The financing need of equitable provision of paid maternal leave in the informal sector in Indonesia: a comparison of estimation methods

Abstract Background Providing an enabling environment for breastfeeding is hampered by the inequitable implementation of paid maternity leave, primarily due to perceived or actual financial costs. To estimate the real cost of paid maternity leave requires using reliable methods. We compared methods utilized in two recent studies in Indonesia. Study A estimated the financial need of providing paid maternity leave in the formal sector with a 10-year forecast at 21% coverage of eligible mothers, while study B estimated similar costs for the informal sector at 100% coverage annually. Results are critical for guiding future application of either method to inform paid maternity leave policies. Methods We compared number of covered mothers working informally, total annual cost, and cost per mother. We modified some parameters used in study A (method A) to be similar to study B (method B) for comparison, namely the period of estimate (annual), coverage (100%), estimate of women potentially breastfeeding, exchange rate, female labor force participation rate, the percentage of women working in the informal sector, and adding administration cost. Results The methods differ in determining the number of mothers working in the informal sector who gave birth, the minimum wage as unit cost, and administrative cost. Both studies estimated the cost at various lengths of leave period. Method A requires more macro (e.g. national/regional) level data, while method B involves (e.g. individual) micro level data. We compared the results of method A with method B, respectively: 1) number of covered mothers working informally were 1,425,589 vs. 1,147,204; 2) total annual costs including administrative costs were US$650,230,167 vs. US$633,942,726, and; 3) cost/mother was US$456 vs US$553. Conclusion Certain flexibilities can be applied to both methods, namely using parameters specific to respective regions (e.g. provincial level parameters), flexible period of analysis, and the use of administrative cost. In a setting where micro data is scarce and not easily accessible, method A provides a feasible approach, while method B will be most appropriate if suitable micro data is available. Future comparison studies in other settings are needed to provide further evidence on the strengths and weaknesses of both methods.