Marital fertility
We next examine the effects of rice prices, using different time lags, on the probability of having a recorded marital birth among women aged 15—49 in the two villages (table not shown).
We find that marital fertility (to be more precise, the likelihood of having a recorded marital birth) was mostly unaffected by annual fluctuations in local rice price, although the effects were always negative, as expected. We also do not see any notable differences in the effects of logged rice price by sex of birth. Our earlier study on marital fertility in the two villages found the widespread use of sex-selective and parity-specific infanticide to achieve a relatively small family size with sex-balanced offspring set (Tsuya and Kurosu 1999). In this earlier study, we also found that the fertility effects of rice prices remain insignificant, even after controlling for many family-related characteristics including the number and sex-composition of surviving children. Altogether, we interpret the lack of response of marital fertility to local economic fluctuations to imply that, in these villages, strong family control was not limited to times of economic hardships, but was practised regularly.27As shown in Table 17.4, the effects of local rice price variations on marital fertility remain insignificant after introducing household landholding (and other covariates) into the model. However, household landholding significantly increases the likelihood of having a recorded marital birth, especially a female birth. This suggests that women in better-off households were more likely to have a recorded female birth than were women in poor households. This in turn implies that living in wealthier households with more resources, women could afford to keep a baby, especially a female baby, without resorting to infanticide.
Finally, we can see that there are significant temporal differentials in marital fertility. The likelihood of a recorded marital birth became significantly higher in the nineteenth century, especially during the last three decades of the Tokugawa era (1840—70). As explained before, the rigid social stratification became increasingly unstable during the last decades of the era, and the gender hierarchy accordingly became more flexible which in turn may have enabled wives/ couples to keep female babies.
5.3