Farmer Value Creation vs Type of Service Provider

Key Messages

Understanding which types of organizations are best able to create economic value for farmers can be hugely insightful. Besides potentially triggering comparisons (and competing claims) between different types of organizations in which models are ‘best’ at helping farmers, it can help donors, investors and support organizations decide which types of models to support in order to generate the most impact for farmers. However, FarmFit data does not show a clear pattern, including by our most rigorous analytical methods as well as our qualitative insights. All else being equal, FarmFit data suggests that the amount of farmer value created is not driven by which type of organization is delivering goods and services to farmers.

While this limits the amount of analysis shown on this page, it can also be seen as a promising insight: all else being equal, FarmFit data suggests that there are no reasons why one organization may be less able to create value for farmers than another. 

Read on below for more detail and (limited) analyses.

Understanding how type of provider relates to farmer value creation

One of the most potentially controversial insights the FarmFit Insights Hub could provide is which types of providers are able to create the most value for farmers. This would be a hugely impactful insight as it would help donors, investors and support organizations decide which types of business models to support if the aim is to create the largest amount of economic value created at farm-level for each dollar in support (or investment) provided. However, FarmFit data does not show a compelling answer to this question.

At first glance, the results appear to show a pattern: farmer value created is higher in models run by specialized services and (especially) by farmer-led organizations compared to global and local off-takers. However, these results are in fact not compelling due to several reasons.

First, as elsewhere in the Insights Hub, the samples of specialized service providers and farmer-led models are too small and internally diverse to be representative of the broader landscape. Compounding this challenge is the high spread of results for these two groups. (Click here to read more.)

Second, when comparing global and local off-takers, for which FarmFit does have a representative sample, the differences are minor with, again, a large spread as well as a similar clustering of results around $500. 

Third, FarmFit’s more rigorous machine learning methods farmer strongly suggest that the type of provider does not play a strong role in farmer value created: the results seen are very small and even directionally the results inconsistent with the data shown above.

Importantly, this doesn’t mean that there is no relationship, but rather that our data does not currently suggest there is a strong relationship. As a result, this page does not provide any implications, nor does it detail opportunities for further research. As we collect and analyze more data, this may change and we also invite input from other organizations if they observe a strong relationship here.

When looking at the relationship between type of provider and the other two outcomes analyzed in the Insights Hub, we find that:

  • For Service Delivery Cost per Farmer, local off-takers invest significantly more than global off-takers, due both to more incentives to invest in farmers and lower efficiencies. Click here for more details 
  • For Direct Cost Recovery, the results mirror those of service delivery cost: in addition to investing more on a per-farmer basis, local off-takers recover a significantly higher proportion of their costs by charging farmers. A large number of global off-takers, in fact, does not charge farmers at all for services provided. Specialized service providers, unsurprisingly, show the highest direct cost recovery, which is unsurprising as their business models rely on direct cost recovery to be viable. Click here for more details

Strength of Relationship 2/5

  • Weak relationship between driver and outcome variables
  • Results are consistent across analytical models used
  • Several limitations regarding sample or indicator