By relying on the number of (surviving) boys per hundred girls observed in the population censuses as a cumulative measure of differential mortality during birth, infancy, and childhood, this paper shows that average Italian child sex ratios (aged 0–4) were abnormally high between 1861 and 1921. Our estimations indicate that unexplained excess female mortality resulted in around 2–3 per cent of ‘missing girls’ during this period. Likewise, by constructing a new dataset on child sex ratios at the provincial level during the same period, this article shows that child sex ratios tended to be higher in Southern Italy, a geographical cleavage that became stronger as time went by. Crucially, the results reported here cannot be explained by registration issues because (1) the analysis holds if we focus on the sex ratios of older children (aged 5–9) and (2) these patterns are also clearly visible using death registers. Unexplained excess female mortality early in life disappeared from the 1920s onwards, thus suggesting that either discriminatory practices gradually vanished and/or that they no longer translated into higher mortality rates due to enhanced living standards.

中文翻译：

---

自由主义意大利失踪女孩，1861 年至 1921 年

本文根据人口普查中观察到的每 100 名女孩中的（存活）男孩数量作为出生、婴儿期和儿童期差异死亡率的累积衡量标准，表明意大利平均儿童性别比（0-4 岁）异常1861 年至 1921 年间这一数字很高。我们的估计表明，不明原因的过高女性死亡率导致这一时期约 2-3% 的“女孩失踪”。同样，通过构建同一时期省级儿童性别比的新数据集，本文表明意大利南部的儿童性别比往往较高，这种地理差异随着时间的推移而变得越来越明显。至关重要的是，此处报告的结果不能用登记问题来解释，因为（1）如果我们关注年龄较大儿童（5-9 岁）的性别比例，则分析成立；（2）使用死亡登记时，这些模式也清晰可见。从 20 年代起，生命早期不明原因的过高死亡率就消失了，这表明歧视性做法逐渐消失和/或由于生活水平的提高，它们不再转化为更高的死亡率。