Criteria of evaluation for organizational choice

The criteria to compare different institutional forms of public good pro-vision are the following: the extent of economies of scale (ES) and the ex-tent of preference adequacy (PA). There are of course other important criteria of institutional choice, but ES and PA are two very important ones in connection with international outsourcing, on which we will fo-cus in Section 4.3. Furthermore, the underlying theoretical concepts of ES and PA are proper starting points for an international comparison.

Other concepts, which are briefly exposed in Section 4.2.1.3, are not very helpful for international comparisons because they very much de-pend on the institutional framework, and one often needs a lot of insti-tutional background knowledge to be able to assess them thoroughly.

Hence, they cannot be properly analyzed across a larger set of countries.

For our theoretical framework we draw heavily on prior work on institutional choice by Bolter (1998), who focuses on outsourcing in

ge-72 The analysis is a first approach and suffers from not being built on one consistent theoretical framework. Nevertheless, it is able to provide valuable results for a better understanding of the net advantages or disadvantages of VSC.

neral, and Gantner and Eibl (1999), who analyze public good provision in Liechtenstein and apply their central concepts to VSC and, especially, to international outsourcing.

4.2.1.1 Costs and economies of scale

So far, the cost-side has been discussed extensively with regard to eco-nomies of scale arguments, which are, without a doubt, the most impor-tant ones. But there are other cost disadvantages of smaller countries, two of which are however, on a closer inspection, associated with dise-conomies of scale:⁷³

– Easterly and Rebelo (1993) find that country size and the structure of tax systems are interrelated. Smaller countries seem to rely more heavily on inefficient taxes (e.g., custom taxes) than larger countries, which normally rely mainly on income taxes. They explain this he-terogeneity by the high bureaucratic and setup cost of an income tax scheme. Hence, we have again a consequence or special case of the economies of scale argument in the context of tax levy.

– It is more difficult and more costly to provide external security in small countries than in larger ones, according to Alesina and Spolaore (1997). Again, if we simply consider «security» as a public good, we have an economies of scale problem. Conversely, one might argue that small countries are often too small or, more preci-sely, too unimportant to constitute a target in a war or conflict.

– Demand or supply shocks may hurt a smaller country more seri -ously, because it is often not able to compensate the affected region or sector with redistribution from the rest of the country or from other sectors, respectively (Sachs and Sala-i-Martin, 1992).

Therefore, «insurance» against shocks may be more costly for small countries, and exposure to uninsurable shocks has more serious consequences.

73 See also Alesina and Spolaore (1997).

4.2.1.2 Preference adequacy

The extent of preference adequacy is a central idea in economics, especi-ally in the economic theory of federalism. It depicts the hitherto neglec-ted demand side in our concept. Figure 4.2 gives an overview of its mean ing by displaying per-capita provision costs of public goods (CO) and the marginal costs of not being able to comply with individual pre-ferences, which we denote non-PA, because it can be viewed as the in-verse of preference adequacy. It is obvious and intuitive that cost opti-mality and preference adequacy point in different directions. Figure 4.2 displays a pure public good, where the theoretically optimal number of consumers with regard to costs is infinite. In contrast, compliance with the preferences of the consumers decreases with an increasing number of consumers. Hence, non-PA is an increasing function of the number of consumers. The point of intersection of the two functions displays opti-mal jurisdiction size. Note that we assume in Figure 4.2 that the two func tions are monotone and differentiable, which does not have to be the case in reality.

On the demand side, smaller countries have the benefit of being able to avoid a set of problems typical of larger countries (especially conges -tion⁷⁴, heterogeneity costs and costs of coordination). Additionally, po-litical decisions are expected to be generally more in line with the elec-torate’s preferences in smaller countries. Assuming that every inhabitant of a country has an individually optimal quantity-quality point on a continuous scale for any public good provided (for any policy pursued by the government), the sum of distances between those optimal points of all inhabitants and the point depicting the political decision(s) made is supposed to be smaller in small countries than in larger ones. To put it differently, «the average cultural or preference distance between indivi-duals is likely to be positively correlated with the size of the country»

(Alesina and Spolaore, 1997, p. 1029). This fact is generally attributed to the lack of cross-cutting cleavages and ethnic fractionalization as well as due to the cultural homogeneity of VSC.⁷⁵

74 City states are a notable exception.

75 Note that Chapter 5 provides contrasting evidence with regard to this proposition.

We, therefore, conjecture in Chapter 5 that the stronger identification with one’s own country in a VSC may be a promising explanation for the smaller distance that is de-scribed by Alesina and Spolaore (1997).

Besides these indicators pointing to a higher preference adequacy in VSC, it has to be taken into account that people in smaller units seem to be more happy, perhaps due to perception of an easy access to public po-sitions, knowing politicians personally or having the impression of being able to personally influence government policies.⁷⁶ To our knowledge, there are no comprehensive comparative studies of happiness across countries, because of severe methodological caveats when comparing happiness indicators across cultural and national borders. The proposi-tion that people are happier in smaller countries can be inferred from results that people are happier in more federalist countries, where the dis -tance between individual preferences and chosen public policies and/or public goods provided is supposed to be shorter as well (Frey and Stutzer, 2000a, b).

The advantage of federalism and/or smaller units like VSC in terms of preference adequacy can – supplementary to Figure 4.2 – be dis played in Figure 4.3. Think of MU_aand MU_bas the marginal utility of two in-dividuals A and B (or regions A and B, equivalently). The optimal pro-vision quantity for A and B would be q_aand q_b, respectively. In autarky

76 See Jonsson and Olafsson (1991) and Kristinsson (2000) on the happiness of Ice -landers.

Figure 4.2: Per capita costs and non-preference adequacy of public good provision

number of «consumers»

CO costs

non-PA

they would clearly provide these optimal quantities. In the case of a cen-tral decision-making authority, where it is only possible to provide a cer-tain quantity of the good, A and B will arrive at a quantity between qa

and q_b, say q_a+_b. q_a+_b is a «compromise» which leaves both individual A and B, with a welfare loss indicated by the lined area. Note that eco-nomies of scale are not existent in the model underlying Figure 4.3. As a consequence, federalism is one possible way for larger countries to com-ply with different non-PA curves within the population and, hence, to diminish preference distances.

4.2.1.3 Other possible criteria

Besides competition, which has already been analyzed in detail in Section 4.1.2, there are two other noteworthy concepts that should be enumerated and explained in brief:

– The extent of institutional congruence is an institutional yardstick with which different organizational units of public good provision are compared. It is intimately related to the concept of fiscal equi-valence. Speaking of institutional congruence means that decision

costs

tax price

quantity of publicly provided good MUb

MUa

qa+b

qa qb

Figure 4.3: Federalism and preference adequacy

makers, tax payers or financiers and beneficiaries of a public good are congruent. If the provision process of a public good or a juris-diction as a whole is organized according to the principle of institu-tional congruence, then it also complies with the criterion of fiscal equivalence, which postulates that in any regional or local unit of a country there should be equivalence between the tax payments of the inhabitants and the value of public goods and services provided (Olson, 1969; Blankart, 1998). The degree or extent of institutional congruence is a good indicator of institutional efficiency. If full congruence is achievable, problems associated with negative incen-tives can be avoided. It is, e.g., often argued, especially in federal countries, that the central government passes laws which impose implementation costs on regional governments.

– Different institutional arrangements lead to different administrative and indirect costs in the production and provision process of public goods. Private production of public goods may be either less costly or more costly than public production, depending on the task.

We do not assess these two additional criteria, because there is no appa-rent reason to believe that there is a systematic difference between VSC and larger countries with regard to them. Furthermore, one would re-quire a case study approach of single countries to be able to draw con-clusions from an analysis of those criteria.