The economic and political consequences of the partition of the Indian subcontinent in 1947
Partitioning states into smaller units is a solution that is often suggested to resolve ethnic conflicts around the world. The appealing elementary school logic of fostering peace by separating groups that are not playing well together continues to be actively discussed by both policymakers and academics, assuming a prominent role in contemporary debates over the continued conflict in Iraq, Syria, and beyond.
Yet, little is known about the conditions under which what appears to be a viable political compromise devolves into a human disaster. Neither are the economic and political consequences of partition well understood. The partition of South Asia on religious grounds in August 1947 has the potential to teach us much about these issues. Seen initially as a viable political compromise, the partition of the Indian subcontinent instead led to one of the largest forced migrations in world history, with an estimated 17.9 million people leaving their homes (Aiyar 1998, Bharadwaj et al. 2008a). Estimates of the number killed between March 1947 and January 1948 range from 180,000 to one million. There were 3.4 million ‘missing’ members of targeted minorities in the 1951 census (Bharadwaj et al. 2008a).
The impact of Partition on the population shares of religious minorities
Figure 1 shows the share of affected religious minorities in the population of pre-Partition India according to the census of 1931, and in India and Pakistan three years after the Partition (in the 1951 censuses). The shares of Muslim populations in Punjab fell to insignificant levels by 1951, as did the shares of Hindu and Sikh populations in West Pakistan. In contrast, in the east, the disappearance of minority population shares was much more muted.
Figure 1 The proportion of target minorities (Muslims in India, Hindus and Sikhs in Pakistan) in 1931 and 1951
Sources: Reproduced from Jha and Wilkinson (2012), based on Bharadwaj et al. (2008a) and relevant censuses.
These falls in the share of minorities in the populations reflect two processes:
- First, there was an outflow of the target minorities due to ethnic cleansing (i.e. forced migration, religious conversion, or death).
- Second, there were inflows – migration from across the border by members of a local religious majority.
The extent and patterns of ethnic cleansing that led to the collapse of minority population shares in then western India was not anticipated. As late as November 1945, the Commander of the Indian Army, Field Marshal Claude Auchinleck, predicted that the “principal danger areas (for ethnic conflict) are likely to lie in the United Provinces, Bihar and Bengal”, these having also been the principal areas where religious rioting had taken place in the years leading up to Independence (Figure 2a, and Jha and Wilkinson 2012). Yet, the patterns of ethnic cleansing were very different (Figure 2b), and the areas that Auchinleck thought would be most susceptible to violence still enjoy sizeable minority populations to this day.
Figure 2a Hindu-Muslim riots in Indian districts, 1850-1942
Figure 2b Ethnic cleansing, 1931-1951
Note: Proportion of minorities missing due to forced migration, conversion, or death.
What explains the patterns of ethnic cleansing?
As the previous figures show, being close to the border made it easier to leave and also easier to compel others to do so. But beyond this, our research points to the crucial importance of the distribution of skills and the presence of inter-group complementarity – gains from trade – of those in the affected districts in changing both the intensity of the Partition and its subsequent economic effects.
An important dimension was the local organisational skills available to different groups to mobilise themselves collectively during this period. Despite the Partition boundaries being drawn by a British civil servant, Sir Cyril Radcliffe, in a cloistered retreat far from the borderlands, many believed that by changing the ‘facts on the ground’ in terms of their relative population share, they could shape the ultimate location of the national borders. Ethnic cleansing meant successfully facing down not just the local police, but the British Indian military presence, such as the Punjab Boundary Force, aimed at maintaining order in these areas.
One such resource was drawing on the organisational skills of local demobilised veterans of what had been the largest volunteer army in world history – the 2.86 million troops India mobilised during the Second World War. Rather than the Partition violence being an uncoordinated series of attacks of neighbour against neighbour, the Punjab Boundary Force “found a countryside not easily intimidated even by seasoned and heavily armed troops like themselves…What seems to have been happening was that the army was being challenged by experts, by people who had formerly served within its ranks…the massacres of Punjab, in all their gruesome efficiency, bore the imprint of military training and organization” (Kamtekar 1988, citing the Rees Report).
Figure 3 Combat experience of Indian Army troops raised in different districts, 1940-45
The ‘baptism of fire’ experienced by combat veterans in particular is often recognised as a training ground for military skills in organisation and improvisation. Jha and Wilkinson (2012) exploit the fact that Indian infantry battalions were recruited from particular martial races from particular districts, but once in the army were designed to be inter-changeable and deployed in response to enemy action and the needs of the moment – such as the sudden entry of Japan into the war – without regard for their districts of origin. They show that districts that raised troops with an additional month of combat experience saw an increase of 17,000 minorities ‘cleansed’ in a district. The effect of raising units with increased combat experience was particularly pronounced in districts where the minority population approached the majority, and thus the potential gains to ethnic cleansing – both in terms of the spoils to be had and the threat minorities might pose to the majority group’s numerical dominance – were also greater.
But veterans also served to mobilise their own communities for defence, and, when needed, to move en masse to a new homeland. Districts where the minority was small but organised by combat veterans saw relatively less ethnic cleansing.
The peaceful migration of Sikhs from Lyallpur
When the minority was large but organised, veterans appear to have facilitated peaceful mass migration. In Lyallpur, in current-day Pakistan, despite having a 62.9% Muslim population, the sizeable Sikh minority had been favoured in military recruitment. Lyallpur had also been included in both Congress and Sikh proposals as part of India. Sikh veterans had enjoyed an average of four months of front-line combat experience. Initially mobilised into defence, Lyallpur had seen little of the ethnic violence of neighbouring districts with similar ethnic composition.
Figure 4 Lyallpur in context
The drawing of the Radcliffe boundary, which left Lyallpur well within Pakistan, changed this calculus. Giani Kartar Singh, the local Sikh leader, decided to implement a mass evacuation of his people. According to Ian Morrison of the Times of London, the movement was:
. . . orderly and well organized. The Sikhs moved in blocks of 40,000 to 60,000 and cover about 20 miles a day. It is an unforgettable sight to see one of these columns on the move. The organization is mainly entrusted to ex-servicemen and soldiers on leave who have been caught by the disturbances. Men on horseback, armed with spears or swords, provide guards in front, behind, and on the flanks. There is a regular system of bugle calls. At night a halt is called near some village where water is available, watch fires are lit, and pickets are posted." (The Times 1947)
The movement of these Sikh columns were carefully targeted towards a particular area: the Sikh princely states and the districts of Gurdaspur, Amritsar, Jalandhar, Hoshiarpur, Ludhiana, and Ferozepore. For the first time in history, Sikhs came to represent more than half of the total population of a sizeable contiguous area (Grewal 1998). Giani Kartar Singh of Lyallpur played a key role in the later secession of these districts, first as PEPSU, a state whose administration was dominated by Sikhs, and ultimately as part of a new Punjabi-speaking state (Grewal 1998). Thus, the local organisational skills available during the Partition led to population concentrations that have shaped state-level politics in northwest India ever since.
Disruptions to trade
Beyond the distribution of skills, the Partition had the potential to disrupt the patterns of trade between groups and regions. In some Indian cities, including modern cities and those with origins as Muslim capitals and patronage centres, members of different religions had competed economically; the Partition provided an opportunity to get rid of minority competitors. In medieval port cities, however, historic Muslim advantages in going to Mecca had led to a long tradition of trade and trust between religious groups (Jha 2013). In these medieval port towns, combat experience led to less cleansing of ethnic minorities (Jha and Wilkinson 2012).
Though the Partition was not as disruptive to trade within towns with histories of inter-religious trade, it did draw a boundary between previously economic integrated regions that had large economic consequences. The case of jute in the east provides a classic case study on a flourishing industry disrupted by the Partition. Bengal was one of the most important centres for growing and producing jute, which at the time was known as the ‘golden fibre’. The boundaries separating West Bengal from East Pakistan (which later becomes Bangladesh) separated the jute growing areas in East Pakistan from the jute mills in West Bengal. As a result, the price of jute in the world markets sky-rocketed (Bharadwaj and Fenske 2012). In subsequent years, independent India started growing jute itself, and the areas in which jute is grown correlate strongly with areas that received migrants from East Pakistan. A likely reason is that refugees from East Pakistan came with knowledge specific to cultivating and growing the crop. Hence, while the Partition lines and a trade embargo disrupted a flourishing industry in the short run, refugee specific knowledge may have mitigated long-term damage to this industry (Bharadwaj and Fenske 2012).
Literacy levels, demographic changes, and economic activity
While the Partition directly affected economic activity by disrupting trade and other essential inputs, the long-run impacts on economic activity can also be viewed as the result of demographic changes that accompanied it. As a result of the population transfer in the west of then India, where the numbers of Hindus and Sikhs entering India were comparable to the number of Muslims who entered Pakistan, districts in India that saw more refugee activity also tended to see increases in literacy (Bharadwaj et al. 2008b). This was in part due to the historical differences in education levels between Muslims in South Asia and their Hindu and Sikh counterparts. The relatively literate Hindu and Sikh minorities that left rural Pakistan replaced the relatively less educated Muslims, increasing the availability of skills in India. In ongoing work, Bharadwaj and Mirza (2017) examine the long-run impacts of such demographic changes on agricultural productivity on the Indian side. The preliminary findings, which analyse district-level agricultural yields between 1956-2009, suggest that places with more activity by displaced persons as a result of the Partition were more likely to experience higher agricultural yields, although such increases are only observed after the advent of the Green Revolution in India (in the late 1960s). One important channel through which this is likely to have occurred is the relative education of the refugees, which made them more likely to take up technologies and new methods of cultivation that were part of the Green Revolution.
On the Pakistan side, the demographic consequences of the Partition look a little different. While the Muslims who left India were relatively more literate than the Muslims who stayed behind, on average, Pakistani districts did not enjoy the gains in literacy of their Indian counterparts (Bharadwaj et al. 2008b). However, there is substantial heterogeneity across districts. Some areas in Pakistan saw a sharp increase in literacy, mainly driven by the presence of Muslim refugees. For example, Karachi received about 600,000 refugees, 52% of whom were literate. Literate refugees in Karachi made up nearly 91% of the total literates in 1951. Displaced Muslims in Pakistan also moved with a better quality of education – while 42% of the resident literates had formal educational qualifications, 57% of the literate refugees had such certificates. Refugees – also known as Muhajirs in Pakistan – came to occupy important positions in the city’s administrative and business sectors, perhaps due to their higher education and skill sets.
The end of colonial rule in India culminated in the birth of two nations, and the forced displacement of many millions. Differences in the distribution of education and organisational skills across communities and the extent to which different communities and regions traded with each other all played a key role in shaping both the bloody nature of the upheaval during the Partition itself, and its long-term political and economic consequences. Would-be map-redrawers, faced with similar challenges as Sir Cyril Radcliffe faced 70 years ago in attempting to bring peace and prosperity through ethnic partitions, could benefit greatly from understanding what the Partition taught us, at great cost, about these linkages and their consequences.
Aiyar, S (1998), “’August Anarchy’: The Partition Massacres in Punjab, 1947,” in D Low and H Brested (eds), Freedom, Trauma, Continuities: Northern India and independence, Walnut Creek, CA: Altamira Press, pp. 15-38.
Bharadwaj, P, A Khwaja and A Mian (2008a), “The Big March: Migratory Flows After the Partition of India”, Economic and Political Weekly 43(35).
Bharadwaj, P, A Khwaja and A Mian (2008b). “The Partition of India: Demographic Consequences,” Harvard Kennedy School; also published in International Migration 53(4) as “Population Exchange and its Impact on Literacy, Occupation and Gender: Evidence from the Partition of India”.
Bharadwaj, P and J Fenske (2012), “Partition, Migration, and Jute Cultivation in India”, Journal of Development Studies 48(8).
Bharadwaj, P and R A Mirza (2017), “Displacement and Development: Long Term Impacts of the Partition of India”, Working Paper, UC San Diego.
Jha, S (2013), “Trade, Institutions and Ethnic Tolerance: Evidence from South Asia”, American Political Science Review 107(4).
Jha, S and S Wilkinson (2012), “Does Combat Experience Foster Organizational Skill? Evidence from Ethnic Cleansing during the Partition of South Asia”, American Political Science Review 106(4).
Grewal, J S (1998), The Sikhs of the Punjab, Cambridge University Press.
Kamtekar, I. (1988), “The End of the Colonial State in India 1942-47,” PhD thesis, Cambridge University.
The Times (1947), “200,000 on the move”, 19 September.