The Recent Political Context

It might be helpful to begin by providing some background on recent Indian politics. From 1989 until 2014, India was in an “era of coalitions,” in which neither the once-dominant Congress Party nor its major challenger the Bharatiya Janata Party (BJP) could secure more than a third of the seats in India’s parliament, and so each had to cobble together parliamentary majorities with a variety of regional parties that demanded their pound of flesh, and then some, in return for support at the center. The dynamics of keeping these Congress and BJP-led coalitions together, when combined with serious leadership weakness at the top of the Gandhi family-controlled Congress Party, which formed governments after the national elections in 2004 and 2009, have prevented major reforms to India’s economy and bureaucracy in the past decade. So although there was a flurry of reforms—initially under pressure from the International Monetary Fund—in the early 1990s, the momentum has gradually wound down. In 2014, a fed-up Indian electorate decisively voted for change and gave a clear majority (282 of 545 seats) to the BJP and its prime ministerial candidate Narendra Modi, with 54 additional seats to the BJP’s allies. The era of coalitions is over, at least for the next few years, even though for the time being the BJP stills needs some support in the upper house, the Rajya Sabha, to push through important measures such as a politically contentious bill to allow land acquisition for economic development.

The May 2014 general election was a disaster for Congress, which emerged with even fewer than the 54 seats necessary to count as the official opposition, its worst showing ever. Modi ran a presidential-style campaign, running on his 14-year record as chief minister of Gujarat, and highlighting what he portrayed as Gujarat’s record of strong growth, decisive decision making, good governance, and efficient infrastructure and power provision. The Congress tried unsuccessfully to defeat Modi by highlighting what it claimed were the economic and social weaknesses of the “Gujarat model,” and in particular the massive anti-Muslim pogroms that took place in Gujarat in 2002, during Modi’s first year as chief minister. But the promise of Mr. Modi, perceived as a strong leader who was capable of grappling with the cumbersome Indian bureaucracy and delivering growth in his state, combined with Congress’s own poor track record and uninspiring leadership, was enough to lead to a massive defeat for Congress.

Since Prime Minister Indira Gandhi’s 1971 “Abolish Poverty” campaign, the Congress Party has increasingly identified itself with making sure that the interests of the poor, women, lower-caste tribes, and religious minorities, especially the 14% Muslim minority, are not forgotten in the rush for economic growth. In the 1990s and 2000s, the party has spent an increasing share of the national budget on a variety of programs, many named after Nehru-Gandhi family members and other Congress icons, that aim to deliver health care, education, housing, food, and other services to “below poverty line” (BPL) families. The largest of these programs, the Mahatma Gandhi National Rural Employment Guarantee program, provides $6 billion a year in workfare to Indians in the countryside.

The BJP, on the other hand, has become more identified with pro-market economic reforms. Its 2014 manifesto emphasized the need to raise growth rates through new national economic development strategies, and the need to improve the environment for investment and technology, rather than a focus on the groups—scheduled castes, minorities, women, tribals, lower castes—to whom these benefits would be provided. Congress’s mantra was to have balanced and inclusive growth with programs and rights guarantees (the right to food, the right to employment, etc.) to ensure that the poor and marginalized groups were not left behind, while the BJP focused on economic growth for all. So the BJP’s 2014 manifesto had lots of mentions of words like “technology,” “growth,” “infrastructure,” “industry,” “investment,” and “governance” to be delivered to the whole nation, while the Congress had more of an emphasis on policies to guarantee work, health, and education to specific groups, such as women (48 mentions), scheduled castes and tribes (51 mentions), and religious minorities (12).

The most serious debate over the rival economic agendas of the BJP and Congress during the 2013–14 campaign was carried out by several of India’s most famous development economists, Jean Dréze and Amartya Sen, on the one hand, and Jagdish Bhagwati and Arvind Panagariya on the other.^{Footnote 2} Each of these pairs published a book during the campaign—An Uncertain Glory and Why Growth Matters—that provided very different perspectives on India’s recent growth performance and social and economic achievements, perspectives that in important ways paralleled the different approaches of the Congress and BJP. Dréze and Sen highlighted the many Indians who have been left behind in the rush for growth, and the many basic needs—in education, gender and caste equality, food, housing, sanitation, and power—that still need to be met for half or more of the population. Bhagwati and Panagariya, on the other hand, emphasized that economic growth has allowed the trends in almost all of these areas to improve over time, and that more of a focus on removing obstacles to economic growth and investment is urgently needed now in order to directly remove people from poverty and provide the money to pay for more social-welfare spending that will help those who need it. Bhagwati and Panagariya made it clear that the BJP and Modi offered the best hope for the country, going forward, while Dréze and Sen signaled their support for the incumbent Congress, or at least for someone other than the BJP.^{Footnote 3} The Indian press followed the controversy extensively, and portrayed it as an intellectual showdown between the Modi and the Congress models of development.

The picture that Dréze and Sen paint is one that balances some celebration of India’s democratic successes and recent economic achievements with a cautionary tale about how much still remains to be done. As they put it, “India has been climbing up the ladder of per capita income while [in relative terms] slipping down the slope of social indicators” (An Uncertain Glory, p. 8) They cite a lot of sobering facts: Compared to other poor countries, India’s levels of life expectancy, under-five mortality, access to good sanitation, access to schooling, literacy, malnutrition, and immunization are all substantially worse than average (ibid., p. 49). And India is also gradually slipping behind its neighbors in South Asia, such as Bangladesh and Nepal, which used to be much worse than India on health and education indicators but are now often better. A few key statistics bring out the enormity of the challenges: Wages for those in agriculture and the informal sector have stagnated even as the economy as a whole has grown (p. 30); 55% of Indians still have to defecate in the open because they lack proper toilet facilities (p. 63); and gender and caste inequality is still widespread, with one recent survey that they cite showing the continued dominance of upper castes in almost every area of life in one important city in north India (pp. 220–21). Much of India’s poor relative performance, Dréze and Sen suggest, is due to lower levels of state expenditure than in other developing countries. For instance, they argue that the lower levels of access to health care in India compared to China is due to the fact that India devotes 1.2% of GDP to state spending on health care while China devotes 2.7% (p. 37).

The solutions Dréze and Sen advocate will not come as a surprise to anyone familiar with their careers or previous work. To begin with, they argue that the whole notion of development is much wider than simply a focus on growth or income, and needs to focus on whether it provides a series of life-enhancing improvements—capabilities, as Sen and Martha Nussbaum have termed them elsewhere—to the lives of individuals and their families (Nussbaum Reference Nussbaum2001; Sen Reference Sen1999). Dréze and Sen cite, once again, the positive example of the southern state of Kerala (which has, overall, the best Indian development outcomes) in demonstrating the vital importance of sustained state action for improving education, health, and mortality outcomes and tackling deep social and gender inequalities (Uncertain Glory, pp. 77–80). They argue that despite the current desire to target social programs at the most needy, some of the most successful social programs in India—such as the national rural employment guarantee program (NREGA) rolled out in 2006, which guarantees a minimum amount of work to all those who want it in the countryside, and public distribution systems in states such as Kerala, Tamil Nadu, and Chhattisgarh—have employed universalistic principles, which both makes the programs bureaucratically easier to administer and makes sure that those who are most able to opt out also have a stake in these programs’ success (ibid., pp. 192–95 and 206–12). They also see the slowness of economic reforms, in some areas, as a simple reflection of the fact that some of these reforms will not benefit or even will hurt the poor, and that in a democracy we should expect pushback from citizens in these situations. In fact, they express the hope that there will be greater mass democratic pushback in future years over the poor quality and delivery of many basic services and that the media will act as a voice for more than just the top 20%, as they claim it does at the moment (pp. 252–57).

The world looks very different to Bhagwati and Panagariya. They argue that far from a focus on growth and reform being the cause of India’s lagging performance, what progress there has been over the past two decades is largely because of the deregulation and the wealth and employment opportunities it has created. A focus on growth is key, they say, because it both 1) directly creates more jobs and income, lifting people out of poverty, and 2) provides (through taxes) the opportunity for the government to spend more to address the basic needs and inequalities that remain (Why Growth Matters, pp. xviii–xxii). They term these two routes to ending poverty Track I and Track II, and argue that the Indian experience has many lessons for other countries to focus first on the benefits of growth and deregulation, and only secondly on redistribution and welfare. The lessons from Modi’s own state of Gujarat, they claim, are particularly relevant, and they argue that the pro-growth policies in Gujarat over the past few decades led to the state’s enjoying significantly above-average improvements in literacy, poverty levels, and other indicators of welfare (ibid., pp. 57, 74, 78).

Bhagwati and Panagariya argue that the Indian and Gujarat experiences provide a powerful justification for the new Indian government to push for further reforms in a whole number of areas: labor market reforms to change laws that currently make it almost impossible to change some work practices and to lay off and fire people from medium and large enterprises; laws to allow increased private-sector provision in education, especially higher education, and health; vouchers for education and health; laws to make it easier to buy and sell land for development and industry; and a general effort to reduce state regulations at all levels, because of the economic distortions they cause and the corruption they bring in their wake (ibid., pp. 110–21, 127, 137).

Bhagwati and Panagariya’s ideas for what India’s future Track II government welfare spending should look like are, not surprisingly, very different from those envisaged by Dréze and Sen. The latter would get rid of subsidies that they say disproportionately benefit the top 20%, such as those on power and fertilizer, and tax breaks on wealthy groups such as jewelers. But they like universal services and public provision, in part because, as Sen has argued in an interview, “What the poor need is better public services and not cash in their pockets. Cash transfer allows them to get certain things—and I’m in favour of it and retaining it, but it is not a solution. This is not the most efficient way of running it. If you are a peasant farmer and have never been to school, your ability to choose on the basis of information is very limited. Given the asymmetry of information, you’ll never be able to get there.”^{Footnote 4} Dréze and Sen have faith in the state, and they want India to model future efforts on places where India’s 29 state governments seem to be working well, such as Kerala, Himachal Pradesh, and Chhattisgarh. Bhagwati and Panagariya, on the other hand, like the idea of putting cash directly in people’s pockets. They want more private provision, more targeting, and more direct cash transfers and vouchers in order to deliver the benefits to those who most need them, and to minimize the chance that a corrupt state apparatus will siphon off the money (Why Growth Matters, pp. 153–62). Anyone who has seen the inefficiencies of India’s government education and social service provision will find it hard not to see that there is promise in many of their suggestions.

Unfortunately, much of this book is devoted less to making a positive case for reform than it is to attacking the “myths” that have allegedly been promoted by others, such as the idea that Kerala is a success relative to other states, or that inequality and corruption might have gone up under the post-1991 reforms. The particular targets of its attacks are Dréze and Sen, and to a lesser extent the Princeton economist Angus Deaton. Some of the points Bhagwati and Panagariya make hit home; for instance, they point to the relatively small yearly fluctuations in inequality in the 1987–2005 period as suggesting that concerns over inequality might have been overstated. But the examples they use to “demolish” other alleged myths are less convincing. For instance, they try to debunk the idea that corruption has increased since the reforms with the assertion that “Corruption broke out not with the liberal reforms of the 1980s, but under the license permit-raj that peaked in the 1970s during the socialist-era policies of Prime Minister Indira Gandhi. With the government controlling the manufacture, distribution, and price of major commodities, bribes became virtually the only means of accessing the latter within a reasonable time frame” (ibid., p. 86). This argument both lacks any hard data—there are in fact no good time series data on levels of Indian corruption—and also ignores the fact that there is plenty of evidence for widespread corruption in India existing well before the Congress reforms of the 1970s.^{Footnote 5}

The tone of the first half of Bhagwati and Panagariya’s book toward those who hold different opinions is often harsh, in contrast to the more measured language in Dréze and Sen’s. Consider, for instance, the following quotation, presumably directed at Sen (who was a contemporary of Bhagwati’s at Cambridge in the 1950s):

The examples of this type of language in the book, as Angus Deaton has pointed out, tend to undercut support for the authors’ arguments rather than to make them stronger.^{Footnote 6}

While taking potshots at Dréze and Sen, Bhagwati and Panagariya tend, however, to overlook any areas of weakness in their own arguments. For instance, the latter rightly point out that one potential problem with inferring that the postindependence Kerala policy model is a good one is that, already at independence, Kerala’s literacy levels were much higher than those in other states (Why Growth Matters, p. 69.). They then explore a few of the reasons why Kerala might have had higher pre-1947 development outcomes, such as greater openness to markets, higher growth, missionary activity, different local traditions with regard to women’s rights and literacy, social activism, and state and private (especially religious and caste association) initiatives in education after the 1860s (ibid., pp. 75–78). From this brief survey they then confidently conclude that the pre-1947 achievements of the territories that form present-day Kerala were largely market and growth driven, rather than the outcome of state activity (p. 77):

So the historical role of the state, after being briefly acknowledged by Bhagwati and Panagariya in their survey of Kerala, is then quickly shunted to the rear so that growth and markets can take center stage. This is a little too convenient, however, because the fact is that historical institutions played a much greater role in explaining today’s variation than they acknowledge. The past two decades of recent research in political science have pointed to long-run patterns of institutional and state development as one of the most important factors in explaining contemporary variation in the state, legal, public goods, and market institutions that promote growth (Acemoglu, Johnson, and Robinson Reference Acemoglu, Johnson and Robinson2001; North Reference North2005; Mahoney Reference Mahoney2010; Slater Reference Slater2010; Treisman Reference Treisman2000).

In India, likewise, there seems to be a clear connection between the strength of colonial and precolonial states and contemporary variation in economic performance. The preindependence princely states of Travancore and Cochin, which constitute most of current-day Kerala, were also particularly strong regimes that devoted unusually high levels of state resources to public goods and education. Travancore, for instance, spent much more on its combined health and education budgets (23.2% of revenue in 1941–42) than other major princely states such as Jammu and Kashmir (13.1%), Bikaner (10.2%), and Indore (12.6%).^{Footnote 7} Only the major state of Baroda (19.5%) in what is now Gujarat came close, which suggests perhaps that historical patterns of state intervention and spending might even play more of a role in the postindependence Gujarat successes than Bhagwati and Panagariya care to admit.^{Footnote 8} The Green Revolution successes of Punjab, likewise, were built on decades of central spending on that state’s canal, road, and rail infrastructure, and also the decades of spending and training for the disproportionate number of Punjabis who joined the army and other national state services. In the half century before World War II, central government investment in Punjab has been estimated as twice as high on a per capital basis as spending elsewhere in the country.^{Footnote 9}

This delivery of services was not just bestowed from on high by paternalist princes; it was also actively demanded by lower and backward caste associations in these states, which pressed the state for greater social equality (e.g., in temple entry for non-Brahmins), as well as greater access to health and education. In fact, we know that even as early as the eighteenth century, the state in Travancore was using some of the surplus generated from its very profitable state monopolies to support a large number of state charities.^{Footnote 10} And both Travancore and Cochin established state assemblies early in the twentieth century that enabled the representatives of excluded and marginalized groups to press for increased access to basic and higher education, health care, jobs, and more social equality (Nossiter, Reference Nossiter1982). So Kerala’s “good start” at independence, far from directly undercutting the Dréze and Sen story in the way that Bhagwati and Panagariya argue, might also reasonably be seen as stemming from just the kind of state intervention, social spending, and active involvement in democratic politics that Dréze and Sen advocate.

The point here is not that the evidence conclusively proves Bhagwati and Panagariya wrong. It is a fact, for instance, as Bhagwati and Panagariya suggest, that before independence Kerala was also more industrialized and involved in external markets than other areas of India, which might support their growth and market openness thesis.^{Footnote 11} The important point, though, is that the historical evidence from the Kerala case supports multiple interpretations, including the ones that Bhagwati and Panagariya say they have definitively proven false. There are also other examples of an overeagerness to seize on confirming evidence and a tendency not to look too hard for disconfirming evidence in the book. The authors use 1991 to 2011 time series data that they claim debunks the argument that “the state achieved superior health and education outcomes through significantly more activist state interventions.” However, they underplay the significance of evidence that does not agree with their thesis (“Excluding Goa, Kerala spends more than its nearest rival state in eleven out of the twenty years. This may give some credence to the Kerala model except that the expenditures themselves are not all that large”; Why Growth Matters, p. 72), and they do not directly measure the impact of social programs such as job and education reservations for lower castes that might also plausibly explain the outcomes they measure. These programs have long been ubiquitous in the south of India, but have only become widespread in the rest of India in the past three decades. Their increasing use probably explains at least some of the education and income progress made by middle and backward castes (and helps explain why Muslims, who suffer from a high degree of market discrimination but have only recently benefited from these programs, have seen much slower progress), but instead almost all of these good results are ascribed by Bhagwati and Panagariya to the effects of growth, markets and deregulation.^{Footnote 12}

Nor is it clear, as Bhagwati and Panagariya claim, that economic growth must generally precede substantial investments and improvement in education and other services. More than two decades ago, the late Myron Weiner wrote an excellent book entitled Child and the State in India—puzzlingly not cited by either Dréze and Sen or Bhagwati and Panagariya—which looked at the reasons why India was not providing basic education to many of its children. Responding to the frequent argument he heard that the lack of state money or parental income was the issue, Weiner showed that many countries just as poor as India at independence—for example, Botswana, Cameroon, and Ivory Coast—had done much better in providing mass education for their children. And he also showed that historically many states, often from religious or ideological motivations (Protestant Europe and North America in the sixteenth, seventeenth, and eighteenth centuries, Japan after 1868, communist China more recently) had acted decisively to raise literacy rates well before they had become industrialized. This increase in literacy, of course, then helped subsequent industrialization and growth, a point made often by Dréze and Sen in Uncertain Glory.^{Footnote 13} Weiner also showed that India’s problem is not just the amount of money it spends on education, but the fact that so much of it is directed toward the better-off groups who benefit from higher education, rather than to primary education where the developmental payoff is greatest.

India versus Pakistan

While the important long-term effects of India’s preindependence institutional development are largely ignored in Dréze and Sen’s and Bhagwati and Panagariya’s dueling books on contemporary economic policy, they are central to Maya Tudor’s welcome new account of the larger democratic divergences between India and Pakistan. The question she addresses—why India and Pakistan, despite their common institutions before 1947, have turned out so differently since independence—has preoccupied many political scientists and historians over the years, with important contributions by Ayesha Jalal, Philip Oldenburg, and many others.^{Footnote 14} These scholars have offered several central explanations for the democratic divergence. First, a large number have focused on the much worse geographical and institutional inheritances of Pakistan compared to India at independence: Pakistan was a geographically divided state, with much greater strategic challenges, fewer administrators, and a much poorer economy, and it was also less lucky in its political leadership, with its moving force and first leader, Muhammad Ali Jinnah, dying of cancer the year after independence (Jalal Reference Jalal1987, Reference Jalal1992; Oldenburg Reference Oldenburg2010). Second, scholars have highlighted the very different role of the military in both countries—interventionist in Pakistan and professional and responsive to civilian oversight in India (Cohen Reference Cohen1988; Jalal Reference Jalal1987, Reference Jalal1992; Oldenburg Reference Oldenburg2010; Weiner Reference Weiner1989). Third, political scientists especially have highlighted the crucial role of a well-institutionalized political party, the Congress, in consolidating India’s democracy, and have negatively compared it to the role of the Muslim League in Pakistan, a much weaker party that had most of its pre-1947 strength in Muslim minority areas that ultimately remained in India at independence (Oldenburg Reference Oldenburg2010; Weiner Reference Weiner1989).

The Promise of Power makes two theoretical contributions in this well-furrowed field. First, most historically grounded and detailed accounts of the India—Pakistan divergence do not typically make much of an effort to seriously engage with larger theoretical literatures on state development, democratic consolidation, and transitions from colonial rule. Tudor, very commendably, is determined to fill this gap, and she draws explicitly on Barrington Moore’s (Reference Moore1966) macro-explanation for the emergence of democratic regimes, in highlighting the key roles of the commercializing middle classes and landed aristocracies in determining democracy of dictatorship (pp. 16–17). Her second main contribution is that she adds a new twist—the theoretical importance of the class coalitions in each party—to the established literature on the importance of party institutionalization in India and Pakistan. She argues (p. 15) that

One of the strongest aspects of the book is that Tudor makes her case for the importance of party weakness with a wealth of evidence on the founding parties in Pakistan (the Muslim League) and India (the Indian National Congress) before and immediately after independence. The broad points she makes about the relative weakness of the League and strength of the Congress are sometimes not new, but they have seldom been supported with the type of rich archival data we see here, from internal party deliberations, proceedings, and fights. She also does a very nice job in linking these party weaknesses to her thesis about the importance of class coalitions within each party, clearly showing, for instance, how the League reflected the interests of landlords while the Congress reflected the interests of farmers and the professional classes.

In highlighting these interlinked factors of class and party, Tudor deliberately de-emphasizes the importance of such factors as bad institutional or economic inheritances, the different ethnic challenges the new states faced, or the different roles of the military in each of the new states. She rejects the idea that the India—Pakistan divergence was due to India’s having inherited a greater proportion of civil service, a more professional portion of the bureaucracy, or a smaller proportion of the colonial army (ibid., pp. 27–30). She argues that ethnicity, rather than being a key independent variable, was fact closely correlated with class identities (pp. 33–34). She also disagrees with Ayesha Jalal and others who have pointed to the strategic weaknesses of Pakistan, which raised the army’s importance, and the huge imbalance in the new state’s army between Punjab and NWFP (35%), which had virtually all of the troops, and the other major provinces, which had none. She points to other states with large strategic challenges and large armies, such as Israel, where democracy seems to have thrived (pp. 27–30). The military is less an independent variable for her than one that follows directly from the importance of political institutionalization (p. 202). She takes at face value the military’s claim that it only finally launched a coup in 1958, despite numerous requests before this date, because of chaos in the country at large. Tudor approvingly cites Samuel Huntington’s view that “military coups do not destroy parties; they simply ratify the deterioration which has already occurred” (cited on pp. 17–18).^{Footnote 15}

As with most revisionist arguments, however, the evidence in The Promise of Power is at times pushed a bit too far. First, it seems hard to me to argue against the conventional wisdom (see Jalal Reference Jalal1992, Oldenburg Reference Oldenburg2010, and Weiner Reference Weiner1989) that Pakistan really was dealt a much worse hand than India at independence. Pakistan inherited the strategic frontier with Afghanistan, but little of the administrative capacity, military infrastructure, supplies, and tax base necessary to defend it, or to build a strong new state. Well before independence, the Indian Scheduled Caste leader Dr. Ambedkar had understood what was likely to happen as a result of the partition of the country into separate Muslim- and Hindu-majority states. Ambedkar pointed out in his 1941 book Thoughts on Pakistan that “it is the money contributed by the Provinces of Hindustan which enables the Government of India to carry out its activities in the Pakistani provinces,” and he predicted that if the Muslim-majority provinces left, India would benefit from redirecting this military expenditure to civilian purposes (pp., 93–95). Ambedkar was right. The massive cost of defending its borders left Pakistan with hardly any money to spend on health or education: 70% of the Pakistani state’s budget was spent on the military (Jalal Reference Jalal1987, 306). And of course the class coalitions that she highlights were themselves underpinned by regional historical patterns of economic growth and class inequality. So were these party issues or regional issues?

Second, by seeing the military intervention as largely epiphenomenal to the strength or weakness of party institutionalization, Tudor likely understates the importance of taking early and decisive steps to limit the military’s ability to launch a coup, and also underestimates the importance of the military as an institution, in terms of the ways in which civil—military arrangements are configured (Wilkinson Reference Wilkinson2015). In India, the Congress Party had been thinking about the problem of the unbalanced military for years before independence: the need to ethnically rebalance the army, control its officer corps, and introduce new paramilitary forces to carry out domestic security tasks and keep the army out of politics. As soon as he took office in India’s last preindependence government, in September 1946, Jawaharlal Nehru wrote a long letter to the British defense secretary and commander in chief outlining the importance of all of these issues if India’s army were to be safe for its democracy. And over the next decade, Nehru carried out all of these steps. In Pakistan, however, Jinnah had thought not at all about the military before independence, and the state missed a golden opportunity, in its first few years, to make the new regime coup proof.^{Footnote 16} The military, far from being a disinterested institution that was forced into acting because of political failure, was also actively destabilizing the political regime long before 1958.^{Footnote 17} Stephen Cohen therefore has it right when he says that the coup in Pakistan in 1958 was due to a combination of political weakness and military ambition (Cohen Reference Cohen1988, 107–17).

Third, in emphasizing the importance of class, Tudor underemphasizes the important of regional and ethnic cleavages to the overall story. In the Promise of Power, religious and ethnic demands tend to be portrayed, instead, as coterminous with or epiphenomenal to deeper class concerns, something with which many scholars would differ (Horowitz Reference Horowitz1985). For instance, in discussing the religious controversies of the 1950s, she says that “it was the opposition of provincial landlords [worried about parity to East Pakistan with its different class coalition] and not the religious leaders in and of themselves, which destabilized the central government” (p. 183). And elsewhere, she interprets the ethnic divide between East and West Pakistan that led to civil war in 1970–71 as largely a struggle between two very different class coalitions with different economic agendas, the “large landed aristocracy in West Pakistan and the cultivating tenantry in East Pakistan” (pp. 174–76). One problem here, of course, is that motivations were likely just as much religious or ethnic as class, and so it is hard to assign primacy to one particular identity or the other. The other is that, to the extent we can tell, support for many ethnic and religious agendas in Pakistan seems to have crossed class lines, in a way that would be inconsistent with a purely class motivation.

Most India—Pakistan comparisons, because they focus on the issue of India’s democratic successes versus Pakistan’s “failures,” probably tend to tend to paint too rosy a picture of the extent to which both states have had shared weak inheritances, or the way in which some Indian inheritances might themselves have been problematic. Both states inherited administrations that, at least below the level of the most senior officials, were large, unwieldy, and usually corrupt. One top police official in India just before independence told an inquiry that only 1% of his force’s police station commanders were “truly honest,” while another officer reported that most policemen had to take bribes just to make ends meet. The same inquiry was told that the salaries paid to officials who recorded all land transactions had long been “inadequate to ensure honest and better work from this machinery.”^{Footnote 18} These shared weak inheritances seem likely to explain at least some of the patterns of poor development and growth performance we have seen since 1947, by comparison with the equally poor but less venal administrations that the colonial powers left behind in such states as Korea and Taiwan.

The cross-class coalitions in Congress that Tudor explores might not have been an unmixed blessing for the new state. The class coalitions in Congress between the middle professional classes and the middle peasantry may have broken the power of the landlords, but they also underrepresented the very poor and the lower castes, which may go part of the way toward explaining the poor postindependence poverty-alleviation outcomes highlighted by Drèze and Sen.