Journal of Institutional Economics (2017), 13: 1, 109–131
C
Millennium Economics Ltd 2016. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence
(http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the
original work is properly cited.
doi:10.1017/S1744137416000254
First published online 20 September 2016
Understanding the economics of limited
access orders: incentives, organizations
and the chronology of developments
BAS VAN BAVEL
∗
, ERIK ANSINK
∗∗
AND BRAM VAN BESOUW
∗∗∗
Economic and Social History, Utrecht University, Utrecht, The Netherlands
Abstract. In ‘Violence and Social Orders’, North, Wallis and Weingast highlight
the need of societies to control large-scale violence. In response to this need, a
variety of social orders has emerged with differing institutional, political and
economic characteristics. One of these social orders is the limited access order that
was prevalent in most of history and still is nowadays. Taking the conceptual
framework of North et al. as a starting point, we make three advances to their
analysis of limited access orders. First, we analyse the incentive structure of actors
involved, using a formal model of the main interactions in a limited access order.
Second, we decompose organizations into two types and analyse their respective
roles. Third, we use insights from historical research to scrutinize the chronology
of the rise of organizations. Jointly, this allows us to refine and substantiate the
insights gained by North et al., highlight the role of organizations and place the
start of relevant developments earlier in time.
1. Introduction
One of the big questions in the social sciences is why societies display such
large differences in economic and social performance, and which distinguishing
features and characteristics at the societal level underlie these differences.
In search of answers, many scholars have focussed on the institutional
characteristics of societies, either being forms of political organization, social
structures or the rules that govern economic life. A large strand of works has
looked, for instance, at the organization of exchange and allocation of goods
and production factors, dominated either by kinship structures, communities,
feudal systems, markets or states, whilst others have primarily focussed on
differences in the institutions that shape political life. To understand how such
institutional characteristics affect economic and social outcomes, we need to
clarify how they determine the incentives of the actors involved at the micro
level. These incentives can be found, for instance, in the desire to generate or
∗
Email: b.j.p.vanbavel@uu.nl
∗∗
Email: j.h.ansink@uu.nl
∗∗∗
Email: b.vanbesouw@uu.nl
109
https://doi.org/10.1017/S1744137416000254 Published online by Cambridge University Press
110 BAS VAN BAVEL, ERIK ANSINK AND BRAM VAN BESOUW
extract rents, find safety, better employ labour or land, or acquire freedom or
political leverage. In addition, we should ask how the institutional structure of
society is itself shaped by these incentives and how and why it develops over
time, for instance, from less to more open systems of exchange or political
interaction.
An important, recent contribution to this field is Violence and Social Orders,
by North et al. (2009, henceforth NWW). They stress violence as the crucial
variable in understanding differences across societies. Specifically, they focus
on the need of societies to control large-scale, organized violence and on their
relative success in doing so. Violence may lead to destruction of lives and capital
goods, and deter interaction, exchange, trade and the benefits of specialization
that come with trade, leading to significant welfare losses (Acemoglu and
Johnson, 2005). This idea is the starting point of NWW in explaining and
understanding the existence of specific social orders, which can be interpreted
as archetypical societies, with specific institutions that emerge because of the
necessity to control violence. They distinguish three social orders: the ‘foraging
order’ that governed human life until the Neolithic Revolution, approximately
10,000 years ago; the ‘limited access order’ that was prevalent since then; and
the ‘open access order’, which developed only some 200 years ago, in a handful
of western countries.
The book by NWW is ambitious – as illustrated by its sub-title A
conceptual framework to understand recorded human history – and it is widely
acknowledged to be an important contribution. Its impact on actual research up
to now, however, remains limited. Possible causes are that the book is conceptual
and highly abstract, and that the economic logic behind some of the mechanisms
of the social orders, and of transitions between orders is unclear. The authors
have decided not to develop a formal model or empirically testable hypotheses
(p. xii). Instead, they provide a conceptual framework wherein violence is linked
to political organization and economic performance, and, importantly, to the
distribution thereof. However, the incentives of actors are mentioned but left
implicit in their framework. In part, this is because they do not discuss the
interaction between production and appropriation in a systematic way. As a
consequence of not specifying incentives, it is difficult to comprehend what
the problems, constraints and strategies of the agents are, as several reviewers
have already remarked (e.g. Bates, 2010: 755). NWW thus present to the
social sciences an encompassing but abstract framework to distinguish between
different societies through time and space, and understand the basic functioning
of each of these societies. In this paper, we argue that the conceptual framework
of social orders can be advanced precisely by specifying explicit relations between
appropriation – based on the distribution of violence capacities – and production,
following van Besouw et al. (2016). Then, using these relations, we further
assess the interaction between violence, political organization and economic
performance. Also, we further specify the role of organizations in this interaction;
https://doi.org/10.1017/S1744137416000254 Published online by Cambridge University Press
Understanding the economics of limited access orders 111
a role that is stressed by NWW (see below), but will be specified here by explicitly
relating it to incentives and production. Doing so for the ‘limited access order’,
we not only sharpen our understanding of the mechanisms of social orders, but
also come closer to understanding societal development and major transitions
of the type discussed by NWW – both political and economic – and to position
them in time more accurately.
Before we present our additions in detail, it is necessary to discuss how NWW
conceptualize the limited access order – alternatively called the natural state. A
limited access order ‘manages the problem of violence by forming a dominant
coalition that limits access to valuable resources – land, labour and capital – or
access to and control of valuable activities – such as trade, worship and education
– to elite groups’ (NWW: 30). Membership of this coalition is, by construction,
limited to individuals with the capacity to muster organized violence. In the
terminology of NWW, they are ‘violence specialists’. Violence specialists form
a subset of a society’s population and are able to use large-scale, organized
violence and to coerce others under the threat of violence. Their ability to do
so is enhanced by patronage networks, social capital, human capital, physical
strength, wealth, status or prestige, and these to varying degrees, depending on
the specific context. The rest of the population, in contrast, has no capacity for
large-scale, organized violence and is therefore in principle not able to join the
elite. Violence specialists within the elite coalition use their power to collectively
extract rents from the rest of the population; rents that are used to hold the
coalition together. Although the coalition utilizes its coercive power against the
rest of society, under the threat of violence, it restricts open violence. The result
is a social order with a strong elite that exercises its coercive power to extract
rents from the rest of society. Although competition amongst violence specialists
for the distribution of rents may exist, membership of the elite coalition entails a
lasting, informal agreement to respect the privileges and rents of other members.
On the other hand, the elite coalition competes, as a group, with violence
specialists outside their coalition. Violence specialists outside the coalition –
termed ‘warlords’ here – are those who refuse to commit to the coalition’s
agreements and those who are not allowed access to the coalition. As violence
specialists, they have the capacity to extract rents from the ordinary population.
They thus compete for control of the society’s rents. In open access societies,
by contrast, the states possess a monopoly on the legitimate use of violence, as
they have consolidated military and police organizations which are controlled
by the political system. All citizens have access to the political and economic
systems, and they have the right to form organizations. These sustain impersonal
exchange and allow all citizens to compete for political control and for economic
rents, which are continuously eroded as a result of this political and economic
competition (NWW: 21–23). By this definition of social orders, we follow NWW
in arguing that almost all historical societies and most contemporary ones can
be characterized as limited access orders.
https://doi.org/10.1017/S1744137416000254 Published online by Cambridge University Press
112 BAS VAN BAVEL, ERIK ANSINK AND BRAM VAN BESOUW
We should note a marked difference between the ‘elite coalition’ in limited
access orders as depicted by NWW, and of ‘the elite’ as set out in the literature
that compares conflict and development in anarchy with some form of hierarchy
(e.g. Acemoglu and Robinson, 2006; Bates et al., 2002; Grossman, 2002). In
the latter, the elite is generally treated as a monolithic entity maintaining order
amongst the rest of the population and levying taxes in return. In the limited
access order, however, the elite coalition emerges from the pool of violence
specialists, and cooperation of violence specialists in this coalition is not self-
evident as there is always a threat of their not joining or of their leaving the
coalition – the latter, in our view, is nothing more than a violence specialist
choosing to no longer obey the coalition’s agreements. Hence, the elite is a
composite entity, and behaviour of individual violence specialists is constrained
by their relations with other violence specialists. This interpretation of the elite
coalition strikes us as a major advance and brings reasoning closer to real-world
situations, both historically and at present. One can, for instance, think of the
situation in Western Europe in the Middle Ages. Here, the elite would be the
feudal elite, with its members competing with each other for rents. Warlords,
too, would be violence specialists, but operate outside the dominant feudal
order as robber barons, captains of roving mercenary troops or noblemen with
independent domains or territories. In the present-day world, in limited access
orders like Burma, Cuba, Mexico, Russia and many sub-Saharan countries,
the elite would be the politicians and officials in power, who use the state
apparatus as a personal fiefdom and compete with each other for rents, whilst the
warlords would be rival factions and rebel leaders (see case studies in North et al.,
2013).
Whilst NWW take it as given that violence specialists prefer to be part of the
elite coalition, here we will endogenize this preference by introducing production
as an important variable. Depending on production levels and the size of the
elite coalition, violence specialists may prefer not to enter the coalition, but to
operate alone as warlords. The advantage of not entering is that warlords are
not bound by any rules of the coalition regarding the use of violence and the rate
of appropriation of production, and may therefore be able to generate a higher
income than if they had joined the elite coalition. This implies that maintaining
a stable elite coalition within limited access order is even more problematic than
NWW already assume. Clarifying the incentives of violence specialists allows
us to identify this trade-off and forms an important addition of this paper. In
section 2, we will explain such incentives in detail.
A second major advance made by NWW is the way they highlight the role
of organizations. In Violence and Social Orders, organizations form a major
element in the transition from limited access to open access orders. As such, they
are analytically separated from institutions, a distinction sometimes considered
as one of North’s main contributions to the literature (Wallis, 2015). In his
earlier work, North had treated organizations as manifestations of institutions,
https://doi.org/10.1017/S1744137416000254 Published online by Cambridge University Press
Understanding the economics of limited access orders 113
which already was an advance over the neoclassical focus on individuals. By
defining institutions as the rules of the game and means of enforcement, and
then separating the rules from the organizations that actually play the game,
it became possible to have a dynamic relationship between the interests and
incentives facing the organizations and the structure of the rules.
This separation, and the resulting role of organizations, is particularly evident
in Violence and Social Orders. The limited access orders are divided there into
three ideal types: fragile, basic and mature orders, each typified by the role and
structure of their organizations. These ideal types together form a spectrum
along which organizations become more durable, more complex, less bound to
personal power, more numerous and less dependent on the dominant coalition
(NWW: 20–21 and 41–49; North et al., 2013: 10–14). Even though there is no
teleological progress, since mature limited access orders can regress and revert
again, the organizations thus form a vital component in development, as they do
also in the transition from mature forms of limited access orders to open access
orders. In the latter, the number of organizations is large, they can be freely
founded by all citizens, and access to them is an impersonal right of all citizens.
An elaborate system of rules, and checks and balances on powerful individuals
and on impersonalized organizations sustain the open access order. The number
of open access orders, by their definition, so far remains small.
Our paper is inspired by NWW’s Violence and Social Orders, but we claim
to make three advances, by further scrutinizing and adjusting important parts
of it. First, we stress the trade-off violence specialist’s face in deciding to join
the elite coalition as a crucial force of inertia in the political and economic
development of limited access orders. Second, we discuss the role and importance
of organizations within this framework. That is, we discuss the importance of
organizations in developing more stable configurations of violence specialists,
and the effect of organization on production – as a stylized representation of
economic development. We do this for what has been in most parts of history,
and in most parts of the world still is, the most widespread order: the limited
access order. To this end, in section 2, we use a formal model, based on van
Besouw et al. (2016) that includes production as a variable, in order to gain
insight into the incentives of the violence specialists. Next, we follow North in
his stress on the role of organizations and assess their role in relation to incentives
and production. As a first step, in section 3, we discuss what these organizations
actually are, using the historical record. This discussion will establish the need to
go beyond the single category of organizations considered by NWW. Whilst the
organizations discussed by NWW are top-down, and dependent on the state, we
suggest including a category of bottom-up organizations. In section 4, we will
use the model to stress the importance of distinguishing between these two types
of organization. In section 5, we will return to the historical record and discuss
what these results imply for the chronology of developments as pictured by
NWW. Our third, and final contribution is that our analysis leads us to suggest
https://doi.org/10.1017/S1744137416000254 Published online by Cambridge University Press