In this draft USAID blog, David Jacobstein, formerly from USAID's Democracy, Rights and Governance Bureau, describes how the DC Thinking and Working Politically Community of Practice assessed the extent to which proposals or tenders received by USAID incorporated political economy and thinking&working politically elements. 

One of the perennial questions in the Thinking and Working Politically Community of Practice (TWP CoP) is how to move beyond doing a single political economy analysis (PEA), or even repeated ones, and see that a political economy “lens” is informing the way programming is conducted. While we know full well that working politically isn’t possible unless programming can be somewhat flexible and adapt to its shifting context, we have not gone as far in defining what “politically savvy programming” is in a positive sense. There are related concepts that we can say frame things more positively - such as development entrepreneurship and its series of small bets, or problem-driven iterative adaptation and its notions of authorizing environment and iteration. And there is a consensus that relationships matter, so politically savvy programming should be attending to network weaving and coalition building.

In 2024, the US-based chapter of the TWP CoP decided to try to examine this in more depth. We gathered almost every contract solicitation the US Agency for International Development (USAID) issued for a three-year period (we are working on assistance next), and attempted to categorize the project opportunity based on solicitation as one that strongly invited TWP, mixed, or not at all. In doing so, we had to define for ourselves a rubric that we could at least attempt to apply across geographies and sectors, which might allow some inter-rater reliability. Notably, there were very few highly-rated solicitations, though there is an increasing attention to aspects of TWP - there is clearly a lot of work to do to further the TWP agenda within USAID!

The process of defining a scoring rubric, and then practicing with various solicitations to see if we could consistently apply it, was interesting, and what I want to focus on here. We fairly quickly found that we needed to divide the solicitations into three areas: how they framed the problem the work was to address; the nature of the planned work they suggested be done; and the description of how they would expect work to be implemented in a savvy way. We divided each into a basic/intermediate/advanced (save the last) and built out sub-criteria for each. For example, basic problem analysis might describe the system in which the solicitation’s topic was nested, while better problem analysis might articulate certain key stakeholders in that system and their interests in it, and advanced analysis might frame the problem around the incentives of actors in the system and feasible paths toward change.

Applying the tool was interesting - it was certainly imperfect, as a number of solicitations used techniques (Statement of Objectives, e.g.) that made it harder to perceive what they had in mind. Notably, we put less weight on the work planning area because the most common approach was to include PEAs, or language around TWP (or related aspects of Collaborating, Learning and Adapting (CLA)/adaptive management) as guiding principles or descriptors of how the program was to be run, along with other language around topics such as gender and inclusion or sustainability and local ownership. It left those sections feeling like add-ons - as though someone had written the core of the scope for the activity, then added some buzzwords in the section around how it was to be done, without the two sections speaking to each other. Still, it was powerful to start thinking less in terms of “will the solicitation embrace CLA/adaptation?” or “do they call for a PEA?” and instead start asking “does their theory of the problem sound rooted in the power dynamics of the context?” and “do their ideas of how to approach the work require action based on past effort and a contextually-rooted understanding of ‘how things work here’?”

Here’s the rubric we used:

Scoring Requests for Proposal (RFP) and Requests for Application (RFAs) on “PEA/TWP-informed”

The purpose of this guidance is to provide “evaluators” with criteria for how to evaluate whether an RFP/RFA is “PEA/TWP-informed”. The criteria are divided in three sections: Problem Analysis (Aware and responding of the political context), Work Planning (describing work streams that are political or politically sensitive in nature) and Working (mandating for politically-focused learning during the program cycle). Each then has three scores (numerically 0-3 for the first two scores and 0-1 for the third). 

1. Problem Analysis (usually in the context section)

• None:
  • Purely technical
  • Focus on official laws and policies
• Basic:
  • Basic governance problems mentioned (corruption, lack of execution/ enforcement, weak government systems and processes - look for language that may be sector-specific)
  • Common stakeholders mentioned (i.e. not just beneficiaries, but stakeholders whose support the activity hopes to influence), with little disaggregation (i.e. just mention citizens or government)
  • Mentions lack of political will
  • No real analysis of any of these areas
• Intermediate:
  • Some analysis of why governance problems exist
  • Some analysis of the social and political context
  • Briefly mentions specific stakeholders and their interests/ some analysis of stakeholders
  • Mentions a window of opportunity with little analysis
• Advanced:
  • More sophisticated stakeholder understanding, identification of a broader set of uncommon yet highly important stakeholders, and their interests
  • Windows of opportunity identified with analysis
  • Appreciation for PEA analytical areas, including history, precedent, informal rules of the game, current events (that provide windows of opportunity), unconventional actors (traditional/ village leaders, community associations, etc.); PEA factors shape the logic of the program itself, rather than PEA as a separate line of work

 2. Work Planning (usually in the objectives, or maybe theory of change)

• None:
  • Purely technical (Training to increase capacity, engineering/ infrastructure solutions, provision of goods, etc.); does not focus on improving country systems.
  • No mention/ appreciation of stakeholders (in terms of supporting an intervention and not as beneficiaries), incentives or accountability
  • Seeks to change and/or create laws and policies
• Basics:
  • Some recognition of the need to engage stakeholders to increase support for the intervention/ issue/ proposed changes (not just gather information for the intervention)
  • Some mention of the need for accountability, media (transparency, if it includes who will use info and how), or coordinating stakeholders
• Intermediate:
  • Work objectives and/ or illustrative activities bring in a political (decision-making and power) perspective
  • The following elements are more central to the proposal, with some analysis of how they fit into a political approach
    • Engage stakeholders to increase support for the intervention/ issue/ proposed changes (not just gather information for the intervention)
    • Communicate with these stakeholders
    • Other means of influencing power/ decision making
    • Accountability, media (transparency, if it includes who will use info and how), coordinating stakeholders
• Advanced:
  • There is a strong connection between the objectives/ illustrative activities and the political context; a similarly strong analysis of how the objectives/ activities fit into a political approach
    • Means of influencing power that is sophisticated/ contextually rooted/ based on a deep understanding of past efforts and the “way things work”
    • Working with the grain, again based on a deep understanding of the context
    • Seeks to transform systems and ways of working with a plausible theory of change
  • The RFP/RFA leaves room for uncertainty and adaptation, does not fully prescribe exactly what is to be done

3. Working (usually in implementation principles and guidance)

• None:
  • No “political”/ contextual learning, no flexibility
• Basic:
  • PEA or another form of political/ social contextual learning is required by the implementing partner

(Note: we dropped the other levels for this aspect, as it was difficult to determine meaningful levels.)