1 Regional Variations in the Relationship of Diet and Income in the United States and Brazil, 1996 Ronald Calitri Dissertation, Economics Department New School for Social Research (6/2002) ABSTRACT Variations in nutritional status across income are inferred from food spending, nutrient demand and dietary sufficiency using local polynomial regression in three data levels. 35 nutritional requirements are evaluated in spatial cross-section for ten income groups in Brazil’s eleven city Pesquisa de Orcamentos Familiares 1995/6. using U.S.D.A. NDSR-13 composition tables adjusted for fortification and Brazilian literature. A wide range of insufficiencies abated most rapidly at lowest and upper incomes accompanied by elevated income elasticities. In temporal cross-section using the U.S. Continuing Survey of Food Intakes by Individuals, 1994-96, more severe seasonal variations at lower incomes compounded the non-linear interactions of incomes, food spending, food and micronutrient intake, 35 intake targets for 26 components, household sizes, food assistance and macroeconomic activity. The food spending process is confirmed in U.S. four-person households, contemporary international and historical data. A law of increasing food spending and dietary satisfaction at upper income levels complements Engel’s Law of increasing food share below mean incomes. Five examples then illustrate the socioeconomic gradient of nutrition and health in the United States. Differentially by adult age, formal education was positive for nutritional sufficiency, obesity, and health from fifth through ninth grade, for eleventh grade, high school or college graduation and graduate schooling, while termination at 10th grade or as college junior was negative. Income was twice as powerful as education in reducing insufficiencies. Indicating the population’s awareness of its absolute and income relative nutritional status, CSFII-DHKS food security questions partially identified severe insufficiencies, and dietary satisfaction and health questions identified mild or moderate insufficiencies. Intra-individual intake coefficients of variation were decreasing and less seasonal with income, smaller than inter-individual coefficients and responsive to age and sex. Pubertal stature change was later and stronger and BMI menarche surge later among U.S. females 10-17 at upper incomes, with greater athleticism and sufficiency of most nutrients diminishing through middle to lowest incomes where menarche surges also retarded. Adult physical inactivity was associated with nutrient insufficiencies and increasing health problems at lower incomes. Insufficiencies in few components strongly improved reported health of older adults. Health problems associated with arachidonic acid intake and total lipids. These studies suggest a global inadequacy of budgetary resources to satisfy nutritional requirements short of upper income group food spending levels.