Population, Fertility and Health 1. Introduction. a) Key facts ο· Improvements in life expectancy Figure 1 ο· Reduction in mortality rates Figure 2 1 ο· Decline in Birth rates lags behind Death rates The demographic transition: ππ = π΅π − π·π Population growth follows an inverted-U relationship (first rises, then declines) Figure 3 Source: Todaro & Smith (2015) b) Relationship between population growth and economic growth In the Solow growth model, a higher population growth rate retards economic growth. c) Causes of high population growth In this topic, we’ll investigate: - What are the principal determinants or causes of high fertility rates in developing countries? - Can these determinants of the "demand" for children be influenced by government policy? 2 To address these questions, we’ll develop two economics models and analyze their economic implications: - Macroeconomic perspective: Malthusian “population trap” model - Microeconomic perspective: Household theory of fertility Finally, we’ll provide insights for the role of government policy from randomized studies: - The effectiveness of financial incentives in improving vaccination rates. 2. Malthusian model. a) Setup: Population growth follows an inverted-U relationship (first rises, then declines) ππ = π + ππ¦ + ππ¦ 2 with the following parametric restrictions π > 0 and π < 0. Output growth must be consistent with the Solow model’s predictions about convergence/divergence: If divergence: ππ = π If convergence towards good equilibrium, where π¦ < π¦ ∗ : ππ = ππ¦, where π < 0. b) Model's predictions: ο· Equilbria At equilibrium, ππ = ππ . Since the ππ and ππ curves cross twice, there are two equilibria. 3 ο· Stability of equilibria and transitional dynamics If ππ > ππ , π¦ declines. If ππ < ππ , π¦ increases. These transition dynamic give rise to one stable equilibrium and one unstable equilibrium. The presence of two equilibria and the transitional dynamics give rise to a Malthusian trap. Figure 4 Source: Todaro & Smith (2015) c) Implications for development: ο· countries in middle-income range may be caught up in a Malthusian trap; ο· rich enough countries escape the Malthusian trap; ο· countries that raise savings rate, productivity may escape the Malthusian trap; ο· effective government policies can only be big push policies. Big push policies must be set such that ππ > ππ at a country’s current income level. 4 3. The Microeconomic Household Theory of Fertility. a) Overview Why do birth rates decline after death rates? ο° Households’ choice of how many children to have. In this section, we will explore whether the Malthusian predictions are consistent with individual behavior. The exposition of this theory is based on Todaro and Smith (2015), pp. 303 – 307. b) Setup ο· Goods Households care about two goods: number of children (πΆ) and other goods (πΊ). ο· Optimization behaviour The conventional theory of consumer behaviour assumes that an individual with given set of preferences for a range of goods (a utility function) tries to maximize their own well-being from consuming these goods subject to their own budget constraint. In the application of this theory to fertility analysis, children are considered as a special kind of consumption good relative to other goods. Parents enjoy benefits of having kids, e.g., caring for them when old. ο· Properties of goods - Why is 'more children' better? A higher probability that at least one child survives until adulthood to take care of their parents when old. - πΆ and πΊ are assumed to be normal goods. 5 ο· Demand function for desired number of children πΆπ = π(πΌ, ππΆ , ππΊ , π) - The higher the household income, the greater the demand for children (ππΆπ = ππΆπ ππ > 0). - The higher the net price of children, the lower the quantity demanded (ππΆππ = ππΆπ πππΆ < 0). - The higher the prices of all other goods relative to children, the greater the quantity of children demanded (ππΆππΊ = ππΆπ πππΊ > 0). - The greater the strength of tastes for goods relative to children, the fewer children demanded (ππΆπ = ππΆπ ππ < 0). c) Model's predictions Suppose that the initial optimal bundle is at pt. f on indifference curve πΌ2 . We will explore two alternatives that differ with respect to the introduction of a government policy that raises the cost of raising children: - Rising income but no change in prices A pure income effect (IE) of rising income is depicted by a parallel shift of the budget constraint from ππ to π′π′. As a result, both πΆ and πΊ increase since the new optimal bundle is at pt. h. This choice implies that higher income increases fertility. - Rising income and increase in the price of raising children A pure substitution effect (SE) resulting from a rising price of raising children is depicted by a pivot of the budget constraint from ππ to ππ". As a result, both πΆ and πΊ decrease since the new optimal bundle is at pt. e. A pure income effect (IE) of rising income is depicted by a parallel shift of the budget constraint from π′π’ to ππ. As a result, both πΆ and πΊ increase at the new optimal bundle at pt. g relative to the optimal bundle at pt. e. 6 Combining the two effects, IE and SE, implies that the πΊ increases but πΆ decreases. That is, an economic policy that increases the price of raising children while incomes are rising implies a declining fertility. Figure 5 Source: Todaro & Smith (2015) 7 Income effect (IE): A higher income implies that parents demand more children. Substitution effect (SE): A decrease in the opportunity cost of children rises (employment opportunities, especially for females, increase), parents demand fewer children. If IE dominates, the number of children demanded increases. ο° This result conforms with the predictions of the Malthusian trap that income growth increases population growth. If SE dominates, the number of children demanded increases. ο° This result implies that an economy might escape the “population trap” without government intervention. d) Implications for development As income rises, the birth rate may even rise. If pension plans are adopted during early stages of development, birth rates are likely to decline faster. 4. Government policy. There are three main directions, in which government policy could help a country escape the Malthusian “population trap”. a) Technological innovation Big push policies in technological innovation are possible in the Ghatak and Jiang model or the RR big push model. b) Demographic restrictions 8 Big push policies that temporarily restrict population growth, e.g. the one child policy in China. To be effective, these policies must target all individuals or at least a critical mass of them. The policy implications of such policies produce benefits for future generations by letting countries escape the Malthusian population trap. However, they come at a cost for current generations subjected to the policy. c) Preventative medical interventions, e.g. vaccines The discussion is based on Benarjee et al (2012). ο· Reducing child mortality rates High mortality rates among children due to easily preventable diseases. Immunization improves substantially the likelihood of surviving until adulthood. “Immunisation is a highly cost effective way of improving survival in children in developing countries. Every year throughout the world, however, an estimated 27 million children and 40 million pregnant women do not receive the basic package of immunisations (as defined by WHO and Unicef), and two to three million people die from diseases that can be prevented with vaccines.” ~ Benarjee et al (2012) While reducing child mortality rates is an end in itself, preventative policies reduce fertility rates over time but may contribute to population increase in the short-run. ο· Policy effectiveness - Availability of services vs. financial incentives Financial incentives, such as in conditional cash transfer programmes, can be effective in promoting the use of certain preventive healthcare services. In settings with reliable immunisation services and a high pre-existing immunisation rate such programmes have little impact on immunisation. 9 In a setting with a low immunisation rate (under 6%), improving the reliability of services modestly improved uptake of immunisation. Small non-financial incentives, combined with improved reliability, have large positive impacts on the uptake of immunisation and are more cost effective. Provision of immunization services, even free of charge, typically results in modest increases. ο· Why randomized studies? There are several sources of selection bias in rural areas. Some of them, listed in the working paper version, include: - Access to roads and proximity to a health post; - Service availability is unpredictable (staff skips work on a regular basis or health post is closed for an extended period of time). How does each of these sources affect selection bias? Example: Service availability ο· Regression analysis Design: 134 villages in rural India were randomized to one of 3 groups: - a once-monthly reliable immunization camp (intervention A; 30 villages); - a once-monthly reliable immunization camp with small incentives (lentils and metal plates for completed immunization intervention B; 30 villages), - control (no intervention, 74 villages). 10 Surveys were undertaken in randomly selected households at baseline and approximately 18 months after the interventions started (at endline). The primary end point was the proportion of village children aged 1-3 at endline who were partially or fully immunized. Participants: 1640 children aged 1-3 at endline. PRF: ππ = π½0 + π½1 π·1π + π½2 π·2π + π½3 π·1π π·2π + π’π ππ − immunization rate π·1π − access to a reliable immunization camp π·1π − financial incentives Results: able 2 | Effects of group allocation (A=immunisation, B=immunisation plus incentive). Numbers are in absolute values Mean (95% CI) Control Difference* (95% CI) A B 379 382 Relative risk† (95% CI) A−control B−control B−A — — — A v control B v control — — BvA End point cohort (aged 1-3 years) No in group 860 — No of immunisations 1.20 2.35 2.85 (0.94 to 1.46) (1.99 to 2.71) (2.44 to 3.25) 1.15 1.70 0.55 (0.95 to 1.35) (1.48 to 1.92) (0.26 to 0.83) 2.14 2.66 (1.84 to 2.44) (2.28 to 3.05) 1.22 (1.08 to 1.36) ≥1 immunisation 0.49 0.78 0.74 (0.40 to 0.57) (0.70 to 0.85) (0.67 to 0.82) 0.29 0.26 −0.03 (0.23 to 0.35) (0.20 to 0.33) (−0.09 to 0.04) 1.59 1.52 (1.35 to 1.83) (1.29 to 1.75) 0.96 (0.80 to 1.11) Has BCG scar‡ 0.28 0.50 0.50 (0.21 to 0.36) (0.41 to 0.59) (0.41 to 0.59) 0.22 0.22 0.00 (0.15 to 0.28) (0.15 to 0.28) (−0.08 to 0.08) 1.76 1.76 (1.41 to 2.12) (1.41 to 2.12) 1.00 (0.79 to 1.21) Completely immunised 0.06 0.18 0.39 (0.03 to 0.09) (0.11 to 0.25) (0.30 to 0.47) 0.13 0.34 0.21 (0.09 to 0.16) (0.30 to 0.38) (0.15 to 0.28) 3.09 6.66 (1.96 to 4.21) (4.53 to 8.80) 2.16 (1.54 to 2.78) Logbook cohort (aged 1-3 years) No in group — — — — — — — Total No of immunisations — 3.70 4.18 (3.39 to 4.01) (3.99 to 4.37) — — 0.59 (0.25 to 0.93) — — 1.15 (1.05 to 1.25) Completely immunised — 0.48 0.67 (0.38 to 0.59) (0.59 to 0.74) — — 0.22 (0.10 to 0.33) — — 1.43 (1.12 to 1.73) 407 725 *Estimated by fitting multilevel mixed effect linear model, with clustering at hamlet and household level. †Estimated by fitting multilevel mixed effect Poisson regression, with clustering at hamlet and household level. ‡For analysis with BCG scar as outcome, there were 790 observations in control group, 334 in treatment A group, and 336 in treatment B group. Source: Banerjee et al (2012) 11 Interpretation: Improving reliability of services improves immunization rates, and small, non-financial incentives have large positive impacts on the uptake of immunization services in resource-poor areas. 12