In my last blog, I provided an update on how investors and companies monitor the performance of smallholders in supply chains. In this one, I ask – what are the barriers and incentives for collecting more relevant data on smallholders who work with agribusinesses?
As part of our research into smallholder supply chains, we engaged with a sample of 18 agribusinesses and financiers and conducted six deep-dive case studies in Ethiopia, Kenya and Malawi. We focused on the leather and meat export segment in Ethiopia; the dairy and maize segment in Kenya, and the mango and tea segment in Malawi; all value chains with very disparate structures, but all engaging with smallholders in various ways.
We came across three common data collection challenges:
1. Through their impact on smallholders’ livelihoods, many agribusinesses are contributing to SDGs 1 and 2, but this is often not being captured at the company or national level.
While many of the companies are working explicitly with smallholders and providing incomes (and processing payments data), more could be done aggregating this data upwards to various SDG indicators and goals – for instance, SDG indicator 1.2.1 – demonstrating incomes leading to (partial or full) poverty alleviation. This data collection is motivated by concerns primarily regarding output and productivity, but at a higher level, if this data was collected more systematically, it could have significantly higher developmental impacts.
This is in contrast to 20 per cent of our sample having no data on the poverty status or income of the farmers they work with. In contrast, 80% of companies held data on output levels or product quality. Interviews suggested key impediments to this included lack of resources and no perceived business value in this form of data collection. Such data was more likely to be collected when there was an ‘accountability relationship,’ such as working relationships with donors on projects or impact-orientated investors (such as Development Finance Institutions).
2. Data quality and use are generally low, so qualitative data is of greater importance.
The questionnaires and case studies revealed a lack of knowledge of best practice in data collection methods, and often low-quality data collected. In the case studies, we found multiple datasets with circular references, significant data gaps and inconsistent units. The research found limited knowledge of best practice data collection procedures, such as the DCED Standard or IRIS indicators. While 11 per cent of actors used some kind of industry standards, more than half applied their own internal standards to data collection, with varying degrees of quality and precision.
The case studies show that where systematic data collection was lacking, a significant amount of qualitative information is collected on smallholders, particularly in value chains with long-term relationships between buying firms and smallholders, such as the tea industry in Malawi. It is here where a much more participative approach to engaging with smallholders was seen. In other areas, qualitative information was used to make up for the lack of systematic data – such as in the leather sector in Ethiopia.
Where information takes the form of feedback from smallholders (such as in the tea and mango value chains), it is an important – or potentially important – way in which accountability can be promoted between buying firms and the smallholders they work with. As well as improving smallholder welfare, maintaining these channels of communication can also be to the advantage of the firms themselves.
3. Value chain characteristics key for determining what data is collected on smallholders
Smallholder relationships with agribusinesses are determined by the nature of the value chain they are operating in. Inevitably, this also impacts the ability of (incentivised) agribusinesses to collect smallholder data. In particular, data collection and quality seemed to be determined by the following factors:
(i) The number of potential smallholder producers is not too large, and supply and demand conditions are relatively tight
a) Data collection on smallholders is very difficult for large-scale maize processors in Kenya, as a result, they have focused their efforts on capacity-building traders.
(ii) Buyers are not distanced from smallholders by many intermediaries
a) The leather segment in Kenya is characterised by a large number of traders acting as intermediaries. This stops interaction between the ultimate end user and means that interventions to improve the quality of the product, as well as the collection of data, are difficult to capture.
(iii) Value chain actors are able to devote sufficient resources to these activities
a) Margins are tight for many agribusinesses, where there is not an efficiency point to make regarding smallholder data collection or a lack of perceived value, it is harder to incentivise such data collection.
(iv) Products vary widely in quality – where the product is relatively standardised in terms of quality, there is less need for M&E in the value chain from a producer perspective
a) To the untrained eye, products that appear standardised may not be – for instance, the presence of aflatoxin in maize. However, in the dairy sector, strong health concerns ensure that more M&E is done to make sure that the cow remains healthy.
(v) Consumers are concerned with smallholder welfare.
a) Consumer groups have worked hard to ensure fair labour rights in certain value chains; such as tea, cocoa, and coffee; typically export plants. Where these occur, there is a greater push for data to drive standards and welfare. This does not appear to be as common for “staple” crops, such as maize.
These findings should not give cause for despair in some value chains(!). Value chain characteristics can be altered, and new accountability relationships can be created. However, for systemic impact so that more data on smallholders can be collected, we believe more should be done to demonstrate the commercial value of closer, longer-term relationships between agribusinesses and smallholders. An innovation fund for M&E or a donor-funded technical assistance facility could ensure that this knowledge is disseminated effectively. That knowledge should include a standard ‘lean’ template for smallholder data collection. This should be aligned with the SDGs, and focus on income and productivity to encourage business participation so that agribusiness can see commercial value in this kind of information. Another route is technology, where we saw examples in Kenya of digital methods to bridge the gap between geographically-diverse smallholders and the agribusinesses they engage with.
