Purpose – The purpose of this paper is to take a forecasting approach to examine the relationship between the US birth rate, marriage rate, and economic conditions (measured by both realized unemployment and expected unemployment). The expectation data come from the Michigan Surveys of Consumers. Design/methodology/approach – Utilizing monthly data, the authors first specify a univariate and three augmented autoregressive integrated moving average forecasting models for 1975-2001. Second, the authors use recursive estimation to generate multi-period forecasts of the birth rate for 2002-2008. Third, the authors employ standard evaluation methods to compare the predictive information content of the forecasts. Findings – First, the birth rate is pro-cyclical. Second, the marriage rate contains useful predictive information for the birth rate. Third, controlling for past information in the birth and marriage rates, both realized and expected unemployment embody useful information for predicting the birth rate. Fourth, expected unemployment is a more informative indicator than realized unemployment. Practical implications – The finding that the birth rate is pro-cyclical emphasizes the importance of economic stability in promoting childbearing, and the authors suggest counter-cyclical macroeconomic policy to shield families from major shocks. A stable economy, and especially one where families are optimistic about the future, promotes childbearing. The results also empower policymakers to analyze systematically the impact of changes to the structure of marriage on childbearing. Originality/value – This appears to be the first study that utilizes a forecasting approach to better understand the complex relationships between childbearing, marriage, and macroeconomic conditions. [ABSTRACT FROM AUTHOR]