Lecture 4- MortalityRiskandHuman Capital Investment- Fortson (2011) Flashcards

1
Q

What is the main goal of the paper?

A

To explain how prevalence of HIV changes as investment in human capital changes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Why might HIV prevalence impact human capital investment (education investment)?

A

Completing more schooling comes at the opportunity cost of not earning money right how. However, after completing further schooling there is greater earning potential.

Thus a trade off between current and future consumption.

Having HIV reduces life expectancy, hence, the future is discounted at an even higher rate and the value of future consumption falls

So as HIV prevalence rises, school attendance falls.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Describe the dataset used by Fortson ?

A

They used cross sectional data from DHS across 15 countries in sub-saharn Africa that gives demographic, economic as well HIV Prevalence information

They used data from 1952 to 1991 and split those born from 1952 to 1979 as being the “unaffected” cohort and those from 1979 to 1991 as being “affected” as HIV become prevalent in late 80s

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Describe the empirical strategy used by Fortson in “Mortality Risk and Human Capital Investment” in 2011?

A

HIV became highly prevalent in the late 1980s.

So, the data from DHS was split into two cohort:
Unaffected cohort: Those born from 1952-1979
Affected cohort: 1980-1991

Those from the affected cohort would have been going to school in when HIV was rife

A difference-in-difference regression was used

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What regression was used by Fortson?

A

Sicr=B0+B1 HIVr*I(c=>1980)+B2 Ficr+B3 Rural,icr+ λc+ar+eicr

Sicr is the educational outcome for individual I in cohort c in region r.
HIV variable shows prevalence of HIV in region r
I is an indicator variable for if the respondent was in the affected cohort (born post 1980)
F is an indicator variable for if individual I in cohort c in region r is female

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is an issue to the dif-in-dif approach in terms of the intensity of HIV prevalence?

A

The dif-in-dif measures HIV prevalence in a binary way as either yes HIV was prevalent post 1980 or no its
wasn’t prevalent for pre 1980.

However, there very likely is a varying degree of HIV prevalence especially post 1980

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What did Fortson do to check to see if the lack of variation in HIV prevalence when usingf dif-in-dif threatened the validity of the results.

A

To test to see if differing birth cohorts had differing prevalence of HIV, Fortson excluded those born from 1971-1984 to see if this made a large impact on the results.

It did not, suggesting that the binary element in the dif-in-dif is not problematic.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What did she do to see if the impact of HIV prevalence on educational outcomes was due to OVB.

A

She tested to see if several variables that weren’t included in the regression had an impact on HIV prevalence and education:

Tested to see if a high amount of those that had HIV orphans? This is important also because orphans are less likely to be enrolled in school (satisfies requirements of OVB).
She found that the effect of HIV on education does not come from orphans alone.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What was the main conclusion?

A

Children living in an area of high HIV prevalence have a shorter time-horizon so value of future consumption falls relative to current consumption and education investment falls as earning income now is viewed as preferable.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly