On the political economy of results terminology

October 04, 2011 | Adrian Gnägi | Learning Elsewhere |


Rating: none

By Adrian Gnägi

Adrian picture for sdclanA few weeks ago I participated in a training course on impact oriented monitoring and evaluation. The course really helped me organizing my thinking on managing development programs for impact. Time and money very well invested, I found. One critical moment for me was when one of the trainers presented an overview on results terminology. Even though her presentation was introduced with a Confucius citation (my translation: “if the concepts are not right, the order of things is lost”), she presented the 4 terminology clusters as “some do it like this, others do it like that”. I felt compelled to explain why I think this free choice of results terminology to be wrong. Since I was struggling to explain it in simple words, I decided to write it up. That’s what this post is all about: why results terminology matters.

The basic theoretical notion this post is built on is that linguistic choices are motivated. There is political economy behind the terminology we use. Concepts (always, necessarily) highlight certain elements and leave others in the dark. What is mentioned or highlighted can easily be related to, what is not said is more difficult to be argued upon. The concepts we use help us to linguistically construct the reality in a certain way. We make our choice of terminology according to our position in society: we use the words that allow us to construct reality in the way we see it. When it comes to mainstreaming terminology, this boils down to political economy. Those able to impose their terminology force others to talk about reality as it looks from their own position.

Result terminology currently – according to Daniela Keller in the above mentioned training course – comes in 4 versions:

  1. the German Ministry of Development Cooperation (BMZ) model that uses 4 differentiated result categories
  2. the OECD DAC terminology that uses one single composite term (“results”), increasingly adopted by development agencies
  3. the standard evaluation research terminology that talks about outputs, outcomes, and impact
  4. a compromise model, used for example by SDC, that distinguishes between outputs and outcomes.

Different versions of development results terminology

What “order of things is lost”, or “what positions are served”, when a choice among the 4 options is made? I suggest looking at the most differentiated terminology (BMZ) first, and thereby analyzing what is more difficult to talk about (what is “lost”) when using more composite terms.

The most unusual term in the BMZ model clearly is “use of outputs”. The term originates in (social science) action theory. It reflects the common-sense wisdom that “one can lead a horse to the river, but one cannot force it to drink”. In BMZ thinking, development interventions are arenas of negotiation between different actors who all have their own agendas. Even when there is agreement to do something together, this agreement never covers 100% of the different agendas. Development interventions produce, through inputs in form of staff, money and knowledge, certain outputs. Those outputs are used or appropriated (by the same or by other actors) as inputs for their own agendas. “Use” always and necessarily means “intended use” (as specified in cooperation agreements), and “non-intended use”. The important thing the concept highlights is that development activities assemble actors with different agendas who may cooperate to produce certain outputs, but who may have highly divergent agendas on how to use those outputs. The aggregation logic from outputs to outcomes is not pre-determined, but constructed by actors as development/history unfolds. In other words: if no distinction is made between outputs and their use, “working misunderstandings” where outputs are used to serve other than the agreed upon ends pass as development results.

The differentiation between outputs and outcomes has been popularized by evaluation theory and nowadays is broadly understood and applied in the development world. Outputs are what can be produced by projects with a certain planning accuracy by controlling context influences (typically understood as what can be logframed easily). Outcomes are aggregations of outputs and other influences happening beyond the control realms of projects, but still with considerable influence by development interventions. The critical issue the distinction between outputs and outcomes allows to talk about is contribution: the development world is marred by projects that deliver on their entire output list, but that do not contribute significantly to intended change. In other words: if no distinction is made between outputs and outcomes, symbolic action passes as development result.   

The distinction between outcomes and impact also has been popularized by evaluation theory, but the impact concept today is not widely applied in the development world. The main difference between outcomes and impact is time/scale. While “outcomes” relate to mid-term (often 3-5 years) and some kind of spatial delimitation (sectors, regions, or actor categories), impact relates to generational societal transformation (like food security, democratization, or peace). The critical issue the differentiation between outcomes and impact allows to talk about is aggregation of change vectors. In other words: if no distinction between outcomes and impact is made, this allows to gloss over policy incoherence, contradictions between policies and market forces, or competing models of what a “better world” means for different actor alliances.

So whose interests are served when using differentiated results terminology, and whose interests are served when using composite terminology?

  1. Differentiated results terminology allows for conceptualizing and managing the entire value chain from inputs to the kind of societal transformations development cooperation promises. This obviously is in the interest of those whom the promises are addressed at: the people normally referred to as “target groups” in the development world. But it also is in the interest of those providing resources under the promise that the resources will bring about those societal transformations: tax payers and private & corporate donors. It also is in the interest of those development actors who signed up for their task because they sincerely want to contribute to making our world a better place to live in.
  2. Composite results terminology allows for glossing over that the value chain from inputs to societal transformations may not be conceptualized or managed well, that inputs provided may not lead to impact promised. This obviously is in the interest of those who are accountable for conceptualizing and managing the development value chain. Those who are caught in the gap between high-flying promises and inadequate resource provision, between conflicting policy objectives, between what seems doable and influences beyond their control. If outputs or outcomes are development results, then designing and managing neat LogFrames is their main task; messy issues like conflicting actor interests, policy incoherence, or emergence are not on stage. And the responsibility query is much more difficult to formulate.

The choice between differentiated or composite development results terminology is between serving the interests of the people whose lives are to be improved through development, or serving the interests of those who conceptualize and manage development interventions. Those interests sometimes converge – not always, though. 

No Comments

Leave a Reply