234 CHILD WELFARE ² Household Resource Allocation in Stepfamilies: Darwin Reflects on the Plight of Cinderella By ANNE CASE, I-FEN LIN, AND SARA MC LANAHAN * The living conditions of American chil- dren have changed dramatically during the past 50 years. In 1950, the vast majority of children were born to married parents and lived with both parents until they reached adulthood. Today, over a third of all children are born to unmarried parents, and over half will live apart from at least one of their par- ents before reaching age 18 ( Larry Bumpass and James Sweet, 1989 ) . A majority of these children will live with a stepparent or foster parent, and a substantial minority will ex- perience multiple family disruptions and multiple changes in parental personnel. Numerous studies have shown that remar- riage is not the panacea it was once thought to be. Children who grow up in two-parent fam- ilies consisting of a biological parent and a stepparent have outcomes very similar to chil- dren who grow up with only one parent, and worse than children who are raised by both of their biological parents. These differences can- not be explained by differences in income, since stepfamilies and original two-parent families enjoy similar levels of income. One explanation for why children in step- families do poorly is that stepparents are less altruistic and, indeed, may be quite hostile to- ward nonbiological children. According to the theory of parental solicitude, such feelings are rooted in psychological mechanisms that have ² Discussants: Julie Nelson, Brandeis University; Robert Pollak, Washington University – St. Louis. * Princeton University, Princeton, NJ 08544. We thank Angus Deaton, Karla Hoff, Jonathan Morduch, Christina Paxson, and seminar participants at Princeton University for helpful comments and suggestions. We gratefully ac- knowledge the support of the John D. and Catherine T. MacArthur Foundation. evolved over time through the process of nat- ural selection. Thus, while household incomes may be similar for biological and nonbiologi- cal children, investments in children may be quite different. This paper examines differences in one input, food consumption, between differ- ent types of two-parent families, using data from the Panel Study of Income Dynamics (PSID). These data contain information on all dyadic relationships within households between 1968 and 1985, and they allow us to identify several types of parent–child re- lationships, including biological, adoptive, step-, and foster relationships. The paper addresses three questions: (i) Does the number of nonbiological children in the household affect food consumption? ( ii ) Do the effects vary by the sex of the non- biological parent? ( Is having a step- mother worse than having a stepfather?) (iii) Do the effects vary by type of non- biological parent? (Are there differ- ences among adoptive, step-, and foster parents?) Psychologists Martin Daly and Margo Wilson (1987) offer a powerful explanation for why nonbiological parents may be less willing to invest in children than biological parents. Drawing on evidence from evolution- ary biology, they argue that parental invest- ments ( time and money ) are costly, and psychological mechanisms have evolved to promote altruism toward one’s own offspring and aggression and hostility toward other peo- ple’s offspring. According to these research- ers, natural selection favors adults who engage in nepotism and who reject ‘‘parasitic’’ efforts