Is complexity theory relevant to poverty alleviation programs?

For me, global economic inequality is a huge issue. A helpful short video describes the problem.
Recently, there has been a surge of interest among development policy analysts about how complexity theory may be relevant in poverty alleviation programs.

On an Oxfam blog there is a helpful review of three books on complexity theory and development.
I recently read some of one of these books, Aid on the Edge of Chaos: Rethinking International Cooperation in a Complex World, by Ben Ramalingham.

Here is some of the publisher blurb.

Ben Ramalingam shows that the linear, mechanistic models and assumptions on which foreign aid is built would be more at home in early twentieth century factory floors than in the dynamic, complex world we face today. All around us, we can see the costs and limitations of dealing with economies and societies as if they are analogous to machines. The reality is that such social systems have far more in common with ecosystems: they are complex, dynamic, diverse and unpredictable. 

Many thinkers and practitioners in science, economics, business, and public policy have started to embrace more 'ecologically literate' approaches to guide both thinking and action, informed by ideas from the 'new science' of complex adaptive systems. Inspired by these efforts, there is an emerging network of aid practitioners, researchers, and policy makers who are experimenting with complexity-informed responses to development and humanitarian challenges. 

This book showcases the insights, experiences, and often remarkable results from these efforts. From transforming approaches to child malnutrition, to rethinking processes of economic growth, from building peace to combating desertification, from rural Vietnam to urban Kenya, Aid on the Edge of Chaos shows how embracing the ideas of complex systems thinking can help make foreign aid more relevant, more appropriate, more innovative, and more catalytic. Ramalingam argues that taking on these ideas will be a vital part of the transformation of aid, from a post-WW2 mechanism of resource transfer, to a truly innovative and dynamic form of global cooperation fit for the twenty-first century.

The first few chapters give a robust and somewhat depressing critique of the current system of international aid. He then discusses complexity theory and finally specific case studies.
The Table below nicely contrasts two approaches.

A friend who works for a large aid NGO told me about the book and described a workshop (based on the book) that he attended where the participants even used modeling software.

I have mixed feelings about all of this.

Here are some positive points.

Any problem in society involves a complex system (i.e. many interacting components). Insights, both qualitative and quantitative, can be gained from "physics" type models. Examples I have posted about before, include the statistical mechanics of money and the universality in probability distributions for certain social quantities.

Simplistic mechanical thinking, such as that associated with Robert McNamara in Vietnam and then at the World Bank, is problematic and needs to be critiqued. Even a problem as 'simple" as replacing wood burning stoves turns out to be much more difficult and complicated than anticipated.

A concrete example discussed in the book is that of positive deviance, which takes its partial motivation from power laws.

Here are some concerns.

Complexity theory suffers from being oversold. It certainly gives important qualitative insights and concrete examples in "simple" models. However, to what extent complexity theory can give a quantitative description of real systems is debatable. This is particularly true of the idea of "the edge of chaos" that features in the title of the book. A less controversial title would have replaced this with simply "emergence", since that is a lot of what the book is really about.

Some of the important conclusions of the book could be arrived at by different more conventional routes. For example, a major point is that "top down" approaches are problematic. This is where some wealthy Westerners define a problem, define the solution, then provide the resources (money, materials, and personnel) and impose the solution on local poor communities. A more "bottom up" or "complex adaptive systems" approach is where one consults with the community, gets them to define the problem and brainstorm possible solutions, give them ownership of implementing the project, and adapt the strategy in response to trials. One can come to this same approach if ones starting point is simply humility and respect for the dignity of others. We don't need complexity theory for that.

The author makes much of the story of Sugata Mitra, whose TED talk, "Kids can teach themselves" has more than a million views. He puts some computer terminals in a slum in India and claims that poor uneducated kids taught themselves all sorts of things, illustrating "emergent" and "bottom up" solutions. It is a great story.  However, it has received some serious criticism, which is not acknowledged by the author.

Nevertheless, I recommend the book and think it is a valuable and original contribution about a very important issue.