In today’s context, fertility rates play a crucial role in shaping population growth and women’s reproductive health. The fertility rate refers to the total number of children a woman would bear by the end of her reproductive period, based on age-specific fertility rates for a given year (The World Bank … 2024). These rates vary both between and within regions, communities, and countries due to a range of influencing factors.

This study examines 25 Indian states and utilizes statistical data from all five waves of the National Family Health Survey (NFHS), conducted between 1992–1993 and 2019–2021. For the first time, the influence of three key factors on fertility across these states has been estimated. Analysing differences across India is essential due to the substantial demographic, social, and economic diversity among states.

According to the Sample Registration System (SRS), fertility indicators such as the total fertility rate (TFR) have shown a marked decline: from 5.2 to 4.5 between 1971 and 1981, and from 3.6 to 2.0 between 1991 and 2020 (SRS 2013; SRS 2020). In rural areas, the TFR decreased from 5.4 to 2.2 between 1971 and 2019, while in urban areas it fell from 4.1 to 1.6. Between 1990–1992 and 2019–2021, the national TFR reached two children per woman – the replacement level of fertility (Our World in Data, 2024), according to NFHS data (IIPS 1995; IIPS 2022).

However, in states such as Bihar, Meghalaya, Uttar Pradesh, and Manipur, the TFR remains above the replacement level (over 2.1), whereas in states such as Assam, Madhya Pradesh, and Rajasthan, it has already fallen below this threshold. Additionally, projections from SRS and IIPS suggest that India’s TFR, which declined from 3.1 in 2001 to 2.0 in 2020, is expected to decrease further to 1.7 by 2041.

This study focuses on five rounds of the NFHS, including the most recent NFHS-5 (2019–2021). Previous research has not conducted a cross-sectional analysis comparing different survey periods across states. In contrast, this study provides a comprehensive and precise assessment of the factors influencing fertility across 25 states.

The objective of this paper is to offer a detailed analysis of fertility differentials and to statistically estimate three key factors contributing to fertility differences across India, using data from all five rounds of the NFHS.

Few general theories have been proposed to explain fertility transition.

In the last 25 years of the 20th century, industrialized countries experienced a new phase of fertility transition, in which birth rates or fertility rates fell below replacement levels, accompanied by a decline in death rates or mortality rates. This phase is known as the Second Demographic Transition Model (Lesthaeghe 1991; Van de Kaa 1987). The model is characterized by changes in variables such as marriage patterns, fertility, contraception, and lifestyle in European countries.

Changes in marriage patterns include the postponement of marriage, an increase in the age at first marriage, a rise in non-marital cohabitation, higher divorce rates, and a growing number of couples having only one child. In terms of fertility, the age at which women have their first child has increased, and the age group with the highest fertility has shifted upward. Greater access to contraception and birth control methods has enabled couples to plan the number of children they have. Finally, changes in lifestyle have been linked to declines in mortality rates.

R. Easterlin developed a model explaining the association between fertility levels and relative cohort size. This relationship is influenced by fertility rates and changes in relative income (such as wages and unemployment). The theory suggests that fertility follows a cyclical pattern driven by both sociological and economic factors, where large birth cohorts tend to produce smaller subsequent cohorts. According to this hypothesis, a small cohort (within a 20-year period) tends to experience higher fertility, better job market prospects, and a higher standard of living. Conversely, a large cohort size, after 20 years, limits intended fertility, delays marriage and childbearing, and ultimately results in lower fertility (Doliger 2004).

Caldwell’s theory of intergenerational wealth flows (Caldwell 1976) highlights the fertility transition from pre-industrial to industrial and post-industrial societies, with a focus on the economic role of children. In less-developed economies, particularly agricultural societies, children are considered economic assets, contributing labor to the family. In contrast, in highly developed countries, the rising costs of raising and educating children make them financial liabilities. This theory also emphasizes how social and cultural norms in traditional and agrarian societies often encourage larger families, whereas modern societies tend to favor smaller families.

As societies transition from pre-industrial to post-industrial stages, fertility rates tend to decline. This shift is particularly evident as education becomes more accessible, especially for women. Educated women often delay marriage and motherhood in favor of career development and personal growth. Fertility rates decrease as women gain greater economic and personal autonomy, along with increased access to family planning. According to this theory, both families and societies worldwide frequently view fertility as a sign of wealth and prosperity. Moreover, the theory provides a valuable framework for formulating policies that promote social justice, sustainable development, and gender equality (Kaplan & Bock 2001).

The quantity-quality trade-off was explained by Gary Becker in his “Becker model.” According to this approach, children are viewed as commodities that provide utility through both their quantity and quality. The “quality” of children refers to the investments and resources parents allocate to them, such as education and upbringing. Wealthier parents can afford to invest more in their children, as expenditures typically increase with income. However, as income rises, the demand for additional children tends to decrease. In the United States, the 20th century witnessed a rise in child quality, driven by growing demand for educated workers. As a result, many parents chose to have fewer children, prioritizing higher quality of upbringing and education over the number of children (Doepke 2015).

Before the onset of industrialization, women in India had, on average, five to six children over their lifetime. Fertility rates remained high into the early 19th century but began to change as fertility behavior varied across states and regions. By the second half of the 20th century, fertility started to decline rapidly. By 2019–2021, fertility rates had dropped to an average of two children per woman (IIPS 2022), well below the replacement level, marking a significant fertility transition during this period.

This decline in fertility rates is linked to several factors, including a rise in the age at first marriage, the prevalence of non-marital cohabitation, higher divorce rates, an increase in the number of couples having only one child, a shift in the age at first birth, and a change in the peak fertility age group from 20–24 years to 25–29 years. Additionally, widespread access to contraception and birth control, along with lifestyle changes, has played a major role in this fertility transition (Lesthaeghe 1991; Van de Kaa 2002).

In pre-industrial India, children were primarily regarded as economic assets, and social and cultural norms favored larger families. However, after industrialization, the cost of raising and educating children increased, turning them into financial liabilities for parents, who consequently preferred to have fewer children. Moreover, as education became more accessible for women, they gained greater autonomy in decisions regarding the postponement of marriage, utilization of family planning, and prioritization of their careers and personal development (Caldwell 1976; Kaplan & Bock 2001).

The analysis of fertility patterns and differentials, along with an examination of factors such as caste, wealth index, and mortality rates influencing these patterns, was conducted using data from all five rounds of the National Family Health Survey (NFHS) in India: NFHS-1 (1992–1993), NFHS-2 (1998–1999), NFHS-3 (2005–2006), NFHS-4 (2015–2016), and NFHS-5 (2019–2021). Missing data from intervening years were estimated through interpolation.

The dependent variable in this analysis is the total fertility rate (TFR), representing the number of children a woman would have by the end of her reproductive period if she were to bear children according to current age-specific birth rates. The explanatory variables include residence, education, religion, contraceptive use, birth order, age at first marriage, birth interval, unmet need for family planning, and sex preference.

The present study analyzes the patterns and differentials of fertility, as well as the influence of key factors – residence, education, religion, contraceptive use, birth order, birth interval, unmet need for family planning, and sex preference – on fertility in the states of India using data from five rounds of the NFHS. According to the latest NFHS data, fertility in India is below the replacement level (2 children per woman) overall, as well as in both rural (2.1 children per woman) and urban (1.6 children per woman) areas. However, in some states such as Bihar, Meghalaya, Uttar Pradesh, and Manipur, fertility remains above the replacement level. Therefore, it is essential to examine the factors influencing fertility patterns and differentials among women of reproductive age.

Fertility among women of reproductive age is higher in rural areas than in urban areas, with the highest levels observed in the 20–24 age group. Fertility declines after the 30–34 age group and approaches nearly zero by ages 45–49. Additionally, the TFR has shown consistent rural–urban differences across India. In 1992–1993, the TFR was 2.7 in urban areas and 3.7 in rural areas, compared to 1.6 and 2.1, respectively, in 2019–2021. These differences are likely associated with variations in the age at first marriage, contraceptive prevalence, desired family size, women’s socio-economic status, participation in non-household employment, and broader social and cultural norms related to childbearing.

Consistent with previous research, the present study also finds higher fertility in rural compared to urban areas. This result aligns with earlier studies conducted in West Bengal, India (Chouhan et al. 2020), Ethiopia (Dana 2018), and Nepal (Adhikari 2010). These disparities may be attributed to the higher financial cost and greater parental investment required to raise and educate children in urban settings compared to rural ones (Adhikari 2010). Women in urban areas generally have better access to services and resources, which promotes higher contraceptive use (Adhikari 2010; Retherford & Thapa 2003). Furthermore, rural women tend to marry at younger ages than their urban counterparts (Zarate 1967).

Women in the reproductive age group of 15 to 49 years with higher levels of education tend to have lower fertility rates compared to those with only primary or no education. Our findings are consistent with previous studies conducted in India, Nepal (Adhikari 2010), and Uganda (Ariho & Nzabona 2019). Educated women are more likely to delay marriage (Reddy 2003), extend the interval between marriage and childbirth (Roy & Hossain 2017), use contraception, and have smaller family sizes compared to uneducated women (Castro Martin 1995). Additionally, education significantly influences women’s fertility behavior (Bbaale & Mpuga 2011), since women who attain secondary education or higher are more exposed to information, have greater autonomy in employment and marriage-related decisions, and are more aware of their own reproductive health and that of their children (Adhikari 2010).

According to previous studies (Balasubramanian 1984; Das & Pandey 1985; Arokiasamy 2002; Mari Bhat&Zaveri 1999; Adhikari 2010), Muslim women had higher fertility than Hindu women. However, our findings reveal that Hindu women exhibit higher fertility rates than Muslim women, although the coefficient for Muslim women is not statistically significant.This fertility rates differ among religious groups and it may be attributed to religious restrictions, cultural conservatism, and a lack of adequate knowledge or misconceptions about modern family planning methods, all of which can influence the acceptance or rejection of family planning (Albsoul-Younes et al. 2003). Moreover, low levels of education among Muslim women have been identified as a key factor contributing to higher fertility (Jeffery & Jeffery 2002). Additional factors such as early marriage, strong son preference, and the practice of polygamy may also contribute to higher fertility among Muslim women, as they can lead to competition among co-wives and consequently increase fertility levels (Sargent & Cordell 2003; Tafforeau et al. 1990).

Globally, contraceptive use has been a key determinant of fertility. However, our study found that the use of contraception has an insignificant effect on fertility. This may be because women who use contraception tend to have fewer children than those who do not, and contraceptive adoption often increases with rising levels of education. Additionally, women residing in urban areas are more likely to use contraceptives due to better access to health-related facilities (Retherford & Thapa 2003).

Another important finding of this study is that birth order is positively associated with fertility. Women with higher birth orders tend to have larger families than women at the first birth order, a result consistent with previous research in India. Children born at higher birth orders are also at greater risk of dying during infancy and early childhood (Determinants and Consequences… 2010; Howell et al. 2016).

The age at first marriage is positively associated with fertility, though the association is not statistically significant. A lower age at marriage tends to extend the childbearing period, and, in the context of lower contraceptive use, this often results in higher fertility rates. Several studies in India have observed a decline in fertility rates as the mean age at marriage increases (Borkotoky & Unisa 2015; IIPS 2007). Conversely, older age at marriage is associated with reduced fertility, as women who marry later are more likely to have attended school, be employed, use contraceptives, and maintain more equitable relationships with their husbands (Som & Mishra 2020; Bbaale & Mpuga 2011). In contrast, early marriage often leads to higher fertility, particularly in developing countries with limited contraceptive use. This situation can reduce access to education and increase the likelihood of high-risk or unwanted pregnancies, negatively affecting the health of both mother and child (Balakrishnan 1987).

Research findings suggest that birth interval has an insignificant direct effect on fertility. However, longer birth intervals are associated with improved child survival rates. A preceding birth interval of less than 18 months doubles the risk of infant mortality compared to intervals of 36 months or longer (Fotso et al. 2013; Potter 1988). Women who already have a son tend to have longer birth intervals than those without a son (Khan et al. 2016). Conversely, the death of a preceding child shortens the interval before the next birth, as women often attempt to conceive again sooner (Khan et al. 2016; Kamal & Pervaiz 2012). Women from high-wealth families tend to have significantly longer birth intervals, influenced by factors such as education and lifestyle (Kamal & Pervaiz 2012). Less-educated women generally have shorter birth intervals compared to their educated counterparts, who are more likely to use contraception to extend birth spacing. This difference is largely due to greater awareness among educated women of the negative effects of short birth intervals and the benefits of smaller family sizes (Som & Mishra 2020; Tulasidhar 1993).

Our study also found that unmet need for family planning has a significant effect on fertility rates. An increase in unmet need for family planning is associated with higher fertility (Akram et al. 2020). The unmet need for spacing is particularly common among younger women, as observed in NFHS-2 (1998–1999) in Maharashtra. This trend may be attributed to a higher proportion of illiterate married women, who often lack access to information and may fear potential side effects. Additionally, barriers related to contraception – such as limited availability and low awareness – along with fertility-related factors such as lactational amenorrhea, desire for more children, and sex preference, also play a role. Opposition from husbands and other family members, religious beliefs, and low socio-economic status further contribute to the persistence of unmet need (Akram et al. 2020; Wulifan et al. 2017; Rahman 2016; Patil et al. 2005; Khokhar & Mehra 2005; Westoff 1978).

With regard to sex preferences, our study found that women who desire more sons than daughters exhibit higher fertility rates. Son preference is particularly pronounced among older women, who often have more children and may have experienced the death of a child (Barman & Sahoo 2021). Sons are frequently viewed as productive assets in family farms or businesses, as providers of security during emergencies and in their parents’ old age, and as successors who carry on the family name and perform various ancestral rites. Additionally, analyses of state- and district-level data suggest that lower son preference in the southern states of India is an important factor contributing to reduced fertility rates in those regions (Drèze & Murthi 2001).

This study revealed significant patterns and differentials in fertility across age groups in India. Fertility differences were observed across multiple demographic categories and were influenced by several key factors. The main determinants shaping fertility patterns and differentials include place of residence (rural vs. urban), education level (no education, primary, high school or above), religion (Hindu), birth order (particularly first birth order), unmet need for family planning (especially unmet need for spacing), and sex preference (a desire for more sons than daughters or more daughters than sons).

Based on the findings of this study, it is recommended to strengthen efforts aimed at reducing fertility disparities among the states of India. Therefore, government agencies and relevant organizations should assess and implement policies that effectively address these differentials.

NFHS: National Family Health Survey

TFR: Total Fertility Rate

ASFR: Age-Specific Fertility Rate

ADF: Augmented Dickey – Fuller

OLS: Ordinary Least Square

FE: Fixed Effect

RE: Random Effect

FGLS: Feasible Generalized Least Square

LM: Lagrange Multiplier