Shantou surrenders its entire fertility premium in four years

Take the Shantou series on its own terms: 5.31 fewer births per 1,000 residents in four years ^[1], which is 40.2 per cent off the 2021 base, or about 12 per cent a year compounded ^[8]. Jieyang lost more in absolute terms, 6.85 per 1,000, but took six years doing it, so its pace of 1.14 per 1,000 a year ran slower than Shantou's 1.33 ^[11]. Shantou, the city with the loudest claim on the clan tradition, fell fastest.

Read the two series against each other and you get a price for the culture. Shantou's 2025 rate is the provincial average ^[1]; Jieyang's 9.35 sits 1.46 per 1,000 above it, a premium of 18.5 per cent ^[9]. That 1.46 is the entire measurable remainder of the clan structure, the ancestral halls and sibling networks, that the South China Morning Post describes as the region's defining feature ^[3], and residents and demographers quoted in the same report say it will not hold without stronger economic support ^[5].

The part that matters for capital is not the birth count. Those sibling networks were the region's informal credit system, the safety net behind family businesses, new ventures and a far-flung overseas diaspora ^[4]. Sherry Chen, 32, told the paper she would have three children if she could afford them and that two is now common among her peers back home ^[6]; going from three to two means roughly a third fewer people per household available to guarantee a loan, cofound a venture or inherit one ^[12], and it arrives in a place where that network was doing work a bank was not.

Now the reason this could be wrong. Everything reported here is a crude rate per 1,000 residents, for 2019, 2021 and 2025, with no total fertility rate and no age structure attached ^[10]. A region that exports its 25-to-34 cohort to Shenzhen and keeps its grandparents will post a collapsing crude rate with no change whatever in how many children a Chaoshan woman has, because the births register somewhere else and the denominator ages in place. The base years also differ between the two cities, 2021 for Shantou against 2019 for Jieyang ^[1][2], so the two declines are not strictly comparable.

This is probably wrong in its strong form, but the useful version is narrower: if the strongest cultural scaffolding in the country bought four years of convergence and nothing more, then culture is not an available lever, and the levers that remain all cost money. Or rather, the more interesting version is a test. If norms and cash are complements, a fixed per-birth transfer should buy more births in Jieyang than in the provincial average county, because the network that supports a third child is still standing. If they are substitutes, the same yuan buys the same nothing everywhere, and the 1.46 gap closes on schedule.

The counter-thesis deserves stating plainly: perhaps clan culture never held fertility up at all, and Chaoshan's premium was just later urbanisation and lower incomes arriving on a lag, in which case convergence is mechanical and implies no subsidy at all. Either way, someone is about to price a birth. The going rate, if you believe Jieyang is the last holdout, is whatever it costs to defend 1.46.