University of Nairobi Economic development of US Essay I will post an essay on the west coast of the United States. The time limit is completed within 24 h

University of Nairobi Economic development of US Essay I will post an essay on the west coast of the United States. The time limit is completed within 24 hours. I uploaded some reading materials but I am not sure if it is needed. The suggestion is that tutor can take a look to save writing time. This time the essay requirement is about 1000 words with the theme of economic development. I set the deadline time to 24hours, I will delay the time according to the newly requested deadline, no need to worry.but please remember that there is only one day to complete this essay. Health and economic development
Measures of health
•
•
•
•
•
Life expectancy
Infant mortality rate
Hunger and malnutrition
Burden of disease
Self reported health and morbidity
Causes of mortality in HICs
Causes of mortality in LICs
Note: Why measurement is important
Note: Why measurement is important
Outline for the week
• Macro health and economic growth
• Population dynamics over time
• Micro economic theories to understand health
investments and outcomes (Dasgupta & Ray
1986, Grossman 1972)
• Examples of research trying to understand sub
optimal health investment and policies in
developing countries [your readings]
Macro health and development
• Population health positively associated with
macroeconomic outcomes.
• What about changes in population level
health and economic growth?
Note on causality
• Channel of causation is not clear (health changes can increase
incomes, increased incomes can improve health)
• One way is to look at global changes in health technology, and
think about places with heavier ex ante disease burdens –
these places should be affected more by a given change in
technology
• Two examples of studies that use this:
– Acemoglu and Johnson (2007): use large scale reductions
from infectious diseases around the world
– Bleakley (2010): uses malaria eradication
The Global Epidemiological Transition
(Acemoglu and Johnson 2007)
Pre-transition
Disease Prevalence
Year, Relative to Transition
Before
After
High
–
+
Low
+
+
Mortality Reductions and Changes in Life Expectancy at Birth
40.0
Change in Life Expectancy, 1940-1980
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
0
0.2
0.4
0.6
0.8
Reduction in Predicted Mortality
1
1.2
Mortality Reductions and Population Growth
8.0
Annual Pop. Growth Rate, 1940-1980
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
0
0.2
0.4
0.6
Reduction in Predicted Mortality
0.8
1
1.2
Mortality Reductions and Economic Growth
12.0
Annual Growth Rate of GDP per capita, 1940-1980
10.0
8.0
6.0
4.0
2.0
0.0
0
-2.0
0.2
0.4
0.6
0.8
Reduction in Predicted Mortality
1
1.2
Studying the Lasting Effects of
Eradication over the Lifecycle (Bleakley 2010)
Pre-eradication
Malaria Prevalence
Childhood, Relative to Eradication
Before
After
High
–
+
Low
+
+
Percent Change in Income, Pre-Eradication Birth
Cohorts vs. Post-Eradication Birth Cohorts
Index of Pre-Eradication Malaria Incidence
Bleakley, AEJ Applied, 2010
Population
• Previous slides and our discussion of Solow
suggest population dynamics are important
• How do we characterize broad changes in
birth and death rates? Demographic transition
• Why does the demographic transition occur?
Population dynamics and
development
Demographic Transition
Stages of demographic transition
1. Birth and death rates are high, somewhat in balance
2. Death rates fall due to improved health, birth rates take longer to adjust
3. Birth rates fall due to reduced child mortality, contraception, increased
urbanization etc
4. Low birth rates and low death rates – birth rates may fall below replacement
levels
Demographic Change in Sweden
Stage 2 – Demographic Dividend
Think of this as a consequence of Stage 2 of the demographic transition
1. Falling mortality and rising birth rates imply an increase in working age
population in the future
2. Outside Africa, the dividend has been good for growth
3. What are some of the conditions required to take advantage of the
demographic dividend?
Stage 3 – when do birth rates fall?
1. Income/wages per capita
1. Income/substitution effect
2. Status of women
1. Delayed marriage – greater schooling (example in slides)
2. Returns to schooling
3. Family Planning
4. Infant Mortality (example in slides)
5. Social insurance (example in slides)
6. Quantity-quality trade off
Ex: Schooling and Fertility
Ex: Schooling and Fertility
Ex: Child mortality and fertility
Ex: Fertility and Old age insurance
Need to plan for old age, but there are no markets for old age insurance
Children very often cited as source of old age insurance
But, women have a limited fertile cycle AND infant/child mortality is high
Let the probability that a child grows up to look after you be p
Let q be the threshold probability that you need for old age insurance (that atleast one
child takes care of you
How many children (n) do you need such that you reach this threshold?
If p=0.5 and q=0.9, n=4!!
1-(1-p)n>q
Stage 4 – very low birth rates
Why is fertility so low in Europe?
Kohler et al 2006
Reasons for Low Birth rates
Kohler et al 2006
Reasons for low birth rates
Kohler et al 2006
Japan and birth rates
What happened in Japan? The sharp drop is not data error.
Auspicious years for fertility in
Vietnam
Do and Phung 2006
Similar norms in China
Micro approach to health and
development
• Health as an input to productivity (think of Solow)
• Health as something we might value (“consuming”
health)
• Health has a fundamental right (Sen’s ”capabilities”
approach)
• Health is salient in developing countries:
– If health drives incomes and incomes drive health,
this could put developing countries in a health
based “poverty trap” if incomes (and health) are
low
Poverty Traps Through Health
• Many mathematical models of poverty traps
• Most have the same basic ingredients.
• Today: a model of subsistence agriculture by
Partha Dasgupta and Debraj Ray.
• Basic idea:
•
Nutrition ↔ Productivity
Food, time t+1
Food, time t
Dasgupta & Ray (1986)
• Illustrates both nutrition-based efficiency wages and poverty
traps more generally
• The origins of this model come from an empirical puzzle, which
is the existence of involuntary unemployment ->
– Why don’t the unemployed undercut the employed and work for a
lower wage?
• Perhaps because employers prefer to hire fewer workers at a
higher wage because it is more efficient?
– What does efficiency mean in this context?
– Health-income-health linkages are one mechanism
• Can this generate a poverty trap?
Dasgupta & Ray (1986) – Model Setup
• Based on relationship between calorie intake, basal metabolic
requirements of calories, and work capacity
• Resting metabolism -> maintaining body temperature,
breathing, heart pumping, kidneys functioning etc
– ~1500 – 2000 Kcal/day
• Energy for work comes over and above this
• Total hours are fixed, but effectiveness is not
• Workers supply “effective labor” or “labor power”, which is a
function of hours and effectiveness/hour
• This is what the employer cares about
• Key is the convexity in the relationship between income and
work capacity (mediated through calorie intake)
Capacity Curves
•
•
•
•
Difference between the two pictures?
What does I* represent?
Note the regions of increasing returns
Can draw the capacity curve as having multiple regions of
increasing returns (hence possibility of multiple equilibria)
Poverty Traps
• Model provides a framework for understanding
persistent disadvantage
• Testing the model: less clarity on this but consider that
– Calories are cheap in poor countries
– Is the relationship between income and calories that steep?
(will return to this in a few slides!)
• Regardless, the model’s logic may be correct in other
applications
– Child investment à intergenerational transmission of
poverty.
– Institutions and macroeconomic growth.
– Many other examples.
Investing in Health
• Health is like other human capital that one can
invest in and can get returns from
• What is the optimal level of health
investment?
• Why don’t people invest optimally?
– Sub optimal demand?
– Poor supply?
Supply of health
• Limited provision
• Low quality
• Drug development in LICs
Limited Provision
Low quality of service
Das, Holla, Mohpal, Muralidharan (AER 2016)
Das, Holla, Mohpal, Muralidharan (AER 2016)
Summary of Results
Drug related R&D in LDCs
• While the benefits to developing countries of pharmaceutical
research is large, there is very little research on LDC diseases
like malaria or tuberculosis (TB)
• Pecoul et al. (1999) report that of the 1,233 drugs licensed
world wide between 1975 and 1997 only 13 were for tropical
diseases. Of these
–
–
–
–
5 came from veterinary research
2 were modifications of existing drugs
2 were produced for the US military
Only 4 out of 1,233 drugs were developed by commercial
pharmaceutical firms for human tropical diseases
• WHO (1996) reports that 50% of global health R&D is
conducted by the private sector, but only 5% is focused on the
diseases of LDC’s
Why is this the case?
• Limited incentives for such research in for profit pharmaceutical
companies due to small effective market size (EMS)
– EMS = number of consumers who have the ability to pay
• Connecticut alone spends more on health care than the 38
LDC’s of sub-Saharan Africa
• Research is very expensive
• Typical new drug costs ~$500M to take to the market
• Hence the focus of the pharmaceutical industry on looking for
“block buster” drugs that will make a lot of money
– Lipitor, Avastin, etc
• ARV’s were first developed for HIV patients in developed
countries – then adapted and subsidized for LDC’s
• But no incentive to come up with a vaccine for malaria/TB etc.
• So what are the policy options available?
Advanced Market Commitments
• An innovative and promising idea
• Instead of spending billions on government directed research that
may or may not be successful, a combination of
donors/governments/development agencies could create an
“advanced market commitment” (AMC) to buying doses of a
successful vaccine at a pre-committed price
• Consider a commitment to buy 500M doses of a malaria vaccine at
$10/each
• Now, there is a $5B market opportunity for private firms to pursue
• Welfare gains of a successful vaccine would far exceed the costs
– Consider the polio, and small pox vaccines
• BUT, you only pay if the drug is successful!!
Advanced Market Commitments (2)
• Markets are great at rewarding innovation and efficiency and
market production is typically more efficient than government
production
• But, the market only cares about you if you have purchasing power
• The use of AMC’s can potentially combine the best of both worlds
• Public financing puts purchasing power behind the needs of the
poor to make sure their needs have a “voice”
• But, competitive private provision can provide better incentives for
efficiency than those found in government provision
• Again – you only pay IF the research is successful
• So, the RISK is with the private sector (as it should be – this is how
all drug development takes place), but there is a big reward
Advanced Market Commitments (3)
• There are some potential limitations:
– Exact outcome has to be specified in advance
– What if successful drug is found that works in 20% of the population?
– What if a clinically successful vaccine is developed but user countries don’t
like it?
• Details of the contract and commitment matter a lot
– But, can be figured out
• Rewarding applied products may not stimulate the required basic
science research that is needed
– AMC’s are not a substitute to government-funded basic research but a
complement that fills a market failure on the “applied research” side
– Academic and reputational rewards combined with public funding seem to
work pretty well for basic research – the gap is in “applied research”
Breakout room Class 1: Practice question on
fertility
Breakout room class 2: Video and
• Watch this video: discussion

• Discussion questions:
– Name any 3 factors that led to the eradication of malaria in
the US?
– Were there any potential downsides to malaria
eradication?
– Suppose you were brought in as a consultant to the CDC to
examine the malaria eradication episode and compute
costs and benefits, how might you go about such a task?
Assume they will give lots of data to do this.
• Define ”benefits” first. What are some of the benefits? What
method from last week might be best suited to analyze this?
• Define some of the costs. How would you compute these?
L2: Breakout room question 1
Applied Impact
Evaluation
Causal Inference and Overview of
Impact Evaluation Methods
Impact Evaluation
Logical Framework
How the program
works in theory
Measuring Impact
Identification Strategy
Data
Operational Plan
Resources
1: Causal Inference
Counterfactuals
False Counterfactuals
• Before & After (Pre and Post)
• Enrolled and Not Enrolled (Apples and
Oranges)
“
Our Objective
Estimate the causal effect (impact) of
intervention (P) on outcome (Y).
(P) = Program or Treatment
(Y)= Indicator, Measure of Success
Example: What is the effect of a Cash Transfer Program
(P) on Household Consumption (Y)?
Problem of Missing Data
Estimate what would have happened to Y
in the absence of P.
We call this the Counterfactual.
od
o
g
a
to
y
is
e
k
n
o
e
i
t
h
a
T
lu
a
v
e
t
impac
d
a vali
l!
a
u
t
c
r fa
e
t
n
u
co
Finding good comparison groups
The treatment and comparison groups should have
identical characteristics, except for benefiting from
the intervention.
In practice, use program eligibility & assignment rules
to construct valid counterfactuals.
n
o
s
i
r
mpa on for
o
c
od
o
eas ween
g
r
a
y
l
With , the on mes bet isons
p
o
r
grou ent outc compa
d
r
diffe ents an tion (P)
m
n
treat interve
e
is th
Case Study: Progresa
National anti-poverty program in Mexico
• Started in 1997
• 5 million beneficiaries by 2004
• Eligibility –based on poverty index
Cash Transfers
• Conditional on school and health care
attendance
Case Study: Progresa
Rigorous impact evaluation with rich data
•
•
506 communities, 24,000 households
Baseline 1997, follow-up 1998
Many outcomes of interest
•
Here: consumption per capita
What is the effect of Progresa (P) on Food
Comsumption Per Capita (Y)?
•
If impact is an increase of $20 or more, then scale up
nationally.
1: Causal Inference
Counterfactuals
False Counterfactuals
• Before & After (Pre and Post)
• Enrolled and Not Enrolled (Apples and
Oranges)
False Counterfactual #1: Before &
After
Y
A
A-C = 2
IMPACT?
C (counterfactual)
A-B = 4
B
T=0
T=1
Baseline
Endline
Time
Case 1: Before & After
What is the effect of Progresa (P) on
consumption
(Y)?
Y
(1) Observe only
beneficiaries (P=1)
(2) Two observations
in time:
Consumption at T=0
and consumption at
T=1.
268
A
α = $35
233
IMPACT=A-B= $35
B
T=1997
T=1998
Time
Case 1: Before & After
Consumption (Y)
Outcome with Treatment
(After)
268.7
Counterfactual
(Before)
233.4
Impact
(Y | P=1) – (Y | P=0)
35.3***
Estimated Impact on
Consumption (Y)
Linear Regression
35.27**
Multivariate Linear
Regression
34.28**
Note: If the effect is statistically significant at the 1% significance level,
we label the estimated impact with 2 stars (**).
Case 1: What’s the problem?
Economic Boom:
o Real Impact=AC
o A-B is an
overestimate
Economic Recession:
o Real Impact=A-D
o A-B is an
underestimate
Y
268
A
Impact?
α = $35
C
?
233
esn’t
Before & After do
mecontrol for other ti
varying factors!
Impact?
B
D?
T=1997
T=1998
Time
1: Causal Inference
Counterfactuals
False Counterfactuals
• Before & After (Pre and Post)
• Enrolled and Not Enrolled (Apples and
Oranges)
False Counterfactual #2:
Enrolled & Not Enrolled
If we have post-treatment data on
o Enrolled: treatment group
o Not-enrolled: “comparison” group
(counterfactual)
Those ineligible to participate.
Those that choose NOT to participate.
Selection Bias
o Reason for not enrolling may be correlated with
outcome (Y)
Control for observables.
But not un-observables!
o Estimated impact is confounded with other
things.
Case 2: Enrolled & Not Enrolled
Measure outcomes in post-treatment
(T=1)
Ineligibles
(Non-Poor)
Eligibles
(Poor)
Not Enrolled
Y=290
Enrolled
Y=268
In what ways might E&NE be different, other than their enrollment in the program?
Case 2: Enrolled & Not Enrolled
Consumption (Y)
Outcome with Treatment
(Enrolled)
268
Counterfactual
(Not Enrolled)
290
Impact
(Y | P=1) – (Y | P=0)
-22**
Estimated Impact on
Consumption (Y)
Linear Regression
-22**
Multivariate Linear
Regression
-4.15
Note: If the effect is statistically significant at the 1% significance
level, we label the estimated impact with 2 stars (**).
Progresa Policy Recommendation
Impact on Consumption (Y)
Case 1: Before &
After
Case 2: Enrolled
& Not Enrolled
Linear Regression
Multivariate Linear Regression
Linear Regression
Multivariate Linear Regression
35.27**
34.28**
-22**
-4.15
Will you recommend scaling up Progresa?
B&A: Are there other time-varying factors that also
influence consumption?
E&BNE:
o
o
Are reasons for enrolling correlated with consumption?
Selection Bias.
Note: If the effect is statistically significant at the 1% significance level, we label the estimated
impact with 2 stars (**).
Keep in Mind
B&A:
Compare: Same individuals Before and After they receive
P.
Problem: Other things may have happened over time.
E&NE
Compare: Group of individuals Enrolled in a program
with a group that chooses not to enroll.
Problem: Selection bias. We don’t know why they are not
Both counterfactuals may
enrolled.
lead to biased estimates of
the impact.
2: IE Methods Toolbox
Randomized Assignment
Randomized Offering/Promotion
Discontinuity Design
Difference-in-Differences (Diff-in-Diff)
Matching (P-Score matching)
2: IE Methods Toolbox
Randomized Assignment
Randomized Offering/Promotion
Discontinuity Design
Difference-in-Differences (Diff-in-Diff)
Matching (P-Score matching)
Randomized treatments and comparisons
1.
Population
2. Evaluation
sample
3. Randomize
treatment
Comparison
Treatment
= Ineligible
= Eligible
External
Validity
Internal
Validity
Unit of Randomization
Choose according to type of program
o Individual/Household
As a rule of thumb,
o School/Health Clinic/catchment area rand
omize at the
o Block/Village/Community/Region
smallest viable unit
of implementation.
Keep in mind
o Need “sufficiently large” number of units to detect
minimum desired impact: Power.
o Clustering reduces effective sample size
o Standard Errors need to be “clustered”
o Spillovers/contamination
o Operational and survey costs
Case 3: Randomized Assignment
Progresa CCT program
Unit of randomization: Community
506 communities in the evaluation sample
Randomized phase-in
o 320 treatment communities (14446 households):
First transfers in April 1998.
o 186 comparison communities (9630 households):
First transfers November 1999
Case 3: Randomized Assignment
320
T=0
T=1
Time
Treatment
Communities
186
Comparison
Communities
Comparison Period
Case 3: Randomized Assignment
How do we know we have good clones?
• In the absence of Progresa, treatment and
comparisons should be identical.
• Let’s compare their characteristics at
baseline (T=0)
Case 3: Balance at Baseline
Case 3: Randomized Assignment
Treatment
Comparison T-stat
Consumption
($ monthly per capita)
233.4
233.47
-0.39
Head’s age
(years)
41.6
42.3
-1.2
Spouse’s age
(years)
36.8
36.8
-0.38
Head’s education
(years)
2.9
2.8
2.16**
Spouse’s education
(years)
2.7
2.6
0.006
Note: If the effect is statistically significant at the 1% significance level, we label the
estimated impact with 2 stars (**).
Case 3: Balance at Baseline
Case 3: Randomized Assignment
Treatment
Comparison
T-stat
Indigenous=1
0.07
0.42
0.07
0.42
-0.66
-0.21
Number of
household members
5.7
5.7
1.21
Bathroom=1
Hectares of Land
0.57
1.67
0.56
1.71
1.04
-1.35
Distance to Hospital
(km)
109
106
1.02
Head is female=1
Note: If the effect is stati…
Purchase answer to see full
attachment