One can therefore draw the overall conclusion that VSC use all three possibilities to keep costs low in the public sector. On the one hand, they try to source out the provision of public goods international- ly whenever feasible. On the other hand, they simply do not provide pub lic goods or «tailor» them to the country size. It is fact that the pub - lic sector of VSC are examples of integration measures within deeply integrated areas, although this finding has hardly been recognized hitherto in the scientific discussion. Take the EU integration, where – with the exception of Luxembourg – only larger countries are involved, as proof. The idea of a common currency, of common embassies in less important countries outside the EU or of a common defense and exter- nal policy are not qualitatively different from international outsourcing performed by VSC. On a larger scale, one can more easily understand the differences between public goods with different levels of publicness. Whereas there are no attempts to create common universities (all coun- tries – with the possible exception of Luxembourg – are large enough to operate full universities nationally), the efforts to create and introduce a common currency were serious. Contrary to the monetary system, uni- versities are far from being a pure public good; a fact which explains the different treatment in EU integration. Finally note that there are of course further criteria involved in or- ganizational choice, some of which are mentioned in Section 4.2. We do not go into detail here with regard to these criteria, because we wanted to focus on international outsourcing. Is a VSC an optimal unit to provide public goods? Although Chapter 4 has shown that there are several widely used options to re duce the negative size effect, the more general result of Chapter 3 prevails of course. One therefore has to conclude that VSC are not optimal units for the provision of public goods from a cost point of view. Higher prefe- rence adequacy and other intangible advantages of VSC may level out this disadvantage, but it is very difficult to quantify the according argu- ments. VSC can however lower costs by organizational options. These options, the associated costs and their constraints have been analyzed in Chapter 4 in detail. To date we still do not know why the number of VSC has been in- creasing during the last decades despite the obvious public sector disad- vantage. The growing number of secessions of countries cannot be explained by the results in the previous chapters. Even more significant- 124 
Very small countries: organizational choice and international outsourcing