A reform proposal can make technical sense, but if it is perceived to threaten the interests of a particular group- or does not take into account the needs of another- the most technically-sound proposal might falter.

Evaluation is often known as a way to determine what has worked and what has not, and, of course, for the ratings that follow from such an analysis. Evaluation is also key to finding explanations as to why things are happening, or not. That is how evaluation can inspire learning and signal when course corrections are necessary.

So what does this have to do with political economy analyses?

Regular readers of our evaluations will have noticed that political economy plays an incredibly important role in our search to explain why things are working or not. In simple terms, political economy takes into account the roles played by different stakeholders; specifically, their interest in­­- and political power over- decisions. The effects can be far-reaching and undermine development efforts and their outcomes.

In a recent paper, we reviewed whether and how political economy analyses affected the outcomes of policy lending. The positive effects were clear: understanding who the stakeholders are, what they have to gain or lose, and how to engage them in the development process all mattered in making an intervention successful.

More often than not, analytical work done at the planning stage tends to focus on technical issues. These are incredibly important, but they are not sufficient. A reform proposal can make technical sense, but if it is perceived to threaten the interests of a particular group- or does not take into account the needs of another- the most technically-sound proposal might falter. For instance, from our work on the financial viability of the electricity sector, it was obvious that setting tariffs at cost-recovery levels is essential for the sector to function, but this course of action is politically sensitive and hard to implement. Understanding the political economy (together with the distributional effects of such changes) can make or break the success of the intervention.

Political economy analyses can also provide a platform for dialogue among stakeholders. The most successful cases involve broad-based consensus building through transparent processes. They build ownership where different points of view are understood and the best possible shared solution is developed. In successful cases, political economy analysis can make implementation easier, because the potential bottlenecks are anticipated and addressed ahead of time. This is particularly so if the analysis provided the aforesaid platform for dialogue and it was used to build broad-based ownership.

An important shift is needed when undertaking political economy analyses to focus specifically to the set of intended interventions. More often than not, analyses are generic for a country as a whole, but not tailored to the situation, issues, and stakeholders that will be concerned with the intervention. Once the specifics are understood, the key to success lies in understanding how the alliances among stakeholders – that already exist or that can be forged from common interests – can support the needed policy changes. For instance, in another review we did on environmental policy loans we found examples where government and industry interests could be brought together around the need to provide greater clarity of regulations rather than lower the bar for environmental standards.

On the other hand, political economy analyses can be highly sensitive. When a few influential stakeholders have strong interests, undertaking or publishing a political economy analysis can be difficult- if not impossible- and potentially counterproductive. In these situations, informal analyses are better suited to help make informed decisions. For instance, we have seen projects that have been stopped, as under these circumstances the necessary reforms would not have been possible.

The ultimate question – which we have not answered in full – is why are political economy analyses not used more often, if they are a key driver of success and failure? It has been a question on my mind in particular in situations of conflict and fragility. Years ago in another role I was associated with an evaluation that highlighted the many political economy analyses that had been done for Southern Sudan (before it became an independent country) and that had gone unused. When strife broke out, it was in the same areas that had been flagged in both the political economy analyses and the evaluation. Could tragedy have been prevented if our analytical work had been heeded?

Read IEG's Report:  The Role of Political Economy Analysis in Development Policy Operations


Submitted by Sylvester O. O… on Fri, 11/18/2016 - 10:38


Thank you very much for this thought provoking piece on the centrality of Political Economy to the success of reforms. It is a very interesting area of study. I am very much interested in reforms - to be more specific public service reforms, and for that reason over time I have been reading IEG Reports on implementation of public service reforms, especially in developing countries. The following factors have featured over time as challenges to successful implementation:
1. Lack of capacity;
2. Approaching reforms as "Technical solutions"
3. Lack of "Political Will" closely tied with Local Ownership of Reforms, local contingent factors etc.
4. Segmentation- ETC.

What you have stated is very crucial and needs to be taken on board for successful implementation. However amidst all these is the challenge of the HUGE divide between researchers and practitioners. There is a huge body of literature on why practitioners don not often rely on findings of research studies. The biggest challenge, I find is that research as currently practiced seeks to establish trends and comparisons from a cross-sectional sample for generalizations. To a practitioner like me however, whereas cross-sectional analysis of a phenomenon might be helpful in establishing the broad principles, my immediate area of concern and what I am likely to immediately take on board is a longitudinal study of my immediate area of concern. A objectives of a researcher and a practitioner are very different even on the same subject. A practitioner is first and far most a solution finder, a researcher looks for 'causes' and may therefore in most cases propose 'several solutions'. In spite of these apparently different views, both need to work closely for sustainability. I have in the course of my work as a public servant, read in highly rated Journals, articles by highly reputable Professors about reforms in Kenya- on some programs and initiatives I was deeply involved in, which interviewed a sizable sample of 'Kenyans, who unfortunately were not either closely involved and therefore did not know the challenges- but at the end of the day, the observations and the conclusions from such articles are what other readers may try to implement- while those issues may not be exactly what happened.

I am sharing this issue of research in relation to political economy, and in this forum and relating it 'evaluation and research' because my PhD Research was in a related area- "Political Influence, Appointments, Public Sector Management and Reforms in Kenya 1963- 2014" I collected data from 100 senior public servants at the level of Directors and Deputy Directors, and also had face to face interviews with 13 Top Managers - CEOs and senior executives in watch dog institutions and university professors in Kenya. When I went back to Newcastle (Australia) my interpretation of this data and that of my supervisor was totally different in areas I considered fundamental to my thesis- such as what made Civil Service so central to successive governments in Kenya and therefore the role of patronage appointments in the civil service . In fact it took me so long and in most of these crucial areas, I had to support my interpretation from other sources to demonstrate that, if I took his interpretation then I would not be representing the trends in Kenya. That data could have very well supported another view in Australia or other developed country.

As for the search for general trends in research and so on. In my thesis because of the crucial role of Development Partners- I dedicated my last Chapter and had specific questions on the role of donors. Beginning from the common position that 'reforms needed to be locally owned' to succeed, I asked whether the fact that reforms were seen mainly as being driven by World Bank and IMF was a reason for their lukewarm reception. The responses I got were amazing- in fact that was one of the most answered questions by over 95%. And the overwhelming response was NO. With most respondents saying that - that argument is advanced by local politicians to frustrate implementation of reforms.

My one cent contribution to political economy of reforms is that - it is crucial but there is need to progress the research approaches that can promote ready acceptance by practitioners, and secondly ensure progressive advancement of similar themes which could have been advanced earlier. In my view there is a close relationship between discussions on- Political Will, Local Contingent Factors and Now Political Economy in the discourse on Public Sector Reforms.

Kind Regards,

Sylvester O. Obong'o Phd,
Deputy Director | Performance Management and Public Service Reforms,
Public Service Commission of Kenya.

Submitted by Thuso Maphaka on Fri, 11/18/2016 - 11:49


Please kindly give me an example of how 'political economy analysis' can be applied in the study of Urban Low Income Housing in the developing world using brief notes and stakeholder power relations.

Submitted by oumar diop Boudouck on Sat, 11/19/2016 - 04:54


in solving the must complex problems, we have to start from the foundation. In history man has always settled where there is land and water.
Lets give the poorest people land and water to take their destiny in their own hands.

Submitted by Alexander O'Riordan on Sun, 11/20/2016 - 07:08


In my experience the biggest challenge with programming the findings of PEAs is that very often the PEAs are done in a way that is disconnected from the programming cycle and constraints thereof. For PEAs to be more applicable they need to be applied in nature especially in focusing directly on the political economy of the particular donor/development partner institution that is commissioning the PEA. Here the real challenge is that the primary incentive in many donor/development partner institutions is for officials to protect and project their reputation of having a moral and technical authority over the partner government officials that they claim to work with but more often on behalf of. The fundamental driver for a PEA is the possibility that donors/development partners have programmed their aid based on what they see as a technical solution but in reality it has political ramifications at best and may be nothing more than a political position at worst. Take for example, putting in place phyto-sanitary certification for trade - they are always presented as a technical good but the specific controls are clearly political in that they are a driven by a specific trade deal that is almost always in place for political reasons (even if there are also economic considerations). Accordingly, PEA findings are often resisted or ignored because they first and foremost explain how what is technical to the donor is political to the partner government. And, this is threatening because it compromises the illusion that donor/development partner officials speak from a moral and technologically superior stand point. If donor/development partner programming is political then its relevance is not determined by it being morally or technologically superior but rather whether it delivers in the same political economy that partner government officials work within. Simply put, PEAs become uncomfortable because they make plain that donor and development partner officials are no less immune to the compromises of political decision making than their partner government officials are.

In this regard, I too can confirm that the political risks identified in South Sudan and acknowledged in many a PEA were as real when the country declared independence as they are today. It was a business as usual approach perhaps justified by the fact that donors and development partners did not have the power to shift the incentives in South Sudan that makes those PEAs appear so prescient.

Submitted by Anonymous on Mon, 11/21/2016 - 07:02


A very useful article.

Agree with the views on why PE analyses are useful and why they should be used more often, especially for policy lending. Understanding the various interests would inform decisions by lenders and possibly also help forge alliances in country between proponents for particular policies that would supported needed change in the country. It is important that the PE analyses are done quickly and results used efficiently. What seems true is that the nature of alliances shifts with the political cycle of a nation. What may have been alliances in one season may shift if their interests are threatened by issues that may be far removed from the policy issue being pursued, but where they feel they need to forge other 'alliances' to protect new or existing interests. Keeping an informal eye on the PE is also useful to be current with the 'mood' in the country.

Why are PE analyses not used so much, it could be that most lending institutions are very technical in their approach, and rightly so. It is difficult to incorporate what appears to be, reading of intentions of players in a technical sphere, without being deliberate about it. Most lending institutions, the World Bank inclusive, carry out fiduciary assessments before any lending for policy reforms or projects is undertaken and these are mandatory. Reviews including country systems, financial management, procurement, environmental and social need to be undertaken and all project preparatory and appraisal teams need to have these skills on board as staff or consultants, possibly if PE analyses are included on the mandatory checklist before approval of a lending or grants, they could be taken more seriously and used more often.

Add new comment

By submitting this form, you accept the Mollom privacy policy.