Demography

1. Introduction

Muslims constitute the second largest religious group in India and thus the largest religious minority. The 2001 census enumerated India’s Muslim population at over 138 million, and by 2006 the Muslim population would be over 150 million. India’s Muslim population is amongst the largest in the world, exceeded only by Indonesia’s and close to the Muslim populations of Pakistan and Bangladesh. Moreover, it is larger than the total populations of most countries of the world.

In India, populations of all major religions have experienced large growth in the recent past, but the growth among Muslims has been higher than the average. Religious differentials in growth were observed in the pre-Independence period as well. The last intercensal decade however, has shown a reversal in the trends in growth; not a negative growth but a decline in intercensal growth for India, from 23.9% during 1981-91 to 21.5% during 1991-2001. This has occurred in both the largest religious communities, Hindus and Muslims, with the latter showing a larger fall from 32.9% to 29.5% or 3.4 percentage points, and the former from 22.7% to 19.9% or 2.8 percentage points. Thus, the growth differential has narrowed and is an early indication of convergence occurring over the medium term.

2. Population Size and Growth

At the beginning of the twentieth century, the Muslim population (in the post-Partition areas) was close to 30 million and grew rather slowly up to 1921 and later moderately, as did the overall population. Partition led to large-scale migration, and in 1961, well after the major Partition-linked migration had ended, India’s Muslim population was enumerated at 47 million, about 10% of the total population of 439 million. The latest census, conducted in 2001, enumerated 138 million Muslims out of India’s total population of 1029 million.

Over the forty-year period 1961 to 2001, the (total) population more than doubled, from 439 million to 1,029 million, an increase of 134%. The rapid growth is attributed to a sharp fall in mortality; though fertility also declined, especially over the later portion of the period, the decline has not matched that in mortality. Population growth has been high for all the major religions over the period with the Muslim population increasing rapidly from 47 million to 138 million. This amounts to an increase of 194%, just short of trebling, and much higher than the average increase of 134%. The Muslim population growth has been close to 30% in each of the four intercensal decades since 1961, with the latest decade showing a fall to a level just below 30%.

The annual growth rate has averaged 2.7% over the period 1961-2001, well above the national average of 2.1%. Hindus and Christians show marginally lower growth, 2.0 percent, Jains even lower, 1.8 percent, and Sikhs and Buddhists, marginally higher, 2.2 percent.

In 1961, the largest group, Hindus, accounted for 83.5% of India’s population followed by Muslims, with 10.7%; other minorities had much smaller shares - Christians 2.4%, Sikhs 1.8%, and Buddhists and Jains accounted for less than 1% of the total population. By 2001, the share of Hindus had fallen to 80.5% and that of Muslims had risen to 13.4%. This rise of 2.7% points between 1961 and 2001 is a consequence of the higher than average growth among Muslims. The shares of other minorities have remained nearly the same, though some small changes, a rise followed by a fall, occurred among Christians and Sikhs. The rise in the share of Muslims has been less than three percentage points over the four decades, that is, less than one point a decade.

3. Spatial Distribution

The Muslims in India reside across the country and yet their concentration varies substantially. Besides, the demographic dynamics have changed over different periods in time and in different regions. The trends in the southern states are quite different from those in the north-central states. The majority of the Muslim population in India are in four states — Uttar Pradesh, West Bengal, Bihar and Maharashtra. Generally, large states also have large Muslim populations, as expected. However, Punjab and Orissa, with populations of over twenty million each, had fewer than one million Muslims.

While the growth has continued throughout the forty-year period 1961-2001, the recent intercensal decade, 1991-2001, has shown a decline in the growth rate of Muslims in most of the states; this is in keeping with a decline in the overall national population. The Muslim population increase was quite modest, below 20%, much below earlier levels in Tamil Nadu, Kerala and Andhra Pradesh.

4. Age-Sex Composition Of Population

Most populations in the world have more women than men. At birth the share of boys is always higher, around 105 boys per 100 girls, but higher mortality among males compared to females leads to a sex composition favourable to females. However, India and some South and East Asian countries differ from this pattern. Female mortality was higher than male mortality in these parts though now this is not the case and the mortality gap is quite narrow. As a result, there are more men than women in India and the sex ratio (females per thousand males) is lower than 1000; for the period 1961-2001 this has hovered around 930. The Muslim population shows a similar pattern yet sustains an increasingly better sex ratio compared with the general population.

An associated indicator which exhibits relative social position in India is the child sex ratio (the number of female children under 5 for every 1000 male children under 5). As is well known, India is one of the few countries in the world to have a child sex ratio that is less than 1000. In addition, the overall child sex ratio in the country has been declining steadily during the last half century. It has declined from 976 in 1961 to 964 in 1971, 962 in 1981, 953 in 1991 and 927 in 2001. The low and falling child sex ratio is the result of two factors: excess female infant mortality (relative to male infant mortality) and female foeticide. Both in turn reflect parental discrimination against girls.

The NFHS data indicate that Muslims have the highest child sex ratio of any social group in the country. For instance, the child sex ratio among Muslims was 986 girls per 1000 boys in the age group 0-5 in 1998-99, significantly higher than the ratio of 931 among SCs/STs, 914 among other Hindus and 859 among other groups.

Surprisingly, even though Muslims already had the highest child sex ratio of any group in 1992-93, they were the only social group to experience a further increase in the ratio between 1992-93 and 1998-99. In contrast, Other Hindus experienced the largest decline (of about 5%) in the child sex ratio despite having the second-lowest child sex ratio in 1992-93.

5. Urbanization

India’s population is predominantly rural. In 2001 only 27.8% lived in urban areas, cities and towns of various sizes, showing a low degree of urbanization. Moreover, the tempo of urbanization has been quite low after 1981, with only about two percentage points rise in the share of the urban population over each decade. The Muslim population is also predominantly rural, but the level of urbanization among them has been higher than the population as a whole. In 1961, while overall only 18.0% of the population lived in urban areas, 27.1% of the Muslim population did so. This substantial gap has persisted and in 2001, 35.7% of the Muslim population was urban compared to 27.8% of the overall population. Muslims have generally been relatively more urbanized even in the past. By and large, India’s Muslim population is less linked to land than the overall population.

6. Demographic Processes

Population change is a product of three processes, mortality, fertility and migration. The higher than average growth rate of Muslims has often raised the question of why this is so. Obviously, one or more of these three factors is different for the Muslim population. We do have information on fertility and mortality by religion and hence can analyse this issue in some detail.

Estimates from different surveys as well as indirect census-based estimates show that infant and childhood mortality among Muslims is slightly lower than the average. Separate estimates for rural and urban areas show that the lower than average child mortality among Muslims is partly on account of their higher urbanization. Within urban areas, Muslim childhood mortality level is very close to the average urban level. While Muslims enjoy some advantage in survival compared to the general population, the mortality among other large minority religious groups, Christians and Sikhs, is even lower than Muslims. Essentially, childhood mortality among Muslims is lower only compared to the Hindus.

Among SRCs, SCs/STs suffer from the highest infant and under-five mortality rate, followed by Other Hindus. Muslims have the second-lowest infant and under-five mortality rate of any SRC in India. This is somewhat surprising, given the economically disadvantaged position of Muslims. In virtually every region, with the sole exception of the Northeast, Muslims have the second-lowest infant and under-five mortality rates of any SRC (after the "Other" group). In the South and West, their relative position is even better than in other regions. For instance, in the South, the infant mortality rate among Muslims is as low as 29 per 1000 live births – significantly lower than the rate of 61 among SCs/STs and the rate of 52 among Other Hindus.

Muslims not only have among the lowest infant and under-five mortality rates of all SRCs in India, they also have experienced some of the largest declines in infant and under-five mortality of any social group during the 1990s. The only states where child mortality among Muslims has worsened – both in absolute terms as well as relative to other SRCs – are Madhya Pradesh and Rajasthan.

Why exactly Muslims should have some advantage in child survival over other SRCs despite their lower levels of female schooling and lower economic status is a question that needs further exploration. For instance, it would be important to know whether the advantage is the result of better infant feeding and care practices among Muslims.

...efforts were made from pooled data of two surveys, the NFHS-1 and NFHS-2, to construct life tables for Hindus and Muslims and the results show that the life expectancy for Muslims is higher than average by about one year. Besides, estimates of maternal mortality also show lower than average maternal mortality among Muslims. Broadly, it could be said that Muslims do have a slim advantage over the average in survival.

The total fertility rate (TFR) is the most widely used summary indicator of fertility; this is the number of live births a woman has on an average during her lifetime, if she goes through the reproductive span, following a given age-specific fertility schedule. Religious differentials in fertility from various sources, surveys (NFHS-1 and 2) and the Census show that among the four large religious groups fertility is the lowest among the Sikhs, closely followed by the Christians and the highest among the Muslims. Various other surveys also corroborate the higher than average fertility among Muslims. There has been a large decline in fertility in all the religious groups; whereas in the pre-transition period the TFR was above 6, in recent years it has fallen below 4. Thus, the process of fertility transition is in progress in all communities. Fertility varies among Muslims according to socio-economic characteristics as well as on the level of the individual and there are large regional variations in fertility in India. While some states have reached a very low level fertility, with TFR close to 2.1, or near the replacement level, the north-central states have moderate levels of TFR, closer to 4. In states that have low fertility, the fertility of Muslims is also low, though higher than average. In fact, Muslims in the southern states have lower fertility than the average in the north-central states. For example, according to the NFHS-2, the TFR for Muslims in Kerala, Tamil Nadu, Andhra Pradesh and Karnataka as well as in Jammu and Kashmir was in the range 2.5 to 2.8, while that for the general population in Uttar Pradesh it was 4.0 and 3.8 in Rajasthan.

The relatively high fertility of a section of the population could be on account of various factors. A low age at marriage obviously is conducive to high fertility. However, data show that Muslims do not have a lower age at marriage than average. A point made on the higher fertility of Muslims was that the proportion of women married in reproductive ages was relatively high, because widow remarriage is well accepted in the Muslim community unlike the Hindus. However, recent data from the 2001 Census show that the marital status distribution of Muslim women is not notably different from that of the general population in the reproductive age groups, the ages that matter for fertility. The other important factor contributing to fertility differential is the use of contraceptives. Data on contraceptive practices (% of couples of reproductive age using contraception) for Muslims and the general population from various surveys, two by the Operations Research Group (ORG) in the 1980s and two by the NFHS in the 1990s, show that the use of contraception is widely prevalent among Muslims but to a lesser degree than the average. In contraceptive prevalence rate, there is a gap of about 10 percentage points between Muslims and the average. A careful examination reveals that it is the use of sterilization that shows a wide gap. Apparently, reversible methods are used relatively more commonly by Muslims compared to others. But sterilization is less popular among Muslims. ‘Unmet need’ for contraception is relatively high amongst Muslims and there is evidence of a large demand for reversible methods.

The facts do not support the common perception that Muslims shun family planning, as over one third of Muslim couples were reported to be using some contraception. Various other surveys also confirm that there is substantial contraceptive practice among Muslims (this is true in India and in several countries with large Muslim populations as well). However, the prevalence of practice is lower among Muslims than other SRCs in India and this is primarily responsible for keeping Muslim fertility above the average level. Use of contraceptives is known to be highly positively related to the level of education. Besides, as the level of education rises, the Muslim-non-Muslim differences narrow down.

Since the growth of the Muslim population has been higher than average in all the recent decades, there is a feeling that there is considerable international migration of Muslims into India. Detailed analyses for the decade 1981-91 showed that part of the higher than average growth of Muslims is accounted for by lower than average mortality, but a major part was explained by higher fertility. The contribution of migration, obtained as the residual, was relatively small, about one sixth of the growth differential between Hindus and Muslims. Other assessments also show that the contribution of migration to the growth differential is small. Thus, while international migration is also responsible for some of the growth in India’s Muslim population, it plays only a minor role; the principal factor is the higher than average fertility.

Demographic transition is the process of shift from a regime of high fertility and mortality to low fertility and mortality; this generally begins with mortality decline and is followed by fertility decline. Most of the developed countries in the world have gone through this and have reached very low mortality and fertility. India too, is in transition, with mortality having fallen considerably, and fertility dipping especially after 1970. Of course, mortality is not yet very low; life expectancy has crossed 60 years but is much less than in the developed world that shows expectancies above 75 years. Besides, the TFR is close to 3, above the value of 2.1 that corresponds to the low replacement level. Hence it could be said that though India is well into transition it is yet to complete the process.

The Muslim population in most states is well into transition. There is a good deal of correspondence between overall fertility and Muslim fertility in the states, although the latter is higher than the average. Generally, Muslim fertility is a notch higher than overall fertility in some states, and in a few others, it falls within the same range. The gap between Muslim fertility and overall fertility is quite low in Jammu and Kashmir, Madhya Pradesh and Andhra Pradesh. Clearly, the Muslim population in India is well into transition, especially in all the large states, though it is behind the average. There is obviously some lag in its transition. Other evidence shows that the lag is of 10-15 years, that is, the fertility of the Muslim population at a point in time is closer to the average fertility 10-15 years ago.

7. Child Nutrition

Another important indicator of social well-being is child nutrition. Child malnutrition significantly increases the risk of infant and child death, with some estimates suggesting that child malnutrition is responsible for half or more of child deaths in the developing world. Unlike infant and under-five mortality, which is lower among Muslims than among most other SRCs, Muslims are worse off than most other groups in terms of child under-nutrition. For instance, Muslims suffer from the highest rates of stunting and the second-highest rates of underweight children among all social groups. In general, though, the differences across the social groups are not overly large, indicating that child malnutrition and low birth-weight are pervasive across all SRCs in India.

The evidence shows that Muslim children are at a slightly higher risk of child malnutrition than Other Hindu children. However, they are less likely to be underweight or stunted than SC/ST children. But in two regions – the North and the East – the rate of low birth-weight babies among Muslims actually increased sharply between 1992-93 and 1998-99, with Muslims performing much worse than the all-group average for the two regions. There is an enigma in the finding that Muslims have an advantage over Other Hindus in infant and under-five mortality but suffer a disadvantage in child nutrition rates. This incongruence is difficult to understand as most factors that are associated with low rates of infant and child mortality (e.g., delivery and utilization of high-quality health services, high female literacy, and good hygiene and child feeding practices) are also typically associated with low rates of child malnutrition.

8. Future Population Growth Prospects

Some projections show that the size of India’s Muslim population would stabilize at about 320 million. Independent projections carried out to see how the results would vary if the replacement level is reached by 2041 yielded a figure of 340 million. Thus, the Muslim population is expected to rise, partly due to higher than replacement level fertility for some time and partly due to population momentum, to a level of around 320-340 million.

The question often asked is whether, and if so, when, will the Muslim population become the largest group? The counter position is that how does it matter which population is the largest. However, given the political and social environment, the debate continues and there is much speculation on this matter. A recent work (Joshi, A.P., M.D. Srinivas and J.K. Bajaj (2003), Religious Demography of India, Chennai: Centre for Policy Studies) examined this issue and by extrapolating the trends of the twentieth century, arrived at the conclusion that in India, the Muslim and Christian populations together would be close to the 50% mark around the year 2050. But this is for India including Pakistan and Bangladesh, that is, the pre-Partition area of India. There are two problems with this exercise. First, it fits a cubic function to the share of population and this is used to extrapolate the share of a community (the authors use the term Indian religionists to include Hindus, Sikhs, Buddhists, Jains as one group and the other group includes Christians and Muslims), and such a curve becomes steeper as time passes. Second, it assumes that the current trends would continue in the future. But now that fertility decline has been established among all communities, the fertility gap is seen as a transitory matter. As the process of fertility transition progresses, fertility would decline in all the large communities; once some communities reach a low level of fertility further decline would be slow, whereas those lagging, such as the Muslims, would catch up. This would thus narrow the gap, and eventually all communities would reach low fertility as has occurred in much of the developed world. Essentially, a convergence is expected and the present gap in fertility and population growth is not likely to persist forever. The question is how long it would take for the gap to close and what would be the growth differential during this period.

In order to project the share of the Muslim population, projections for the total population are required. Earlier projections assumed that Muslims would reach replacement level fertility ten years later than other communities. The projections further showed that the share of the Muslim population in India would rise somewhat, to just below 19% (320 million Muslims in a total population of 1.7 billion) and then stabilize at that level. If it should take a longer time for the gap to close, the share of the Muslim population would be correspondingly higher. Alternate projections on the assumption that both the Muslim and non-Muslim fertility would reach the replacement level but the former would take 10 or 20 years longer showed that by 2101 the Muslim population may reach around 320-340 million in a total population of 1.7-1.8 billion and the Muslim population share would be between 18 and 19 percent. Broadly, one could say that the Muslim population share is expected to rise from the current level but not expected to be much above 20 percent by the end of the century.

Recent experience of European countries shows that fertility in many populations has fallen well below replacement level and population sizes have begun to fall rather than stabilize at some ‘ultimate level’ as was presumed in the past. At this time, it is difficult to say whether this would happen for India as well in this century. If it does, the population sizes of the total and Muslim populations at the end of the century would be lower than those given by the projections cited above (results of some alternative projections indicate that the total population could be below 1.5 billion and the Muslim population below 300 million by the end of the century).

The pace of convergence depends on a number of socio-economic, political and programme factors, and the process will be hastened with the spread of mass education especially amongst women and girls and a sustained reduction in poverty across all population groups in India. While religion is an important element influencing the lifestyles of sizeable segments of citizens, its impact on regulating the human fertility of Muslims is not strong. For example, the contraceptive prevalence rate among Muslims, an overt expression of acceptance of the modern concepts of family planning, has been increasing in recent years nearing, 40%. Over 20 million Muslim couples currently use modern contraception practices and this number will grow if quality and choice based reproductive health care services are made accessible to Muslims across India. However, the relatively higher incidence of poverty and the widening gap in literacy between the Muslims and other comparable SRCs, particularly among women at young ages, could in fact impede the decline in Muslim fertility. Excepting Kerala, other states in advanced stage of fertility transition such as Tamil Nadu, Karnataka and recently Andhra Pradesh have achieved noteworthy declines in fertility without major improvements in human development parameters. But practically all well-designed research across the world has pointed out that improvements in female education associated with declines in poverty levels will facilitate a faster decline in human fertility and improvement in life expectancy. Both the above factors are important as exclusive goals to be achieved.