A missing middle for impact measurement?

Written by Ellie Turner, Head of Agriculture for 60 Decibels.

The “missing middle” refers to small and medium enterprises (SMEs) in the Global South that need capital (ranging from $50K to $2M) to grow, but are too large for microfinance and too small or risky for traditional bank loans or equity. The obstacles in financing this market segment are well-documented–including high risks and transaction costs–and many blended finance solutions have emerged to overcome them. 

Meeting the $106B ag-SME financing gap has the potential to positively impact millions of smallholder farmers. 

But, much like the financing itself, measuring impact in the missing middle is challenging. 

There are an estimated 220,000 SMEs that need financing. Funders know that they can’t finance one at a time. To close the gap, initiatives need to address dozens, if not hundreds of SMEs. 

So how do you measure impact for an initiative of this scale?

Of course, if an intervention is big enough and geographically focused, like Feed the Future – you can use population based studies in your zone of influence, tracking population level indicators like poverty levels and food security over time.

Or, if it’s small enough, like a handful of SMEs, then you can survey the participating farmers as the AINFP program did here. 

But what about funds of funds, or the types of initiatives aimed at closing the ag SME finance gap at scale? They typically finance SMEs that are distributed across different business environments, face different climate risks, and have different business models. Some of the same characteristics that make SMEs hard to finance make it hard to measure the impact of financing them at scale.

• These initiatives cannot know who their target beneficiaries are ex ante. They can set parameters–like priority technologies, geographies, crops–but until an SME applies for and receives the financing, there is no way to know which farmers will be impacted. 
• Similarly, because they don’t focus on narrowly defined sub-populations, population-level impacts aren’t relevant. 
• Because of the nature of the transactions, there are often several layers involved – from donors and implementing partners to fund managers and local lenders, who have direct relationships with SMEs and can do mid-sized transactions at scale. 
• SMEs are extremely heterogeneous in their business models, impact theses, and impact timelines.  

So how should we think about measuring impact? (Without reverting back to output metrics like ‘farmers reached’ or ‘hectares covered’ that merely capture scale?)

First, let’s not forget to start with the most direct impact: the impact on the SME that got the financing (the missing middle!) They’re also the easiest group to reach, and they can tell you exactly how their business and the services they offer farmers have changed. See an example from Aceli Africa – more than 9 in 10 SMEs reported increased revenues and the number of farmers they serve thanks to the financing they received. 

Second, you can still capture farmer impacts. There are a few keys to making this work. 

1. Forget the baseline (gasp!) In SME financing, you simply cannot know what baseline will be relevant before you know which SMEs you are going to finance. And it’s okay. Use an ex-post design, and ask farmers to reflect on the changes they’ve experienced since working with the SME, or since the SME received the financing. Farmers know the most about their situation, and they will tell you. 
2. Define your beneficiary base. These will most likely be the farmers that work with the ag SMEs. This can be the hardest part of the exercise, but if you design for it up front and incorporate it as a financing requirement for the SMEs, it works well.  
3. Standardize. With this number and variety of SMEs, it’s inefficient to dig into the impact thesis of each one and reinvent measurement tools. Use metrics and questions that work for any agricultural intervention, anywhere, and can be aggregated and compared across different contexts. 
4. Collect data remotely. The logistics of in-person data collection are cost and time prohibitive, particularly for multi-geography impact studies. Save yourself the headache and use phone calls. 

60 Decibels is proud to serve as a learning partner to Aceli Africa, with whom we have tested and refined our methods for listening to SMEs and the farmers they serve at scale. We are also supporting the Huruma Fund, which targets a slightly different segment of agricultural finance but faces similar considerations in measuring impact at scale. We’ve learned a great deal through these partnerships, and are keen to learn more from others doing lean impact measurement!