THE CHALLENGE | Impact modeling can have a profound effect on social impact funders’ strategy and results. So why is it not yet common practice in the social sector?

SOCIAL IMPACT FUNDERS, LIKE BUSINESSES, MUST FIND WAYS TO ALLOCATE CAPITAL TO MAXIMIZE RETURNS AND MEET THEIR GOALS. But, for social impact funders (e.g. donors, governments, development agencies, impact investors), those decisions can be more complex. For one, defining returns in the social sector is often challenging. Objectively measuring, forecasting, and comparing returns across investment opportunities proves more difficult still.

Also known as social impact analyses, impact modeling tools can estimate the socioeconomic, health, or environmental impact of investment projects through the logical arrangement of key qualitative and quantitative data and assumptions. Employing social impact models is challenging, but the outcome can save or improve millions of lives.

Camber has used impact modeling to help its partners

Uncover the previously underestimated health burden of cryptosporidium, a pathogen that thanks to impact modeling we now know is driving nearly 25% of diarrheal disease and mortality in sub-Saharan African children.

Reveal the impact of achieving full-scale coverage of 15 basic preventative (vaccination) and therapeutic interventions, an effort with the potential to avert 950,000 annual childhood pneumonia deaths.

Show the poverty-reducing effects of investing in different agricultural commodities, revealing that many funders have been overinvesting in new maize varieties when other commodities offer far higher income impact for poor farmers.

Impact modeling is a powerful tool for SOCIAL IMPACT funders but is not yet commonplace due to three major barriers:

  1. In the cases where relevant data exist, they are frequently not available in clean, simple, and useful forms.
  2. The number of stakeholders comfortable working with those data is generally small (and well below the level of senior decision-makers), leaving much of the available data excluded. 
  3. Finally, even when used, data are rarely organized, understood, and presented in a straightforward impact model, meaning they are rarely applied to the high-level strategy decisions where they would have the greatest impact.

The agricultural development Revenue and Income Impact Model case we expand on below offers an example of how social impact funders can navigate the data and impact modeling process and apply those inputs to key strategy and funding decisions to improve their ultimate impact.


THE CASE  |  Our Revenue and Income Impact Model (RIIM) revealed many major agricultural development funders are investing in commodities, farmer segments, and geographies that have far less impact on poverty reduction than readily available alternatives.


Camber identified a gap in funders’ poverty reduction goals and their investment focus areas and delivered a decision-making tool to better ensure investments in agricultural development will have the largest impact on poverty reduction. With funders’ help, our team shifted the decision-makers from a simple focus on whether or not a new investment offered large yield increases to an estimation of actual income impact of many different investments, which revealed that funders’ primary focus on yield had been leading them to support less effective interventions.  
 


THE OUTCOME

Impact modeling uncovered a misalignment between the typical commodity focus (selected based on traditional relationships, expert opinion, and experience) and the commodity focus suggested by the more consistent, data-backed RIIM. Maize, for example, perceived as a good bet for poverty reduction, has been a primary investment focus for most social impact funders in the agriculture sector. Maize has a relatively high intervention cost and low sale price, however, so despite its ubiquity on smallholder farms, it offers markedly smaller impact.

Impact modeling showed that current investments are not structured to reach future goals. Separating smallholder farm households into distinct segments based on land agricultural potential and household proximity to input (e.g. fertilizer) markets revealed differences in realized yields. Often, we found that every segment would need to experience yield increases in order to reach overall target goals, which was unlikely given the difficulty and higher costs of reaching more remote farmers. Current investments will likely fall short of their hoped-for impact.

Impact modeling indicated that downstream investment in the supply chain may be as or more important than R&D. Baseline data analysis indicates that use of potential interventions, such as fertilizer tends low and is costly for farmers. Historically prioritized by funders, R&D technologies aiming to increase yields have not yet generated significant productivity gains for farmers nationwide. Low adoption rates are more likely the barrier to the realization of expected yield increases.


THE BOTTOM LINE  |  Impact modeling gives social impact funders access to the same optimization tools employed by the private sector, offering funders a better way to predict and compare the effects of investments before they are made.