Development Economics

1
Development Economics

• Population
• Health Nutrition

2
World population
3
Source Demographic Yearbook, 2004, United Nations
4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
Demographic transition

• Mortality decrease
• Infant mortality first, then adult mortality
• 3 main factors, not independent from each other
• Health technologies (infrastructures, drugs,
  global public goods)
• Mothers education (quality of nutrition and
  child care)
• Income increase (nutrition).
• Fertility transition
• Progressive adjustment to the number of children
  actually wanted (strategic complementarities)
• Quantity-quality trade-off drawn by the increased
  survival of children
• Further decrease in the number of birth due to
• Lowering uncertainty on life expectancy of
  children
• Change in the economic and social condition of
  women
• Strong variations of the dependency ratio over 30
  to 50 years
• Problem labour market absorption capacity.

10
(No Transcript)
11
(No Transcript)
12
Fertility choices

• Demand for children (Becker - 1960)
• Expected benefits
• Missing markets
• Old age support
• Insurance
• Child labor
• Costs
• Child rearing (direct and indirect costs)
• Uncertainty
• Child survival
• Child earnings in adulthood
• Child attention to its parents in adulthood.
• Increases the number of desired children
• Can be reduced through changes in mortality rates
  through information

13
Income improvements and fertility choices
14
Is fertility too high?

• Information
• relative to current mortality rate
• Uncertainty
• overshooting
• Externalities
• Within societies
• Within families
• Social Norms
• Television.

15
Population issues

• Negative impact
• Malthusians models
• Other models with capital dilution (Solow,
  Ramsey)
• Impact of black plague and influenza / Imapct of
  AIDS ?
• Examples of organized lowering of fertility
  (China, Tunisia)
• Positive impact
• Models with technical progress depending from the
  population (Boserup, Simon)
• New consensus non significant impact ? health
  issues

16
(No Transcript)
17
(No Transcript)
18
Health and Nutrition

• Descriptive elements
• Measures
• Relation between health/nutrition and income
• Difficulties to identify causal relations
• Impact of income on health/nutrition
• Impact of health on income
• Theory of efficiency wage
• Empirical results

19
Nutrition and income
20
Famines

• Not always due to insufficient availability of
  food resources at the country level
• Sen Problem of access to available resources
  trade entitlement failure.
• Articulation between initial endowments and
  economic and social environment.
• Ex of Bengal in 1943. Relative price of rice
  shoots up. Terms of trade labour/rice decrease
  from 100 in 1939/40 to 32 in 1942/43 households
  who suffered most fishermen and agricultural
  wage workers. (6,4 of sharecroppers and small
  farmers died while 40,3 of agricultural wage
  worker did).

21
Health and income Strong correlation between
income and nutrition, that diminishes with the
countrys standard of living
22
During the war, the increase in the stature of
Vietnamese men was brought to a halt.
23
The health/wage correlation is not entirely due
to difference in education.
24
Measuring health

• Health is multidimensional
• Several measures
• Each dimension can have a different impact on
  productivity and/or labour supply.
• Measures of health
• Measures de nutrition

25
Measures of health

• General Health Status Self-evaluation of health
  status on a scale from 1 to 5.
• Doesnt capture the multiplicity of health
  dimensions
• How to interpret the answer  good health?
• What is the norm? Comparability issue.
• Individuals with low access to the health system
  over-estimate their health status (they assume
  they are ok as long as they dont have any
  contrary information)
• Measurement error systematically correlated with
  income under-estimation of the impact of health
  on wage.

26
Measures of health (2)

• Symptoms
• Diarrhoea, fever, respiratory difficulties
• Same measurement error problem than GHS
• Days where normal activity could not be carried
  out.
• Individuals for whom the opportunity cost of time
  is the highest loose less days.
• Norm is adapted to physical condition.
• Capacity to undertake certain efforts
• For the younger individuals, there is very little
  variance.

27
Nutritional measures.

• Calories availability
• Food expenditures (including self-consumption)
  are converted into calories.
• Suppose no losses
• Does not take into account meals taken outside
  the home or offered to visitors.
• Calories intake
• Weighing before and after meals.
• 24 hours recall
• The difference between availability and intake is
  positively correlated to income.

28
Nutritional measures.(2)

• Anthropometric Measures
• Size long term indicator, determined by
  nutrition in infancy.
• Body Mass Index (weight in kg / (size in m)²)
  indicator of current nutritional status. Higher
  mortality risk for BMIlt18 or gt30.
• Arm circonference
• Measurement errors are not systematically
  correlated to individual characteristics.
• Can be used as control when using calories
  consumption data (needs are correlated to BMI).

29
Relation income/nutrition

• A priori, it exists both ways
• Income ? nutrition/health through food
  consumption and demand for health care.
• Nutrition/health ? income through its impact on
  productive capacity.
• Serious identification problems.

30
Theoretical framework

• Production function for health
• HH(N,L A,B,D,?,eh)
• where N health inputs and Llabour supply
• A individual characteristics, B family
  characteristics, D health environment.
• ? unobservable, e measurement error.
• Wage function
• ww(H A,S,B,I,?,ew)
• S education, I community variables
  (infrastructures)

31
Identification Problems

• If ? and ? have common components, the estimated
  effect of health on wage will be over-estimated.
• Simultaneity
• cpo
• ? If N increases w, via H, hence the implicit
  price of N decreases and its consumption
  increases.

32

• Reduced form for nutrition demand
• Nj Nj (pn , pc , A,S,B,V,D,I, ? ,? ,?)
• ? also appears in the wage function (hence the
  simultaneity issue.
• Possible instrumental variables conditionnally
  to H and I, pn and D could be valid instruments
  for health in wage functions. Often distance to
  health infrastructures, quality, food price
• Problem in case of endogeneous placement.

33
Impact of health on labour supply

• Same difficulties
• In reduced form, impossible to say whether imapct
  of health on labour supply is due to productivity
  (income effect and/or price effect) or to the
  direct effect of preferences.
• Identifying the structural modelrequires to find
  the appropriate instruments.
• Pitt, Rosenzweig Hassan 1990 estimate a
  structural model.
• Foster Rosenzweig 1994 Healthiest individuals
  choose pice-rate remuneration or
   self-employed  jobs. Daily wage workers put in
  less effort.

34
Impact of income on nutrition

• Elasticity of calories demand with respect to
  income
• If high, there exist substitution possibilities
  with other consumption if calorie cost increases.
  
• From the nutritional point of view it is not
  necessarily desirable.
• Very varied results from 0,86 in Sierra Leone
  (Strauss 1984), to 0,01 in Indonesia
  (PittRosenzweig 1985)

35
Impact of nutrition on productive capacity
efficiency wage theory

• Leibenstein (1957) at low levels of nutrition,
  the production capacity is weak. When nutrition
  improves, productive capacity increases, first at
  increasing rate and then at decreasing rate.
• (Stiglitz, Dasgupta, Ray)

36
There exists a minimum wage ( efficiency wage )
under which productivity jumps discontinuously to
very low levels.
37
Other consequences

• In equilibrium, involuntary unemployment affect
  individuals in poor health.
• Non-wage inequalities are transmitted to the
  labour market.
• Long term labour contracts should be observed
  because of future returns to current nutrition.
• Intra-family disparities
• No empirical results to confirm convincingly the
  existence of the poverty trap suggested by
  theory.
• Pitt, Rosenzweig Hassan 1990 intra-family
  inequality, but taxation of the better endowed.
• Rural wages in India gtgtgt nutritional minimum
  (Swamy, 1997)
• Daily labour contracts.

38
Impact of nutrition on productive capacity

• But, despite empirical difficulties, results
  point to a rather positive effect
• Strauss 1986, Sierra Leone, elasticity of
  agricultural production to calories intake0,33,
  i.e. a worker who consumes 1500 calories a day is
  0,6 times less productive than the one who
  consumes 2400.
• Sahn Alderman, 1988, Sri Lanka, elasticity of
  male wage to calorie consumption0,2.

39
Impact of health on income

• Stronger for the poor, not very big
• Brazil (ThomasStrauss) not be illiterate
  30cm
• In general it disappears with instrumentation.
• Thomas Strauss, Brazil
• Size has the strongest effect
• BMI increases male wage, mainly for the less
  educated
• Calories intake increases wage at very low level
  of consumption
• Protein intake has higher returns for highest
  consumption levels.
• Schultz Tansel, impact of illness
• Côte dIvoire men having lost 1 day of work
  during the month have a wage lower by 18 no
  effect in Ghana no effect for women.

40
Controlled experiments

• Cure against schistosomiases (sugar cane
  plantations)
• Tanzania increase the income of infected workers
  but not enough to catch up with non-infected
  workers.
• Cameroon no effect, but no difference ex-ante
  either.
• Iron supplementation (rubber - Indonesia)
• Anaemic workers (20 less productive than others)
  , catch up with others after 60 days of treatment
  (about 300 obs.,).
• Caloric supplementation
• Small positive imapct of hourly productivity
  (Kenya, 47 obs. )

41
Other impact of health

• Impact on education
• Glewwe Jacoby, Ghana impact of undernutrition
  on age of entry at school
• Miguel Kremer, Kenya, Deworming increases
  attendance but does not improve the results at
  cognitive tests.