In Why Nations Fail (New York: Crown Business, 2012), Daron Acemoglu and James Robinson argue that sustained growth depends primarily on “inclusive institutions” that bring pluralism into government. It is these political and economic structures, more than resource endowments, knowledge, financing, foreign aid, or a variety of other factors, that the authors believe account for a country’s prosperity or poverty. This paper looks at Pakistan through their lens. There is good evidence that Pakistan lacks the kind of institutions the authors associate with lasting prosperity. However, Pakistan’s troubles are compounded by some factors that the authors do not take into account, such as stubborn insecurity and ethnic fragmentation. Encouraging entrepreneurship and expanding the middle class may be the most effective long term remedy.
At the heart of Acemoglu and Robinson’s theory is the distinction between “extractive” institutions, designed to extract wealth from a country and its people and concentrate it in the hands of a small group of elites, and “inclusive” ones, which bring into the economic and political governance of a country a variety of interests capable of balancing one another. The pluralism resulting from this competition of interests, they believe, will naturally allow a broader distribution of prosperity. They argue, based on examples from all over the world, that sustained economic growth is associated with inclusive institutions, both economic and political. Lasting growth also requires an environment that fosters innovation and accepts Schumpeterian “creative destruction” of enterprises whose technological time is past.
Periods of high growth, in the authors’ view, can occur where such institutions are lacking and pluralism is absent, such as the Soviet Union in its first half-century, provided the government, however autocratic it may be, allocates resources efficiently and enjoys sufficient “centralization.” They argue, however, that these conditions rarely encourage innovation or creative destruction, and such growth is unlikely to endure (again, the Soviet Union provides an example). The authors seem to use “centralization” to mean basic governmental authority. It does not seem to refer to the situation of countries with strong local sub-units (provinces, prior kingdoms that merged, and the like), a condition that clearly obtains in Pakistan.
Looking at Pakistan:
Pakistan is highly diverse. Punjab dominates its political life and, importantly, the army. It accounts for 57 percent of Pakistan’s people and economy. Sindh is home to about 22 percent of the population and some 27 percent of the economy, with the country’s major industries heavily concentrated in and around Karachi. Khyber Pakhtunkhwa (formerly Northwest Frontier Province) and Balochistan are much more sparsely populated; Balochistan is home to Pakistan’s natural resources, especially natural gas. All provinces except Punjab have spawned separatist movements. The Acemoglu/Robinson template does not address this kind of geographic disparity; we will discuss this along with some other weaknesses in their approach later on.
Politics: Several political parties have held or shared power in Pakistan. They generally do not, however, represent distinct economic or social bases, so their “inclusion” does not necessarily enhance economic pluralism.
Pakistan’s major political parties are largely family affairs. It is hard for new leaders to rise from the ranks unless they are designated tokens – in other words, the parties are more “extractive” than “inclusive.” The Pakistan People’s Party (PPP), which leads the current government, is built around the Bhutto family. It has some presence all over the country but is strongest in two very feudal areas: rural Sindh, home of the Bhuttos and the Zardaris, and southern Punjab, where the Bhuttos developed powerful alliances early on. The Pakistan Muslim League (Nawaz), powerful in Punjab but nowhere else, relies heavily on Nawaz Sharif and his family. The smaller parties that have filled out governing coalitions for both these parties, and the “rentable” political followers who make up what Pakistanis call the “King’s Party,” include notables with groups of followers, either based on rural social and land-use relationships, or deriving from the notables’ role in local shrines or other institutions.
The army, always a critical political actor, is the country’s one strong institution, heavily Punjabi, a “family” all its own. During its periods at the helm, it maintained alliances with feudal barons and industrial leaders – provided these allies recognized that they were junior partners. As retired army officers and the institutions that serve them have become more prominent in the economy, there is increasing evidence of “extractive” behavior, for example in the way land grants and cushy civilian jobs are used to reward retiring military officers.
Three political forces stand outside this model. The Muhajir Qaumi Movement (MQM), the party of those who came from what is now northern India, has built its Karachi powerhouse on ethnic solidarity and provision of urban services. The multi-ethnic Karachi enjoys a kind of pluralism, but its competitions tend to play out in deadly serious competitions over land and turf – not in electoral contests of the sort Acemoglu and Robinson describe from the industrial revolution in England. More fundamentally, the MQM has emerged in national politics as an ethnic spokesman rather than a representative of urban economic interests.
The largest of the Islamic parties, the Jamaat-e-Islami, is the most internally democratic of Pakistan’s political parties, a force on the streets of Punjab but less of a factor in parliament. And former cricket star Imran Khan plays to packed rallies in cities and towns around the country, with an astonishingly diverse array of followers. In the two elections he has contested, however, he has won a total of one seat. Inevitably in Pakistan, he is dogged by rumors that the army is making him into “their” candidate. He does not represent a particular economic or social “base.” We will see in the next election whether he can win a significant number of parliamentary seats, and if so whether his popular appeal, and his appeal to urban populations, can bring politically marginalized economic constituencies into the power structure.
Economics: Pakistan’s economic structure has both inclusive and extractive elements. Agriculture accounts for 21 percent of gross domestic product and 45 percent of the labor force. The crops that make the most important cash contribution to the economy, cotton and rice, come disproportionately from the extractive and feudal environments of Sindh and southern Punjab. Its industrial production – another 25 percent of the economy –relies heavily on textiles, mainly produced from the cotton crop.
During Pakistan’s heady growth days in the 1950s, the “22 families” that controlled industry were the stuff of legend. Most were also large land-holders. This part of Pakistan’s industrial base has often exhibited the rent-seeking behavior the two authors associate with extractive institutions, though the business community’s recent support for expanding trade with India suggests that its leaders may be starting to take a more inclusive view. Interestingly, Pakistan’s days of highest growth came in the late 50s/early 60s, when Ayub Khan’s autocratic government combined with a comfortable relationship with the captains of industry – the authors’ “centralization model” – but the textile industry failed to innovate, and the growth of those years ran out of steam.
Pakistan has important entrepreneurial elements as well. The All-World Network recently brought to Washington a group drawn from Pakistan’s fastest-growing entrepreneurial businesses, some of them now quite substantial in size. About half the participants had started their own businesses; the rest were carrying on a family business started a generation or two earlier. The group that best matches the inclusive upstarts who transformed British politics during the industrial revolution is perhaps the export-oriented sporting goods and surgical instruments producers in the towns of northern and central Punjab. These small producers have a huge interest in lowering fuel prices and stabilizing erratic electricity supply. The political pressure that led Pakistan to cancel IMF-required price increases, however, came mainly from the dog-eat-dog politics of Karachi – and was precipitated by a near-breakdown of the governing coalition. In other words, the small industries of Punjab have not yet become a major influence in politics. Another example of grass-roots, non-agricultural policy impact was the response to President Musharraf’s effort to oblige all retail traders to register with the income tax authorities. The resulting protest shut down commerce within Pakistan for about three days. This showed considerable power – but it was being exercised out of suspicion toward the government, and it did not translate into more positively oriented political influence.
If Pakistan’s struggling middle class has not yet found an effective outlet in politics, has the country created an environment in which more people can take the entrepreneurial path to prosperity? The success of the sporting goods and surgical instruments industries shows that entrepreneurial talent exists and finds outlets. But it’s a tough road. The World Bank’s Ease of Doing Business index lists Pakistan at number 105 out of 183 in 2012, down from 96 a year earlier. This score is about at the midpoint of other countries in the region, but it suggests that entrepreneurs require more than the usual grit and determination.
Comparisons: Pakistan’s political inclusiveness is weaker than most of its comparable neighbors – India and Sri Lanka, and to some extent Bangladesh. The family glass ceiling in politics and the link between political parties and economic elites are weaker in both India and Sri Lanka than in Pakistan; Bangladesh has problems similar to Pakistan’s, though not as severe. None of these three countries has an economic structure that encourages extractive behavior in the way Pakistan’s cotton economy does. And while there are serious concerns about the leaching of criminal behavior into politics in both India and Sri Lanka, both benefit from a much more solid history of regular elections, and both have used elections to change governments. Pakistan does better than either India or Bangladesh on the Ease of Doing Business index. And India and Sri Lanka have states or provinces with strong identities, along with related governance challenges.
Pakistan’s political and economic structures are problematic, from the perspective of the Acemoglu/Robinson template. Its governance is weak; assuming that they intend “centralization” as a proxy for governance, they would find today’s Pakistan a poor candidate for “growth without inclusion.” And the difficulty of doing business there, coupled with the very thin layer of well educated people, suggests that Pakistan is not yet set up to base its growth on innovation.
I am deeply skeptical of single-factor explanations of something as complex as economic growth, but this analysis certainly presents powerful evidence that inclusiveness belongs on the short list of important ingredients. However, it has left out some important elements, most of which complicate the task of achieving sustained growth in Pakistan.
First, the authors do not address security. Even with stronger governance than Pakistan’s, war in Afghanistan and a virtual civil war inside Pakistan cast their shadow on its economic prospects. The place where political violence has reigned for the longest time, Karachi, is also the major port handling exports. These security problems also reinforce the role of the army in the country’s politics – which pushes away from the kind of inclusiveness the authors support, and discourages steady economic growth.
Second, despite their bow toward “centralization,” the authors do not address the problem of countries with highly diverse ethnic structures or provincial systems. Pakistan faces the twin challenge of taking advantage of very unequal sources of growth in particular regions and of weaving them together into a national whole. Punjab and parts of Sindh probably have the best shot at growth – the largest middle class, the greatest number of “urbanizing” towns, substantial populations that are increasing their literacy, a large number of successful small farmers. Khyber Pakhtunkhwa has the most egalitarian social structure (at least for men), the Pashtun tribal councils. However, those councils work at the level of the tribe, and do not necessarily provide a vehicle for junior members of the tribe to supplant their elders, nor do they provide for “pluralism” in the sense of an alternative to the basic tribal structure. Balochistan has been in more or less open revolt from the Government of Pakistan for four decades; whatever “centralization” means, it does not apply here. Moreover, Balochistan and Khyber Pakhtunkhwa are provinces where the central government is held in contempt – hardly a recipe for peaceful participation in inclusive rule-setting.
Third, some economic structures make it hard to avoid “extractive” institutions. In Pakistan’s case, a major crop, cotton, is easily grown on huge landholdings, and provides the major raw material for the country’s principal industry. This does not invalidate the authors’ analysis, but it begs a further question – how does one organize such an economy for “inclusion”?
Fourth, the authors reject the idea that ample resources lead to sustained growth in the absence of inclusive institutions, “centralization” and innovation-friendly rules. However, resources could certainly be an important complement to inclusive institutions, and a some of the cases they cite are countries whose investment was powerfully assisted from abroad (e.g. Korea, which received generous U.S. aid and substantial private investment). They concede the importance of efficient resource allocation (e.g. Russia). This leaves me with the question of how important resources and their allocation are to enabling inclusive institutions to generate growth.
More broadly, the Acemoglu/Robinson analysis does not address the interaction between different factors in facilitating or discouraging growth. To take one more example: education by itself may not deliver sustained growth (the Soviet Union and, even more, the countries of Central Asia are good illustrations) – but how important is education in taking advantage of the assets a country has – including inclusive institutions?
Implicit in this analysis is the importance of entrepreneurship and the middle class. There may be some good news here for Pakistan. Rural areas now dominate politics. They are tailor made for the politics of stasis, with age-old land-use relationships and often feudal landholding structures. While Pakistan’s immigrants to towns and cities initially take their rural relationships with them (as described by Lieven), there are good reasons to expect that this will eventually shift. If it does, the people living, not just in Lahore and Karachi but in the smaller cities of Punjab and Sindh may be ripe for new kinds of economic organization as well as political mobilization.
The book does not present policy recommendations. A few things come to mind as implications for Pakistan. First, what Pakistanis do matters vastly more than what Americans do. Second, encouraging entrepreneurship, facilitating trade, and expanding the middle class should be priorities. Third, basic governance and policing are hugely important, but foreign assistance in these areas will fail unless Pakistani political leadership is fully committed. An issue outside this book’s framework is how to encourage a dynamic future for Pakistan’s cities. Growing cities might be the most promising way to encourage economic and social pluralism.
 Population figures: Federal bureau of Statistics, Pakistan Demographic Survey 2006, table 1, http://www.pbs.gov.pk/sites/default/files/population_satistics/publications/pds2006/tables/t01.pdf .
 The best description of this setup is in Anatol Lieven’s Pakistan: A Hard Country, New York: Public Affairs, 2011.
 See Ayesha Siddiqa, Military Inc.: Inside Pakistan’s Military Economy, Ann Arbor: Pluto Press, 2007.
 See Steve Inskeep, Instant City: Life and Death in Karachi, New York: Penguin, 2011.
 See, for example, Vali Nasr, Forces of Fortune: The Rise of the New Muslim Middle Class and What It will Mean for Our World, New York: Free Press, 2009, pp. 203-232, and Anita Weiss, “Population Growth, Urbanization and Female Literacy,” in Stephen Cohen, ed., The Future of Pakistan, Washington: Brookings, 2011.
 The standard reference works, including Pakistan Demographic Survey, all cite Pakistan’s urbanization 32-35 percent. However, a number of economists who have specifically looked at the problems of cities believe that a more comprehensive definition of “urban population” that includes medium-sized towns would encompass over half Pakistan’s population, and would better reflect the transition away from rural life. Author’s conversation with Nadeem ul-Haque, July 20, 2012.