What is the optimal size of government? This is an issue which has interested political scientists, economists, psychologists, anthropologists and more for many years. For some anarchists the ideal size of government is zero; in contrast, the totalitarian perspective argues for a society where the State controls all [1]. It is argued that anarchists sometimes err in a conflation between government and the state [2]. A contemporary quirk is that most contemporary "anarcho-socialist" perspectives argue for the abolition of the state yet for democratic governance of social property, in contrast to the "anarcho-capitalist" perspective, which argues for the abolition of democratic governance of social property in favour of numerous private states.

Starting from an examination of the need for governments of different scope, the following looks at some of the major theories concerning the optimal size of different types of governance, heavily informed from public economic theory. In doing so, it is hoped to overcome some of the shortcomings of theories of this sort, but also to provide some steps towards resolving what are often well-meaning, but extremely flawed, perspectives in political theory and voluntary association.



The Scope of Governance

One of the successes of modernity and liberalism has been the political assertion of the ontological category of autonomy of the self founded in the individual scope of sensory experience. Kant argued for autonomy as a personal moral right, consisting of "auto" (negative freedom) and "nomos" (moral principles) [3]. Development psychologists, especially Jean Piaget and Lawrence Kohlberg, have examined the neurological and communicative basis from the transition from childhood to adulthood where the sense of autonomous reasoning and post-conventional morality becomes evident [4]. This process is graduated, and with different rights applied at different ages, although with a general orientation towards the mid-late teen-aged years. Individual differences are screened out in favour of the efficiency of procedure.

In general, the principle that adults with adult reasoning govern themselves, with reciprocity (others govern themselves), and with an inter-subjective agreement to actions through informed consent, is widely and increasingly universally accepted. This is perhaps the greatest political success of modern liberalism which has a sound ontological basis - everyone has this same experiential sense of autonomy. Sociologists and anthropologists can and do investigate both the variance and sameness of preferences and even question to what degree supposedly individual choices come from socialisation (far more than individuals believe they do, obviously). But as a political right even self-destructive rights (e.g., voluntary euthanasia) is widely accepted, albeit in a carefully managed fashion and only in limited circumstances.

But even from this sense of individual autonomy, and governance of the self, it can be shown how the assumption of perfectly discrete "self-regarding actions" come into conflict with a reciprocal sense and become "other-regarding actions", to use the categories famously employed by Mill [5]. For example, the medical principle of autonomy states that adults of adult reasoning are entitled to determine whether or not they receive a vaccination - however, if one does not do so, the public is within their defensive rights to limit the contact that the individual has, even against their will. After all, they would be potentially infecting others against their will and the degree of prevention is in proportion to likelihood and seriousness (a person with a common cold probably should stay at home; a person with ebola will be in quarantine). What is important here is the application of governance from outside of the normal body when an act moves from being self-regarding to other-regarding without informed consent.

It might be less comfortable for some, but this truth continues to apply from individuals to groups, to local regions, etc. A society cannot function, for example, if there is not some means of governing access to resources, independent of whether the resource is subject to capitalist, socialist, or any other system of resource management. If a good (e.g., bread) has a production time of P_t, and the time access and deplete the good is the good is T_t, and P_t gt&; T_t, then very quickly, without rules of access, everyone will be consuming a good without producing the good - and production will crash [6]. It is no good to assume that universally virtuous behaviour will apply; indeed, it is a defensive reason to assume that it will not apply. Access rules (sometimes called "property rights") are required. Access rules also imply exchange rules, of which price signals derived from competition are currently considered a relatively efficient means for the distribution of commodities in an economy with scarcity and imperfect information, and importantly, a means of determining the maximum satisfaction of utility of multiple interacting consumers.

However, arguments for a minimal social government elaborate from this principle to cover defence, contracts, law enforcement, and exchange. However such "minarchism", or Night-watchman state ("Nachtwächterstaat", disparagingly coined by Lassalle in 1862), popular among anarcho-capitalists [7], displays a great level of naivety. Left to its own devices, the minimal state with market relations is still subject to less than optimal matters of market failure. Specifically, the economically damaging effects of imperfect competition, the inability to supply public goods, a lack of mitigation of negative externalities or provision of positive externalities, and the inefficiencies that arise from asymmetric information. These are therefore also candidates for public intervention from an external body, whose provision aids development [8]. Thus "self-governance" (whether individual, local, etc) is limited to the extent that it refers only in proportion. As a whole, each level has autonomy but also as a "bundle of rights", to use the common term in property law [9].

Raising these issues in no way suggests that the only solution is via an external (e.g., public) industry, but it does suggest the need for additional governance above and beyond the minimal state, and more importantly, a governance above the institution that has most access rights to the resource. This is of course, balanced by concerns of "government failure" [10], such as the excess growth of bureaucracy, targeted rent-seeking through legislation, and market capture - all leading to aggregate inefficiencies. If a government is to intervene in market failure, the outcome must be better, not worse. Partiality is important here - health and education, for example, provide benefits which are both personal and social. The typical strategy here is minimum social standards with elective choices according to personal desires.

The Scale of Governance

In general, where a good has a relatively low-level of complexity in the production and distribution process efficiencies can be gained with a high level of planning (using tools such as simultaneous equations and an input-output model [11]). In contrast, where the good has a high-level of complexity, the more efficient means of production and distribution is via price signals. Where a good is relatively discrete in its production and distribution, with a low-level of possibility for market failure, small-scale and decentralised governance (e.g., through private firms, cooperatives, etc) is most appropriate. In contrast, where the good is prone to market failure (e.g., the opportunity for large positive externalities) an additional level of governance is required. Of course, most goods fits within the continuum of these idealised dimensions, and could be expressed in terms of variables if necessary. This sketch provides hints of not only where public interventions are required (high levels of market failure) but also what sort of intervention is required (market or plan).

Market Failure and Complexity	Low Complexity	High Complexity
Low Market Failure	Localised planning	Localised Market
High Market Failure	Generalised planning	Generalised Market

At a high level is the provision of public goods, that is, those goods which have the characteristics of non-excludability and non-rivalry. This is an idealised expression public goods can suffer congestion effects and a degree of exclusion, making most closer to "club goods" instead. The provision of goods with positive externalities (e.g., healthcare, education) is also recognised for its benefits, but less obvious are those where there are opportunities for synergistic effects between different goods, but whose production is limited to satisfy local returns (e.g., the famous example of a neighbouring orchid and a beehive producing suboptimal, local-return, levels).

It is tempting to suggest that, assuming that proposed government interventions have a net positive, that the roles of government could be carried out by a single body across the widest possible jurisdiction. After all, there are clear advantages for economies of scale, there is co-ordination advantages, and there is the normative principle that citizens are deserving of equal rights across a society. However, as mentioned, like markets governments too are prone to failure. An intriguing suggestion of the Tiebout model [12] argues that with decentralised club goods an optimal implementation is evident according to where people move to, or at least want to move to (as transaction costs of getting from A to B still apply), thus suggesting that a decentralised application of public resources (although not necessarily revenue collection) is appropriate in many cases. One may further add suggestions such as Dunbar's Number [13], which crosses anthropology and social psychology, which would further suggest a cognitive limit to effective community size.

Public Finance

If there is going to be expenditure by a public body for the purposes of welfare and intervening in market failures, then those expenditures have to be paid for. Various models are in use around the world, including various income taxes, company taxes, sales taxes (and other variants), property taxes, and so forth. Public economists are invariably very concerned with the distorting effects on various attempts to raise public finance, mainly because most taxes distort consumer or producer decisions and come with a dead-weight loss on trades, reducing aggregate utility [14]. This is all in addition to any administrative costs.

This is because such taxes are are a cost on activity. A sales tax on a chair will raise the cost of the chair; the producer will attempt to pass the cost on to the consumer with a degree of success depending on elasticity of demand. Assuming anything less than perfectly inelastic demand (which a producer would be already charging a maximum price for), consumers will be less inclined to purchase the good than previously. With a decline in demand less chairs are produced and provided than what would be optimal. The same applies to any other good or service, to wages, etc. Payroll taxes, despite a low administrative overhead, are a particularly onerous example, literally being a tax on jobs.

However, not all goods are produced at optimal levels. Goods that have negative externalities (such as pollution) where the cost is not captured in the price are too cheap (and which the "tragedy of the commons" [15] is a well-known end result). The generally accepted solution to these goods is to a apply a Pigouvian tax of some sort [16], which raises the cost of the good with revenues being returned to the public body and with general levels of production of the good is declining. Most recent economic research suggests that the information problem of exactly what rate such a Pigouvian tax should be applied. A variant alternative is to employ marketable licenses which come with administrative advantages over direct taxes [17] but also lead to an identical income in conditions of certainty.

Further, there are goods that are fixed by nature in production. Such "economic land" has been recognised by economists since Adam Smith onwards as one which public appropriation of rents is not just a right and proper thing to do, but even a necessity. Because unless economic rents are charged private investors seek monopolisation in the most literal sense of speculative "rent seeking". To put bluntly, the economic class of landlords, who contribute nothing to aggregate wealth, must be taxed out of existence. Public economists would prefer to derive public income where there is no change in a consumer's behaviour, and various resource rents fit under this category; supply is unchanged and because the rent is paid to the public rather than private expropriation, consumer behaviour is unchanged because the rental-price is unchanged - all that is removed is the speculative acquisition.

One hypothesis that is logically consistent and with a good level of empirical evidence is "Henry George Theorem" [18], named after the advocate for public appropriation of economic rents. Investment in public infrastructure increases rental values at the same value that the investments cost. Unlike a system where the benefit is acquired by landlords, if the rents are acquired by a public body which re-invests them into public infrastructure (physical and social) the cycle continues. In other words, the most efficient and just public financing is where there is an abolition of taxes that punish useful productive activity, but where the public receives monies from the rent of economic land, such as through licenses or land tax, etc. Increasingly arguments for a "social dividend" to fund welfare in a manner that is technologically-enhancing also find a voice in the concept of a social collection of economic rents.

Concluding Remarks

The introductory sketch requires significantly further elaboration. Nevertheless, core principles have been identified which provide the means for the highest levels of individual and aggregate liberty, both in the positive and negative sense, utility, and normative justness. To re-iterate these core principles:

* Adults with adult reasoning provide the normative foundation for autonomous choices. The transition from childhood to adulthood occurs with both neurological development for cognitive operations and socialisation for moral reasoning. Particular political rights with equivalent responsibilities are socially granted during young adulthood in a manner for legal expediency which can err due to individual temporal variation.

* Individuals have the right to self-governance and, through reciprocal relations, to establish consensual and informed agreements with others. Where acts do not have the agreement of others and the act is other-regarding, the recipient party can justly call upon an external body for the protection of their moral rights, including prevention where it is not expedient to do otherwise.

* Individuals and organisations have varying production, distribution, exchange, and access rights to goods and services, according to normative and efficiency decisions. In the course of carrying out these rights the possibility of suboptimal and unjust results may arise, which it is inexpedient for the individual or organisation to mitigate. Thus, in cases of such failures proportional and additional governance is required to orientate the activity towards a more optimal result. The possibility is raised that the external governing intervention may also produce additional suboptimal results.

* Goods and services with low levels of market failure are less likely to require external government intervention and are best managed by localised bodies, from individuals to incorporated bodies. Goods with and services with low levels of complexity are candidates for planning rather than market mechanisms. The inverse applies appropriately. Whilst the the collection of resources can be efficiently carried out on a larger scale, the application of such public and club goods is often most effectively carried out by bodies to ensure responsiveness to local needs (large scale infrastructure, the environment etc are exceptional candidates).

* Public finance when applied to production (either directly, or indirectly as a sales tax) comes with an administrative cost and a dead-weight loss with an overall loss of utility. At times (e.g., to goods with negative externalities) this can have a beneficial result, although licenses have administrative advantages. The ideal source of public finance is economics rents, of which land is the most well known example as it is non-distortionary, redirects investment to productive activity and has a high level of normative justness from the perspective of political economy. Resource rents are an excellent source of funds for a universal basic income.

Two major omissions are noted here for future articles. First, an exploration of the optimal manner to elect public representatives, especially in light of median voter theorem, the advocacy of majority rule in May's theorem, and the general challenge to social choice theory expressed by Arrow's Impossibility Theorem. Secondly, an exploration of the most optimal manner to manage localised economic bodies both from an organisational perspective to a socially optimal perspective, looking at capitalist and socialist models as well as partial and variant.

References

1] See for example, Giovanni Gentile, Origins and Doctrine of Fascism, 1929 ("Everything within the state, nothing outside the state, nothing against the state.") or Joseph Stalin, Political Report of the Central Committee to the Sixteenth Congress of the CPSU(B), June 27,1 1930 ("The highest development of state power with the object of preparing the conditions for the withering away of state power)

2] In Against the Class State and for Democratic Government (July, 2012) the following observation is made:

Proudhon makes his famous oft-quoted remarks of what it means for a person to be governed in The General Idea of Revolution, concluding with "That is government; that is its justice; that is its morality". Bukinin in Statism and Anarchy says "[w]e declare ourselves the enemies of every government and every state power, and of governmental organization in general", and Kropotkin defines anarchy as "the name given to a principle or theory of life and conduct under which society is conceived without government" [in Anarchism].

3] Immanual Kant, Grundlegung zur Metaphysik der Sitten, (Groundwork of the Metaphysic of Morals), FP 1785

4] Lawrence Kohlberg, Essays on moral development (2 volumes), Harper-Row, 1985

5] C.L. Ten, Mill on Self Regarding Actions, Philosophy, Vol 43 No 16, pp. 29-37, 1968

6] D. Usher, Theft as a Paradigm for Departures from Efficiency, Oxford Economic Papers New Series, Vol. 39, No. 2, pp. 235-252, 1987

7] Robert Nozick, Anarchy, State, and Utopia, Basic Books, 1974

8] Joseph E. Stiglitz, Markets, Market Failures, and Development, American Economic Review, 79(2), pp. 197–203, 1989

9] Daniel B Klein, John Robinson. Property: A Bundle of Rights? Prologue to the Property Symposium, Econ Journal Watch 8(3): 193-204, Sept 2011

10] Barack Orbach, What Is Government Failure. Yale Journal on Regulation Online, 30, pp. 44-56., 2013

11] W. Paul Cockshott, Allin Cottrel, Towards a New Socialism, Spokesman, 1993

12] Paul W. Rhode, Koleman S. Strumpf, A Historical Test of the Tiebout Hypothesis: Local Heterogeneity from 1850 to 1990. NBER Working Paper No. 7946, 2000

13] R.I.M Dunbar, Neocortex size as a constraint on group size in primates. Journal of Human Evolution. 22 (6): 469–49, 1992

14] James R. Hines, Jr. Three Sides of Harberger Triangles. Journal of Economic Perspectives. 13 (2): 167–188, 1999

15] G. Hardin. "The Tragedy of the Commons" (PDF). Science. 162 (3859): 1243–1248, 1968

16] The alternative Coase theorem - not really a theorem - requires an implausible level of equality of bargaining power, symmetric information, and low transaction costs - does not require further exploration here.

17] Lawrence H Goulder, Ian W.H. Parry , Dallas Burtraw, "Revenue-raising versus other approaches to environmental protection: The critical significance of preexisting tax distortions," The RAND Journal of Economics, 28(4): 708–731, 1997

18] Richard J. Arnott, Joseph E. Stiglitz, Aggregate Land Rents, Expenditure on Public Goods, and Optimal City Size, Quarterly Journal of Economics. 93 (4): 471–500, 1979



Cross-posted at the usual suspects.