Jensen - Returns to Education (2010)

Question 1

Why? Motivation?

Answer 1

How important are returns to education in determining schooling decisions?
Do students have accurate information when they decide whether to continue schooling?
Is education inefficiently provided in the Dominican republic?

Question 2

Why Dominican Republic?

Answer 2

80% complete pre school but only 25% complete secondary school, but mean earnings are 40% greater for those who complete secondary school. This gives the impression that students are underestimating the returns to education.

Question 3

Explain the Methodology

Answer 3

• RCT using non-rural household surveys with 150 sampling clusters from the 30 largest urban areas
• Selected schools where students attended 8th grade (final year of pre-school)
• First round
• Then randomized at the school level and separated into treatment and control groups
• treatment given true(estimated) ROE to update their perceptions on the returns to education
• second round interview
• third round interview

Question 4

What was the first round?

Answer 4

15 8th graders randomly selected and interviewed on perceived RoE

Question 5

What was the second round?

Answer 5

Students interviewed about ROE again to see if their perceptions had changed

Question 6

Third round?

Answer 6

Students interviewed to find out the total number of years of education completed

Question 7

Explain the estimation strategy

Answer 7

• Used a DiD approach to compare both groups of students
• allowed author to evaluate treatment impact on perceptions and years of schooling whilst controlling for time trends and institutions etc
• Also split treatment group further into poor and least poor to identify heterogeneity in impact of perceptions updates

Question 8

Main Results

Answer 8

• Found students overestimate returns to primary and underestimate the returns to secondary educ => explains the data found in DR as a whole
• The kids who were informed about true ROE had increased perceived returns to education after treatment
• Gained 0.2 years more schooling over the next four years
• The largest effect was on the least poor students, this implies that credit constraints were an important factor reduced secondary education for poorest children

Question 9

Evaluations

Answer 9

• intervention is possible, but it must provide more information to students and also reduce the barriers to attendance
• CCTs like Progresa are potentially more impactful and cheaper
• If it is a problem of credit constraints then can we use microsavings instead (ashraf), this could help provide funds for education