DETER MINANTS of HE ALTH An Economic Perspective MICHAEL GROSSMAN D E T E R M I N A N T S O F H E A LT H Determinants of Health An Economic Perspective Michael Grossman COL UM BIA UN I V ERS I TY P R ES S New York Columbia University Press Publishers Since 1893 New York Chichester, West Sussex cup.columbia.edu © 2017 Columbia University Press Library of Congress Cataloging-in-Publication Data Names: Grossman, Michael, 1942- author Title: Determinants of health : an economic perspective / Michael Grossman Description: New York : Columbia University Press, 2017 | Includes bibliographical references and index Identifiers: LCCN 2016046853 | ISBN 978-0-231-17812-9 (cloth : alk paper) | ISBN 978-0-231-54451-1 (e-book) Subjects: | MESH: Health Status Indicators | Socioeconomic Factors | Risk Factors | Models, Economic | Collected Works Classification: LCC RA418 | NLM WA 900.1 | DDC 362.1—dc23 LC record available at https://lccn.loc.gov/2016046853 Columbia University Press books are printed on permanent and durable acid-free paper Cover design: Jordan Wannemacher For my wife, Ilene, my daughters, Sandy and Barri, my sons-in-law, Steve and Dave, and my grandchildren, Zack, Ben, Alise, and John for increasing the efficiency of my health production, lowering its price, and causing me to demand a large quantity of quality-adjusted life years Contents Foreword by John Mullahy Introduction and Acknowledgments PART The Demand for Health: Theoretical Underpinnings and Empirical Results Introduction to Part 1 On the Concept of Health Capital and the Demand for Health The Human Capital Model Afterword to Part PART The Relationship between Health and Schooling Introduction to Part The Correlation between Health and Schooling An Exploration of the Dynamic Relationship between Health and Cognitive Development in Adolescence Parental Education and Child Health: Evidence from a Natural Experiment in Taiwan Women’s Education: Harbinger of Another Spring? Evidence from a Natural Experiment in Turkey Afterword to Part ix xv 42 111 121 122 128 190 212 249 286 CONTENTS | V I II PART Determinants of Infant Health with Special Emphasis on Public Policies and Programs Introduction to Part Variations in Infant Mortality Rates among Counties of the United States: The Roles of Public Policies and Programs Determinants of Neonatal Mortality Rates in the United States: A Reduced Form Model Birth Outcome Production Functions in the United States 10 Unobservables, Pregnancy Resolutions, and Birth Weight Production Functions in New York City 11 The Impact of National Health Insurance on Birth Outcomes: A Natural Experiment in Taiwan Afterword to Part PART The Economics of Unhealthy Behaviors Introduction to Part 12 The Effects of Government Regulation on Teenage Smoking 13 Beer Taxes, the Legal Drinking Age, and Youth Motor Vehicle Fatalities 14 Effects of Alcoholic Beverage Prices and Legal Drinking Ages on Youth Alcohol Use 15 Rational Addiction and the Effect of Price on Consumption 16 An Empirical Analysis of Cigarette Addiction 17 An Empirical Analysis of Alcohol Addiction: Results from the Monitoring the Future Panels 18 The Demand for Cocaine by Young Adults: A Rational Addiction Approach 19 An Economic Analysis of Adult Obesity: Results from the Behavioral Risk Factor Surveillance System 20 Fast-Food Restaurant Advertising on Television and Its Influence on Childhood Obesity 21 Food Prices and Body Fatness among Youths Afterword to Part Reflections Index 295 296 305 331 357 377 402 450 467 468 482 509 534 562 570 604 620 666 694 716 752 772 779 Foreword John Mullahy H ad Michael Grossman never conceived and written his seminal paper, “On the Concept of Health Capital and the Demand for Health,” it is unlikely that we—certainly economists, but in reality a much larger “we”—would understand the determinants and consequences of human health as well as we today Grossman’s paper, which has received to date more than five thousand Google Scholar–reported citations, may not have technically launched the field of health economics,1 but it certainly cemented the field’s standing and relevance as a science oriented toward understanding and improving people’s health Indeed, only a rare student of health economics has not been exposed at some point in their training to “the Grossman model” of health developed in the paper, which appears as chapter of this book Assembling Grossman’s most important work in this one volume and reissuing alongside it his classic monograph, The Demand for Health: A Theoretical and Empirical Investigation, is not only a fitting testimonial to him and the importance of his scholarly work, but also serves to guarantee that current and future scholars in this field will have ready access to the ideas of one of the finest minds—and finest people—in our discipline There is considerable value added by the commentary Grossman provides in the introductions and INDEX | 792 instrumental variables procedure: in healthschooling studies, 124–25, 216–19, 249; parent’s schooling and, 126; schooling coefficient and, 92; for self-selection correction, 299, 378–79; for United States neonatal mortality study, 343; variability of, 290 intelligence: Aviation Cadet Qualifying Examination and, 141–42; demand curve for, 181n17; early childhood environmental factors in, 130, 135; genetic factors in, 130, 135, 140, 191; low birth weight and, 190 See also mental ability intelligence quotient (IQ): achievement and, 201–2; in Terman Life Cycle Study of Children with High Ability, 287–88 See also Wechsler Intelligence Scale for Children interest rate See rate of interest intergenerational relationships, time preference hypothesis and, 98 intertemporal complementarity, 475, 605, 621, 624 intertemporal utility function: of health demand, 9–11; in human capital model, 46–47 investment, gross See gross investment investment commodity, health as, 8, 130–31 investment in health: cost of health capital and, 22; education and, 95–97; marginal monetary rate of return on, 14, 59; monetary rate of returns of, 17; motive for, 86; parametric differences in marginal cost of, 55–56; pure investment model, 14–18; timing of, 286–87 investment in human capital model, 7, 17–18, 43 investments in human capital, 17; time preference hypothesis and, 92–93 job satisfaction, in NBER-Thorndike sample, 148, 150–51; current health and, 155–56; estimated bias in schooling coefficients in, 153–54, 154; as indirect component, 159, 159–60, 160; in life-cycle model, 185n70; maximum likelihood estimates of dichotomous logit survival functions, 173; means and standard deviations of, 177; ordinary least square estimates of, 151–53, 152; ordinary least square estimates of dichotomous EXCELL in, 161, 161; simple correlation coefficient matrix of, 178–79; two-stage squares estimates of health functions with, 167 joint production: in human capital model, 47; negative family income elasticity and, 75–76, 105n39 Journal of Health Economics, demand function for health in, 111 junior high school openings under Taiwan 1968 education reform: allocation of, 226–27; family planning program and, 227; increase in entrance with, 221, 221–22; numbers of, 219–21, 220 Kessner index, 454 Kids Walk-to-School Program, 741 knowledge, stock of: equilibrium, 134; after formal schooling, 136; health stocks and, 114; productivity and, 7, 43; school investment and, 133–34; student health and, 129 knowledge capital: nonmarket efficiency and, 47, 66; produced per year of school attendance, 138; productivity and, 137 Kraft Foods, 696 Labor Insurance: coverage of, 406; NHI coverage compared with, 420, 421; other coverage offered by, 446n4 labor market productivity: adult’s health and, 163–64; childhood obesity and, 717 labor market time, adult obesity and, 667, 670 labor supply, dynamic model of, death and, 56 lagged alcohol consumption, 606 lagged consumption coefficients, 591, 623 lagged health variables, 86 lagged mortality rate: direction of bias on, 353n15; in United States neonatal mortality study, 343–44 Laporte, Audrey, on health dynamics, 116–17 Laspeyres index, of food price, 723–25 latent health variable: in conditional demand function for physician visits, 80–81; gross investment as, 83; in multiple indicatorsmultiple causes model, 77, 80 late spontaneous abortions, infant health production function, 386, 387, 390 least squares estimates: biases with, 86; for firstdifference models, 86; for fixed-effects model, 86; of health production functions, 73–74 legal drinking age: alcohol consumption and, 610–12, 613; changes in, 470, 473, 534–35, 557n30; in cocaine demand study, 629, 631, 636, 638, 639, 643, 645, 647 legal drinking age and motor vehicle accident mortality study, 510–11; analytical framework for, 513; beer cost in, 517–18; border age omitted, 522–23, 523; border phenomenon in, 517; dataset for, 514, 515, 516; dichotomous variables for, 518–19; drinking sentiment in, 518–19, 524; evasion and, 511–12; motor vehicle mortality rate in, 519; outcome measures for, 516; predicted effects of increases in, 525–28, 527, 533n39; regressions for, 514; regression with state dummies, 523, 524–25; simulation for, 525; state specific research on, 512; weighted least-squares regression estimates of logit motor vehicle accident mortality equations, 519, 520–21, 522 legal drinking age and youth alcohol use study: alcohol use measurement for, 538–41, 793 540; beer frequency equation in, 547–48; beverage-specific drinking frequencies in, 539, 540, 546–47, 557n23; data and subsample selection for, 537–38; drinking sentiment in, 543–44; estimation techniques for, 546–47; evasion and, 536; federal excise tax and, 536; independent variables for, 541–44; liquor demand functions in, 549; maximum likelihood estimates of multinomial logit beer frequency equations, 544–46, 545; price elasticities in, 537; religion as regressors in, 548; short-run changes in, 536–37; simulations in, 549–50, 551–52, 552–53; summary for, 553–54; underreporting in, 539, 541 legalization of banned substances and consumption, 562; addictive capital in, 564, 564; cocaine demand study and, 621, 651–52; debate on, 620; federal excise tax and, 761– 62; implications of, 566–67; instantaneous utility function for, 563; price and, 621; price elasticity and, 621; reinforcement in, 563–64; temporary policies and, 567–68, 652; total cost and, 565–66; unsteady states and, 564–65 length of life: Ehrlich and Chuma on, 54–55, 114; endogenous, 117; exogenous variables in, 55, 115–16; gross investment production function and, 116; health capital and, 137; initial assets and, 55–56, 116; in intertemporal utility function, 9, 46; iterative determination of, 51, 115; job satisfaction and, 150; marginal cost and, 55–56, 116; optimal, 50–56; rate of depreciation and, 55, 115; Ried on, 51, 55–56, 115; total, 35; wealth and, 115 See also life expectancy; longevity life cycle variations: diminishing returns and, 115–16; in goods and time allocation, 101n10; in gross investment, 20–21, 63; lagged responses to, 114; in rate of depreciation, 18–22, 19, 63; in stock of health, 21, 62 life expectancy: college graduation and, 287–88; health status and, 134; schooling and, 123; in Terman Life Cycle Study of Children with High Ability, 287–88 life length See length of life lifetime utility, maximization of, 52–54 lifetime utility functions: alcohol addiction and, 606; with alternative horizons, 55; in cocaine demand study, 623 lifetime wage variable, 80 liquor legal drinking age, in youth alcohol use study, 540, 541–42 logit function, 162, 184n61, 519 long-distance smuggling measure, 578, 579–80, 597–99 | INDEX longevity: health capital and skill capital and, 114; income and, 776; optimal health paths and, 57 longitudinal data set, causality and, 191 long-run price elasticities, 590, 591 low birth weight: abortion and, 453–54; BCHS projects and, 369; in birth outcome production function, 363; brain function and, 190; Canada NHI and, 460; family planning clinics and, 369; as first-order effect, 450; infant mortality and, 297; Medicaid and, 459–60; mother’s schooling and, 252, 290; neonatal health and, 334; neonatal mortality and, 297, 359–60, 361, 362, 364; WIC program and, 458–59 low birth weight, in Taiwan NHI study, 411; difference-in-differences estimation, 420, 421, 422; full interaction model of, 430, 431–33; lagged variable estimates of, 434, 435–36; mean of postneonatal rate and, 418, 419; medical technology and, 425, 426–29; propensity score weighted regressions of, 441, 442, 443; by time period, 418, 420, 420 Low-Income Households’ Health Insurance (LII), 406 malnutrition, in health production, 151 marginal benefits: discounted, 49–50, 54, 58–59; of gross investment, 58; of health stock increase, 87; optimal length of life and, 54, 116; present value of, 49 marginal cost: education and, 26; length of life and, 55–56, 116; in monopoly pricing model, 593; parametric differences in, 55–56; productivity change and, 34–35 marginal cost of gross investment, 17, 36, 49; discounted marginal benefits of, 58; marginal utility of lifetime wealth and, 77; MEC schedule and, 64; optimal stock of health capital and, 59–60; percentage rate of change over time and, 57–58; rate of depreciation and, 132; stock of health and, 61; wage rate and, 65, 68 marginal disutility, of sick time, 14–15, 180n3 marginal efficiency of health capital, 15, education and, 26–27, 27; elasticity of, 28, 132; human capital and, 66; linear production process of, 57; optimal stock of health capital and, 28; stock of health and, 60–61, 61 marginal monetary rate of return on investment in health, 14, 87; exogenous variables and, 59–60; opportunity cost of capital and, 60 marginal productivity, health capital and, 16, 16, 39n35 marginal product of health capital, 12, 36; changes in, 22–23; education and, 26, 39n35; health stock and, 52; of individuals no in labor force, 17; MEC schedule and, 16, 16, 61, 61, 64; production function and, INDEX | 794 marginal product of health capital (continued ) 39n27; with travel and waiting time, 78; undiscounted, 52; wage rate and, 64–65 marginal products: in gross investment production, 10; of human capital, 17–18; of medical care, 35 marginal rate of return: health capital cost and, 18–19; with risk-averse behavior, 88 marginal rate of substitution: health stock and, 67; of healthy time, 86; wage rate and, 56 marginal revenue, in monopoly pricing model, 593 marginal utility: cigarette addiction and, 573; of healthy time, 12, 14–15, 36, 49, 86, 180n3; of lifetime wealth, 77; of wealth, 12, 36, 49, 574 marginal utility of health capital: health demand function and, 80; monetary equivalent of, 67, 104n25 marijuana consumption: cocaine consumption and, 647; price elasticities of, 759–60 marijuana decriminalization, in cocaine demand study, 629, 631, 636, 638, 639 marijuana legalization: alcohol-related traffic fatalities and, 761; motor vehicle accident mortality and, 760–61 market efficiency: determinants of, 85; education and, 27–28; environmental variables and, 25; marginal efficiency of capital schedule and, 22–28; pure investment model of demand for good health and, 64–67; wage rate and, 23–24 market goods: in health production function, 47; scarcity of, 10, 47–48; wage rate and, 23–24 market productivity, schooling and, 155, 158, 306, 333 marriage age: in Turkey, 254, 279; Turkey Compulsory Education Law study of, 262, 272, 273, 274 maternal age: infant health and, 407; in Taiwan NHI study, 415–16, 417, 420, 421, 422 maternal and infant care project, 325n2; female poverty and, 369; introduction of, 332; neonatal mortality and, 334, 362; site selection for, 343; in United States infant mortality study, 313, 314–15, 322 mathematical terms, 35–36 mechanical ability, in NBER-Thorndike sample, 148, 149, 149–50; current health and, 158; maximum likelihood estimates of dichotomous logit survival functions, 173; means and standard deviations of, 177; ordinary least square estimates of, 151–53, 152, 153; schooling and, 184n53; simple correlation coefficient matrix of, 178–79 marginal efficiency of capital schedule, 15, 15; education and, 26–27, 27; elasticity of, 19, 19–22, 24, 24, 36, 62, 75; for healthy time production function, 63–64; human capital in, 25–28, 27; marginal cost of gross investment and, 64; marginal efficiency of health capital, 60–61, 61; marginal product of health capital and, 16, 16, 22–23, 61, 61, 64; market and nonmarket efficiency and, 22–23; rate of return on investment, 60–61, 61; wage effects on, 23–25, 24 Medicaid: abortion and, 452–53; in birth outcome production function, 365; birth probability function and, 394–95; birth weight and, 461; child mortality and, 461–62; cigarette smoking and, 461; college graduation and, 462; first-time pregnancies and, 314, 316–17, 319, 326n11, 340, 365; high school graduation and, 462; infant mortality and, 297, 306, 461–62; introduction of, 325n2, 332, 352n3, 403; low birth weight and, 459–60; neonatal mortality and, 298, 317, 318, 334, 460; pregnancy coverage by, 314; prenatal and perinatal care services of, 309, 459, 461; private insurance switch to, 403–4; unhealthy behaviors and, 468; in United States infant mortality study, 313, 314–15, 317, 318, 319, 320, 322; in United States neonatal mortality study, 338, 339, 345, 346, 347–49, 349 medical care availability, in Taiwan NHI study, 423–24 medical care coefficient, production function and, 113 medical care services: age and, 75; birth control cost and, 309; in conditional output demand function, 77; demand for, 8, 43, 113; demand function for, 73, 75; doctor visits as indicator for, 82; education and, 26–28, 27; education and demand for, 66–67; gross investment share of, 72; health impact of, 132; health insurance coverage and demand for, 105n36; in health investment production function, 80; in health production function, 47, 78, 82; in household production functions, 9–10, 47; insurance coverage in Taiwan for, 406–7, 408; in intertemporal utility function, 9; marginal products of, 10; partial relationship with health, 79; price of, 35; rationing of, 65; reduced form demand function for, 71; schooling outlays compared with, 129; shadow price of health and, 6, 9; sick time and, 64; total consumption of, 46; wage elasticity of, 65, 75, 78; wage rate and, 24, 24–25, 28 medical inputs, in Denmark, 78 medical technology in Taiwan NHI study: equations for, 423–24; estimates of, 424–25, 426–29; full interaction model, 430, 431–33; infant hospital bed numbers in, 424, 428–29, 430, 433; lagged variable estimates of, 434, 435–36; obstetrical department number in, 424, 426–27, 430, 431; pediatric department number in, 424, 427–28, 430, 432 795 Medicare, unhealthy behaviors and, 468 medicine complaints, in MIMIC model, 77 mental ability: schooling and, 134, 140; wage rate and, 72 mental well-being: job satisfaction and, 150; physical health and, 139; physical illness and, 145 metropolitan statistical area, 700, 713n5 Mexico, sugar-sweetened beverage tax in, 758–59 Midlife in the United States survey, 289 Mills ratio, for infant health production function, 383–85, 395 minimum legal drinking age, 516, 631 Ministry of National Education’s National Education Statistics, 257 mixed investment-consumption model: comparative static age, schooling, and wealth effects in, 86; exogenous variables in, 106n44; uncertainty in, 89 monetary rate of returns of investment in health, 17; cost of capital and, 22–23; in pure consumption model of demand for good health, 60, 67; stock of health and, 60–61, 61 See also marginal monetary rate of return on investment in health monetary value: of future health consequences of food choice, 738–39; of marginal utility of healthy time, 86, 104n25; of negative health effects, 472; of time lost, 131 Monitoring the Future research program, 608–10, 620–22, 625–26, 632–33, 663n8 monopoly power, cigarette addiction and, 572, 592–94 monopoly pricing model, 593 monopoly states, 531n25 Mormons, in vehicle accident mortality study, 515, 518, 524; logit motor vehicle accident mortality equations, 519, 520–21, 522; regression with border age omitted, 522–23, 523 Mormons, in youth alcohol use study, 540, 543–44; multinomial logit beer frequency equations for, 544–46, 545; as regressors, 548 mortality: body mass index and, 690n2, 691n18; state excise tax on cigarettes and, 753; trends in, 773–74; in twin studies, 289 See also child mortality; death mortality experience of NBER-Thorndike sample: nature of analysis, 170–72, 172; results of, 172–74, 173 mortality rate: cigarette price and, 469; motor vehicle, 519 mother’s schooling: adolescent childbearing and, 252–53; antenatal visits and, 290; birth weight and, 139, 216–18, 222, 239–40, 240, 252, 290; child mortality and, 217, 222, 261–62; children’s health and, 213, 224–25, | INDEX 251; fertility and, 218, 252–53; infant health and, 252; infant mortality and, 239–40, 240, 290; neonatal mortality and, 239–40, 240, 252, 347, 349; postneonatal mortality and, 443; poverty and, 308; premature-births and, 252; program intensity measure for Taiwan 1968 education reform study and, 227, 228, 229–30, 230, 236–39, 237; public policy and, 205; in Taiwan education reform, 214; teenage smokers and, 506n34; in twin studies, 289 mother’s schooling, in Taiwan NHI study, 415–16; characteristics of, 417, 418; difference-indifferences estimation, 420, 421, 422; full interaction model of, 430, 431–33; lagged variable estimates of, 434, 435–36; mean of postneonatal rate and, 418, 419; medical technology and, 425, 426–29 mother’s schooling, in Health Examination Survey: characteristics of, 206; regression coefficients of, 202, 203, 204 motor vehicle accident mortality rate: antidrinking campaign and, 535–36; campaign for reduction of, 509; federal excise tax and, 473–74; legal drinking age and, 470, 473; marijuana legalization and, 760–61 See also beer excise taxes and motor vehicle accident mortality study; legal drinking age and motor vehicle accident mortality study multiperiod mixed investment-consumption model, uncertainty in, 89 multiple indicators-multiple causes (MIMIC) model, 76–77; demand for health model and, 81 municipal excise tax, on cigarettes, 596 myopic addiction models, 571, 576–77; of cigarettes, 580–82, 581; for cocaine demand study, 642, 647; cocaine frequency structural demand function for, 642, 645–46, 647; cocaine participation structural demand function, 642, 643–44, 647; first stage regressions, 660–61; rational models compared with, 605, 621 narcotics, infant health production function, 386, 392 Nash-bargaining, 113 National Bureau of Economic Research, 695; NBER-Thorndike sample of, 142–43 National Center for Health Statistics (NCHS): family planning clinics, 314–15; Mortality Tape, 310, 336; Natality Tape, 310, 336; NHANES II of, 535; NHANES I of, 94–95 National Child Development Study, 218 National Cooperative Medical System, in rural China, 404–5 INDEX | 796 National Health and Nutrition Examination Survey I: obesity trends and, 668, 668–69; schooling in, 94–95 National Health and Nutrition Examination Survey II, 558n35; beverage-specific drinking frequencies in, 557n23; cigarette addiction and, 592; obesity trends and, 668, 668–69; youth alcohol consumption and, 535, 537–38, 541 National Health and Nutrition Examination Survey III, obesity trends and, 668, 668–69 National Health and Nutrition Examination Survey 1999–2000: childhood obesity and, 720; obesity trends and, 668, 668–69 National Health and Nutrition Examination Survey 2001–2002, childhood obesity and, 720 National Health and Nutrition Examination Survey 2003–2004, childhood obesity and, 720 National Health Examination Survey I, obesity trends and, 668, 668–69 National Health Insurance: in Canada, 403, 460; in Taiwan, 301–2 See also Taiwan, National Health Insurance National Health Interview Survey, US, of 1989, 76 National Highway Traffic Safety Administration, 516 National Institute on Alcohol Abuse and Alcoholism, 471 National Longitudinal Survey of Young Men, schooling in, 94–95 National Longitudinal Survey of Youth, 671, 694, 696, 699–700, 705–6 National Longitudinal Survey of Youth, of 1997, 694, 696, 699–700, 705–6 National Opinion Research Center, 1963 survey of, 74, 143, 144, 171, 182n34 National School Lunch Program, 741 National Science Foundation, 471 NBER-Thorndike sample, 125–26; ability variables in, 149, 149–50; current health in, 157–58; decomposition analysis in, 158–60, 159, 160; elasticity of health in, 157; estimates of excellent health functions, 161, 161–62, 163; health and schooling regressions with, 147–49, 148; health increases in, 155–57; health measurement in, 143–47, 144; health status in, 143, 144; mortality experience of, 170–74, 172, 173; ordinary least squares estimates of health functions in, 151, 152, 153; ordinary least squares estimates of schooling functions in, 151, 153, 153; past health proxy of, 147; schooling coefficients in, 153–55, 154; variables in, 148, 148 See also recursive health-schooling system; simultaneous-equations health–wage model estimations negative health effects: monetary value of, 472; rational addiction and, 474 negative legal drinking age coefficient, 532n35 neonatal health, low birth weight and, 334 neonatal intensive care: availability and use measures for, 374n8; availability of, 342–43; in birth outcome production function, 363, 364; insurance coverage in Taiwan for, 406–7, 408; neonatal mortality and, 337, 338, 339, 346, 347–48, 349, 362, 364, 370 neonatal mortality, 223; abortion and, 298, 320, 321, 327n12, 336, 347–50, 349, 361–62, 368–69, 370, 371; BCHS projects and, 362, 364, 370; in birth outcome production function, 363; birth weight and, 450; causes of, 336; cigarette smoking and, 359–60, 362; community health centers and, 334, 341, 362; contraceptive use and, 322, 361; equations for, 364; factor contribution to reductions in, 349, 369–71, 370; family planning program and, 320, 321, 362, 370; father’s schooling and, 239–40, 240; female poverty and, 346, 347, 349; health endowment and, 360; high-risk women and, 362; low birth weight and, 297, 359–60, 361, 362, 364; maternal and infant care projects and, 334, 362; means and standard deviations of, 234; measurement of, 335–37; Medicaid and, 298, 317, 318, 334, 460; mother’s schooling and, 239–40, 240, 252, 347, 349; neonatal intensive care and, 337, 338, 339, 346, 347–48, 349, 362, 364, 370; perinatal care and, 359–60; physician visits and, 327n14; premature-births and, 297; prenatal care and, 359–62, 368–70, 370; production function, 359–60; race-specific regressions for, 310, 332–33, 336–37, 338, 339, 343; rate decline of, 357–58; rate of, in United States neonatal mortality study, 338, 339, 346; schooling and, 328n17, 348–49, 349; survival probability and rate of, 333–34; in United States, 332; WIC program and, 298, 334, 362, 364, 368, 370, 370 See also United States, neonatal mortality study Netherlands Health Interview Survey, of 1984, 82 net investment in stock of health, 46–47, 130; in Ehrlich–Chuma model, 58; positive, 58 neuroeconomics, 764–65 newborn intensive care hospitals/1000, in United States neonatal mortality study, 338, 339 New York City pregnant women, 379, 385, 386–87, 395, 397n7 Nielsen Retail Scanner Data, 756, 759 Nigeria: female schooling in, 218, 222; Universal Primary Education Program in, 253 797 | INDEX Nigerian Demographic Health Survey, of 1999, 218 noncognitive skills, 287 nondurable household commodities, schooling and, 112 nonearnings income of family, in NBERThorndike sample, 148; estimated bias in schooling coefficients in, 153–54, 154; indirect components and, 160, 160; in lifecycle model, 185n70; maximum likelihood estimates of dichotomous logit excellent health functions, 162, 163; means and standard deviations of, 177; ordinary least square estimates of, 151–53, 152; simple correlation coefficient matrix of, 178–79; two-stage squares estimates of health functions with, 167 nonmarket efficiency: determinants of, 85; education and, 25–26, 40n37; environmental variables and, 25; knowledge capital and, 47, 66; MEC schedule and, 22–28; pure investment model of demand for good health and, 64–67 nonmarket productivity: schooling and, 155, 158, 306, 333 See also health productivity non-medical inputs, in production of health and, 80 nonnegative gross investment, 38n17, 38n26 number of healthy days per unit stock, 35 numerical ability, in NBER-Thorndike sample, 148, 149, 149–50; current health and, 158; maximum likelihood estimates of dichotomous logit survival functions, 173; means and standard deviations of, 177; simple correlation coefficient matrix of, 178–79 nurture, in cognitive development, 205 opportunity cost of capital, marginal monetary rate of return on investment in health and, 60 optimal control techniques, 16–17 optimal gross investment, 49–50; control techniques and, 16–17; indeterminacy problem of, 57–59; longevity and, 57; optimal length of life and, 51–52 optimal length of life, 50; algorithm for selection of, 54; discounted marginal benefits of investment and, 54; fixed terminal time problems for, 51; indeterminacy in, 115–16; issues with, 51; iterative process for, 51–53, 115; marginal benefits and, 54, 116; maximization of, 52; preliminaries for, 51–52 optimal stock of health capital, 14–15, 15; demand for health and, 50; determination of, 60–61, 61; first-order conditions for, 52; gross investment in, 16, 50; healthy time and, 16, 16; initial assets and, 88; in last period of life, 60, 104n22; length of life and, 56; marginal cost of gross investment and, 59–60; marginal efficiency of health capital and, 28; optimal control techniques for, 16–17; optimal length of life and, 53; in period t, 85; selection of, 51 Oregon Health Insurance Experiment, 446 organized bootlegging, 597 output demand functions, for infant health, 380 overweight, 479n1; as dependent variable, 709, 710, 711; means and standard deviations for, 704, 705–6, 707; parents, 702–3 See also obesity ovulation cycle knowledge: in Turkey, 279; women’s education and, 249 oxycodone, 773–74 OxyContin, 774 obesity, 479n1, 696; determinants of, 477; in health production, 151; of mother, 702, 706; rates of, 477; self-rated health status and, 287; television fast-food restaurant advertising and, 478; twin studies and, 288; unhealthy food and, 756–57 See also adult obesity; body mass index; childhood obesity; food prices and obesity study obesity, in Heath Examination Survey, 195, 196, 198; high blood pressure and, 202; lagged coefficients of, 199, 200, 201; oral health and, 202; regression coefficients of family background effects on, 202, 203, 204, 210n15 occupation, schooling and, 89–90 Office of Economic Opportunity, 352n3 one-tailed test: age coefficient and, 105n38; regression coefficients and, 245n15 on-the-job training: human capital and, 181n15; in wage function, 163–64 opium demand, 622 pain, relief of, 43 Panel Survey of Income Dynamics (PSID), 1986 wave of, 95 parental consent laws, 453 parent’s assessment of youth’s overall health, in Health Examination Survey, 195, 197, 198; lagged coefficients of, 199, 200, 201; regression coefficients of family background effects on, 202, 203, 204, 210n15 parent’s income: children’s health and, 135; human capital investments and, 134; in third variable explanation, 139 parent’s schooling, 123; children’s health and, 125–26, 135, 214, 216–17, 251; fertility and, 216–17; human capital investments and, 134; infant health and, 241; instrumental variables and, 126; program intensity measure for Turkey Compulsory Education Law study and, 259; socio-economic status and, 251; of spouse, 140–41; Taiwan education reform INDEX | 798 parent’s schooling (continued ) of 1968 and, 229–30, 230; teenage smokers and, 506n34; in third variable explanation, 139 See also father’s schooling; mother’s schooling parent’s utility function: maximization of, 333; survival probability and, 359 parity: in infant health production function, 386, 387, 390; in Taiwan NHI study, 416, 417, 420, 421, 422 past consumption: in B-M model of addiction, 571, 605, 621; cigarette consumption and, 572, 581, 581; coefficient of, 590, 613, 614, 641–42, 648–49, 651; current consumption and, 574–76, 583, 612, 622, 641; in myopic cigarette addiction, 580, 581; past prices and, 607; in rational addiction, 600; two-stage least squares coefficients of, 589 past prices: current consumption and, 571, 574–75, 585, 607, 624; past consumption and, 607; in rational cigarette addiction model, 583, 584 peer effects, time preference and, 99 per capita cigarette consumption: addiction and, 570–71, 577, 578, 579; cigarette price and, 571; Fairness Doctrine and, 484; pro-smoking advertising and, 492, 501 per capita Medicaid payment, in United States neonatal mortality study, 338, 339, 346 percentage body fat: body mass index compared with, 716, 719, 742; construction of measures of, 720–23; errors with, 727; food prices and, 478, 716; for food prices and obesity study, 719, 729–30; gender and, 728; race/ethnicity and, 728 percent high school educated, in United States neonatal mortality study, 338, 339, 346 percent poor, in United States neonatal mortality study, 338, 339, 340, 346 perinatal care services: infant mortality and, 309; of Medicaid, 309, 459; neonatal mortality and, 359–60 perinatal period, 352n2 periodontal index, in HES, 195, 196, 197–98, 209n8; high blood pressure and, 202; lagged coefficients of, 199, 200, 201; obesity and, 202; regression coefficients of family background effects on, 202, 203, 204, 210n15 physical ability: in demand curve for health, 140; schooling and, 134 physician-prescribed medicine complaints, in MIMIC model, 77 physician visits: as medical care services indicator, 82; in MIMIC model, 77; neonatal mortality and, 327n14; to obtain medical care, 78 pill See contraceptive use pill mills, 774 Planned Parenthood, 315, 325n2, 352n3 planning horizon, optimal length of, 51 postneonatal mortality, 310; mother’s schooling and, 443; Taiwan NHI and, 302, 410–11, 411, 413, 418, 419, 421, 422; in Taiwan NHI study, 425, 430, 443; in United States, 332, 450–51 post-shock health, earnings function dependency on, 106n48 poverty: addiction and, 567; in birth outcome production function, 363; infant health production function and, 386, 389; schooling and, 308 See also births to women in poverty; female poverty pregnancies: first-time, 314, 316–17, 319, 326n11; insurance coverage in Taiwan for, 406–7, 408; Medicaid coverage of, 314; probability of, 379; quitting smoking and, 753; in Turkey, 279; women’s education and, 249 pregnancy resolution: abortion price and, 300; birth probability function and, 397n4; birth weight production function and, 397n4; contraception price and, 300; decision for, 380; fertility control and, 377–78; in infant health production functions, 377; selfselection, 299, 378 premature-births: causes and prevention of, 450, 458; infant mortality and, 297; mother’s schooling and, 252; neonatal mortality and, 297 premature-births, in Taiwan NHI study: difference-in-differences estimation, 420, 421, 422; full interaction model of, 430, 431–33; lagged variable estimates of, 434, 435–36; mean of postneonatal rate and, 418, 419; medical technology and, 425, 426–29; propensity score weighted regressions of, 442, 443; by time period, 418, 420, 420 premature deaths, obesity and, 668 prenatal care clinics, infant health production function and, 386, 387, 389 prenatal care demand function: estimates of, 388, 389–90; in infant health production function, 380–81, 385, 387 prenatal care services: abortion price and, 396n3; in birth outcome production function, 363, 366, 367, 368; birth weight and, 455; in birth weight production function, 384, 388, 454–55; delay of, 386, 392, 394; demand functions of, 379; efficacy of, 378; infant health and, 407; in infant health production function, 385; infant mortality and, 309; insurance coverage in Taiwan for, 406–7, 408; mean values of, 398n10; of Medicaid, 309, 459, 461; neonatal mortality and, 359–62, 368–70, 370; teen family planning and, 369 prescription drug abuse, 774 799 prescription opioids, 773–74 price elasticity: of adult’s health, 157; of alcoholic beverages, 537, 614; in B-M model of addiction, 621; of children’s health, 157; of cigarette demand, 485–87, 493–94, 494, 501–2, 566, 572, 575, 589–90, 590, 752–53; of cocaine consumption, 621–22, 624, 647–48, 651; in cocaine demand study, 641–42; of cocaine participation, 759–60; of fast-food, 739–40; of health, 183n44; of heroin, 622, 759–60; of illegal drug use, 759–60; legalization of banned substances and, 621; long-run, 590, 591; of marijuana, 759–60; rational addiction and, 476, 566–67, 763; short-run, 590, 591; of teenage smoking, 472 price of health, in Taiwan education reform, 215–16 price of medical care services, 35, 48; wage rate and, 56 price of schooling, in Taiwan education reform, 215–16 primary schools: Indonesia construction program of, 217, 222, 252; Turkey enrollment in, 255–56, 257 principal components analysis, of nonchronic health problems, 76 private insurance: health stock and, 89; Medicaid switch from, 403–4; in Taiwan NHI study, 414–15, 420, 421, 424, 426–29; unhealthy behaviors and, 468 probit-tobit models, 82 production function: education and, 73; estimation of, 73–75; of gross investment in health, 57, 65–66; of health capital, 8; of human capital, 181n11; least squares estimates of, 73–74; marginal product of health capital and, 39n27; medical care coefficient and, 113; neonatal mortality, 359–60; survival probability, 306, 333, 359; two-stage least squares estimates of, 74 See also birth outcome production function; birth weight production function; infant health production function production function of healthy time, 16, 16; in demand curve for health, 131; equation for, 21, 61, 61–64; gross investment function and, 71; health status scales and, 144, 145 productive efficiency: allocative efficiency and, 243n1; future preference and, 97; of schooling and health, 90, 124, 214 productivity: education and, 26–27; in gross investment production function, 70, 112; human capital and, 34–35; knowledge capital and, 137; stock of health and, 7; stock of knowledge and, 7, 43 See also market productivity; nonmarket productivity program intensity measure for Taiwan 1968 education reform study, 225–26; father’s | INDEX schooling and, 227, 228, 229–30, 230, 237–39, 238; for females, 233–34, 235, 236; by gender, 230, 231–32; for males, 234, 235, 236; means and standard deviations of, 234; mother’s schooling and, 227, 228, 229–30, 230, 236–39, 237; precision of, 229; treatment group and, 228–29 program intensity measure for Turkey Compulsory Education Law study: coefficients of interactions between age and, 267, 268; parent’s schooling and, 259 pro-smoking advertising: ban on, 469, 482–85, 500, 501; outlays on, 507n40; per capita consumption of cigarettes and, 492, 501; regression coefficients for, 499–500 Protestants, in vehicle accident mortality study, 515, 518, 524, 543–44; logit motor vehicle accident mortality equations, 519, 520–21, 522; regression with border age omitted, 522–23, 523 Protestants, in youth alcohol use study, 540; multinomial logit beer frequency equations for, 544–46, 545; as regressors, 548 psychic rate of return on investment in health, 36; in pure investment model, 60 psychomotor control, in NBER-Thorndike sample, 148, 149, 149–50; current health and, 158; maximum likelihood estimates of dichotomous logit survival functions, 173; means and standard deviations of, 177; ordinary least square estimates of, 151–53, 152, 153; simple correlation coefficient matrix of, 178–79 public goods, 700 Public Health Cigarette Smoking Act of 1970, 469, 482 public policy: for childhood obesity, 717, 719; on children’s health, 213, 402–3; on cigarette smoking, 503n1; in developing countries, 213; on fast-food restaurant advertising, 696, 711–12; infant mortality and, 297, 306, 317, 318; mother’s schooling and, 205; schooling and, 90, 123; unhealthy behavior and, 468, 478; unintended consequences of, 758; of United States, 129, 213 See also antidrinking campaign; antismoking campaign; Fairness Doctrine; legal drinking age; maternal and infant care project; Medicaid; Medicare; National Health Insurance; Special Supplemental Food Program for Women, Infants, and Children pure consumption model of demand for good health: age-related depreciation rate effects, 67–68; data and results for, 74–76; demand for good health in, 73; demand function in, 77; empirical formulation of, 70–71, 73–74; health relation to wealth in, 68; human capital parameter in, 69; lifetime wage effect INDEX | 800 pure consumption model of demand for good health (continued ) in, 81; monetary rate of returns of investment in health, 60, 67; rate of interest and, 69–70; reduced form health demand function in, 80; schooling in, 112; uncertainty in, 88, 106n49; wage elasticity in, 68–69; of Wagstaff, 78 pure investment model of demand for good health: data and results for, 74–76; demand for good health in, 14–18, 73; depreciation rate effects in, 62–64; empirical formulation of, 70–71, 73–74; equilibrium stock of health in, 131–32; exogenous variables and, 62; initial assets and, 89; lifetime wage effect in, 80–81; marginal disutility of sick time in, 180n3; marginal monetary return on investment in, 87; marginal product of health capital in, 61, 61–62; marginal utility of healthy time in, 180n3; market efficiency and, 64–67; NBER-Thorndike sample and, 156–57; nonmarket efficiency and, 64–67; optimal stock of health capital at age t, 60–61, 61; predictions from, 131; psychic rate of return on investment in, 60; reduced form of, 78–79; schooling in, 112; uncertainty in, 87–88; wage coefficient and, 75; of Wagstaff, 78–79; wealth elasticity and, 73 quantity-quality interaction model of family decision making, 113 Quetelet’s index See body mass index quitting smoking: allocative efficiency and, 96; BMI and, 757; cigarette price and, 753; pregnancies and, 753; schooling and, 95–96; Wu-Hausman endogeneity test and, 95 race/ethnicity: abortion and, 301; birth weight production function and, 393–94; BMI and, 681, 685, 685–86; childhood obesity and, 706, 717; in food prices and obesity study, 727, 734, 735–38, 736; infant health and, 327n14; infant health production function and, 395, 397n7; neonatal mortality and, 310, 332–33, 336–37, 338, 339, 343; percentage body fat and, 728 RAND Health Insurance Experiment, 403 Rapid Coverage of Compulsory Education Program, 255 rate of depreciation: age and, 18–19, 19, 28, 45, 47, 51–52, 67–68, 106n45; birth defects and, 135; corner solutions and, 59; death and, 103n18; education and, 39n35, 70, 83, 104n31, 112; genetic factors and, 135; gross investment and, 22, 63–64; health capital and, 20–21, 38n19, 67–68, 73; health stock and, 51–52, 60–61, 61; length of life and, 55, 115; life cycle variations in, 18–22, 19, 63; marginal cost of gross investment and, 132; negative, 36n2; net investment and, 46; pure investment model of demand for good health and effects of, 62–64; rate of disinvestment and, 71; shadow price and, 42, 45; at young ages, 134, 181n12; at zero, 57 rate of disinvestment, rate of depreciation and, 71 rate of interest, 35; in cigarette addiction, 573; in cocaine demand, 623; pure consumption model of demand for good health and, 69–70; stock of health and, 60–61, 61; time preference and, 98, 103n20; wealth and, 69 rate of investment, infinite, 57 rate of return, wage rate and, 65 rational addiction: alcohol consumption and, 566, 763–65; cigarette smoking and, 96, 476, 566, 763; cocaine consumption and, 634–35; for cocaine demand study, 620, 642, 647; cocaine frequency first stage regressions, 657–59; cocaine frequency structural demand function for, 635, 638–39, 639; cocaine participation first stage regressions, 654–56; cocaine participation structural demand function, 635, 636–37, 639; consumer behavior and, 106n43; estimates of rational demand functions of cigarette addiction, 585–87, 586; estimates of rational models of cigarette addiction, 582–83, 584; exponential discounting in, 764; frequency and, 632; future preference and, 98–99; gambling and, 566–67; habit formation and, 556n19; myopic addiction and, 577, 605, 621; negative health effects and, 474; participation and, 632; past consumption in, 600; price elasticity and, 566–67, 763; time discount factor in, 476; unstable steady states and, 564–65, 577 See also Becker-Murphy model of addiction; cigarette addiction rational system, 764 real beer tax, in vehicle accident mortality study, 515; logit motor vehicle accident mortality equations, 519, 520–21, 522; regression with border age omitted, 522–23, 523 real income, in vehicle accident mortality study, 515; logit motor vehicle accident mortality equations, 519, 520–21, 522; regression with border age omitted, 522–23, 523 real price of beer, in youth alcohol use study, 540, 543; multinomial logit beer frequency equations for, 544–46, 545 real price of liquor, in youth alcohol use study, 540, 543 recuperation time, 22 recursive health-schooling system: decomposition analysis in, 158–60, 159, 160; empirical implementation of, 141; 801 empirical results of, 174–75; estimates of excellent health functions, 161, 161–62, 163; health measurement in, 143–47, 144; interpretation and estimation of, 138–41; NBER-Thorndike sample for, 141–43; ordinary least squares estimates of health functions in, 151, 152, 153; ordinary least squares estimates of schooling functions in, 151, 153, 153; regression specification, 147–51, 148, 149; schooling coefficients in, 153–55, 154 reduced form demand function: good health in, 77; for health and medical care, 71–72; for health capital, 132–33; schooling in, 81; for stock of health, 58 region and city size, in NBER-Thorndike sample, 165, 165–66; means and standard deviations of, 177; simple correlation coefficient matrix of, 178–79; two-stage squares estimates of wage functions with, 168 regression analysis, 186n77; for United States, infant mortality study, 307–8 regression coefficients, two-tailed tests and, 245n15 reinforcement: addiction and, 562–63, 575; instantaneous utility function of, 563–64 relative price difference, in teenage smokers, Fairness Doctrine, and cigarette price study, 490–91; quantity-smoked regression coefficients of, 497–98; regression coefficients of, 493–95, 494; sample means of, 490; smoking-participation regression coefficients of, 496 relief of pain, 43 reside in south, in NBER-Thorndike sample, 165, 165; means and standard deviations of, 177; simple correlation coefficient matrix of, 178–79; two-stage squares estimates of wage functions with, 168 residents of wet counties, in vehicle accident mortality study, 515, 518; logit motor vehicle accident mortality equations, 519, 520–21, 522; regression with border age omitted, 522–23, 523 restaurants See fast-food restaurants; full-service restaurants reverse causality bias, 344 Ried: on death, 56; on length of life, 51, 55, 115; on parametric differences in marginal cost of investment in health, 55–56 Right to Life movement, 323 risk-averse behavior, for uncertainty in demand for health models, 87–88 Roe v Wade, 451–52 School Breakfast Program, 741 school entry age laws, 218 | INDEX schooling: adult obesity and, 672–73; adult’s health and, 180n2; BMI and, 681, 682; causality from, to health, 90–92, 123, 129, 133, 213–14; causality from health to, 91–92, 123–24, 129, 213–15; children’s health and, 218; in cocaine demand study, 629, 631, 636, 638, 639, 643, 645, 647; compulsory, 123, 213–14, 216; demand curve for, 133–34; direct efficiency effect of, 97–98; disentangling time preference effect from, 94–95; earnings and, 124, 217; future time preference and, 96–97; general intelligence and, 150; good health correlation with, 89–90, 122–23, 128–29, 214; health productivity and, 158, 251; in household production function model, 129; income and, 89–90, 126; knowledge stock after, 136; life expectancy and, 123; market productivity and, 155, 158, 306, 333; mechanical ability in NBER-Thorndike sample and, 184n53; mechanism of health influence, 291; medical care services outlays compared with, 129; mental ability and, 134, 140; neonatal mortality and, 328n17, 348–49, 349; nondurable household commodities and, 112; nonmarket productivity and, 155, 158, 306, 333; poverty and, 308; present time preference and, 96–97; public policy and, 90, 123; in pure consumption model of demand for good health, 112; in pure investment model of demand for good health, 112; quality of, 291; quitting smoking and, 95–96; smoking and, 93–96; by spouse, 140–41; third-variable explanation of, 91, 124, 129, 139–40, 213, 215, 287–88; twin differences in, 291; in wage function, 217; wage rate and, 155 See also education; healthschooling relationships; parent’s schooling; Taiwan education reform, of 1968; time preference schooling, in Taiwan education reform, 215–16; means and standard deviations of, 234 schooling coefficient, 72; ability hypothesis and, 92; in demand function for medical care, 75; functional limitation and, 94–95; instrumental variables procedure and, 92; in NBER-Thorndike sample, 153–55, 154; time preference hypothesis and, 92–93 schooling variable: in demand function for medical care, 78–79; disability index and, 95 school-loss weeks, in NBER-Thorndike sample, 147 school quality, wage rate and, 72 self-financed, infant health production function, 386, 389 self-prescribed medicine complaints, in MIMIC model, 77 INDEX | 802 self-rated health status: alcohol, excessive use of and, 290; education and, 287–88; mortality and, 170; in NBER-Thorndike, 143–45, 144; obesity and, 287; smoking and, 287, 290 self-selected resolution of pregnancies, instrumental variables procedure for, 299, 378–79 sensitivity analysis for Taiwan NHI study: robustness checks: aggregated data, 434, 437, 438–39, 440–41; robustness checks: propensity score weighted regressions, 441, 442, 443 serial correlation bias, 343–44 service flow per unit stock, 35, 46 shadow price of health, 6; health demand curve and, 8–9, 45; health demand quantity and, 70; in human capital model, 42, 45 “Shadow Prices, Market Wages, and Labor Supply,” 156 share of depreciation in cost of health capital, 36 short-distance smuggling, 597–98 short-run price elasticities, 590, 591 “Should the Grossman Model of Investment in Health Capital Retain Its Iconic Status?,” 116–17 sick leave, 81 sick time, 36; disutility of, 14–15, 130; in goods budget constraint, 10; in gross investment function, 10–11; health capital and, 14–15; healthy time relationship to, 14; medical care services and, 64; own time input and, 64 See also time lost significant abnormalities, 195, 196, 197–98; lagged coefficients of, 199, 200, 201; regression coefficients of family background effects on, 202, 203, 204, 210n15 simultaneous-equations health-wage model estimations: direct, indirect, and total effects of schooling on health and wage rate, 169–70, 170; with NBER-Thorndike sample, 163–64; results of, 175; structural equations for, 164–66; supplementary variables for, 165, 165; two-stage squares estimates of health functions, 166–69, 167; two-stage squares estimates of wage functions, 166, 168 single agent decision-making process, 764 skill capital, longevity and health capital and, 114 social insurance, health stock and, 89 Social Security Act: maternal and infant care projects in, 325n2, 352n3; 1967 amendments to, 352n3; 1972 amendments to, 325n2, 352n3 socio-economic status: abortion and, 455; parent’s schooling and, 251 soda taxes, 742, 758–59 Southern Baptists, in vehicle accident mortality study, 515, 518, 524, 543–44; logit motor vehicle accident mortality equations, 519, 520–21, 522; regression with border age omitted, 522–23, 523 Southern Baptists, in youth alcohol use study, 540; multinomial logit beer frequency equations for, 544–46, 545; as regressors, 548 Special Supplemental Food Program for Women, Infants, and Children (WIC program): in birth outcome production function, 363; birth weight and, 458; infant mortality and, 297; introduction of, 332, 352n3; low birth weight and, 458–59; neonatal mortality and, 298, 334, 362, 364, 368, 370, 370; in United States neonatal mortality study, 338, 339, 340–41, 345, 346, 347, 349, 349 specification of risk, 88 spontaneous abortions, infant health production function, 386, 387, 389 spouses: schooling by, 140–41; strategic behavior between, 113 Stanford Heart Disease Prevention Program, 1979 interviews by, 93 state excise tax: on alcoholic beverages, 470–71, 510, 517–18, 535, 755–56; on cigarettes, 470, 486, 578, 579, 596–97, 753; legalization of banned substances and, 761–62 stock of health See health, stock of stock of human capital See human capital stock of knowledge See knowledge, stock of strategic behavior, between spouses, 113 structural demand function: for cocaine demand, 623; for consumption, 606; for stock of health, 58 structural infant health equations for Taiwan 1968 education reform study, 236, 237 substitution effect, 69 substitution in consumption: elasticity of, 70–71, 104n27; time preference and, 98 substitution in production, elasticity of, 72 Surgeon General’s Report on Smoking and Health, 469, 484, 503n1, 572 Survey of Income and Family Expenditure, in Taiwan, 416 Survey of National Opinion Research Center and Center for Health Administration Studies of the University of Chicago, of 1963, 74, 143, 144, 171, 182n34 survival probability: abortion and, 306–7; contraceptive use and, 306–7; infant mortality rate and, 306; neonatal mortality rate and, 333–34; parent’s utility function and, 359; production function of, 306, 333, 359 Swedish twin registry, 289 System to Retrieve Information from Drug Evidence, 620, 626, 759 Taiwan, National Health Insurance: infant mortality and, 302; introduction of, 301–2; 803 before NIH introduction, 405–6; postneonatal mortality, 302 Taiwan, National Health Insurance study: conceptual framework for, 407–9; coverage after NIH, 407, 408; coverage before NIH, 406–7, 408; data and sample for, 409–12, 411; difference-in-differences for, 404; discussion of, 443–45; endogenous inputs in, 409; farm sector employment in, 414–15, 421, 422; fertility and, 416; government sector turnover in, 414; infant health and, 404; low birth weight in, 411, 418, 419–21, 420, 422, 425, 426–29, 430, 431–33, 434, 435–36, 441, 442, 443; marriage for insurance in, 413–14; maternal age in, 415–16, 417, 420, 421, 422; mother’s schooling in, 415–16, 417, 418, 420, 421, 422, 425, 426–29, 430, 431–33, 434, 435–36; parity in, 416, 417, 420, 421, 422; postneonatal mortality rates in, 410–11, 411, 418, 419, 421, 422, 425, 430, 443; predictions about impacts of, 409; prematurity in, 418, 419–21, 420, 422, 425, 426–29, 430, 431–33, 434, 435–36, 442, 443; private sector employment in, 414–15, 420, 421; regression framework for, 404 See also difference-in-differences approach to Taiwan NHI study; medical technology in Taiwan NHI study; sensitivity analysis for Taiwan NHI study Taiwan education reform, of 1968, 213–14, 252; age and program intensity in, 227–28, 228; allocation of junior high schools in, 226–27; compulsory attendance and, 229; Cox’s modified logistic specification for, 232–33; data for, 223–25; dependent variable in, 226; discussion on, 241–42; equations for, 215–16; estimates of effects of, 233–36, 235; family planning program and, 227; junior high school openings under, 219–22, 220, 221; means and standard deviations for, 233, 234; parent’s schooling and, 229–30, 230; returning students in, 230–31; treatment group and program intensity in, 228–29 See also program intensity measure for Taiwan 1968 education reform study Taiwan Panel Survey of Family Dynamics, 223 Taiwan Survey of Income and Family Expenditure, 416 taste hypothesis, allocative efficiency and, 91 Tax Burden on Tobacco, The, 490–91 taxes: for antidrinking campaign, 535; fat, 742; on food, 743; per calorie, 479; soda, 742, 758–59; on unhealthy food, 479 See also beer excise taxes and motor vehicle accident mortality study; federal excise tax; municipal excise tax; state excise tax | INDEX teacher-child ratio, in Turkey Compulsory Education Law study, 257; Young, 265, 266, 267 teenagers: alcohol, excessive use by, 473; antidrinking campaign and, 511, 535–36; marijuana use by, 761 See also alcoholic beverage price and youth alcohol use study; childhood obesity; legal drinking age and youth alcohol use study teenage smokers: by age and sex, 504n10; bandwagon effects and, 487; cigarette demand of, 470, 472, 486; cigarette price and, 485; government effect on, 482; parent’s schooling and, 506n34; trends in, 483, 483–84; underreporting of, 490 teenage smokers, Fairness Doctrine and cigarette price study: cigarette demand function for, 486–89; HES sample for, 485–86; percentagepoint impact of Fairness Doctrine, 500, 500–501; price and related regression coefficients for, 493–95, 494; procigarette advertising and, 492; quantity-smoked regression coefficients of, 497–98; smoking behavior measures in, 489, 490; smoking-participation regression coefficients of, 496–97, 499 teen family planning: in birth outcome production function, 363; prenatal care and, 369 television, in teenage smokers, Fairness Doctrine, and cigarette price study, 491–92; causality of, 506n37; Fairness doctrine and, 492–93; quantity-smoked regression coefficients of, 497–98; sample means of, 490; smokingparticipation regression coefficients of, 496–97, 499 television fast-food restaurant advertising, obesity and, 478 Terman Life Cycle Study of Children with High Ability, 287–88 terminal time problems, optimal length of time and, 51 Thailand, 30 Baht program in, 404 third-variable explanation of schooling and health, 91, 124, 129, 139–40, 213, 215, 287–88 Thorndike, Robert L., NBER-Thorndike sample of, 142–43 time: inputs in household production functions, 9–10; marginal products of, 10; in meal production, 672; scarcity of, 10, 47–48 See also sick time time budget constraint, 10–11, 48 time cost variable, in demand function for physician visits, 79 time discount factor: in alcohol addiction, 606; food prices and obesity and, 738; in rational addiction, 476 time input in gross investment function, 35, 47; in health production function, 78; sick time and, 64 INDEX | 804 time lost, 36; in goods budget constraint, 10; monetary value of, 131; stock of health and, 10–11 time preference: ability hypothesis and, 92; in cigarette addiction, 573; cigarette smoking and, 93–94; in demand function for health, 98; disentangling schooling effect from, 94–95; in Dutch DNB Household Survey, 287; endogenous determination of, 96–98; food prices and, 743; for future, 96–97; in health-schooling studies, 124, 215; infant health and schooling and, 241; intergenerational relationships and, 98; investments in human capital and, 92–93; for present, 96–97; quitting smoking and, 95; rate of interest and, 98, 103n20; schooling and, 92–93, 124; schooling coefficient and, 92–93 tobacco, infant health production function, 386, 392 Tobacco Tax Council, 578, 596 tobit models, 82 tolerance: addiction and, 475, 562–63; instantaneous utility function of, 563 total length of life, 35 travel time effect, 79 Turkey: Bylaw of Primary Education in, 255; child mortality in, 254; marriage age in, 254; women in, 250; women’s employment in, 254 Turkey Compulsory Education Law, 249–51; introduction of, 255; primary school enrollment under, 255–56, 257; teacher hiring under, 255, 256 Turkey Compulsory Education Law study: antenatal visits and, 262, 272, 273, 274; coefficients of interactions between age and program intensity, 267, 268; completed years of schooling for, 262–64, 263, 264; data for, 256–58; education and, 262; education in pre-policy period, 264, 264; fertility design and, 262; fertility results and, 269, 270–71, 272; first birth age and, 262, 272, 273, 274; first-stage regression model for, 258–59; gender equality and, 262, 274, 275–76, 277; geographical unit for, 260–61; marriage age and, 262, 272, 273, 274; mother’s education and child mortality in, 261–62; outcome variables in, 260; probability of years of schooling, 267–68, 269; regional level analysis of, 265, 266, 281n12, 281n13; schooling effect of compulsory education design, 258–61; schooling effect of compulsory education results, 262–68, 263, 264, 266, 268, 269; standard errors in, 261 Turkey Demographic and Health Surveys: contraceptive use and, 254; data of, 256–58; descriptive statistics, 279; previous research with, 254–55 twin differences: health-schooling relationships and, 288–89; in schooling, 291 two-stage least squares: for cocaine demand study, 642; of elasticity of cigarette consumption, 583–84, 585; exogenous variables and, 86; for fast-food restaurant advertising study, 711; of future consumption, 589; of past consumption, 589; production function estimates with, 74; of rational cigarette addiction, 583–84, 584 two-tailed tests, regression coefficients and, 245n15 uncertainty: in human capital model, 86–87, 112; impact of, 88; in multiperiod mixed investment-consumption model, 89; in pure consumption model of demand for good health, 88, 106n49; in pure investment model of demand for good health, 87–88; specification of risk and, 88 unemployment, adult obesity and, 670–71 unhealthy behaviors, economics of, 468, 752 unhealthy food: obesity and, 756–57; regulation of, 741; taxes on, 479 unhealthy goods: behavioral underpinnings of, 468–69; demand for, 469 United Kingdom, compulsory school reform in 1947 in, 218 United States: abortion rate in, 311; infant mortality in, 331–32; Medicaid in, 403; mortality trends in, 773–74; neonatal mortality in, 332; 1970 Census of Population, 310; postneonatal mortality in, 332, 450–51; public policy of, 129, 213 United States, infant mortality study: abortion in, 317, 318, 319, 320, 321; abortion reform and, 311, 313, 315, 317, 323; analytical framework for, 306–9; Area Resource File for, 309–11; county data for, 310–11; crosssectional regression analysis in, 307–8; data and measurement for, 309–11; independent variables in, 312–16, 313; maternal and infant care projects in, 313, 314–15, 322; Medicaid in, 313, 314–15, 317, 318, 319, 320, 322; observed rate decomposition, 308–9; ordinary least squares regressions of neonatal mortality rates, 316–17, 318; race-specific regressions for, 310; rate of decline in, 305; selected factor contribution to reductions in, 319, 320, 321; time period for, 311 United States, neonatal mortality study: abortion in, 338, 339, 341–42; analytical framework for, 333–35; BCHS projects in, 338, 339, 341, 345, 347–48, 354n16; community health centers in, 338, 339, 346, 349; county data for, 336–37; equations for, 334; factor contribution to reduction in, 349; family 805 planning clinics in, 338, 339, 340–41, 346, 347, 349; independent variables in, 340–43, 354n16; lagged mortality rate in, 343–44, 347–48, 353n15; linear regressions in, 345; means and standard deviations of variable for, 339; Medicaid in, 338, 339, 345, 346, 347–49, 349; neonatal mortality measurement in, 335–37; ordinary least squares regressions of, 345, 346, 347–48; production functions for, 335; public program availability in, 344–45; race-specific regressions for, 332–33, 336–37, 338, 339, 343; variable definitions for, 338; WIC program in, 338, 339, 340–41, 345, 346, 347, 349, 349 Universal Primary Education Program in Nigeria, 253 University of Michigan’s Institute for Social Research, 608, 621, 625 unobserved stock of health: in 1976 Danish Welfare Survey, 76–77, 83; in 1986 West German Socio-economic Panel, 81 unstable steady states, rational addiction and, 564–65, 577 utility maximization: continuous time, 30–31; discrete time, 29–30 Van Doorslaer, dynamic demand for health model of, 82–83 vehicle accident mortality study See beer excise taxes and motor vehicle accident mortality study; legal drinking age and motor vehicle accident mortality study vehicle miles traveled, in vehicle accident mortality study, 515; logit motor vehicle accident mortality equations, 519, 520–21, 522; regression with border age omitted, 522–23, 523v visual perception, in NBER-Thorndike sample, 148, 149, 149–50; current health and, 157; estimated bias in schooling coefficients in, 153–54, 154; indirect components and, 160, 160; maximum likelihood estimates of dichotomous logit excellent health functions, 162, 163; means and standard deviations of, 177; ordinary least square estimates of dichotomous excellent health functions, 161, 161; simple correlation coefficient matrix of, 178–79; two-stage squares estimates of health functions with, 167 wage coefficient: in demand function for medical care, 75, 78–79; in pure investment demand function, 78 wage elasticity: consumption model and, 73; of health, 68, 78, 185n69; investment model and, 73; of medical care, 65, 75, 78; wage function and, 185n70 wage function: on-the-job training in, 163–64; schooling in, 217; wage elasticity and, 185n70 | INDEX wage rate, 35; cigarette smoking and, 94; cost minimization and, 32–33; demand curve for health and, 185n67; discounted, 12, 49; elasticity of, 39n32; in energy balance equation, 672; food prices and obesity and, 738, 743; in goods budget constraint, 10, 48; gross investment production function and, 24, 32–33; health change and, 68–69; health demand and, 28, 70, 75; healthy time and, 23–24, 64, 81; human capital stock and, 137–38; length of life and, 56; lifetime wealth and, 77; marginal cost of gross investment and, 65, 68; marginal product of health capital and, 64–65; MEC schedule and, 23–25, 24; medical services demand and, 28; mental ability and, 72; parametric variations in, 68; rate of return and, 65; schooling and, 155; school quality and, 72; stock of health and, 24, 24–25, 44, 61; Wagstaff and, 78; in Zweifel– Breyer critique, 81 See also simultaneousequations health–wage model estimations wage rate in NBER-Thorndike sample, 148, 149; current health and, 156–57; indirect components and, 160, 160; maximum likelihood estimates of dichotomous logit excellent health functions, 162, 163; means and standard deviations of, 177; ordinary least square estimates of dichotomous excellent health functions, 161, 161; simple correlation coefficient matrix of, 178–79; structural equation for, 164–65; two-stage squares estimates of health functions with, 166–69, 167; two-stage squares estimates of wage functions with, 168 Wagstaff: demand function for physician visits in, 79; dynamic demand for health model of, 83–84; estimate of reduced form of pure investment model, 78–79; human capital model differences with, 76–77; initial assets in, 78; principal components analysis by, 76; reduced form health demand function in, 80; wage rates in, 78 Wald test, 587 War on Drugs, 761–62, 775 war on obesity, 742 War on Poverty, 352n3; infant mortality and, 297 wealth: elasticity of demand for health and, 70–71, 73, 105n34; health relation to, 68; health stocks and, 87; interest rate and, 69; length of life and, 115; marginal utility of, 12, 36, 49, 574 See also full wealth constraint wealth effect, 69 Wechsler Intelligence Scale for Children, in Health Examination Survey, 195, 196, 209n7; lagged coefficients of, 199, 200, 201; regression coefficients of family background effects on, 202, 203, 204, 210n15 INDEX | 806 weeks spent in hospital, in MIMIC model, 77 weight difference, in NBER-Thorndike sample, 148, 151; current health and, 155–56; estimated bias in schooling coefficients in, 153–54, 154; as indirect component, 159, 159–60, 160; in life-cycle model, 185n70; simple correlation coefficient matrix of, 178–79; two-stage squares estimates of health functions with, 167 weighting, in regression analysis, 186n77 West German Socio-economic Panel, of 1986, 81 WIC center, infant health production function and, 386, 387, 389 Wide Range Achievement Test, in Health Examination Survey, 195, 196, 209n7; lagged coefficients of, 199, 200, 201; regression coefficients of family background effects on, 202, 203, 204, 210n15 Women of Childbearing Age Tape, 310 women’s education: contraceptive use and, 249, 253–54; in Turkey, 249 See also mother’s schooling women’s employment, in Turkey, 254, 280n8 women’s empowerment, in Turkey Compulsory Education Law study, 258 work, hours of, 36; in energy balance equation, 672–73; in full wealth constraint, 48; in goods budget constraint, 10, 48 work-loss days due to illness, as indicators for health stock, 82 work-loss weeks: health status and, 145; in NBER-Thorndike sample, 144, 144–45, 183n38; in 1963 Survey of National Opinion Research Center and Center for Health Administration Studies of the University of Chicago, 144, 144–45, 183n38; production function of healthy time and, 145–46; variations in, 146–47 Wu-Hausman endogeneity test, quitting smoking and, 95 Wu’s T2 statistic, 373n5 years of experience in labor force, in NBERThorndike sample, 165, 165; means and standard deviations of, 177; simple correlation coefficient matrix of, 178–79; two-stage squares estimates of wage functions with, 168 years of formal schooling, good health and, 89 years of formal schooling completed, in NBERThorndike sample, 148; indirect components and, 160, 160, 173; means and standard deviations of, 177; past health and, 153, 157; simple correlation coefficient matrix of, 178–79; two-stage squares estimates of health functions with, 167; two-stage squares estimates of wage functions with, 168 years of formal schooling completed by father, in NBER-Thorndike sample, 148; estimated bias in schooling coefficients in, 153–54, 154; indirect components and, 160, 160; means and standard deviations of, 177; simple correlation coefficient matrix of, 178–79 years of formal schooling completed by mother, in NBER-Thorndike sample, 148; estimated bias in schooling coefficients in, 153–54, 154; indirect components and, 160, 160; means and standard deviations of, 177; simple correlation coefficient matrix of, 178–79 years of formal schooling completed by wife, in NBER-Thorndike sample, 148; current health and, 155–56; estimated bias in schooling coefficients in, 153–54, 154; as indirect component, 159, 159–60, 160; indirect components and, 160, 160; in lifecycle model, 185n70; means and standard deviations of, 177; simple correlation coefficient matrix of, 178–79; two-stage squares estimates of health functions with, 167 Young, in Turkey Compulsory Education Law study: in mother’s schooling and child mortality, 261–62; in schooling effect, 258–61; teacher-child ratio and, 265, 266, 267 young adults: addiction and, 567; alcohol abuse by, 605; alcohol addiction of, 605–6; antidrinking campaign and, 511, 535–36; cocaine consumption by, 622; mortality trends in, 775 See also alcoholic beverage price and youth alcohol use study; legal drinking age and youth alcohol use study; teenagers youth alcohol use study See alcoholic beverage price and youth alcohol use study; legal drinking age and youth alcohol use study Zweifel–Breyer critique: finding used by, 79; negative relationship between health and investment, 113; non-medical inputs in production of health and, 80; schooling effects in, 81; wage effects in, 80–81; wage rate in, 81 ... for health policy analysts aiming to better understand the determinants of the health outcomes and behaviors in Grossman’s studies This book’s title, Determinants of Health, is well chosen Health. .. measures of schooling attainment and of healthiness or health capital As Grossman notes in the closing words of this book, The positive relationship between more schooling and better health is one of. .. literature on determinants of the health status of the population, only one determinant of which . . is medical care (p 1).” As I rapidly approach the end of my fifth decade as a professional economist,