Impact evaluation of the conditional cash transfer program for secondary school attendance in Macedonia

Publication Details

Armand, A and Carneiro, P, 2018, Impact evaluation of the conditional cash transfer program for secondary school attendance in Macedonia, 3ie Impact Evaluation Report 69. New Delhi: International Initiative for Impact Evaluation (3ie)

Link to Source
Alex Armand and Pedro Carneiro
Institutional affiliations
None specified
Grant-holding institution
None specified
Education, Multisector, Social Protection
Conditional Cash Transfers
Gender analysis
Conditional Cash Transfers
Gender analysis
Equity Focus
Evaluation design
Difference-in Difference (DID), Instrumental Variables (IV), Randomised Control Trials (RCT)
3ie Series Report
3ie Funding Window
Open Window Round 4


This study presents findings from a randomized controlled trial (RCT), with results verified using additional methodologies, looking at the alternative ways to encourage secondary school attendance through conditional cash transfers (CCTs) in Macedonia. The paper explores different methods of delivering the CCT, comparing equal quarterly payments to smaller quarterly payments and a larger final payment. The researchers also explore the influence of varying the recipient of the CCT between the household head and the mother of the child.


Macedonia has recently experienced both relatively high unemployment rates and high poverty head count rates. Secondary school education rates of around 80% also trail regional comparisons by around 10 per cent. Given this situation, the Government of Macedonia worked with the researchers to design a nationwide CCT program to encourage secondary school enrollment, attendance and graduation figures.

Research questions

This evaluation exploited the roll-out of the CCT programme to test if the recipient of the payments (head of household vs. mother) or timing of the CCT payments (spaced payments vs. lump sum) influenced the main outcomes of interest. 


The researchers combine an RCT approach with a difference-in-difference analysis and an instrumental variables (IV) estimations. As the program was simultaneously distributed throughout the country, randomization only influenced what type of treatment recipients received. Treatment households were identified through being eligible for social financial assistance (SFA). A difference-in-difference (DID) estimation was conducted, comparing SFA households with households eligible for child allowance (CA), another government scheme for households with secondary school age children.

Theory of change

The researchers assert that receipt of the CCT should increase household income, which will encourage secondary school age children to shift their time from the workplace to the classroom. The conditions of the treatment will require the children to enrol and attend secondary school.

The researchers targeted women in the household to receive payment of the CCT because the literature generally believes that women spend more money on household goods. By receiving the money, these mothers also might further encourage their children to attend school. As a secondary effect, the money also might alter the decision making power within the household.


The researchers collected three waves of data, which allows them to compare short and medium term impacts of the different CCT treatment arms.

Main findings

  1. CCT payments increased school enrollment but not school attendance.
  2. Intermediate incomes show increased household nutrition when CCT payments sent to mother, but results don’t hold up in the second round of follow up data.

Additional publications


Scroll to Top