(Mis)perceptions about children
Under review
I quantify how early childhood teachers' perceptions of developmental delays vary with their reference group. Using standardized non-cognitive and cognitive development measures for Australian children aged 4--5, I find that, conditional on a child's own development, teachers in neighbourhoods with lower average non-cognitive development are less likely to perceive delays. The estimated neighbourhood gradient is large enough to reverse the apparent ranking of neighbourhoods. Neighbourhoods with higher measured non-cognitive development have higher reported non-cognitive delay rates, but lower predicted rates when neighbourhood-average development is held at a common value in the fitted model. Teachers' perceptions are reflected in mothers' beliefs, which in turn predict uptake of therapy. Finally, teachers with university degrees are more likely to report delays among children with low measured development in both domains, while more experienced teachers are more likely to report non-cognitive delays among children with low measured non-cognitive development.
The evolution of prices and quantities of occupational human capital
Accepted at the Journal of Human Capital
I examine the roles of inter-cohort human capital growth and technical change in the rising U.S. wage inequality between 1970 and 2022, estimating the evolution of prices and quantities of human capital in occupational groups specializing in abstract, routine, and manual tasks. To quantify these changes, I use a flat spot price identification method that exploits wage changes for workers approaching retirement to infer shifts in prices of occupational human capital. Importantly, this method allows both the level and career accumulation profile of human capital to differ flexibly across cohorts. The results show that the price for abstract occupations has increased relative to manual and routine occupations, while prices for manual and routine occupations have declined in absolute terms. These price changes are consistent with the predicted role of technical change in increasing the relative demand for abstract tasks and decreasing the price of routine tasks. The relative wage growth of manual workers compared to routine workers has been driven primarily by the growth in the quantity of human capital linked to educational upskilling in manual occupations among younger cohorts rather than by an increase in the relative price of manual tasks.
Female-favourable gender gaps in multiple measures of academic achievement, among children and young adults, have increased over recent years. These disparities have been linked to deficits in boys' literacy, which manifest early in childhood and accumulate over time. I investigate the impact of parental time investment decisions on the widening of literacy gaps between boys and girls. I estimate a model of the mother's time investment and child skill accumulation, allowing for gender differences in the literacy production function, initial endowments, and parental preferences for children's human capital. My analysis centers on the development of skills in children aged 6 to 15 as observed in the Longitudinal Study of Australian Children. I document that mothers tend to allocate more time to their daughters, and although my point estimates suggest that these differences in maternal time investment are mainly explained by the mother's preferences for literacy between boys and girls, these estimates are not statistically significant. Overall, my findings suggest that the role of mothers' aggregate time investments in the expansion of literacy deficits may be limited, and this expansion is driven by productivity differences unrelated to aggregate time investment.
The cost of investing in children shapes how much families invest in their skills. Measuring this cost is challenging because child investment combines multiple inputs, including parental time, goods, and childcare, each with its own price, and families rely on different mixes of these inputs depending on the technology through which investments produce skills. We develop an approach for combining large, disjoint datasets, the Consumer Expenditure Survey, the American Time Use Survey, and the American Community Survey, to estimate relative demand for these inputs and construct the unit price of total investment for American children from 2005 to 2019. We document how this price, and the investment it buys, varies across families and over time.