By Gideon Obare
Director, Tegemeo Institute of Agricultural Policy and Development, Egerton University, Kenya.
Here's a startling reality: After three decades of structural adjustment programs (SAPs) across
Africa, only two of numerous neoliberal policy promises have been consistently delivered:
higher real interest rates and increased foreign direct investment. Everything else? The evidence
suggests these policies may have made things worse.
This finding comes from a fascinating new preprint study by Karangwa, Rwamihigo, and Su
(2025), which empirically examines how 29 African economies responded to what they term
"radical neoliberal policies" from 1990 to 2022. Having co-authored a paper on
Lessons from Structural Adjustment Programmes and their Effects in Africa, I found their analysis both
compelling and concerning, not just for what it reveals about policy impacts, but for how it's
presented.
Why academic tone matters more than you think.
Let me be direct: this paper's most significant weakness isn't methodological but rhetorical. The
authors use terms like "hypocrisy", describe policies as transforming economies into "jungles,"
and characterize neoliberalism as fundamentally exploitative. While their findings may justify
these feelings, such language creates an immediate barrier to engagement.
Here's why this matters: When a World Bank economist, IMF official, or multinational
corporation executive sees inflammatory rhetoric in an abstract, they often stop reading.
Unfortunately, the authors present empirical evidence that these stakeholders need to see. We're
not just talking about academic citations here but about policy influence that could affect
millions of lives.
The solution isn't to abandon critical perspectives but to let evidence speak louder than emotions.
Instead of "policy hypocrisy," discuss "unintended consequences".Rather than "exploitative by
design," focus on "misalignment with local contexts." It isn't about being soft; it is about being
strategic.
The data doesn't tell the whole story.
The authors use a Vector Error Correction Model (VECM) to analyze long-term relationships
between policies and outcomes. It's a solid choice, but it concerns me: they have put all their
analytical eggs in one methodological basket.
The robustness question: Why not compare results with alternative approaches? Dynamic
System GMM estimators could better handle the endogeneity issues inherent in policy analysis.
Synthetic control methods might provide more explicit causal identification. The authors justify
their choice but don't adequately explore what other approaches might reveal.
The proxy problem: Some variable choices seem questionable. Using CO2 emissions as a proxy
for deregulation conflates environmental outcomes with regulatory frameworks; these aren't the
same. Similarly, treating FDI as a proxy for privatization merges two distinct economic
phenomena.
These aren't fatal flaws, but they represent missed opportunities to strengthen what could be
groundbreaking analysis.
What's missing from this analysis?
Here's the most significant methodological concern: the treatment effect problem. The authors
focus exclusively on the post-1990 period - the structural adjustment era - without adequate
comparison to pre-policy performance. This is like trying to assess the effectiveness of a new
medication by only looking at patients after they've taken it without knowing their baseline
health.
Think about what else was happening to African economies during 1990-2022:
-
-
-
- Massive shifts in global commodity prices
- Regional conflicts and political instability
- Technological changes affecting traditional export sectors
- Climate variability impacting agricultural productivity
- Evolving global trade patterns
Without proper before-and-after comparison, how can we isolate the effects of structural
adjustment from these other factors? A difference-in-differences approach comparing countries
with varying policy experiences could provide more substantial causal evidence.
This isn't just academic nitpicking; it is about ensuring policy recommendations are based on
solid causal identification rather than correlation patterns.
Beyond the "anti-everything" trap
While thoughtful, the authors' policy recommendations risk falling into the "anti-everything"
trap, advocating for the opposite of current policies without fully considering implementation
realities. This creates a "he who pays the piper calls the tune" perception problem.
What's needed instead:
-
-
-
- Recognition of successful market-oriented reform cases (yes, they exist) and analysis of
what made them work
- Deeper engagement with why well-intentioned policies often fail in practice
- A more nuanced understanding of the vast differences between, say, Rwanda's experience
and Chad's experience with similar policies.
The paper treats 29 diverse African economies as relatively homogeneous, but this diversity
could be the key to understanding when and where different approaches might work.
What the authors got right.
Despite these concerns, let me emphasize what impressed me about this work:
Empirical commitment: Too much development discourse remains stuck at the ideological
level. The authors' quantitative approach represents precisely the kind of evidence-based analysis
we need more of.
Context sensitivity: Their emphasis on local conditions and development stages resonates with
growing recognition that one-size-fits-all approaches don't work.
Challenging orthodoxy: Academic progress requires scholars willing to question established
paradigms. The authors' systematic critique contributes to necessary scholarly debate.
Power dynamics: Their attention to how global economic structures constrain African policy
space raises vital questions about development sovereignty.
A path forward for better policy research.
For the authors, I'd suggest:
-
-
-
- Expanding robustness analysis with alternative model specifications
- Incorporating pre-SAP data for proper treatment effect analysis
- Moderating language to focus on empirical findings rather than ideological
characterizations
- Developing more nuanced policy recommendations that acknowledge implementation
challenges
For the broader research community, this paper highlights the need for:
-
-
-
- More sophisticated causal identification in development policy research
- Greater attention to heterogeneity across African countries
- Balance between critical analysis and constructive engagement with policymakers
The stakes are too high for poor communication.
Here's the bottom line: Africa's development trajectory is too essential for academic research to
be dismissed due to presentation problems. The authors have produced valuable empirical
insights about policy effectiveness that deserve serious consideration across the policy spectrum.
With methodological strengthening and more strategic communication, findings like these could
contribute much more effectively to the evidence-based policy discussions that Africa's future
depends on. The fundamental questions about policy effectiveness, local ownership, and
appropriate development strategies aren't going away—and neither should rigorous efforts to
answer them.
The challenge isn't just doing good research—it's ensuring that good research gets heard by those
who need to hear it most.
What do you think? Have you observed similar patterns in your country's experience with
structural adjustment? How can researchers better bridge the gap between empirical findings and
policy influence?
Read the original paper: Unpacking the African Response to Radical Neoliberal Policies.
