The two National Family Health Surveys have shown a dramatic change in fertility data in Manipur. The total fertility rate in the state was the sixth lowest in the first survey while it was the seventh highest in the second survey. Almost all known major determinants of fertility do not support this dramatic change in fertility. Hence, the fertility rates in the state shown by NFHS II seem to be highly overestimated.

W KUMAR SINGH

T

Manipur is one of the smaller states which lies in the north-east corner of India. The demographic problem in the state is almost unexplored. If one looks at the population problem in the state carefully one can see the real picture quite differently from what it seems. The state has a 22 lakh population according to the 2001 Census. Though it is very small, the percentage of population in the state (out of the total population in the country) increased from

0.22 in 1991 to 0.23 in 2001 while that in Kerala it declined from

3.44 to 3.10 over in the corresponding censuses. The density of population in the four valley districts is so high (Appendix Table A) that their average density of population exceeds those in 448 districts (78.5 per cent of all districts) of the country and the density of population of all districts of Jammu and Kashmir, Himachal Pradesh, Uttaranchal, Rajasthan, Sikkim, Arunachal Pradesh, Nagaland, Mizoram, Tripura, Meghalaya, Assam and Chhattisgarh.The growth rate of population increased from 29.29 per cent in 1981-91 to 30.02 per cent in 1991-2001 which is the third highest in the country.

There has been a controversy in respect of the growth rate of population in the two hill districts – Senapati and Chandel – during 1991-2001. The decadal growth rates in these two districts were 81.96 per cent and 72.80 per cent while that in the state was 30.02 per cent. In some villages, for example, Paomata, Pural and Chakpikarong the growth rates were more than 100 per cent. But the growth rates in the three valley districts – Imphal East, Imphal West and Bisenpur were less than 20 per cent. Surprisingly, the district level total fertility rate (TFR) estimates (reference period – 1994-2001) given by Guilmoto and Rajan (2002) reveal that Senapati which had the highest growth rate had one of the lowest TFR (2.2). Imphal too had a TFR of 2.2. The TFR in Chandel was also lower than that in Ukhrul and Thoubal. To examine the possibility of reviewing the census report, a central team including the additional secretary and director of home affairs and the registrar general of India visited Manipur on September 15, 2003 and discussed the matter in the state capital (Thoudang,1 September 15-16, 2003). Accordingly, the government of Manipur started a renumeration/resurvey from October 7 to October 12, 2003 in the villages which were considered to show abnormal growth (Thoudang , October 5, 2003). But the people of the concerned districts did not cooperate with the resurvey.

The increase in growth rate in the state is also quite contradictory to the trends of the birth rate. The birth rate per 1,000 population given by the registrar general of India has declined continously from 29.1 (1984) to 19.4 (1990) and further to 18.6 (1999) [Directorate of Economics and Statistics: 40]. Such contradiction is also found in the case of National Family Health Surveys (NFHSs). In NFHS I the TFR in the state was very low (sixth lowest in the country). The other high performer states in respect of TFR were Goa (1.90), Kerala (2.00), Mizoram (2.30), Tamil Nadu (2.48) and Andhra Pradesh (2.49). The TFR in India was 3.39, while that in the state was only 2.76. On the other hand, there was a drastic change in fertility during 1990s. That means that in NFHS II, the TFR in the state ranks the seventh highest among the 25 states of the country. The highest TFRs were found in Meghalaya (4.57) and followed by UP (3.99), Rajasthan (3.78), Nagaland (3.77), Bihar (3.49), MP (3.31) and Manipur (3.04) respectively. In this second survey, the TFR in the state is higher than that in the country (2.85). Since solving the population problem highly requires reliable data on population, and fertility is the major determinant of population growth, the present paper is concerned only with the drastic change in fertility data of NFHSs.

This paper attempts first to explain the levels, trends and patterns in fertility in Manipur. Secondly, it tries to examine what are the factors responsible for this dramatic change in fertility in Manipur. Available secondary data mainly from the NFHS, the census and statistical abstract of Manipur are used. The study consists of four sections. The first section is the introduction. The second section analyses the trends and patterns of fertility. The third section examines the determinants of fertility. The last section presents the concluding remarks.

The fertility trends in the state along with the three southern states – Andhra Pradesh, Kerala and Tamil Nadu and India are shown in Table 1. The TFR (2.76) in the state is not considerably

Economic and Political Weekly April 15, 2006 different from that (2.59) in Andhra Pradesh. Similarly, the crude birth rates (CBRs) in both states are very close. These figures are below the national figures – 3.39 and 28.7, respectively. But in NFHS II, the CBR and TFR in southern states and India have declined while those in Manipur have increased. The gap of TFR between Manipur and Andhra Pradesh has increased from 0.17 to 0.79. Similarly, the gap between the state and Kerala has increased from 0.76 to 1.08.

Fertility by background characteristics reveals that the increase in mean number of children ever born (CEB) in rural areas is more than that in urban area (Table 2). CEB has increased in all educational levels whereas it has declined in the case of India. Both TFR and CEB are negatively related to the standard of living and educational level. The higher the standard of living and educational level, the lower the TFR and CEB. It implies that both educational level and standard of living have a depressing effect on fertility. Total fertility is the summation of age-specific fertilities. The most productive age group is in 19-24. But the age-specific fertility in this age group is lower than even that in Kerala which has generally been accepted as a model state for demographic study (Table 3).

Kerala has the lowest fertility rate in India which is even lower than some of the advanced countries. In higher age groups, the age-specific fertility in the state is higher than those in the southern states. Moreover, all age-specific fertilities except in the age group 20-24 have increased. It indicates that any policy attempting to reduce fertility in the state should target the higher age group, particularly in the age group 25-34. The increase in

Table 2: TFR for the Three-Year Preceding the Survey and Mean Number of CEB to All Women Age Group 40-49 by Selected Background Characteristics, Manipur and India

for the income elasticity for number of children is negative. If the net price of child variables such as the opportunity cost of the wife’s time in child rearing, cost of rearing and cost of child education are controlled (i e, included in the statistical model, the income elasticity became positive [Borg 1989].

While the Chicago-Columbia school stressed the demand side, some studies [Easterlin 1975; Easterlin, Pollack and Watcher 1980; Rosenzweing and Schultz 1985; Wolpin 1984] emphasised the importance of the interactive process between demand for and the supply of births. Supply-side of fertility is influenced by fecundity, child survival rates, female age at marriage, duration of lactation and mother and child healthcare which are again influenced by income and many other factors. For example, the higher the income, the higher the fecundity. Child survivals are likely to increase through health as higher income enables parents to better access to pre-natal and post-natal cares and general healthcares. The equality between supply and demand are caused by the use of contraceptives and other birth controls.

Social development theory stressed the role of social development [Caldwell 1980; Krishnan 1976]. Among the social development factors, female literacy (which has a significant influence on age at effective marriage) and health status as represented by IMR are emphasised. Since improvement in health status is considered to be a necessary condition of fertility decline even by the demographic transition theorists, the crux of social development theory was the importance of the female education. Another important school of thought which explains fertility change is diffusion theory which is also known as ideational theory as enunciated by Cleland. Demographers of this school [Cleland 1985; Retherford 1985; Coale and Watkinson 1986 and Cleland and Wilson 1987] tried to explain the fertility transition through the diffusion of information and new social norms about birth control. The diffusion of ideas and process come through social interactions and the spillover effect of communication, information, social network and dissemination. Moreover, diffusion is more effective within a linguistic boundary. Diffusion can take place through education, mass media and extension programme of development. Diffusion takes place under a wide variety of social, economic and demographic conditions and hence the decline in fertility. According to them the fertility transitions in Europe took place because of diffusion of knowledge of contraceptives rather than the result of socio-economic changes.

After surveying various literature on fertility transition Manson (1997: 444-45) rightly remarked that none of above theories provided a complete explanation for all known fertility declines. On a decadal scale, the classical transitional theory is frequently contradicted, despite being plausible on a millennial or centennial scales. In both developed and developing countries, the correlation between the level of urbanisation or industrialisation and the decade in which the nations or provinces first experience a fertility decline are weak. Even when in economic modernisation, a shift in values toward individualism and self-fulfilment that occurs with rising affluence and secularisation, the theory explains the transition in Europe but not in several developing countries. The neoclassical theory provides a quantifiable framework for examining fertility change, but as a theory it ignores the environmental and institutional conditions that change costs, income and preferences which also influence fertility decline. In the same way, the supply and demand theory framework contains few ideas about the institutional determinants of fertility decline. Like other theories the diffusion theory is also incomplete [Manson 1997: 444-45]. It is highly difficult to quantify the diffusions and hence still remains at the theoritical level. It tries to explain when a change has already occurred. But it cannot convincingly predict why such phenomena do not continue to occur in other parts or regions [James and Subramanian 2003:1222].

Thus a decline in fertility can take place under different socioeconomic and demographic conditions [Coale and Watkins 1986] and the decline in fertility may be attributed to different factors. The determinant varies from one state (region) to another state (region). The decline in fertility occurred along with high infant mortality rate of 221 per cent in Germany in 1890 and with low literacy rate of 40 per cent in Bulgaria in 1912 [Knodel and Walle 1979]. The achievement of fertility in Kerala was non-concurrent with industrialisation, increase in PCY and standard of living. Female literacy rate emerged as the single most important factor in explaining the variation in fertility in Kerala [Krishnan 1976 and Bhat and Rajan 1990]. However, the demographic transition in India could take place through accelerating the process of diffusion by making use of formal and non-formal communication channels [Bhat and Rajan 1990]. In Tamil Nadu the quick fertility declined without any decline in infant mortality rates, improvement in literacy rates and the bettering of the status of women. This fertility change was explained by better rural-urban linkage through road transport and poverty. The rural-urban connectivity through better transportation changed the attitude of rural people, without migration to urban area, to educate them and to realise that they could achieve higher standard of living and could cope with the increasing price situation by reducing the family size [Savitri 1994].

The reduction in fertility in Andhra Pradesh did not conform to the classical theories which necessitated considerable improvement in an economic or social development for reducing fertility. Rather, the reduction was the result of an effective governance of family planning programme and generous welfare measures (such as ICDS, public distribution of food which had reduced number of population below the poverty line dramatically) [James 1999 and James and Subramanian 2003]. An able government’s

Table 3: Age-Specific and TFR and CBR, according to Selected States, India

policies and efficient administration have a significant role to play in effective use of contraceptive and decline in fertility [James 1999]. The state level analysis revealed that in India there was a strong association between fertility and female education and mother and child health (particularly reproductive and child healthcare [Manna 1998]. On the basis of experiences of the Asian countries as well as the review of literature concerned, it is found that there was a significant effect of education on fertility [Tilak 1992]. In India modernisation or westernisation helped lower the traditionally high level of fertility. Westernisation proceeded through high level of literacy, urbanisation and industrialisation and a rise in the status of women. More recently, the spread of transport and communication had also been identified as a key factor influencing fertility decline. Among the factors, female literacy, urbanisation, the percentage of male workers engaged in non-agricultural activities, infant mortality rate and percentage of women classified as economically active, only female literacy rate was statistically significant. Female literacy was so important that if educational difference between the religions could be controlled, the inter religious group differences would quite disappear. The spread of TV had widened the awareness about higher standard of living enjoyed by the better-off sections which had also been one of the key factors leading to fertility decline [Visaria and Visaria 1994: 3285-91]. The states or regions lagging behind in fertility reduction seemed to be socially and economically backward [Shariff 1996]. The study of the fertility decline in Bangladesh revealed that it did not seem to be any unique relationship between the incidence of poverty and the direction of fertility decline. After a stage, poverty had a depressing effect on fertility. But the decline did not occur until critical parametric changes took place in terms of the land-man ratio, urban growth, literacy rate, production and family organisations, women’s position and health conditions and medical technology [Adnan 1998]. In terms of reducing total fertility, the income effect was greater in India than in other low or middle income countries. Though the income effect was significant for reducing fertility in India, it was less important as compared to non-income factors

– higher levels of female education and access to good healthcare [Measham et al 1999]. Thus no threshold and no panacea could be prescribed for all states or regions for a sharp reduction in fertility. The solution to the problem varies depending on the state, country or region-specific factors.

Thus, the cause of fertility change in Manipur is also to be investigated in its own way. Infant and child mortality rates are generally considered important factors in determining fertility. If these rates are high, parents are uncertain about the number of their children and they try to compensate the expected loss of infants and children by an increasing number of births. However, that the overall infant and child mortality has declined during the 1990s indicates there is no reason to expect that the parents would like to increase the number of children (Table 4). Infant mortality rate has declined from 42.4 to 37.0. Similarly, under five mortality has declined significantly from 61.7 to 56.1. This conclusion is also supported by Table 5 which shows the per cent distribution of currently married women by contraceptive method used. The percentage of currently married women who used any method of contraception increased from 34.9 in 1992-93 to

38.7 in 1998-99. Similarly, any modern method also increased from 24.1 to 25.9 in the corresponding periods. In both cases the increase is more in the rural than in the urban.

The percentage of female sterilisation has increased drastically from 10.9 to 14.4 while that of males declined. The increase in female sterilisation and decline in male sterilisation are more significant in urban areas. Female sterilisation has increased from

in the model state, Kerala. There, higher fertility is found in the higher age group. But, the increase in literacy rate is higher in the higher age group (Table 7). This contradicts the common accepted norm that higher literacy creates lower fertility. In other words, there is hardly possible an increase in fertility in the higher age group in the state. Moreover, the increase in the illiteracy in the lower age group (6-19) is also questionable in the context of increasing importance and recognition of universal literacy parts of the country and world. However, the present paper does not go in that direction.

The singulate mean age at marriage (SMAM) for males increased from 26.4 in 1971 Census to 28.3 in 1993 (NFHS I) and further to 28.6 in 1999-2000 (Table 8). Such increase is more significant in the case of females, where it increased from 22.2 to 25.0 and further to 25.4 in the corresponding years. According to SMAM measures, females in the state tend to marry about three years late in 1999-2000 as compared to the females in 1971. During the period between NFHSs, the increase in the age at marriage is about five months. The recent increase in the SMAM declines as compared to the previous decades. In spite of it, the increase in female SMAM does not support the increase in fertility.

It has already been noted that the demand for children may be increased in two ways – quantity and quality. With the increase in income, parents try to substitute the number of children for child quality and hence in the number of the children. Such a negative relationship is found in the same survey as mentioned elsewhere but not over the period of time. The growth rate of PCY increased almost continuously during 1980s and 1990s (Table 9). The average annual growth rate of PCY at current prices increased from 10.35 per cent in the first period (1983-91) to

10.95 per cent in the second period (1992-98). The average annual growth rate at constant prices also increased from 1.43 per cent to 3.13 per cent in the respective periods. It implies that with an increase in PCY, the fertility in the state would decline rather than increase.

The rural-urban connectivity through better road transportation and reduction in poverty is much related to the size of expenditure on the rural development. But the expenditure on rural development in the state has a declining trend. The peak expenditure of Rs120.89 lakh in 1990-91 declined drastically to Rs 9.90 lakh in 1998-99 (Table 10). It seems to have led to an increase in fertility. But if the size of population is taken into account such inference cannot be drawn. According to the directorate of rural development and panchayati raj, the population served by the rural development and cooperative was 13,76,814 [Directorate of Economics and Statistics 2001]. If this population is taken into account the per capita expenditure on rural development for the peak expenditure year, 1990-91 is less than Rs 9. Such a meagre amount cannot influence the variation in fertility in the state.

Status of women is also one of the important determinants of fertility. Usually the states where women play an important social and economic role have experienced rapid progress in improving health and reducing mortality and fertility. Kerala and Manipur are two striking examples. The empowerment of women has had different basis: the early promotion of female literacy played a crucial role in Kerala while other sources of female emancipation (including economic roles of women) have been the central role in the case of Manipur. Women in these two states have played a more active role than their sisters in the large north Indian states [Dreze and Sen 1999: 176]. The higher the status of women, the lower the fertility rate. The status of women in the state is high as compared to those in other states. Women in the state have been participating in most affairs of life. They are not confined to the kitchen and homestead only. The Nupi Lan (women’s agitation) in 1939 is well known for the importance and leading role played by women of Manipur in the emergence of a new Manipur after the end of the second world war. The movement was based on genuine fears and apprehension resulting from uncontrolled export of rice from the state. This led to a scarcity and an increase in price of rice by 200 per cent. The movement was originally confined to banning of rice export, but subsequently changed its character to become a movement for constitutional and administration reform in the state [Yambem 1976]. Meira Paibi (the torch bearer) and Nisha Band and many other

Table 7: Percentage Distribution of the Household Population Age 6 and Above by Female Illiteracy, Manipur

Note: *Calculated a age group 20-24 to 45-49 of NFHS-I are adjusted to make comparable.

Source: Same as 2 and 4 of Table 2.

Source Singulate Mean Age at Marriage (SMAM)

1971 Census 26.4 22.2 4.2 1981 Census 27.3 23.4 3.9 1991 Census 28.1 24.7 3.4 1993 NFHS-I 28.3 25.0 3.3 1999-2000 NFHS-II Urban 28.7 25.9 2.8 Rural 28.6 25.0 3.5 Total 28.6 25.4 3.2

Note: The table is based on the de jure population. The male SMAM minus the female SMAM may not correspond excactly to the difference in SMAM due to rounding.

Source: Same as 4 of Table 2.

Note: *Computed. Sources: (i) Directorate of Economic and Statistics, 1992.

Economic and Political Weekly April 15, 2006 women organisations have participated in maintaining law and order, banning of intoxicants and improving other socio-economic development in the state.

Meira Paibi has fought against the excess of security forces, pleaded for human rights, participated in saving and protecting the territorial integrity of the state [Singh 1999-2000]. In respect of women, Ima Keithel (mother’s bazaar) is in the heart of the capital city of Manipur which is hardly found in other states of India. In most markets business (mostly petty business) are run by women. For maintaining their family they run their business in any market whether it may be in rural or urban areas. Their petty fish business includes – clothes, vegetables, ornaments, fish, fruits, toys, etc. Like many other states women have participated in politics and become elected members particularly in municipal council, gram panchayats and zilla parishads. In every part of the state local Nupi Marup (women’s financial association) helps to improve their families’ financial position. Through such ‘marup’ they help themselves in buying jewelleries and household durables, taking mortgage of agricultural lands, construction of houses, etc. Sometimes they can even buy landed property with the help of marup. Most of dowry for their daughters are paid by joining such financial associations. At the state level there is also one women cooperative bank in the state capital, Imphal. Table 11 reveals the autonomy of women.

The demographic problem of Manipur is an unexplored area. It is because of the law and order problems, lack of reliable data and absence of a good transport and communication system. For a proper understanding of the population of the state, reliable data on demographic variables are required. But demographic data in the state is very poor. The recent two NFHSs have made

Note: – Not available. Source: Same as Table 10.

a good supplement to the existing data supplied by the sample registration system and census. The fertility trend in these two surveys has shown that the country as well as all states except Rajasthan and four north-eastern states – Manipur, Meghalaya, Mizoram and Nagaland – have a declining trend. The TFR in the state was the sixth lowest in the first survey while it was the seventh highest in the second survey in the country. In this context the present study mainly attempts to examine the reason for the dramatic change in fertility in the state during the two surveys.

Almost all known major determinants of fertility in theory and in the experience of other states and countries do not support the dramatic change in fertility in the state during the 1990s. Infant and child mortality rates during 1990s declined from 42.4 to 37.0 and from 20.2 to 19.9 respectively. Under five mortality has also declined significantly from 61.7 to 56.1. The percentage of currently married women who used any method of contraceptive has increased considerably. Similarly the use of any modern method also increased from 24.1 in 1992-93 to 25.9 in 1998-99. The percentage of female sterilisation increased significantly from

Appendix Table A: Ranks (by Descending Order) of District byDensity of Population, Geographical Area, Percentage ofGrowth Rate of Population, Literacy Rate, Percentage of Childand Size of Population, 2001

District Rank Rank Rank Rank Rank Rank (Density (Geogra-P C of (Literacy (P C of (Size of of Popu-phical Growth Rate) Child in Population lation per Area Rate of the Age in (000)) Sq Km) Sq Km) Population Group of 0-6)

Note: * Valley District. Source: Registrar General of India, 2001.

District Literacy Rate *(PC) 1991 2001** P M FPMF

Bisenpur 54.94 68.59 41.13 71.59 82.25 61.09 Chandel 46.68 57.39 34.00 57.38 66.12 48.57 Churchandpur 58.17 66.38 49.30 74.67 84.98 64.40 Imphal East 60.65 80.65 54.99 76.38 86.44 66.30 Imphal West 73.01 84.63 61.12 80.61 89.40 72.24 Senapati 46.04 55.26 36.13 50.47 56.39 40.04 Tamenglong 50.16 59.92 39.68 58.46 67.04 49.11 Thoubal 52.47 68.33 36.31 67.90 80.50 55.34 Ukhrul 62.54 72.11 51.57 68.96 75.40 61.91 Manipur 59.89 71.63 47.60 68.87 77.87 59.70

Notes: * Literacy rate is the percentage of literate to population aged seven years and above. ** Provisional.

Sources: (a) Directorate of Economics and Statistics, 1998.

Economic and Political Weekly April 15, 2006 10.9 to 14.4. The female literacy too increased from 34.61 per cent in 1981 to 47.60 per cent in 1991 and further to 59.70 in 2001. Singulate mean age at marriage for female has also increased marginally. The average annual growth rate of PCY at constant prices increased from 1.43 per cent (1983-91) to 3.13 per cent (1992-98). Rural connectivity through rural development expenditure in the state has no role to play in the state as per capita expenditure in the state has been less than Rs 9. Manipuri women have participated in all aspects of socio-economic activities and in maintaining law and order of the state. The Nupi Lan (women’s agitation) in 1939 has shown the historical participation of women in the state issues and movements for constitutional and administration reform even during the British period in India. They have played an important role in banning of Nisha (intoxicants) and maintaning law and order of the state and economic improvement of their families through various associations. The percentage of women beaten or physically mistreated since age 15 in the state is lower than those in Andhra Pradesh, Tamil Nadu and India. The percentage in Tamil Nadu is more than twice that in the state. The percentage not involved in any decision-making in the state is less than half those in Andhra Pradesh and Kerala. The percentage in the state is about one-third of the percentage in the country. The percentage of women with access to money in the state is higher than that in Andhra Pradesh and Kerala. The diffusion of knowledge of family planning has also increased during the two surveys. All these reasons suggest that the fertility figures in the state shown by NFHS II seem to be large overestimates. The NFHS data are to be used with care. Future surveys should take into account these shortcomings.

mr'

Adnan, Shapan (1998): ‘Fertility Decline under Absolute Poverty: Paradoxical Aspects of Demographic Change in Bangladesh’, Economic and Political Weekly, No 22, pp 1337-48.

Becker, G S (1960): ‘An Economic Analysis of Fertility’ in Demographic and Economic Change in Developed Countries, Universities-National Bureau Conference Series No 11, Princeton University Press, Princeton.

Becker, G S and H G Lewis (1973): ‘On the Interaction between the Quantity and Quality of Children’, Journal of Political Economy, 81, pp S279-88.

Bhat, Mari P N (1994): ‘Levels and Trends in Indian Fertility – A Reassessment’, Economic and Political Weekly, Nos 51-52, pp 3273-80.

Bhat, Mari P N and S Irudaya Rajan (1990): ‘Demographic Transition in Kerala Revisited’, Economic and Political Weekly, September 1-8, pp 1957-80.

Borg, Mary O'Malley (1989): ‘The Income-Fertility Relationship: Effects of the Net Price of a Child’, Demography, Vol 26, No 2, pp 301-10.

Bose, Ashish (1992): Population of India – 1991 Census and Methodology, BR Publishing Corporation, Delhi.

Caldwell, J C (1980): ‘Mass Education as a Determinant of Timing of Fertility Decline’, Population and Development Review, Vol 6, No 2, pp 225-55.

Cleland, J (1985): ‘Marital Fertility Decline in Developing Countries: Theories and the Evidence in J Cleland and J Hobcraft (eds), Reproductive Change in Developing Countries: Insights from the World Fertility Survey, Oxford University Press, Oxford, England.

Cleland, J and C Wilson (1987): ‘Demand Theories of the Fertility Transition: An Iconoclatic View’, Population Studies, Vol 41(1), pp 5-30. Coale, A J and Sushan Watkinson (1986): The Decline of Fertility in Europe, Princeton University Press, Princeton.