Published on Jobs and Development

DFIs should work together to measure job impacts

This page in:
Image The creation of jobs and the promotion of economic transformation are the main development challenges in low and middle income countries. Development finance institutions (DFIs) support private sector activities through finance and technical assistance. These are key instruments in stimulating private sector-led job creation. However, without active collaboration amongst DFIs it will be difficult to fully understand their impact. For instance, collective action through the Let’s Work partnership has the potential to enhance the evidence base on the impact of DFI-supported firms on job creation.
 
Measuring the impact of DFI-supported operations on job creation improves the accountability to stakeholders. Explaining the impact is one important component in the value for money agenda: how many jobs are created as a result of an intervention. A useful starting point is to agree on a common definition of what is a job in a DFI-supported firm. Then come more tricky questions about job quality and the indirect impacts via growth and economic transformation. Unfortunately DFIs, and especially the smallest among them, lack the capacity and information to precisely estimate the jobs they create - both directly and indirectly.
 
However, there are ways to approximate the impact. These methods range from tracer studies to value chains and econometric studies. It is understandable that DFIs undertake their own evaluations, but given the complexity and duration of these, it is also common sense to collaborate. For example, it takes a lot of effort to examine the full impact of tourism investment through the value chain. Yet this is needed to inform sceptical commentators who might be under the impression that these investments just create jobs for a few, rich, skilled workers. Lack of collaboration yields unsatisfactory results. In the UK for example, the fragmented nature of impact assessment studies among the NAO, IDC and ICAI suggest they can only highlight small parts of the impact chain, with little new evidence. Collaboration at scale could alter this.


 
Collaboration can also help steer investment decisions ex-ante by providing practical tools and knowledge to investment officers at the right time. It is fantastically useful to have a basic understanding of the employment potential of investments, sectors and countries, under specific circumstances. For example the IFC jobs study of 2013 argues that indirect jobs created can be six times higher than the direct jobs. Work by the Supporting Economic Transformation programme in Tanzania suggests that the job multiplier in some sectors is three times higher than in others. The Let’s Work Partnership could help by compiling knowledge on which type of investments in which countries have the greater transformational and job creation potential. This information can be directly used in investment project sheets.
 
The pay-off from collaboration will be greater depending on the type of study and sector. DFIs will continue to collect information on investee companies and perhaps use individual tracer studies. They may also be able to use easy to employ input-output models to assess ex-ante job impact. But they have a lot to gain from collaboration on ex-post job impact in the wider economy. This could include econometric studies or value chain studies in sectors of interest.
 
The sector also matters for choosing between techniques and the level of collaboration. Investments in infrastructure are often large, lumpy and transformational, and an understanding of their impact needs to be undertaken through in-depth modelling studies and the development of on-line tools that can benefit a number of DFIs operating in the sector. On the other hand, small scale manufacturing investments do not require the CGE modelling attention, but might benefit from value chains studies around agro-processing, for instance. Finally the monitoring activities from DFIs would benefit from common, workable definitions and standards.
 
This leads us to suggest the following elements of a joined-up policy research agenda
  • GVC studies for textiles, agribusiness , leather, logistics and retail could inform and steer future DFI operations in these sectors
  • Econometric studies that measure the job impact of financial services could be used for ex-ante decisions
  • CGE models and econometric studies could be used for large scale investments e.g. in infrastructure
  • Standards for monitoring and evaluation that drive comparability across DFIs.
This is a rich agenda that the Let’s Work partnership is already focusing on. It is ramping up its knowledge work in terms of developing some of these tools and approaches. Three working Groups have started working on GVC, tracer studies and CGE models for different sectors and findings from these studies will further generate development of operational tools, briefings and other forms of knowledge management. 

The Program Coordination Unit of the Let’s Work partnership is housed within the World Bank Group and current  partners of Let’s Work include the African Development Bank Group (AfDB), Asian Development Bank Group (ADB), Austrian Federal Ministry of Finance (BMF),  Department for International Development (DFID),  European Investment Bank (EIB), 15 European Development Finance Institutions (CDC, DEG, etc.,), Inter-American Development Bank (IDB), International Finance Corporation (IFC), International Labor Organization (ILO), International Youth Foundation (IYF), Islamic Development Bank (ICD), Ministry of Foreign Affairs of Netherlands, Overseas Development Institute (ODI), Private Infrastructure Development Group (PIDG) , Swiss Secretariat for Economic Affairs (SECO), World Bank Group, and World Business Council for Sustainable Development (WBCSD).

Authors

Dirk Willem te Velde

Director of Supporting Economic Transformation at the Overseas Development Institute

Join the Conversation

The content of this field is kept private and will not be shown publicly
Remaining characters: 1000