew Potoski, and Mildred Warner. We are thankful to Laia Domènech for very helpful research assistance. Comments from two anonymous reviewers and the editor have been extremely helpful.

Notes

Source: Authors' from Survey on Local Services.

Note: CR1 is the market share of the leading firm in the market. CR4 is the aggregated market shared of the four leading firms in the market.

Note: Data for number of firms refer only to 157 observations (141 observations refer to metropolitan areas, and 16 observations refer to non-metropolitan areas).

Note: Data for number of firms available only for 154 municipalities (157 observations).

Source: Authors' from Survey on Local Services.

Note: Data for number of firms refer only to 157 observations.

Note: Standard errors in parentheses (robust to heteroscedasticity and clustered by province). Significance at 1% (***), 5% (**), 10% (*), 11% (⁺). Marginal effects are evaluated at sample means.

Note: Standard errors in parentheses (robust to heteroscedasticity and clustered by province). Significance at 1% (***), 5% (**), 10% (*). Changes from from 0 to −1 (major to minor firm): 3 observations. Changes from 0 to 1 (minor to major firm): 17 observations. No change: 66 observations.

Note: Standard errors in parentheses (robust to heteroscedasticity and clustered by province). Significance at 1% (***), 5% (**), 10% (*), 15% (+).

Note: Standard errors in parentheses (robust to heteroscedasticity and clustered by region). Significance at 1% (***), 5% (**), 10% (*). Changes from 0 to −1 (major to minor firm): 3 observations. Changes from 0 to 1 (minor to major firm): 17 observations. No change: 66 observations.

A recent and comprehensive review of empirical evidence on motivations of privatization of local services can be found in Bel and Fageda (2007).

Recent surveys do not find a systematic superiority of private production (Hirsch 1995; Boyne 1998; Hodge 2000; Bel and Warner 2008; Bel, Fageda, and Warner 2010).

For some empirical evidence on that issue, see Bel and Costas (2006) and Dijkgraaf and Gradus (2007a).

Empirical evidence about the relationship between privatization, competition, and costs in the delivery of solid waste collection is scarce. However, some studies have found empirical evidence of the effect of competition on costs for the United Kingdom (Domberger, Meadowcroft, and Thompson 1986; Szymanski and Wilkins 1993; Symanski 1996; Gómez-Lobo and Symanski 2001). Other studies for the Netherlands and Spain have examined how competition conditions influence local services costs. Dijkgraaf and Gradus (2007a; 2007b) show that high levels of concentration imply higher costs in the delivery of local services in the Dutch market. Finally, Bel and Costas (2006) show that contracting out in Spain is a process that converges to a bilateral monopoly to the extent that costs in cities with recent privatization are lower than costs under public delivery, but no significant differences in costs between cities with old privatization and those using public delivery are found.

This is so since markets for local services may be more competitive in those municipalities than in rural areas (Morgan, Hirlinger, and England 1988; Ferris and Graddy 1988; Stein 1990; Benton and Menzel 1992; Miranda 1994; Hirsch 1995; Greene 1996; Nelson 1997; Kodrizcky 1998).

In publicly owned firms (public firms henceforth) the government has control over the organization of the service delivery. However, public firms are managed and organized under private commercial law rules, which means they have wide flexibility regarding crucial factors such as labor organization and inputs purchases. Hence, the autonomy of managers is much greater with a public firm than under bureaucratic delivery. Mixed public-private firms are firms where ownership is divided between the government and the private sector. Usually, the government retains a control stake in the firm, but the firm operates under private commercial law. The private partner tends to be a large firm with a solid position in the market for private production of the particular local service. Warner and Bel (2008) provide a detailed analysis of these organizational forms.

Data on the cities that contributed to the 2000 Survey was used in previous studies (Bel 2006b; Bel and Costas 2006).

From the 207 municipalities with private production that answered the questionnaire, seven (3%) did not include the name of the firm holding the contract. Hence, we have information on the private firm holding the contract for 200 municipalities. In addition, there are 46 municipalities with private production (22% of all municipalities with private production that responded the survey) that did not report information on the number of competitors in the bids for contracts. Non-respondent cities do not look systematically different from respondent cities in the rest of crucial variables, so we believe our survey does not have non-response bias. Overall, we have information on the name of the firm and on the number of firms participating in the last bid for the contract for 154 municipalities.

It is important to note that the number of observations used in the empirical analysis increases in three units since the largest municipality of the sample, Barcelona, has divided the delivery of the service in four districts. Because of this, 200 municipalities in 2006 generate 203 observations. Concerning the comparison between 2000 and 2006, 103 municipalities generate 106 observations.

Some of them are active players in foreign markets like the UK (Davies 2007).

In our analysis, we have used the seven regions used by the government of Catalonia for purposes of regional planning and policy implementation. Using counties as variable for geographical environment is not possible because Catalonia has 41 counties, and our sample is not large enough to be representative at the county level in all cases, especially in those cases where the county is formed mainly by very small cities and towns. Furthermore, using provinces (NUTS 3 according to the statistical definition of Eurostat) would not allow us to have enough variability for this variable since there are just four provinces in Catalonia.

These regional parties are: the regionalist—CiU, the pro-independence party—ERC, and the Eco-socialist party—ICV. Besides these region-wide parties, other municipalities in the sample have a mayor belonging to a strictly local political group.

Among other factors that could induce relationships of the type national firm–national party and regional firm–regional party, we can think of issues related to electoral campaign financing, party organization financing, or sharing information on firms by local politicians within the same party. Carr, LeRoux, and Shrestha (2009) emphasize the importance of communication networks created through institutional ties.

Some municipalities that were using a single major firm in 2000 are using a “temporary joint venture” (UTE) of firms in 2006. We have been careful to consider as “major” those UTEs led by a major firm in 2006.

Choices of one municipality concerning contracts may influence choices made by other municipalities in the same neighborhood, so that random shocks may affect in a similar way municipalities within the same province.

The explanatory power obtained in empirical studies on the determinants of local privatization (literature closely related to our work) typically lies between 0.10 and 0.15 as measured by the pseudo-R ² (Bel 2006a). Recall that we are using the logit technique due to the binary nature of the dependent variable.

Note that Table A.1 in the Appendix shows the results of the estimates of an equation of these analyses in which the dependent variable is the number of firms that have participated in the bid for the contract, while the explanatory variables are the same explanatory variables considered in equation (1). Here the estimation is made using an ordered logit. Results of this estimate shows that the number of firms that participate in the bids is higher in municipalities of larger size and located in urban areas. This estimation is not aimed to fully account for all the determinants of the number of firms that opt to the contracts but just to show that the competition environment seems to be different in large and urban municipalities in relation to small and rural municipalities.

We present results of the move from major to minor firms in the Appendix since only three municipalities made that move in the considered period.

Additional information

Notes on contributors

Germà Bel

Germà Bel ([email protected]) is a professor of economics at Universitat de Barcelona and guest professor at Barcelona Graduate School of Economics. He received his PhD in economics at Universitat de Barcelona. His current research examines public sector reform, with special emphasis on privatization, regulation, and competition in local services and transport infrastructures (http://www.ub.edu/graap/beling.htm).

Xavier Fageda

Xavier Fageda ([email protected]) is an associate professor of economics at Universitat de Barcelona, where he received his PhD in economics. His current research examines privatization and competition in local public services and air transportation (http://www.ub.edu/graap/fagedaing.htm).