The Asian Development Bank (ADB) Corporate Results Framework 2019−2024 sets out indicators to assess the impact and value addition of its operations. In this framework, one measure of operational effectiveness is the number of jobs created as a result of ADB support. A review of selected country portfolios and jobs data for 2010−2019 reveals that the total job creation impact of ADB operations is underreported.

In this study, input–output (IO) and computable general equilibrium (CGE) models are used to develop a customized approach for ADB operations that would improve the fit and predictiveness of jobs impact assessments. In the first phase of this study, a methodology was developed to estimate potential direct, indirect, and induced job creation through ADB project and program portfolios in Fiji, Indonesia, and Thailand during 2010−2019. It used the 2021 ADB multi-regional input–output tables (MRIOTs), national statistics, and data from the International Labour Organization (ILO) and Global Trade Analysis Project (GTAP). These three upper-middle-income countries were selected as pilot cases for this study due to the representative characteristics of their portfolios for the rest of ADB developing members.

In the second phase, model-generated estimates were compared with reported jobs data in operations documents to identify any discrepancies between estimated and reported data, and means to address the discrepancies. Finally, a robustness check was done by applying IO and CGE models to selected projects and programs in different sectors. Findings indicate that the linear, simpler to construct and use IO models may be preferable for projects that support substantial local purchases of non-construction capital goods and services (as compared to imported goods); while the more comprehensive, data-intensive CGE models can help capture the broader-based impact of programmatic policy support and large-scale projects. It is expected that as the model is further tested in selected projects and programs under design or implementation, the fit would further improve. In addition, improving data capture and systematic monitoring can complement quantitative assessments and support poverty reduction impact assessments. The model—with robust data—can be applied to assess and enhance the operational effectiveness of ADB, and to inform government investment decisions and policies to promote inclusive growth.