The demographic change story is misplaced

Fertility has collapsed across every religion in India. Geography now explains more than religion does.

An explainer showing that fertility has declined sharply across all religious groups in India between 1992 and 2021, with the Hindu-Muslim gap narrowing from 1.1 to 0.42 children per woman, and state-level variation now exceeding religious variation in fertility.

The claim that India faces a religion-driven demographic upheaval no longer matches the data. Every religious group's fertility has collapsed. The Hindu–Muslim gap has more than halved in thirty years. The community the story is most often told about — Muslims — has recorded the sharpest decline of any group.

What remains is a regional story, not a religious one. A Muslim woman in Kerala has fewer children than a Hindu woman in Bihar. Education, income, and access to healthcare explain what religion no longer can.

One technical note before the data, since it often comes up: the fertility rate is measured per woman, not per household. A man taking multiple wives does not multiply it — each wife's fertility is counted independently. Polygyny itself is rare and declining in India. NFHS-5 (2019–21) puts it at 1.4% nationally, lower among Muslims (1.9%) than among Christians (2.1%), and driven overwhelmingly by tribal communities in the northeast.

Every religion is now at or below the replacement rate

Total Fertility Rate (TFR) — average children per woman — by religion across five rounds of the National Family Health Survey, 1992–93 to 2019–21. Only Muslims remain marginally above the 2.1 replacement line, and are converging fast toward it.

Muslim Hindu Christian Sikh
Source: National Family Health Survey, rounds 1 through 5 (1992–2021). Buddhist TFR fell from 2.13 (NFHS-2) to 1.39 (NFHS-5); Jain TFR fluctuated between 1.20 and 1.90. Both groups are well below replacement.
Muslim TFR decline
4.41 → 2.36
46.5% drop — sharpest of any religious group over the five NFHS rounds.
Hindu–Muslim gap
1.11 → 0.42
Gap has closed by more than half. On current trajectory, projected to converge by around 2030.
India TFR
3.39 → 2.00
Below replacement nationally. Only five states still above 2.1: Bihar, Meghalaya, UP, Jharkhand, Manipur.

A Kerala Muslim has fewer children than a Bihar Hindu

Within any one state, Muslim TFR does remain higher than Hindu TFR — but the within-state gap is dwarfed by the gap between states. Religion is a weaker predictor than geography, income, and education.

0 1 2 3 4 replacement (TFR 2.1) 1.53 Hindu 2.25 Muslim Kerala 2.88 Hindu 3.63 Muslim Bihar Kerala Muslim (2.25) is lower than Bihar Hindu (2.88)
Source: NFHS-5 state reports, 2019–21. The within-state Hindu–Muslim gap (0.72 in Kerala, 0.75 in Bihar) is much smaller than the between-state gap for a single religion — Bihar Hindu TFR exceeds Kerala Hindu TFR by 1.35 children.
Within-state gap
~0.7 children
Hindu vs Muslim, in either Kerala or Bihar — roughly the same small margin.
Between-state gap, same religion
~1.4 children
Bihar Hindu vs Kerala Hindu — nearly twice the religious gap within a state.
Education premium
2.8 → 1.8
TFR among Indian women with no schooling vs 12+ years. Bigger than any religious gap.

Hindus remain a strong majority. Muslim share plateaus near 18%.

Pew Research's demographic projection takes the fertility-convergence trajectory and extends it to 2050. The Sachar Committee, using different assumptions, reached a similar conclusion for 2100. Both projections assume trends continue — and as the first chart showed, they have.

2011 2050 2100 Census Pew projection Sachar (upper) Hindu 79.8% 14.2% 6.0% Hindu 77% 18% 5% Hindu ~75% ~21% ~4% Hindu Muslim Christian, Sikh, Buddhist, Jain, other
Source: Census of India (2011); Pew Research Center, The Future of World Religions (2015), projection for 2050; Sachar Committee Report (2006), long-run stabilisation estimate. A Hindu majority persists in every published projection.
Hindu share in 2050
~77%
Pew projection. Hindus remain a ~3 in 4 majority — essentially unchanged from 2011.
Muslim share in 2050
~18%
Up from 14.2% in 2011 — driven primarily by younger median age, not higher fertility today.
Long-run Muslim share
17–21%
Sachar Committee estimate for stabilisation by 2100, once fertility fully converges.

Every religious group's population peaks this century, then declines

A population below replacement fertility does not shrink instantly — a young age structure provides decades of momentum. But it does shrink. The UN projects India's total population peaks around 2064 at 1.7 billion, then falls to 1.5 billion by 2100. Every group's curve follows the same arc. Muslims peak later than Hindus, but they still peak.

Hindu Muslim
Populations in absolute millions. Derived by distributing UN World Population Prospects 2024 (medium variant) India totals across Pew Research religious shares, interpolated linearly between published waypoints. Even at its peak, the Muslim population is roughly a quarter the size of the Hindu population. Smaller groups — Christians, Sikhs, Buddhists, Jains — follow similar arcs and peak earlier, since their TFRs are lower.
Hindu
Peaks ~2060
Grows from 1.12 B to ~1.30 B, then declines to ~1.13 B by 2100. Net change over the century: essentially flat.
Muslim
Peaks ~2075
Grows from 199 M to ~327 M, then declines to ~302 M by 2100. Peaks later than Hindus — still peaks well below Hindu population.
Christian
Peaks ~2060
With a TFR of 1.88, Christian population grows modestly, peaks, then drifts down. Projected share: ~2% throughout.
Sikh, Buddhist, Jain
Already peaking
All three well below replacement (TFR 1.39–1.61). Populations plateau near-term and decline through the rest of the century.

How high would Muslim fertility have to be to overtake Hindus?

Move the sliders. The model compounds the two fertility rates across generations, starting from the 2011 Census shares (Hindu 79.8%, Muslim 14.2%), and calculates when, if ever, the Muslim population would exceed the Hindu population. The assumption is deliberately ungenerous to the convergence argument: rates never change once set.

Muslim TFR 2.36
Current: 2.36 (NFHS-5) · Historical peak: 4.41 (NFHS-1, 1992–93)
Hindu TFR 1.94
Current: 1.94 (NFHS-5) · Historical peak: 3.30 (NFHS-1, 1992–93)
Scenarios:
At these rates, Muslim population would overtake Hindu population in 264 years — around the year 2285.
That requires 8.8 consecutive generations at this exact fertility gap, with neither rate ever moving. Both have moved every single NFHS round since 1992.
Hindu share of Hindu+Muslim total Muslim share of Hindu+Muslim total
0% 25% 50% 75% 100% 50% — crossover threshold 2021 2071 2121 2171 2221 2271 2321 84.9% 15.1% 44.2% 55.8% Crossover: 2285
Share of Hindu+Muslim combined population (other religions held separate). Generation length: 30 years. Starting from 2021 Census shares. Model assumes fertility stays constant — the scariest possible scenario for replacement-anxiety framings. In reality, both rates continue to fall.

Fertility in India is not a religious phenomenon

Every religious group's fertility has collapsed. Muslim fertility has fallen further and faster than any other's, and the Hindu–Muslim gap has more than halved since 1992. Within any given state, a small Hindu–Muslim gap persists — but that gap is consistently smaller than the gap between the same religion in a poor state and a rich state.

What drives fertility in India is what drives it everywhere: female education, household income, healthcare access, age at marriage. Kerala crossed replacement fertility in 1988, when India's TFR was still 4. That order of magnitude — the lead a well-governed state can have over a badly-governed one — is larger than any gap religion has ever produced.

The "demographic replacement" story requires every group's behaviour to diverge from the trajectory it has been on for thirty years, and to keep diverging for centuries. Run the model above — even scenarios designed to be hostile to the convergence play out past the horizon of any living human. The actual data shows the opposite is happening: every religious group's population peaks this century and then declines. The majority stays a majority. The minority converges. The real fertility questions now — about ageing, labour force participation, and care — are not about religion at all.