Reforming institutions: where to begin?

Khawaja, M. Idrees ; Khan, Sajawal

Institutions promote growth--this view now holds firm ground. The task then is to 'engineer' growth promoting institutions. Endogeneity characterises institutions, for example, groups enjoying political power influence economic institutions but political power itself is a function of wealth. The question then is: what to reform first? History stands witness that generally the societies with extreme inequality and a heterogeneous population tend to evolve institutions that restrict access to economic opportunities for the poor which in turn constrains economic development. On the other hand societies with greater equality and homogeneous population typically enjoy growth-promoting institutions. (1) Institutional reforms should therefore begin with institutions that serve to create or perpetuate inequality and heterogeneity in the society. We argue that the four different kinds of educational systems in operation in Pakistan are a major source of creating and perpetuating inequality and heterogeneity in the population. Access to a single and common educational system will open-up similar opportunities of higher education and job attainment tar all the citizens, thereby reducing inequality. Diverse educational systems promote different sets of beliefs while a uniform system forges belief-convergence in the society that in turn facilitates agreement on a common set of institutional reforms. Therefore it is the educational system that should be the first to reform. We also argue that in Pakistan, unlike some European countries in the 17th century, neither commercial interest nor fiscal constraints can force the de jure power to reform institutions. Typically, large commercial interests in Pakistan have thrived on favours from the de jure power and therefore have no interest in changing the system. Foreign aid eases the fiscal constraints from time to time relieving government of the need to reform institutions. The thought of a revolution of some kind is still a far cry, the society having no such inclination. The alternative then is the gradual approach preferred by North, Acemoglu and Rodrik. (2) This gradual approach suggests the area of educational reforms.

JEL classification: D02, D03, P16

Keywords: Institutional Evolution, Institutional Change, Human Behaviour

1. INTRODUCTION

The role of history in shaping economic outcomes is being increasingly examined. (3) One view is that important events in the history of a nation shape its institutions that in turn determine its economic performance. A country endowed with poor institutions, performs poorly. The question is how a country can break loose of the historical factors to begin the process of institutional reform and thus place itself on the track of economic progress. What we need to know is that whether it is possible to reform all institutions in one go--the big bang approach, or if the institutions can be reformed only gradually--one, or at best, few at a time. If one favours the gradualist approach, then the obvious question is what to reform first, that is, where to begin'?

Concisely speaking institutions represent 'rules of the game' or "humanly devised constraints that shape human interaction" [North (1990)]. They have also been defined as "actual organisational entities, procedural devices, and regulatory frameworks" [WEO (2003)]. The most widely cited definition in literature is again from North (1981)--% set of rules, compliance procedures, and moral and ethical behavioural norms designed to constrain the behaviour of individuals in the interests of maximising the wealth or utility of principals". He terms formal rules, informal constraints and the enforcement characteristics of the two as the complete set of institutions.

The view that institutions represent the rules of the game holds firm ground. (4) No society is devoid of institutions, however many have poor institutions. Then how does a society gets institutions that promote economic growth? Acemoglu, et al. (2005a) argue that institutions are endogenous--political institutions influence economic institutions and vice versa. For example, political institutions, whether democratic or autocratic, determine who enjoys political power. Who gets access to economic opportunities--masses or the elites, is determined by the political power and hence political institutions. However who makes it to the echelons of power, especially in developing countries, is in part determined by wealth, and therefore economic institutions. Given the endogeneity, an attempt to move from one set of institutions to another, for example, from autocracy to democracy, may be successfully thwarted by the would-be losers. For example monopolies (economic institutions) supported by the autocrat may thwart market oriented reforms, if the monopolist or the autocrat himself is deriving rents from their prevalence. The endogeneity problem tempts one to suggest that institutions can only be reformed with a big bang--reform all institutions in one go, perhaps through a revolution. However this leaves us with the problem of how to stage a revolution. Successful revolutions typically are preceded by a certain thought-process [Masood (1991)] which at times may spread over a century. For example, the European enlightenment thought, beginning as far back as 16th century, preceded the revolutions of UK (1688), US (1787) and France (1789). Even when it becomes possible to stage-manage a revolution, the post-revolution institutional changes may not be too revolutionary. North (1990) has quoted examples from history to show that post-revolution institutional changes exhibit the legacy of the past.

If one were to practice gradualism, reforming institutions one by one, the question arises, what to reform first? What conditions should an institution satisfy to top the agenda of institutional reform? To prescribe such conditions the knowledge of the historical sources that had constrained the development of growth-promoting institutions is essential. Based on implicit evidence for India, (5) with whom Pakistan shares a common colonial heritage, we subscribe to the Engerman and Sokoloff (2005) view that initial inequality and population heterogeneity are the sources of path-dependence exhibited by the institutions.

In this context, Gazdar (2004) explains how the land tenure arrangements put in place during the colonial rule over the areas that now form Pakistan served to create inequality. He convincingly argues that first the land tenure arrangements like Royatwari in Sindh and Mhalwari in Punjab (6) sought to create a landholder-advantage and then the canal colonisation highly skewed the power configuration in favour of the landlords. Ali (1988) also provides exhaustive evidence that canal colonies developed during the colonial rule over India in western Punjab, now the most populous province of Pakistan, served to create inequality and heterogeneity in population. Later on, the development policies pursued in the 1950s and 1960s not only served to perpetuate but further widen the income inequality that prevailed then. To understand how inequality and heterogeneity is casting an adverse influence on development, one has only to look at how influence has been very recently used to divert the natural flow of flood waters to save the agricultural land and residential estates the of landed elites. (7) The endless controversy over construction of Kalabagh dana, presence of regional political parties with votes in specific communities and host of religious parties drawing inspiration from different factions of Islam, are sources of heterogeneity, to name a few.

With this background in mind, we can lay down the criteria for the choice of the institution to be reformed, first and foremost. Our criteria are: (i) Inequality and heterogeneity in population being the source of path-dependence, the institution to be reformed first should serve to reduce inequality and heterogeneity in the population; (ii) the institution selected to be the first should be the one that would face relatively lesser resistance from other institutions or whose reform will not be constrained by the absence of some other institution; (iii) its impact should be all encompassing and long-lasting. Regarding condition (ii), we emphasise at the outset that the condition of 'relatively lesser resistance' by no means implies that we expect to find an institution that will meet little resistance from the stakeholders--the relative nature of the phrase should not be lost sight of. For example suppose that the level of discontent with the de jure power is such that to thwart an attempt by the citizens to secure a change in power structure, the existing de jure power, must do one of the two: curb rent-seeking or reform the educational system to adequately groom the populace. Which one would the rulers choose; naturally the latter. Why? The former would hurt them now while the latter would hurt them, at best, a generation-hence. Path dependence being an essential feature of institutions, these are difficult to change. Given the difficulty, the cost of change is high. Only an all-encompassing and long-lasting impact would justify the costs involved. Hence the condition (iii) prescribed above.

The Paper is organised as follows: In Section 2 we review the works of Douglas North, Darron Acemoglu and Dani Rodrik. Section 3 examines the comparative experiences of institutional change (or non-change) of 17th century Britain and Netherlands versus France and Spain, 19th century Britain and Germany versus Austria and Russia, 18th and 19th century North America versus South America, and Korea and Taiwan versus Congo in the 20th century. Section 4 contains a 'brief' on enlightenment era, the objective being to show to what extent the institutional evolution, has benefited from the thoughts of enlightenment philosophers. Based on the lessons drawn from the theories discussed in Section 2, the historical experiences discussed in Section 3, and the thoughts of enlightenment philosophers reviewed in Section 4, the discussion in Section 5 is devoted to the primary objective of the paper--where to begin the process of institutional reform? Section 6 concludes the paper.

2. THEORIES OF INSTITUTIONAL CHANGE

1. Douglas North (8)

The key elements of North's theory of institutional change are: (i) The process of development of human perceptions and beliefs; (ii) those whose beliefs matter; (iii) intentionality and comprehension of the issue by those whose beliefs matter; and (iv) path dependence exhibited by the institutions. These are only the building blocks in the process of institutional change. The element, in North's framework, that triggers the change in institutions is the change in bargaining strengths of the parties to the contract.

2.1. Process of Institutional Change

To understand the process of institutional change let us begin from the state of institutional equilibrium. Institutions being rules of the game reflect a contract between two parties. The institutional equilibrium prevails when parties to the contract do not want to alter the terms of the contract [North 1990)]. The state of institutional equilibrium does not essentially imply that the parties are satisfied with the terms of the contract, rather, it only reflects that given the costs and benefits involved in altering the terms of the contract, the parties do not consider it worthwhile to devote resources towards changing the terms. To illustrate this, assume that the majority of the populace of a country feels that the de jure power has persistently failed to enforce the terms of the contract, in letter and spirit, i.e., it has failed to implement the constitution. Given this failure, the public wants a change in the de jure power. Further, assume that the desired institutional change is possible only if the masses rise against those who currently wield the de jure power. This will require some sacrifices on the part of the masses and may entail retaliation as well from the de jure power. Sacrifices involve putting in one's time, effort and money. The retaliation may take the form of arrests, loss of government job, and in extreme cases, getting injured or even losing one's life in a violent protest. Given this scenario, the citizens will devote resources towards institutional change only if the perceived benefits from the change are greater than the costs involved [North (1990)]. For example, if the citizens subscribe to the view that a change in de jure power will not affect their lives or, at best, the effect would be cosmetic, then they will not strive for a change in de jure power--masses in Pakistan, who, despite being dissatisfied with the performance of the wielders of de jure power have not actively worked for change, seem to subscribe to this view.

A noteworthy element of North's framework is that it is only the perception of costs and benefits of bringing about an institutional change that matter--the agents' decision does not depend on that, which can be observed, though only ex post. As such, the costs and benefits have to be assessed. This implies that agents can be lured to undertake efforts towards institutional change by exaggerating the expected benefits and underplaying some of the expected costs. The exaggeration of costs on the other hand would discourage the agents to work for a change.

So far we have determined that institutional change is a function of change in human perceptions that ultimately translate into beliefs. Therefore to manage an institutional change by design, it is the beliefs system that should be influenced in a manner which is conducive to achieve the desired institutional change. The crucial question here is if it is possible to influence the belief system of a people, and if yes, to what extent and how quickly. This brings us to the second key element of North's theory of institutional change: the process of belief formation.

2.1.1. Formation of Perceptions and Beliefs

North argues that "institutions impose constraints on human behaviour", therefore, a theory of institutional change will focus on human behaviour. North (2005) rightly delves deep into psychology to understand the process of belief formation. He concludes that human perceptions transform into beliefs, but perceptions themselves depend upon learning. North draws upon the work of a number of psychologists to understand the learning process. One view is that the learning process is guided by epigenetic rules--the development of an organism under the joint influence of heredity and experience. However the exact composition of genetic predisposition and experience remains a moot point. A similar view is that three sources, viz. genetics, cultural heritage and environment contribute to learning. The role of these sources in the process of learning is discussed below:

The genetic predisposition of an individual is composed of what North (2005) calls the artifactual structure (i.e. foundation) which is transmitted from generation to generation. North views the informal norms to be the most important carrier of this artifactual structure, though the structure comprises formal rules as well. He suggests that as changes occur in the human environment, these are gradually assimilated into the socio-cultural-linguistic inheritance and are embodied in the foundation.

According to Hayek (1960), cultural evolution, the second source in North's process of learning, consists of intergenerational transfer of knowledge, values, and attitudes etc., that have accumulated through the Darwinian process of evolution. Thus a society's culture incorporates the distilled experience of the past, more than what a single person can accumulate in his life time. Given the contribution of past knowledge, values and attitudes to the prevailing culture, a cultural change would be very difficult to bring about. Culture can be manipulated by design only to the extent of what the present day knowledge and experience can contribute to it. A certain fractional change in culture will occur in a generation's time depending upon the kind and quantity of knowledge that the society chooses to gain and the experiences that it has to pass through today, or by the act of others or by the will of nature. The process of cultural change is therefore, without doubt, highly incremental.

As to the contribution of human environment, the third source of learning, North again prescribes a slow evolutionary process. He says that "if the mind has been programmed by millions of years of hunter/gatherer tradition then the flexibility to adjust to a very different modern world may be very limited, as implied by evolutionary psychologist. The reason why change in environment is a slow evolutionary process is that millions of years of hunter/gatherer tradition cannot be altered by one-off experience--a steady stream of experiences is required to affect the change". However given John Locke's view on empiricism (htrp://www.wsu.edu:8001/-dee/ENLIGHT/), the environment can be influenced through education or, to speak more broadly, by creating the desired kind of awareness, even if Locke's stipulation about human mind being tabula rasa (i.e. erased board) at birth does not hold true.

Thus, institutional change being a function of change in beliefs, in order to design a conscious institutional change, we shall have to influence what a person learns. Therefore education is at the heart of the matter. No wonder that the countries that boast of good institutions today have been placing emphasis on education for long. For example, in the United States over 40 percent of the school-age population had been enrolled in schools and nearly 90 percent of the white adult males were literate by around 1850. Similarly schooling was also widespread in Canada by early nineteenth century [Engerman and Sokoloff (2005)]. The influence of education on institutional change is discussed more comprehensively later on.

2.1.2. Dominant Beliefs

North (2005) emphasises time and again that institutions depend upon beliefs or the subjective mental constructs that the agents possess. He asks upon whose beliefs the choice of institutions is incumbent, and answer himself, that it is the dominant beliefs, the beliefs of those who are in a position to enact institutional change, that matter. North's view that it is the 'dominant beliefs' that matter, implicitly builds upon his own earlier view [North (1990)] that the change in relative prices alters the bargaining strength of the parties to the contract. The party enjoying greater bargaining power attempts to alter the contract. This is to say that the beliefs of the dominant players matter.

2.1.3. Intentionality and Comprehension of the Dominant Players

North (2005) argues that it is not just the dominant beliefs that matter but the intentionality, and comprehension of the issue, of the dominant players, i.e., the mental construct of the players also matters. He goes on to suggest that the world economic growth has remained sporadic throughout history because either the players' move was never intended to maximise social welfare or the flawed comprehension of the issue has caused the results to deviate from intentions (North, 2005).The rise and fall of the socialist Soviet Union is a case in point where perhaps the intention were correct but the dominant players failed to comprehend the issue in its totality. The case of intentionality can be seen in Pakistan's domestic environment. Laws have been enacted in the recent past to grant independence to SBP. However the tenure of its governor has been fixed at three years that is renewable for another term of three years. The point to note here is that the tenure of the government is five years. How can a governor who must seek renewal of tenure for another term from an incumbent government show independence in policy making? Another case in point is the ongoing debate over whether or not the Chief Justice of Supreme Court, deposed by former military dictator, be restored. (9)

2.1.4. Path Dependence

The most important element in North's theory of institutional change is path dependent which is the resemblance of today's institutions to yesterday's. To reiterate, it's the beliefs system that decides the kind of institutions that a society will choose. Given the painfully slow learning process, described above, that influences the belief system, it is only natural to expect that institutions will exhibit, what the literature on institutional economics terms path dependence--resemblance to the institutions of the yesteryears. There are three important sources of path dependence; (i) increasing returns to scale; (ii) informal rules; and (iii) the organisation's that owe their existence to existing institutional arrangement.

North (1990) argues that institutions exhibit increasing returns to scale which makes the change in institutions difficult. He explains that three sources make the returns to institutions increasing in nature: (i) Initial set up costs, (ii) coordination effects, and (iii) reduction in uncertainty. North explains that when institutions are created de novo, organisations incur costs to learn and adapt their behaviour to the existing institutional framework. Overtime, the organisations learn and evolve to take advantage of the opportunity set offered by the existing institutional framework. This learning and adaptation, cuts down the unit cost of operating within the current institutional framework. Secondly, there are positive coordination effects, directly through contracts with other organisations, and indirectly through investment in complimentary activities by the State. Finally, contracting more and more under specific institutional framework reduces the uncertainty about the permanence of the rule. This makes the parties to the contract more comfortable with the existing institutional matrix. These three elements jointly make the returns to institutions increasing in nature. The increasing returns to institutions in turn create organisations and interest groups that enjoy a stake in maintaining the existing institutional matrix because the change would affect them adversely.

Besides the increasing returns, another source of path dependence is the informal norms, an important component of the institutional matrix. While the formal rules can be changed with a stroke of the pen, informal rules are more difficult to change. Pejovich (2006) eloquently lays down the formation process of informal rules. He argues that as human beings interact to survive, some interactions are repeated over and over again, not the least because the public understands their utility but simply because these have worked. Eventually the interactions that pass the test of time are institutionalised into taboos, traditions, moral values, beliefs etc. To explain the process of change in informal institutions, Pejovich argues that when a person or a community develops a new idea, this enlarges the opportunity set of human interaction. If the new exchange opportunities call for a behaviour which is not in conformity with the established ethos, the community would consider the behaviour of those exploiting the opportunities as sub marginal, and therefore, the community may react with sanctions like ostracism etc. However, if the returns are high enough to sustain a large number of repeated interaction (between more and more groups) relative to costs (including sanctions) the success of new activities would force adjustment in the set of informal institutions. Such adjustment may include the addition of new norms to the set of informal institutions, change in an old norm or simply ignoring an otherwise established norm. It is the painfully slow process of change in informal rules that makes the overall institutions path dependent. The process of formation of informal rules laid down by Pejovich confirms the path dependence argued by North and gradualism in institutional evolution favoured by Rodrick (2006).

Finally, institutions may exhibit path dependence because some of the organisations born out of existing institutional matrix (the combination of formal rules, informal constraints and enforcement characteristics of the two) may owe their very existence to that specific institutional arrangement; and a drastic change in such an institutional arrangement may sound a sudden-death for the organisation. Therefore existing organisations will attempt to block institutional change.

To sum up, the increasing returns to institutions, preferences of the organisation born out of current institutional matrix and the informal rules together conspire to make the change in institutions highly incremental and the institutions path-dependant. North cites various examples to support his views on path dependence e.g., the US constitution, Common Law and the North West Ordinance in the US. In Pakistan we refer to a number of institutions e.g., Land titling [Kardar (2007)] and Civil Service [Haque and Khawaja (2007)] as legacy of our colonial past.

2.1.5. Lessons from Douglas North

The key lesson from North is that path dependence makes it difficult for institutions to change and that any long lasting change must be incremental. North's emphasis upon institutions being a function of belief system provides room for designing an institutional change by influencing the belief system. His view that beliefs and the dominant players 'ability to comprehend an issue matter, calls for influencing their beliefs and improving their comprehension. However, since it is difficult to predict who would be the dominant players a generation hence, a long-lasting institutional change would call for influencing the beliefs and improving the abilities of all and sundry to correctly comprehend an issue at hand. More importantly, influencing the beliefs of all, rather than a few, is required to secure homogeneity in the population. This homogeneity would in turn facilitate agreement on a common set of institutional reforms [Egerman and Sokoloff (2005)]. How is homogeneity in beliefs to be secured, is the subject matter of Section 5.

2.2. Darron Acemoglu

Different set of institutions may induce a different kind of resource allocation; some institutions would allow competitive forces to play their role while others would promote rent seeking. So for individuals to prefer one set of institutions over another is but natural." Acemoglu, et al. (2005a) argues that given the preference of different individuals over different set of institutions, the group with the greater political power is likely to secure the institutions of its choice. (This is similar to North's viewpoint that belief of the dominant players matter or that the bargaining strengths of the players matter).

Acemoglu, et al. (2005a) argues that an ideal course for the groups with conflicting interests would be to agree over the set of institutions that maximise aggregate growth and then use their political power to determine the distribution of gains. In practice, groups with conflicting interests do not follow this course. The reason is that there are commitment problems inherent in the use of political power i.e., a monarch or a dictator cannot credibly commit against use of power to his advantage. A monarch or a dictator enjoying absolute power may promise today to respect property rights but in furore nothing would restrain him to renege on his promise. Citing the case of UK, Acemoglu, et al. (2005a) states " institutional changes in England as a result of Glorious revolution, of 1688 were not simply conceded by the Smart kings. James II had to be deposed for the changes to take place".

Acemoglu, et al. (2005a) suggests that the distribution of political power in society is endogenous. It is the political institutions, for example, monarchy or democracy, that determine who holds the de jure power. However, some individuals or groups, though not allocated power by political institutions, may still enjoy de facto power because of their ability to revolt, hold strikes (by trade bodies), hold protests (peaceful or violent), use military power, clergy power or mercenaries etc., to impose their will upon the society. The de facto power of a group largely depends upon the economic resources that it enjoys, which determines its ability to use force and influence the de jure power. It is often the de facto power that forces a change in de jure power. Acemoglu, et al. (2005a) asks why the de facto power does not settle for getting institutions of its choice from the de jure power but insists on changing the de jure power itself. Drawing upon the works of Lichbach (1995), Tarrow (1991) and Ross and Gurr (1989), the authors answer that de facto power is often transitory in nature. Not being sure that its power will continue unabated, it wants to transform the de jure power in a manner that it will continue to work in conformity with the beliefs of the de facto power even after it has ceased to exist.

2.2.1. Lesson from Acemoglu

The lesson then from Acemoglu is that change depends upon the relative bargaining strengths of the de jure and de facto powers. Suppose that the bargaining strength of de jure power is greater and the existing institutions are poorer, then in this case the institutions will remain poor. However if the bargaining strength of the de facto power is greater and the existing institutions are poorer, the de facto power then will force the de jure power to provide institutions of their choice. The de jure power will either yield in favour of institutional change or will be replaced, no matter what modus operandi is adopted by the people who share the beliefs of the de facto power. The bottom line then is that institutional change will have to wait for the emergence of de facto power that can force the de jure power to yield. The question then is, can the emergence of the requisite de facto power be designed. We take up this question in Section 5.

2.3. Dani Rodrick

Rodrick illustrates the process of institutional development by equating institutions with technology that transforms primary endowments of a society into a larger bundle of outputs. He explains that the requisite technology could be either general purpose or highly specific to local needs. He further argues that if the technology (institution) is general purpose in nature and is easily available on the world market, then it can be adopted by simply importing a blueprint from the developed countries (or any country whose institutions are considered good). However, if the technology is specific to local conditions, which is more often the case, then technology would evolve by trial and error. This suggests that a society is able to build institutions, only gradually. Rodrick argues that one reason why gradualism prevails over the blue print approach is that much of the technology is tacit and therefore not available in black and white. This makes the blue print highly incomplete and of little use to the importers. However, Rodrick feels that imported blue prints can prove useful for some narrowly defined technical issues, but large scale institutional development, by and large, calls for discovering local needs and developing rules that serve such needs.

2.3.1. Lesson from Dani Rodrick

Rodrik's emphasis upon gradualism is akin to North's path dependence. Secondly, Rodrik's view, that imported blue prints have limited usefulness and that for large scale institutional change to happen, local needs must first be discovered, tells us that foreign consultants charged with suggesting reform of local institutions may not be ideally suited to do the task.

3. INSTITUTIONAL CHANGE: HISTORICAL EXPERIENCES (10)

3.1. 17th Century Britain and Netherlands versus Spain and France

The institutions in Britain and Netherlands on the one hand, and Spain and France on the other hand took divergent paths in late 17th century--while Britain and Netherlands moved towards institutions that promoted commercial activity, Spain and France moved towards extractive institutions. Acemoglu, et al. (2005a) argues that whether or not the institutional change occurred depended upon how powerful the groups demanding institutional change were?

The rise of the constitutional monarchy in Europe is instructive. The following scene prevailed in the early sixteenth century UK. From 1603 onwards, England was ruled by Stuarts who continuously had revenue problems. To generate revenue the Crown sold lands, extended monopoly rights, seized private property and defaulted on loan repayments. The Parliament, though in existence, enjoyed little say in affairs of the country and the Crown could dissolve the assembly even upon minor differences with the Parliament. Supreme judicial power rested with the Star Chamber, which held legislative powers too, and primarily represented the Crown's interests. This was Britain prior to the Civil War of 1646. The Civil War and then the Glorious Revolution of 1688 led to sweeping changes in institutions; the Star Chamber was abolished, restrictions were placed on monopolies, cases involving property were to be tried under Common Law and the Parliament was to have regular standings. The Parliament gained a central role in financial matters with exclusive powers to raise taxes. This also gave more security to property rights of all and sundry, especially to the rights of those with financial and commercial interest. In sum, UK was transformed into a parliamentary monarchy with powers of the Crown significantly trimmed. The question that begs the answer is, how could the commercial interest become so strong in Britain. Acemoglu, et al. (2005a) argues that the Lords had gained a stronger position during the 14th and 15th century and were able to force the creation of the Parliament, to put limits to the authority of the Crown (but certainly not to protect the commercial interests). The Lords forced the Crown to 'live on his own' with strict restrictions on expanding his revenues. Perhaps these restrictions later on enabled the commercial interests to become stronger and demand more rights.

The 16th century Netherlands was the most important commercial area of Europe. The powerful groups in the country were for encouragement to commercial activity and enforcement of property rights. Netherlands, being under Spanish control then, provided substantial revenue to the Spanish Crown. Economic development in Netherlands threatened the interest of Spain. The towns in Netherlands, under the leadership of William of Orange, rebelled against Spain, leading to Dutch independence in the 16th century. What is important is the fact that the merchants of Netherlands wholeheartedly financed the rebellion.

An explanation put forth by Acemoglu, Johnson, and Robinson (2005b) for the transformation of Britain and Netherlands is that in the 16th century the opportunities generated by 'Atlantic trade' had increased the wealth and therefore the political power of the commercial interests. This enabled them to demand and obtain more rights.

This brings us to the question as to why, out of the countries involved in 'Atlantic trade' only the commercial interests in Britain and Netherlands were able to enrich themselves from the opportunities generated by the trade, while the commercial interests in France and Spain could not exploit such opportunities. Acemoglu, et al. (2005b) provides the answer. The authors explain that in England and Netherlands the trade was mostly carried out by individuals and partnerships, while in France and Spain, trade was primarily under the control of the Crown. The differences in organisation of trade in turn reflected the different political institutions of these countries. Grant of trade monopolies used to be an important source of fiscal revenues for the Crown; the more powerful monarchs could increase their revenues by granting trade monopolies or by directly controlling trade while for weaker monarchs this was a luxury they could not afford. At the beginning of the fifteenth century, the Crown was much stronger in France and Spain, than in Britain and Netherlands, and this was the most important factor in the difference in organisation of trade in these countries. Consequently, in England and Netherlands, and not in France and Spain, a new class of merchants arose with interests directly opposed to the interests of the Crown. The new class of merchants later on played an important role, as described earlier, in subsequent political changes.

3.1.2. Lessons from the Institutional Evolution in Britain and Netherlands versus Institutional Evolution in Spain and France

Two lessons are apparent from the historical comparison of Britain and Netherlands with France and Spain. One, strong commercial interests hold the potential to emerge as de facto power that may successfully challenge the de jure power if the latter fails to provide the institutions that commercial interests require. We learn from the European history that more often than not, the de facto power that emerged in the form of commercial interest had to force a change in the de jure power to acquire the institutions of its choice.

The second lesson is that the fiscal constraints may force the authorities to strike a bargain with the citizens with the effect that the public provides for the fiscal needs of the government which in turn provides good institutions, the institutions that the public prefers. This incidentally is the thesis of Moore (2002) who argues that nations that enjoy recourse to unearned income (i.e. income from natural resources and foreign aid) typically have to put up with poor institutions while the countries that rely mostly on earned income (from taxation) have relatively good institutions. To account for the difference, Moore argues that to induce the citizens to pay taxes the authorities have to provide them with good institutions and the citizens view taxes as the cost of such institutions. However, since the rulers of the nations with unearned income do not have to lean on citizens for revenues, therefore, they are not constrained to provide good institutions.

3.2. 19th Century Britain and Germany vs. Austria-Hungary and Russia

During the 19th century, Britain and Germany went through rapid industrialisation in contrast to the industrialisation process in Austria-Hungary and Russia. To account for the difference, Acemoglu (Lecture Notes, p. 200) argues that the elites in Britain had relatively more to gain from industrialisation than those in Austria-Hungary and Russia. Besides, while the landed aristocracy in Britain enjoyed relatively secure position and was less threatened by the process of industrialisation, the aristocracy in Austria-Hungary and Russia stood to lose more rents if they lost political power.

The lesson from the above is all too familiar--the rent-seekers will thwart institutional change with success depending upon the bargaining strength that they enjoy.

3.3. North vs. South America in the 18th and 19th Century

In the 18th century, some Caribbean and Latin American countries were richer than North America. However, while North America industrialised rapidly in the 19th century, the Caribbean Islands and much of South America stagnated during the period. Acemoglu (lecture notes) argues that the powerful groups in North America generally favoured policies that encouraged commercial interests and industrialisation, while in the Caribbean and South America the groups in power opposed industrialisation.

To account for the difference in institutional and economic development of North and South America, Engerman and Sokoloff (2005) argue that it is the initial conditions or endowments of a country that play a fundamental role in determining the long run paths of development of a country. The basic import of their thesis is that the colonies in the Americas that began with extreme inequality and population heterogeneity, developed institutions that restricted access to economic opportunities and contributed to lower rates of public investment in schools and infrastructure, thus beginning a vicious circle of underdevelopment. The authors argue that the climate and the soil of the colonies like Brazil and the Caribbean were suitable for growing cash crops like sugarcane. These crops enjoyed large scale economies and were most efficiently grown using slave labour. The colonial masters in these countries imported slave labour from the international market for slaves and thus the population of these countries came to be dominated by slave labour. This led to highly unequal distribution of wealth, human capital and political power. South America was attractive for European colonisers because of the potential huge return that use of slave labour afforded. On the other hand, the areas that now constitute North America and Canada, were not very attractive to the Europeans when they began to colonise the New World (Americas). This was because the climate and soil of the areas was suitable only for the production of grains and livestock that involved small scale economies and used few slaves. When the opportunities in the South were close to exhaustion, the Europeans began to colonise North America and Canada. Since the land was abundant and labour scanty, the colonisers offered various incentives to encourage the migration of European citizens to the United States and Canada. Engerman and Sokoloff identify three historical institutions that were designed to attract European settlers to the areas. These included adult male franchise, schooling and ownership of land. Accordingly much greater percentage of the rural population in United States and Canada owned the land that they cultivated and landholdings were typically smaller. Similarly a far greater percentage of the population enjoyed access to schooling in the United States and Canada than in Latin America. Thus, argue the authors, the institutions that promote growth were necessitated by the homogeneous character of the population in the United States and Canada.

To further argue for belief-homogeneity as a facilitator of institutional change, we refer to Collier (2007). The author stresses that even autocracies are less stable in ethnically diverse societies. The reason is that, in ethnically diverse societies only one of the many groups will be aligned to the autocrat. Given the narrow support base of the autocrat, his support group can engage in rent-seeking or this may induce the autocrat to dole out favours to the opposition. All this constrains institutional change.

The comparison reveals that initial endowments of a country or region influence institutional evolution. The societies that begin with extreme inequality and population heterogeneity tend to have institutions that restrict access to economic opportunities, while the societies with relative equality and population homogeneity are more likely to facilitate the evolution of growth enhancing institutions.

The comparison of North and South America also highlights the primacy of economic interests and also, as to who enjoys power--those with interest in rent-seeking or those with interest in secure property rights. If the de jure power is with the aristocracy, it will not establish good institutions on its own. The good institutions must be forced from the de jure power by some group deriving de .'facto power from one or the other source. The analysis of the consequences of colonisation of North America and South America also confirms Olson's (2000) 'Roving and Stationary Bandits' thesis: when a bandit (the ruler) is out there for a short time, he attempts to extract all that he can (and therefore establishes institutions with the extraction objective in mind); whereas if the bandit is in there to settle down, he extracts only part of the income of his subjects--the intact earning capacity of the subjects allows the stationary bandit a steady stream of extraction, now as well as in the future.

3.4. Korea and Taiwan vs. Congo (Zaire)

In South Korea and Taiwan the leaders pursued developmental policies while General Mobutu made Congo the most kleptocratic regime. Acemoglu explains the reasons for the difference in choice of the rulers. He believes that the explanation lies in 'constraints'--while Mobutu faced little constraints either from its neighbours or from the existing institutions, South Korea and Taiwan faced severe threats of communism via a revolution or invasion.

It was the threat of communist revolution from outside as well as inside that forced General Park Chung Hee in South Korea and Kuomintang regime, led by Chiang Kaishek, in Taiwan to pursue developmental policies. Acemoglu believes that the primary motivation for investment in education and the institution of land reforms in Korea was the containment of unrest. The Kuomintang regime, the rulers of China before the revolution, despite having a history of being corrupt, predatory and rent seekers, were also forced to pursue the industrialisation path to avert the threat of communism in their new shelter--Taiwan.

The situation in Congo was very different from Taiwan. In Congo, General Mobutu, the then Army chief, took over power shortly after independence. Mobutu dismantled the judiciary, removed the already weak institutional constraint, bought political support using state resources and proceeded to accumulate wealth. There were effectively no property rights and the GDP of Congo declined at the rate of 2 percent a year. How could Mobutu get away with this'? To ward off any threat to his rule, Mobutu bought off political support using money provided by US, IMF and World Bank as developmental aid which, in fact, were payments to Mobutu to keep Congo noncommunist.

The lesson from the experience of Korea and Taiwan is that the threat of a revolution may force the authorities to reform. Especially, the threat of an ideological change may induce the authorities to practise the ideology in vogue with more vigour thereby reforming the institutions as a consequence. While the lesson from the experience of Congo is that if the stronger world powers have common interest with the rent seekers, it would constrain institutional development. Put differently, this may also imply that if the world powers have some strategic interests in a country, then it might be easier for them to deal with a single person rather than a democratic regime. That single person, drawing legitimacy from foreign powers rather than the citizens of the country, will not be too bothered to facilitate growth-enhancing institutional change.

4. ENLIGHTENMENT ERA

The 17th century is generally referred to as the European enlightenment era. In the context of institution-building the single most important contribution of the enlightenment thought is its successful attack on absolute monarchy. The thoughts of the enlightenment philosophers seem to have influenced institutional change in a number of countries, especially the framing of the constitution in the US seems to have benefited from the teachings of enlightenment philosophers like Hobbes, Montesquieu and Locke. A brief on the thoughts of the enlightenment philosophers is presented in Box 1.

Box 1

The Enlightenment Era

Hobbes (1588-1679) was probably the first to argue that monarchs
ruled not by the consent of Heaven, but by the consent of the
people. Hobbes held that all human beings, being selfish will fight
for resources. Therefore to protect individuals from each other,
humanity at some early point agreed to a 'social contract' that
specified the rules, individuals would live by. Hobbes reasoned
that as human beings cannot live by their agreements, therefore
authority was created to enforce the terms of the 'social
contract'. By authority Hobbes meant 'monarchy'. For Hobbes,
'humanity is better off living under the circumscribed freedoms of
a monarchy rather than the violent anarchy of a completely equal
and free life'. However, later on, in a twist of fate, his methods
of inquiry as well as his basic assumptions became the basis, for
arguments against absolute monarchy. Marquis de Montesquieu
(1688-1755), a judicial official as well as a titled nobleman, was
amongst the earliest critics of absolute monarchy. Montesquieu's
classic, The Spirit of Laws (1748) recognises geographic influences
on political systems, advocates checks and balances in government
and defends liberty against tyranny in an uncompromising manner.
Baruch Spinoza (1632-1677) held the view that human beings'
inability to preserve themselves forced them to form societies. In
doing so, the individuals surrendered their 'individual right' to
'common right'--a notion very similar to Hobbes' 'social contract'.
Spinoza held that an inverse relationship existed between the power
of an individual and the power of the State. Given this view
Spinoza argues for democracy to create a balance of power between
the state and the 'individual'. John Locke (1632-1704) views human
mind as completely empirical, rather, he argues that the only
knowledge is empirical knowledge. He also held that human mind at
birth is a tabula rasa (erased board). His empiricism coupled with
the notion of tabula rasa meant that moral as well intellectual
outcomes in human development can be altered to societal advantage
by changing the environment through education. Locke proposed an
extension of education to every member of society. His view of
education dominates the western culture even to this day. Voltaire
(1694-1778) popularised Newtonian science, fought for freedom of
the press, and actively crusaded against the Church. In his
endeavours he turned out hundreds of plays, pamphlets, essays and
novels. He wrote around 10,000 letters to different people in
advocacy of his convictions. Even in his own time, he enjoyed the
reputation of a legend, among kings as well as literate commoners.

5. FROM WHERE TO BEGIN?

The discussion in the foregoing sections was meant to draw lessons for our main task, from where to begin the process of institutional reform? Before we present our own arguments for Pakistan, it will prove useful to briefly recap the lessons that we have learnt from historical experiences, of different countries, discussed in Section 3.

Countries that experienced           Lessons from Historical
institutional change versus          Experiences
countries that (with similar
circumstances) that did not
experience institutional change

17th century Britain and             Fiscal constraints and then
Netherlands versus Spain and         commercial interests forced the
France                               crown in LJK to yield good
                                     institutions. Similarly in
                                     Netherlands commercial interests
                                     emerged as the de facto power
                                     that forced the change upon the
                                     rulers.

19th century Britain and Germany     Rent-seekers will thwart
versus Austria-Hungary and Russia    institutional change, with
                                     success depending upon the
                                     bargaining strength that they
                                     enjoy.

18th and 19th century North          1. Institutions are a function of
America versus South America         initial endowments of a nation.
                                     Extreme initial inequality and
                                     populations heterogeneity leads
                                     to development of institutions
                                     that restrict opportunities for
                                     the poor and thus constrain
                                     growth.

                                     2. Economic interests enjoy
                                     primacy, if the de jure power is
                                     with the elites; it will not
                                     reform on its own. Some de facto
                                     power must emerge to force change
                                     onto de jure power

Korea and Taiwan versus Congo        1. Lesson from the experience of
(Zaire)                              Korea and Taiwan: the threat of a
                                     revolution may force the
                                     authorities to reform.

                                     2. Lesson from the experience of
                                     Congo: If the world powers have
                                     some strategic interests in a
                                     country then it would be easier
                                     for them to buy-off/install some
                                     rent-seeking rulers in the
                                     country concerned rather than a
                                     democratic set up. This may
                                     constrain institutional
                                     development in the country
                                     concerned.

5. 1. Can Commercial Interests in Pakistan Force the Dejure Power to Change Institutions for the Better?

If the institutions are poor and the de jure power is not willing to reform institutions on its own or has been held hostage by some de facto power that stands to gain from maintaining the status quo, then some other de facto power must emerge that can force change upon the incumbent de jure power. This is what we learned from Acemoglu, et al. (2005a). The experience of institutional change in 17th century UK and the Netherlands as well as the significant difference between the institutional evolution of North and South America in the 18th and 19th century corroborates this stance.

How will the de facto power that may force the de jure power to enact institutional change emerge? This is the issue. Can the conmaercial interests, in Pakistan, emerge as the said de facto power'? (As has happened in the 17th century UK). This is unlikely because a rent-seeking culture has characterised the economy through much of its history. For example, in the 1950s, the trade policy relying on high tariffs and quantitative restrictions conferred windfall gains on a small group of import licensees [Hussain (1999)], while in the 1960s, the import substituting industrialisation and the export bonus scheme allowed the exporters to amass wealth at the expense of other segments of the society. (11) In the 1980s and 1990s, the bureaucratic and the political elite, and those who could afford to buy-off bank officials, benefited from bank loans that in essence were mostly not repaid. Given that large commercial interests in Pakistan, have prospered by way of rent seeking (and are used to securing favours from the de jure power), it is difficult for such interests to stand up against the de jure power to reform institutions. After all one does not bite one's own hand.

5.2. Can Fiscal Constraints Force the de jure Power to Strike a Bargain with the Citizens for Taxation in Exchange for Good Institutions?

Pakistan has faced fiscal constraints in the past and the situation is no different today. Will the fiscal constraints force the de jure power, as these had forced the de jure powers of UK and Netherlands, in the 17th century, to strike an implicit bargain with the citizens--taxation revenues in exchange for good institutions? Again this is unlikely. The times when the fiscal constraints could force the de jure power to strike a bargain with the citizens was when access to funds was neither available through borrowing from the country's central bank (money creation) nor through foreign aid. Now the instrument of money creation has enabled the governments to delay the day of reckoning till the people burdened with inflation decide to revolt against the government (which does not happen too often). Second, Pakistan because of its geo-strategic position, had enjoyed access to sufficient foreign aid for the better part of its history. Given the present geo-political environment, the trend is likely to continue--foreign aid will alleviate the fiscal constraint and the de jure power will not be too pushed for taxation revenues. The implicit bargain i.e., taxation revenue in exchange for good institutions will not materialise.

5.3. Strategic Interests of Foreign Powers: A Constraint to Institutional Development

Will the Congo-like situation prevail in Pakistan, that is, will the strategic interests of foreign powers constrain institutional development in Pakistan? In fact, a Congo-like situation has prevailed in Pakistan for the better part of its history. It goes without saying that foreign powers, especially the United States, do have strategic interests in Pakistan and the population of Pakistan in general does not feel pressed to pursue the strategic interests of foreign powers. Therefore it is in the interest of the foreign powers to buy off and even install an autocrat or at best a sham democracy, and ensure continuity of such rule. During its history of 61 years, Pakistan has witnessed four military regimes. Three of the four military rulers ruled for almost a decade each, with implicit or explicit support of the United States. US' support to the military regimes in Pakistan, despite its avowed criticism of dictatorship, bears testimony to the 'buy-off and rule' strategy. The regimes that derive legitimacy from foreign powers rather than from the natives is not pushed to pursue institutional reforms, especially when it means shooting one's own self in the foot, e.g., judicial independence.

5.4. Will a Revolution Bring about Institutional Change in Pakistan?

Revolutions are not spontaneous. All revolutions have their thinkers whose thoughts ignite the revolutions [Masood (1991)]. The monarchy in Europe did not collapse overnight. Around the time of the Glorious Revolution (1688) in UK and much before the French revolution, the enlightenment philosophers, like Montesquie, Spinoza and Voltaire had launched a strong attack against monarchy with their pen and voice. The thoughts of people like Allama Shariati and Ayatollah Mutahiri had provided the fodder for the Iranian revolution of 1979 [Masood (1991)]. To stage a revolution that ends up in long-lasting institutional change rather than chaos, not only the society should have developed sufficient apathy with the present rule but it should also have at least some idea of how to proceed after the revolution. Above all, if the human capital required for carrying out the institutional change is not available, even a revolution may fizzle-out or turn into chaos. To conclude, to stage a successful revolution the belief system of the society must be sufficiently influenced so that the society can perceive what wrong is being afflicted upon it and how it can remedy the situation. The question is, how can the beliefs be influenced?

We have shown that fiscal constraints and commercial interests may not prove very effective in securing an institutional change in Pakistan. Besides, given the strategic interest of foreign powers in Pakistan, the possibility of foreign powers thwarting an institutional change cannot be ruled out, if the change is likely to compromise their interests. We have also discussed that given the obtaining intellectual thought process and the state of the human capital, the society in Pakistan may not be ready as yet to stage a revolution that ends up in a meaningful institutional change. How to go about institutional change then? The option that remains is the gradualist approach, strongly advocated by Douglas North and implicitly evident in the works of Darron Acemoglu and Dani Rodrik, to name a few.

5.5. The Gradual Approach

One of the key elements of North's theory of institutional change is path dependence exhibited by institutions. This implies that a quick-fix solution to poor institutions is not possible. We want to re-emphasise here that revolutions that appear to have reformed institutions with a big-bang, were rooted in the thought process that in some cases had begun almost a century before the revolution actually materialised, for example the influence of 17th century enlightenment thought upon the French revolution and the framing of the US constitution. Other key elements of North's theory that (i) institutions are influenced by beliefs, (ii) that dominant beliefs matter, (iii) the role of intentionality, and (iv) comprehension of the dominant players, provide hope that institutional change can be designed, but only with the process of change extending over sufficient length of time. We consider below, whether North's theory of institutional change can be put to practice by way of reform of the educational system in Pakistan.

Institutions that get established, according to North, are a function of beliefs of the society. To design an institutional change, the task then is to influence the societal beliefs. The belief formation, we have learned is a function of genetics, culture and human environment. To recap, beliefs can be influenced only to the extent that today's learning and experiences influence the culture and human environment. Thus the beliefs that are conducive to desired institutional change can be developed, by providing to the citizens, the education and human environment which is conducive to the preferred institutional change.

But it is not just the individual beliefs that matter; rather it is the beliefs of the society that count. This means that in a more homogeneous society the task of securing an institutional change would be relatively less difficult. Here we need to recall the Engerman and Sokoloff (2005) thesis, reviewed in section 3.3, that the homogeneous societies that the US and Canada had facilitated the development of pro-growth institutions in these countries while in South America the presence of heterogeneous societies furthered the development of such institutions that constrained opportunities for the poor and hindered economic growth in consequence. The task then is to forge greater homogeneity in society, which is secured by forging convergence in beliefs amongst the individuals and the various sections of a society.

The question is how to forge belief-convergence? Beliefs, being a function of learning and human environment, the answer lies in providing a uniform learning system and environment for the whole society. How to do that? The answer is: a uniform and universal education system for all during the formative years of human life i.e., childhood and adolescence. To design an institutional change then, the first and foremost requirement is to have a single system of education for all segments of the society, up to a certain minimum level, say till, Grade 12. By a single system, we mean that not only the curricula should be the same, but the environment in schools and colleges should also be more or less similar. Two students reading the same material and sharing the same environment are more likely to have the same beliefs as well. Individuals of a nation, who acquired the same education and have experienced similar environment at schools, are more likely to forge a homogeneous society--a pre-requisite for developing growth promoting institutions.

Is Pakistan's prevailing educational system capable of facilitating the development of a homogeneous society? To answer this question, we examine below the educational structure in Pakistan.

At the school/college level, Pakistan follows four different regimes that include: (i) the O/A level Cambridge system: the schools and colleges that use this system follow the curricula prescribed by the authority which manages the O/A level system in UK, (ii) the English-medium private and public schools which follow the curricula prescribed by the government, (iii) the Urdu-medium government school system, that also follows the syllabi prescribed by the government, but the courses here are taught in the national language--Urdu, and (iv) the madressah system. The curricula of the madressah system are primarily focused on religious education and little effort is made to impart knowledge of secular subjects like science, mathematics and the arts. Besides, the medium of instruction is mostly the national language, Urdu. All the state-owned schools, that offer education at a negligible fee primarily serve the poor and invariably use urdu as the medium of instruction. It is note worthy here that higher education (i.e beyond grade 12) is offered mostly in English language which is also the working language in offices, whether in public or private sector.

Thus the poor, having gone to Urdu medium, state-owned schools are at a disadvantage; their education makes them unfit for the job market. Government's education policy-2009 acknowledges that white collar jobs seem to be reserved for the graduates of English medium schools. This, coupled with the fact that the majority still goes to government Urdu medium schools, is bound to perpetuate inequality which in turn facilitates the development of institutions that ensure elite dominance and constrains economic opportunities for the poor. It is also obvious that the population which is the product of diverse educational systems, like that of Pakistan, is likely to develop beliefs that stand apart* The society will be heterogeneous, rather than homogeneous required for institutional change. To visualise how heterogeneity may constrain institutional change, assume that all MPs in the national parliament are educated till say grade 14, with the 50 percent of the MPs coming from the Cambridge (O/A level) system and the remaining 50 percent from madressahs. Will the majority of MPs in this kind of educated parliament, share views on many issues'? It is no coincidence that a more or less similar education for all, up to a certain grade, by and large, is the norm ill the developed world that boasts of good institutions. To make our case for common educational system stronger, we again lean on North (2005):

"The process of learning is unique to each individual but a common institutional/educational structure will result in shared beliefs and perceptions".

Our case for reform of the educational system also finds support in studies like Rajan (2006) and Azfar (2006). Rajan argues that strengthening the institutions like property rights etc. may help jump-start the economy for a while but the lack of endowments, like education, will leave the poor unprepared for reforms. He cautions that in this situation placement of pro-market institutions may fail to do the trick. Azfar (2006) argues that the shared belief system, which a universal educational system shall produce, will help bring about a consensus among the population, about the acceptable and unacceptable behaviour of the rulers and will therefore force-in an honest government.

The hardest to reform amongst the educational systems being practised in Pakistan is the madressah system. Madressahs are believed to inculcate the so-called orthodox beliefs in pupils, (the perception may or may not be true) and therefore attempts have been made, under foreign pressures, to reform the system. Such attempts have not borne fruit. The reason is that the objective has been to find a quick-fix solution. Unfortunately such a solution does not exist--the clergy that enjoys enough de facto power is not willing to yield. To address the issue one has to account for who goes to a madressah and why. Is the enrolment there by choice or is forced by circumstances. The madressahs in Pakistan, not only impart religious education, free of cost, but also offer food and shelter to the pupils. (The madressah system has been termed as the biggest NGO in Pakistan). The madressahs are apparently funded by charity money. Anecdotal evidence suggests that mostly the wards of the poor are enrolled there. For the poorest of the poor, this is the easiest way to feed their children. In Pakistan, with around 30 percent of the population living below the poverty line, the enrolment on this count is not likely to be small. So the solution lies in addressing the overall issue of poverty, which in any case is not an easy one to tackle, before a number of institutions have been reformed. An alternate is to enforce compulsory enrolment in the formal school system, other than the madressahs. This again involves the cost of enforcement, compensating the parents for whom the nonschool going child is a bread-earner and of course tackling the opposition from the clergy.

The purpose of the foregoing discussion is not to offer a solution but only to provide a glimpse of the hurdles involved, when one attempts to reform the educational system.

The proposed common educational system also takes care of the next element in North's theory: dominant beliefs matter. If all the subjects of a country have gone through the same kind of education and have faced more or less similar human environment, at least at schools and colleges, then belief-convergence between dominant and non-dominant players is likely. Still the beliefs of the dominant players would matter but given convergence, the preferences of the non-dominant players would be automatically, taken care of.

North's argument that comprehension of an issue of the dominant players determines the kind of institutions that will be developed to confront the issue, again provides room for the education to influence an institutional change, because it is the education, and of course the right kind of education, that would influence a person's ability to correctly comprehend the issue at hand.

That the dominant beliefs matter and that the intentionality of the dominant players matters as well calls for choosing such people (through electoral process etc.) to hold de jure power, who share the beliefs of the society and who intend to allow the kind of institutions that the society prefers. The beliefs and intentions of the candidates aspiring for the de jure power can be tracked from a run down of the personal profile of the aspirants. For example, if the candidate or a political party is running for a second term, the performance in the previous term serves as a guide to judge the beliefs and intentionality of the players. However, for the constituents to correctly perceive the beliefs and intentions of the players, they must possess some education, whether formal or informal.

But this is a truth-judgment kind of a thing to say that to reform institutions to begin with, the educational system should be reformed. The issue is who would bell the cat? The natural candidate, in this context, is the de jure power. But the question is what motivates the de jure power to do this. The reform of the educational system, we expect, would reduce the voters' ignorance and thereby lead to all-round institutional reform, including the change in the very structure of the de jure power or the change in the de jure power itself. Given the damage that the reform of the educational system can inflict upon the rulers, why would the de jure power shoot at its own foot'? So, it is difficult to believe, if not naive, that the de jure power will undertake the reform of the educational system on its own.

To reform, the pre-requisite implicit in Acemoglu, et al. (2005b) is that some de facto power must force the de jure power, to reform institutions. This begs the question how such a de facto power will emerge. What incentive mechanism will facilitate the emergence of a de facto or de jure power, that may push for the reform of the educational system? This is a difficult question to answer; the popular print and electronic media may create awareness about the need for a common educational system. But the question then is; what motivates the media to do this'?

We have groped in the dark, perhaps without success, to find out as to what, and who would trigger the reform of the educational system. However one thing is for sure. Reform of the educational system would meet lesser resistance as compared to reform of the other institutions. For example, an attempt to begin the process of institutional reform with the change in the structure of de jure power or the change in the de jure power itself will, in all likelihood, be resisted tooth and nail, by those who currently wield de jure power. Moreover if the change in de jure power is likely to adversely influence interests of strong foreign powers, then securing a change would become all the more difficult. The support extended to the kleptocratic regime of General Mobutu in Congo to thwart communism is just one example of how and why foreign powers may support a corrupt regime rather than encourage growth-conducive institutional change. The case of Pakistan is no different. To further their own strategic ends, the foreign powers, that matter, have comfortably co-existed with at least three military regimes, in Pakistan. Similarly an attempt to establish institutions that do not allow rent-seeking, again may not be successful if the de jure power itself is deriving rents. It is noteworthy that all the reforms referred to above will adversely influence the de jure power today.

In contrast, given path dependence, the reform of the educational system will, at best, influence the de jure power a generation hence. Typically, politicians being myopic, with the vision extending only up to the next elections, are not likely to be as scary of the reforms in educational system than they would be of the change in de jure power today, or the reform of any other institution that adversely influences their fortunes in the near-term.

Therefore the educational system, with its all-encompassing influence, global emphasis and relatively lesser resistance from the de jure power stands as the best candidate to begin the process of institutional reforms. The reform of the system and the increase in literacy rate will in all likelihood lessen, if not altogether eliminate, voters' ignorance and misperceptions while voting. This would raise the possibility of choosing the right kind of people to hold de jure power. Secondly given the voters' improved ability to choose, they are more likely to choose the ones who share their belief system. Thirdly, the rulers, having passed through the same educational system as available to the subjects, are more likely to carry the same beliefs as held by the subjects. It is the shared belief system that will facilitate reform of the remaining institutions.

To understand why educational reforms should enjoy primacy over other reforms, let us look at the case of United States--one of the countries that can boast of good institutions today. Perhaps United States had the highest literacy rate in the world at the beginning of 19th century. The common school movement that began in 1820 did such good that by the middle of 19th century nearly 40 percent of the school age population had been enrolled and nearly 90 percent of the white adults were literate [Engerman and Sokoloff (2002)]. United States is perhaps one country where making of the constitution was debated by way of writing as many as 89 academic papers--now referred to as the 'Federalist Papers'. This was in 1870s. The relationship between literacy and institutional change, evident from the constitution making process of United States, is too apparent to be missed out.

A more recent evidence of the relations between literacy and reforms is furnished by Paul Collier in 'Bottoms Billions'. Collier (2007) argues that "countries need a critical mass of educated people in order to work out and implement a reform strategy" and substantiates it with the case of China and Tanzania. The author suggests that China and Tanzania both failed under Mao Ze Dong and Julius Nyerere respectively but given the critical mass of educated people, China was able to rethink its development strategy while Tanzania was not fortunate enough to have that critical mass.

We have determined that in a heterogeneous society, the elite-dominated de jure power will not facilitate the development of growth-promoting institutions. Despite some useful debate, we failed to conclude that how such a de facto power may emerge, that can force the de jure power to reform institutions. Given the inconclusive debate on the emergence of the requisite de facto power, the question arises, do we gain anything from the simple awareness that the educational system should be the first one to be reformed, if the society cannot force the de jure power to reform the system? The answer is, yes. Suppose that the discontent in a country has reached a point where for the rulers to remain in power they must agree to one or the other institutional change demanded by the society, otherwise they face the threat of a revolution. It is at this point that the society should have a clear idea as to what kind of institutional change to demand. If the de jure power is deriving rents from a host of avenues and the society demands an immediate end to rent-seeking then the probability of acceptance of the demand is rather low as this will affect the fortunes of the de jure power of today. But if the society demands that all children aged five should receive similar education, then the possibility exists that the ruler, being myopic and faced with discontent and threat of a revolution, even though weak, will yield. History stands witness that a couple of Pakistan's rulers, faced with public discontent had, in their twilight days, made an attempt to strike a bargain with the populists if not the public. It is at this moment in time that society should be aware of what it should demand? The demand should be: introduce a uniform educational system, for all and sundry.

6. CONCLUSION

We set ourselves the task of finding answers to two questions. One, is it possible to reform institutions by design and if yes which institution should be chosen to be the first one to be reformed. Given the path dependence exhibited by institutions, it is not possible to reform institutions with a big bang i.e. in one-go. This leaves us with the alternative of practising gradualism in reforming institutions--the alternative preferred by North, Acemoglu and Rodrik. Once we decide to adopt the gradual approach, the immediate issue that comes to the forefront is what to reform first? Hence our second question, i.e., from where to begin?

We excluded the possibility of commercial interests, fiscal constraints and a revolution forcing an institutional change in Pakistan. Commercial interests in Pakistan have typically thrived on favours from the de jure power and are therefore unlikely to emerge as a de facto power against its patron. Theoretically, fiscal constraints may encourage the government to strike a bargain with the citizens i.e., taxation revenue in return for good institutions. But in practice, the de jure power will enter into bargain only if funds from other sources are not available. Given the strategic interests of foreign powers, foreign aid will alleviate the fiscal constraint and the rulers-citizens bargain will fail to materialise. The country does not seem ready for a revolution either. The thought process that typically precedes revolutions seems to have barely begun. The alternative, that remains then is the gradualist approach preferred by North, Acemoglu and Rodrik.

Based on North's theory of institutional change we took the position that institutions can be reformed by conscious design. North holds that institutions are a function of the beliefs of the society and that beliefs among other things are a function of one's learning and experiences. He also holds the view that it is the beliefs of those in a position to enact institutional change that matter. Thus it is possible to mould one's beliefs by influencing what a person learns and what he experiences. Change in beliefs, would then induce an institutional change. The notion of the human mind, at the time of birth, being tabula rasa (erased board) put forward by the enlightenment philosopher, John Locke, also supports our stance that education can shape beliefs to suit one's end. Therefore we concluded that institutions can be reformed by conscious design.

The answer to our second, but the main question is that among the list of institutions that call for reform, the reform of educational system should top the agenda. Educational system as the top-most candidate for reform meets the three point criteria laid down in Section 1 of the paper. The current diverse educational systems serve to create and perpetuate inequality and population heterogeneity. We proposed the introduction of a common educational system, for all and sundry, up to a certain minimum level, say grade 12. The argument being that the introduction of a common educational system will reduce inequality and foster homogeneity in population, which in turn will facilitate development of growth-promoting institutions. In a relative sense, the resistance to reform of the educational system i.e. resistance to establishment of a common education system, up to grade 12, is likely to be lesser than a direct attack on rent-seeking of the de jure power. The former would affect those who wield de jure power, a generation-hence, while the latter will adversely influence them today. The politicians, being myopic, will opt for the former rather than the latter.

The educational system, as the top-most candidate for reform, also lives up to the second and third element of our criteria of choosing the first institution to be reformed. The common educational system will bring about convergence between the beliefs of the masses and those in position to enact institutional change. With the rulers and the subjects sharing beliefs, bringing about a change in the remaining institutions shall be less difficult. Thus the introduction of a common educational system will not only have an all-encompassing influence but the impact will be long-lasting as well.

We learned that a de facto power must emerge to force the de jure power to reform but failed to conclude how the said de facto power will emerge. Given that we do not know as to who will bell the cat, does it pay to be aware that the educational system should be the flag bearer of all institutional reforms. The answer is; yes it pays. Occasions do arise, when even a powerful dictator stands weakened and can see the eminent threat to his rule. It is at times like these that if the society has a clear idea of what kind of institutional change is necessary, it can make the ruler yield. The demand of the society at moments like these should be for establishing a common educational system, for all and sundry, up to grade 12.

To conclude, we want to re-emphasise that we do not expect the reform of the educational system to be impediment-free. One reason why we recommend educational system as the foremost candidate for institutional reform is that we expect such reforms to face relatively lesser resistance. While evaluating our recommendation, the relative nature of the word "lesser' should not be lost sight of.

We have identified 'commercial interest' and the 'need to generate revenue' as the inappropriate levers of institutional change in Pakistan. More such areas can be explored, e.g. social protection and greater federalism. We leave this for future works.

Authors' Note: We are grateful to Aurangzeb Hashmi, Lubna Hasan, Zafar Mahmood, Mahmood Khalid, Musleh ud Din, Rizwana Siddiqui, Usman Qadir, Nasir Iqbal and other participants of the seminars on this subject at PIDE and at a departmental meeting in PIDE, for offering very useful suggestions and comments. We are also thankful to the two anonymous referees whose valuable comments helped improve the paper. All errors and omissions are ours.

REFERENCES

Acemoglu, Daron, Simon Johnson, and James A. Robinson (2001) The Colonial Origins of Comparative Development: An Empirical Investigation. American Economic Review 91:5, 1369-1401.

Acemoglu, Daron, Simon Johnson, and James A. Robinson (2002a) Reversal of Fortune: Geography and Institutions in the Making of the Modern World Income Distribution. Quarterly Journal of Economics 118, 1231-1294.

Acemoglu, Daron, Simon Johnson, and James A. Robinson (2005a) Institutions as the Fundamental Cause of Long-run Growth. In P. Aghion and S. Durlauf (eds.) Handbook of Economic Growth. North Holland.

Acemoglu, Daron, Simon Johnson, and James A. Robinson (2005b) The Rise of Europe: Atlantic Trade, Institutional Change and Economic Growth. American Economic Review 95: June, 546-579.

Acemoglu, Daron, Simon Johnson, and James A. Robinson (2003) Lecture Notes for Political Economy of Institutions and Development, 14.773, http://web.mit.edu/14.773/www/2003%20Class%20notes.pdf

Ali, Imran (1988) The Punjab Under Imperialism. 1885-1947. Princeton: Princeton University Press.

Amjad, Rashid (1982) Private Industrial Investment in Pakistan 1960-1970. Cambridge: Cambridge University Press.

Azfar, Omer (2006) The New Institutional Economics Approach to Economic Development: A Discussion of Social, Political, Legal and Economic Institutions. The Pakistan Development Review 45:4, 965-80.

Banerjee, Abhijit and gakshmi Iyer (2005) History, Institutions and Economic Performance: The Legacy of Colonial Land Tenure Systems in India. American Economic Review 95: 4, 1190-1213.

Collier, Paul, (2007) The Bottom Billion." Why the Poorest Countries are Failing and What Can Be Done About It. Oxford: Oxford University Press.

Dollar, David and Aart Kraay (2003) Institutions, Trade and Growth. Journal of Monetary Economics 50:1, 133-62.

Easterly, W. and R. Levine (2001) What have we Learned from a Decade of Empirical Research on Growth? It's Not Factor Accumulation: Stylised Facts and Growth Models. World Bank Economic Review 15, 177-219.

Engerman, S. and K. Sokoloff (2005) Colonialism, Inequality, and Long Run Paths of Development. (NBER Working Paper 11057).

Engerman, S. and K. Sokoloff (2002) Factor Endowments, Inequality, and Paths of Development among New World Economies. (NBER Working Papers 9259).

Gazdar, H. (2004) The Fourth Round, and Why They Fight on: An Essay on the History of Land and Reform in Pakistan. Available at: http://www.panossouthasia.org/PDF/ Politics%20of%20Land%20Reform%20in%20Pakistan.pdf

Hall, Robert E. and Charles Jones (1999) Why Do Some Countries Produce So Much More, Hover Institution, elsa.berkeley.edu/~chad/pon400.

Hasan, Lubna (2007) Myths and Realities of Long-run Development: A Look at Deeper Determinants. The Pakistan Development Review 46:1, 19-44.

Haque, Nadeem U1 and Idrees Khawaja (2007) Public Service: Through the Eyes of Civil Servants. Governance Series. Islamabad: Pakistan Institute of Development Economics.

Hayek, F. A. (1960) The Constitution of Liberty. Chicago: University of Chicago Press.

Hussain, Ishrat (1999) Pakistan the Economy of an Elitist State. Karachi: Oxford University Press.

Lewis, Hackett (1992) The Age of Enlightenment: The European Dream of Progress and Enlightenment. history-world.org/age_of_enlightenment.htm

Lichbach, Mark I. (1995) The Rebel's Dilemma. Ann Arbor: University of Michigan Press.

Locke, John (n.d.) The European Enlightenment: Seventeenth Century Enlightenment Thought. http://www.wsu.edu:8001/~dee/ENLIGHT/ Masood, Mukhtar (1991) Loh-e-Aayam. Lahore: Sang-e-meel Publication.

Montesquieu, Baron de (1748) Spirit of the Laws. http:// plato.stanford.edu/entries/Montesquieu.

Moore, M. (2002) 'Bad Governance' or the Development of Political Underdevelopment. Paper presented at the conference 'Towards a new Political Economy of Development: Globalisation and Governance' at the Political Economy Research Centre, University of Sheffield.

North, Douglass C. (1981) Structure and Change in Economic History. New York: Norton.

North, Douglass C. (1990) Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.

North, Douglass C. (2005) Understanding the Process of Economic Change. Princeton: Princeton University Press.

Nunn, Nathan (2009) The Importance of History for Economic Development. NBER Working Paper 14899).

Olson, Mancur (2000) Power and Prosperity: Outgrowing Communist and Capitalist Dictatorships. New York: Basic Books.

Pejovich, Svetozar (2006) The Effects of Interaction of Formal and Informal Institutions on Social Stability and Economic Development. In Kartik C. Roy and Jorn Sideras (eds.) Institutions, Globalisation and Empowerment. Elgar, UK.

Rajan, Raghuram G. (2006) Competitive Rent Preservation, Reform Paralysis, and the Persistence of Underdevelopment. (NBER Working Paper 12093).

Rodrik, Dani, A. Subramanian, and F. Trebbi (2002) Institutions Rule: The Primacy of Institutions over Geography and Integration in Economic Development. (NBER Working Paper 9305).

Rodrik, Dani (2006) Institutions for High-Quality Growth: What They Are and How to Acquire Them. In Kartik C. Roy and Jorn Sideras (eds.) Institutions, Globalisation and Empowerment. Elgar, UK.

Ross, Jeffrey I. and Tedd R. Gurr (1989) Why Tenorism Subsides: A Comparative Study of Canada and the United States. Comparative Politics 21, 405-426.

Tarrow, Sidney (1991) Aiming at a Moving Target: Social Science and the Recent Rebellions in Eastern Europe. PS. Political Science and Politics 24, 12-20.

World Bank (2003) World Economic Outlook. Washington, DC.

Zaidi, S. Akbar (2005) Issues in Pakistan's Economy. Karachi: Oxford University Press.

(1) See Engerman and Sokoloff (2005).

(2) See North (1990), Acemoglu (2005a) and Rodrick (2006).

(3) See Nunn (2009) for an excellent survey of the subject.

(4) Enormous literature including, but not limited to, Hall and Jones (1999), Acemoglu, et al. (2001, 2002), Easterly and Levine (2001), Dollar and Kraay (2003), and Rodrik, et al. (2002) have shown that institutions matter in economic growth. For an exhaustive survey of literature on the relationship between institutions and economic growth, see Acemoglu, et al. (2005 a) and Hasan (2007).

(5) See Benergee and lyre (2005).

(6) Sindh and Punjab are the two provinces of Pakistan.

(7) Only newspaper reports are available as reference on the subject. The events are too recent to have found mention in journal papers and other reports.

(8) For exposition of North's theory of institutional change, we draw heavily upon North (1990, 2005).

(9) The Chief Justice was deposed for not yielding to the wishes of the former dictator and his restoration is now being popularly considered as a symbol of allowing the judiciary to function independently.

(10) For this section we draw upon Acemoglu, Lecture notes.

(11) For an exhaustive account of rent-seeking reading through the host of books written on Pakistan economy is essential. These include Zaidi (2005), Amjad (1982), Hussain (1999).

M. Idrees Khawaja <khawajami@yahoo.com> is Associate Professor at the Pakistan Institute of Development Economics, Islamabad and Sajawal Khan <lodhiphd@yahoo.com> is Junior Joint Director, State Bank of Pakistan, Karachi.