International development is a complex global goal that faces massive coordination barriers. The difference in income between rich and poor has expanded over the years from a four to one factor to a hundred to one. Where there once were only a handful of development agencies, thousands have now emerged. The system that connects donor agencies, recipient countries and development challenges is extremely complex and should not be managed with a top-down approach… The Aid Explorer was developed as a tool to facilitate better aid coordination. The Aid Explorer enables users to understand what issues face which countries and which aid organizations are aligned to address these issues.
Some specific pointers:
The Aid Explorer’s Profile pages enables us to see which issues face which countries and which organisations are best aligned to address them
The Network maps can be used to explore how issues, countries, and organizations relate to each other
The Rankings presents the findings and the best alignments of countries, issues and organizations
Many of the grand challenges that confront humanity—problems as diverse as climate change, the stability of markets, the availability of energy and resources, poverty and conflict—often seem to entail impenetrable webs of cause and effect.But these problems are not necessarily impenetrable. Powerful new tools have given scientists a better understanding of complexity. Instead of looking at a system in isolation, complexity scientists step back and look at how the many parts interact to form a coherent whole.
Rather than looking at a particular species of fish, for example, they look at how fish interact with other species in its ecosystem. Rather than looking at a financial instrument, they look at how the instrument interacts in the larger scheme of global markets. Rather than think about poverty, they might look at how income relates to conflict, politics and the availability of water. Whatever the object of study happens to be, complexity scientists assemble data, search for patterns and regularities, and build models to understand the dynamics and organization of the system. They step back from the parts and look at the whole.
This kind of thinking is a major departure from traditional science. For centuries, scientists have worked by reducing the object of study down to its constituent components. Complexity science, by contrast, provides a complementary perspective by seeking to understand systems as interacting elements that form, change, and evolve over time.The multiplicity of ideas, concepts, techniques and approaches embodied by the science of complexity can be applied to people, organizations and society as a whole, from economies and companies to epidemics and the environment.
The aim of this paper is to raise awareness about this new science and its ability to bring clarity and insight to many of the complex problems the world faces today.
In this post I want to argue that there is a middle ground between the unthinking mantras that are increasingly peddled by agencies and the growing number of entirely justifiable critiques.
However, with a growing chunk of economic growth being driven by industries and products that in fact didn’t exist ten years previously, such dismissals seem increasingly Luddite.
While clarity and precision in thinking about innovation is all-important, it is far from easy. We are not helped by the fact that many innovation stories are in fact apocryphal – retrospectively woven to lend the star protagonists much more agency and awareness than in fact they possessed. This is true of even the best known innovation stories. Take Alexander Fleming’s infamous and much-lauded discovery of penicillin – Fleming himself used to describe the conventional account of his contribution as the ‘Fleming Myth’.
Typically in business the market is the ultimate arbiter of innovation, and as we know, most products fail. In aid, however, the market does not provide an adequate indication of what is successful and what is not. This is largely done by the aid system itself. Bill Easterly made much of the fact that the market could get Harry Potter books anywhere there was a demand, which he compared unfavourably to the inability of the aid system to get simple treatments like vaccines to where they were most needed:
There was no Marshall Plan for Harry Potter, no International financing Facility for books about underage wizards. It is heartbreaking that global society has evolved a highly efficient way to get entertainment to rich adults and children, while it can’t get twelve-cent medicine to dying poor children.”
the disparity in the results is indeed heartbreaking… [but] there is a radical difference… between the enterprise of supplying “what is in demand” — which is integrally linked to the buyers’ ability to pay — and that of supplying needed goods and services to people whose income and wealth do not allow a need to be converted into a market demand.
While Sen’s point applies more broadly to aid delivery, it is also relevant to new ideas and innovations within aid.
In any case, whether because of market failure or the wilful self-interest of aid agencies, innovation – which is an ambiguous enough concept in the business realm – becomes very murky territory indeed in development. It is hard to say what innovation actually is, what it generates, and for whom. Like the famous ruling about pornography, many are of the view that ‘I know it when I see it.’
Such vagueness is the ideal seeding ground for development fads, and indeed, innovation is fast becoming the latest ‘fuzz-word’. Everything is being labelled ‘innovation’: as one blogger memorably put it, we seem to be suffering from Innovation Tourette’s.
Little wonder that growing numbers of thinkers and writers see the need to beat innovation with a big snarky stick. These criticisms play a vital role in highlighting the risks and downsides of all shiny new aid agendas – and innovation is no exception.
Having observed such trends in the past, I think there is a danger that between the rise of the fad and the indignant reaction to it, we lose sight and sense of why the issue is question is actually important. Specifically, we risk learning the wrong lessons about what innovation actually is, and the potential it has to add to our work.
What we need is a more precise and accurate way by which to separate the innovation wheat from the faddish chaff. This was in fact one of the key motivations of a study I co-authored with Kim Scriven and Conor Foley while at ALNAP back in 2008-9.
So what did our study suggest in terms of getting more precision in innovation? We found it useful to ask some key questions to identify whether a particular idea or approach was in fact innovative.
Q1. Is the idea being proposed a new Product, a new Process or service, a new way of Positioning aid, or a new Paradigm or mental model? Or is it some combination of the four? (see more here)
Q2. What are the origins of the idea, and what does it aim to do differently to what is already out there? Where, exactly, is the novelty – is it a whole new thing, or is it new combinations of existing things? What, in partciular, are the implications for relationships with aid recpients? Did the idea involve re-thinking that age-old and much-critiqued relationship?
Q3. How disruptive is the innovation? Is it transactional, in that it enables existing efforts to work; incremental in that improves these efforts, or transformational in that it radically changes these efforts?
Q4. What precisely are the expected benefits the idea should confer? Can these benefits be framed in terms of existing evaluation criteria of enhanced relevance, efficiency, effectiveness, impact or sustainability of aid? Or are there other, newer, criteria that matter? How can the benefits be measured – qualitatively, quantitatively, or some blend thereof?
Q5. What are the potential risks and downsides of the idea for all parties – especially aid recipients – and how will these be mitigated against?
Q6. Where can the idea be located in an overarching innovation process? Is it at the early stages of recognition and invention, is it in need of development and implementation, or has it been tested and is now ready for wider diffusion? (see more here)
Q7. What are the networks and relationships that will support and facilitate the innovation process? What capacities and competencies are necessary? Are these in place? How can they be built?
Q8. What is the potential scope of the innovation in terms of wider diffusion? Who might benefit, and in what ways? What is the route to scale, and who needs to be engaged to get there?
There are no doubt many more questions that could be asked, but the above provide a good starting point for what might be termed ‘innovation due diligence’. The key, in my opinion, is to use these and others questions to develop more honest, rigorous stories about ongoing and historical innovations: about how they came about, why, and with what benefits. Such questions are useful because they help us look at innovation from a more systemic perspective: looking not just a idea, but the overall social, technological and institutional context from which it has emerged.
These questions should be relevant whether you are a donor bombarded with new proposals and ideas, an operational aid worker seeking to get funding for your exciting new idea, a blogger wanting to shine a light on the depressing excesses of innovation-speak, or a researcher wanting to investigate an supposed innovation in a systematic fashion (in fact I think we need far more of the last category, but that’s another story).
We will need to keep wielding the big stick as necessary, to curb against such excesses of aid ‘innovation-speak’. But we may also at times need a magnifying glass and ruler – metaphorically speaking – and asking these kinds of questions could help with this. My $0.02 is that if a would-be innovator can’t take a reasonable stab at these questions, they aren’t working hard enough, or they are over-selling something. A lot is spoken about creativity in innovation, but recent work suggests – in echo of the old 1% inspiration, 99% perspiration line – that the larger part of innovation lies in the proper execution of the idea.
I think there is a special role for the aid blogging community in asking such questions and demanding answers. We have seen in the past few years how bloggers have mobilised in a largely self-organised fashion to push back against various poorly considered ideas.
I know there are many bloggers who want to engage with innovation in a serious fashion, and who are dismayed by the current hype surrounding it. We should be able to highlight the good and bad of what we see emerging from the aid innovation agenda. And aid agencies should be willing to open their ideas up to the views and scrutiny of this emerging, globally networked, community of thinkers and analysts. This kind of effort has, in other distributed sectors, developed into new crowd-sourced marketplaces for innovation such as Innocentive. There’s no reason why the same couldn’t happen in our sector.
Our ultimate goal, I’d argue, should be to work to bringing the perspectives of aid recipients into the mix as part of our standard operating procedures. Now that in itself could be seen as a real innovation.
The need for such engagement goes beyond mere niceties. The most effective ideas we uncovered in our 2009 study were precisely those that re-thought and re-formulated this core aid relationship: cash, community approaches to malnutrition, transitional shelter. Put simply, these were the innovations that we found to be most worthy of the term.
This guest post by Andy Sumner andSergio Tezanos Vázquez explores new approaches to classifying developing countries, based on a new IDS Working Paper published last week. In the paper, they develop a more precise and accurate classification system for low and middle income countries, and suggest that this can support a more complex, non-linear appreciation of development trajectories. This post concludes with the intriguing notion of ‘development landscapes’ (building on a recent CGD blog post written by Owen Barder and myself).
[t]he typical case is a largely unindustrialised economy, the foreign trade of which consists essentially in selling primary products for manufactures. There are about 100 identifiable economies of this sort, covering the great majority of the world’s population (1963, p. 80).
The questions to ask about a country’s development are therefore: What has been happening to poverty? What has been happening to unemployment? What has been happening to inequality? If all of these three have become less severe, then beyond doubt this has been a period of development for the country concerned […] If one or two of these central problems have been growing worse, especially if all three have, it would be strange to call the result ‘development’, even if per capita income has soared (1969:24).
Our new paper continues this tradition with what might be described as a complex systems twist. The paper challenges the continuing use of income per capita to classify developing countries as low income countries (LICs) or middle income countries (MICs), given that most of the world’s poor live in the later group (see here,here, here, here).
Further, it highlights the ambiguity over the usefulness of the MIC classification given the diversity in the group of over 100 countries that includes Ghana and Zambia, as well as India, China and Brazil.
Finally, it points to a new way of framing development, which moves beyond the linear league table approach which tends to imply a certain shared pathway (cf critiques of HDI as ranking countries on a scale of 0 to Denmark). In closing, we touch on one example of what this might look like below: there are no doubt others.
We used a cluster analysis to identify five types of developing country using a set of indicators for 2005-2010 covering definitions of development based on the history of thinking about ‘development‘ over the last 50 years from four conceptual frames:
development as structural transformation;
development as human development;
development as democratic participation and good governance;
development as sustainability.
This is what we found:
Our development taxonomy differs notably from the usual income classification of GNI per capita (Atlas method) used to classify LICs and MICs. Notably many countries commonly labelled “emerging economies” are not in the emerging economies clusters because they retain characteristics of poorer countries.
Our analysis generated 5 clusters as follows:
Cluster 1: High poverty rate countries with largely traditional economies. Those countries with the highest poverty and malnutrition headcounts, who are also countries with low productivity and innovation and mainly agricultural economies, with severely constrained political freedoms. This cluster includes 31 countries, some of them might be surprising: Pakistan, Zambia, Nigeria, and India.
Cluster 2: Natural resource dependent countries with little political freedom. Those countries with high dependency on natural resources, who are also countries with severely constrained political freedom and moderate inequality (relative to the average for all developing countries). This cluster includes 9 countries, such as Mauritania, Vietnam, Yemen, Cameroon, Congo, Swaziland and Angola.
Cluster 3: External flow dependent countries with high inequality. Those countries with high dependency on external flows, who are also countries with high inequality, and moderate poverty incidence (relative to the average for all developing countries). This cluster has 32 countries, such as Senegal, Ghana, Indonesia, Thailand, Peru, Colombia, and Panama.
Cluster 4: Economically egalitarian emerging economies with serious challenges of environmental sustainability and limited political freedoms. Those countries with most equal societies, who are also countries with moderate poverty and malnutrition but serious challenges of environmental sustainability and –perhaps surprisingly– limited political freedoms. This cluster has 15 countries, including China, Azerbaijan, Belarus, and Kazakhstan.
Cluster 5: Unequal emerging economies with low dependence on external finance. Those countries with the lowest dependency on external finance and who are also countries with the highest inequality. This cluster includes 14 countries, such as South Africa, Botswana, Costa Rica, Malaysia, Argentina, Mexico, Brazil, Turkey, Chile and Uruguay.
Two-thirds of the world’s poor – not surprisingly given the characteristics noted above – live in Cluster 1 countries though this is largely due to the inclusion of four populous countries (Bangladesh, India, Pakistan and Nigeria –and one should remember a third of world poverty is accounted for by India). About a quarter of world poverty is situated in Cluster 3 and Cluster 4 countries and the remaining 5% live in Cluster 2 and Cluster 5.
What is most striking – and of particular relevance for readers of this blog – is that we find that there is no simple “linear” representation of development levels (from low to high development countries), as found in the Human Development Index or its variants.
We find instead that each development cluster has its own and characteristic development issues. Building a development classification is not a simple task: once we overcome the over-simplistic income-based classification of the developing world, we find that there is no group of countries with the best (or worst) indicators in all development dimensions.
It thus would be more appropriate to build “complex” development taxonomies on a five-year basis than ranking and grouping countries in terms of per capita incomes or other indicators, as this will offer a more nuanced image of the diversity of challenges of the developing world and policy responses appropriate to different kinds of countries.
In keeping with a growing consensus in development thinking, this way of thinking points to an understanding which is less about linear progression, and where change is path dependent, and development policy is best seen as an evolutionary process of learning.
What might such a complex taxonomy look like? A fruitful avenue to explore (following the recent post by Ben Ramalingam and Owen Barder on complexity and results) might be the notion of fitness landscapes from evolutionary biology, which illustrate how different species attain biological fitness through processes of variation, selection and amplification.
Erik Beinhocker and Tim Harford have argued that growth and innovation can be likened to a evolving search of a dynamic fitness landscape. “Development Landscapes” might provide an analogy and a model for explaining the distinct but related positions of countries in different development clusters.
This would enable analysts and policy makers to examine the different pathways the cluster members have been on, and to explore the space of possibilities for future change.
About the guest post authors:
Sergio Tezanos Vázquez is an associate professor at the Economics Department and a research fellow at the Iberoamerican Research Office on International Development & Co-operation at the University of Cantabria.
Andy Sumner is Co-Director of the newly established, King’s International Development Institute, King’s College London.
Earlier this month, Nature published a piece by Daniel Sarewitz on emerging challenges faced in science and research, which has some useful lessons for the aid system.
The greatest threat to science is not due to the usual suspects of “inadequate funding, misconduct, political interference”, etc, etc. Instead, according to Daniel Sarewitz, the problem is more fundamental and relates to a widespread bias towards over-simplified models and positive results.
Bias is an inescapable element of research, especially in fields… that strive to isolate cause–effect relations in complex systems in which relevant variables and phenomena can never be fully identified or characterized…”
The field Sarewitz is writing here about is biomedicine, but he could easily be describing development or humanitarian work. The fundamental problem, as he sees it, is that biases are not random but systemic: “if biases were random, then multiple studies ought to converge on truth [but] evidence is mounting that biases are not random.”
This claim is not new, of course. As the piece argues, systematic positive bias was identfied in clinical trials funded by the pharmaceutical industry back in the mid-1990s. More recently, reviews of so-called ‘landmark’ studies in fields such as cancer research has shown that positive results could only be replicated in a minority of cases.
However, these previous assessments tended to assume that the problem was not with science per se, but rather with those forces that sought to co-op it: industry, government, special interests, and so on. Reduce the influence of these interests, the argument went, and you would eradicate such biases.
But it is now emerging that there are some serious underlying problems within science itself. The cases are wide-ranging across biomedicine: “evidence of systematic positive bias [is] turning up in research ranging from basic to clinical, and on subjects ranging from genetic disease markers to testing of traditional Chinese medical practices.”
The two major faultlines, according to Sarewitz, are the methodological narrowness of the approaches employed to generate evidence, and the culture and incentives of scientists and science funders.
The first one is pertinent for readers of this blog. Researchers seek to reduce bias “through tightly controlled experimental investigations. In doing so, however, they are also moving farther away from the real-world complexity in which scientific results must be applied to solve problems.” Ironically, “the canonical tenets of ‘scientific excellence’” are threatening to undermine the whole enterprise. One rather shocking (for me, at least) example relates to the latest developments in research on mice, where a lot of resources and funds have been poured into the cloning of genetically identical animals, in order to enable fully controlled, replicable experiments and rigorous hypothesis-testing. Any sense of moral repugnance aside, perhaps the worst thing about this endeavour is that the findings of the research subsequently undertaken have turned out to be useless when applied in the real world.
Sarewitz also writes about the lack of incentives to ‘report negative results, replicate experiments or recognize inconsistencies, ambiguities and uncertainties’. There are also challenges around the various cultural and attitudinal positions taken toward science among funders, scientists, the media and the public at large. Sound familiar?
It should – such issues are not a problem for biomedicine alone:
[they are] likely to be prevalent in any field that seeks to predict the behaviour of complex systems — economics, ecology, environmental science, epidemiology and so on. The cracks will be there, they are just harder to spot because it is harder to test research results through direct technological applications… and straightforward indicators of desired outcomes…
Sarewitz closes with one potential solution, which may also be of relevance for work in development and humanitarian fields:
Scientists rightly extol the capacity of research to self-correct. But the lesson coming from biomedicine is that this self-correction depends… on the close ties between science and its application that allow society to push back against biased and useless results.
So what can we in the aid sector do about such bias, if indeed it is present in our work?
The first idea is the one that Sarewitz suggests: “societal push back”. Sadly, despite the rhetoric and growing practice of participation, the scope for Southern stakeholders – especially aid recipients – to ‘push back’ against useless results in development and humanitarian research is still severely limited. This doesn’t mean we should stop the effort, however, and perhaps new technologies and feedback processes can help us here.
The second strategy might be to address issues of the incentives and cultures which perpetuate such biases. But we seem to be far too concerned with developing country actor incentives and motivations to look at those in our own organisations. As one participant at a recent ODI event put it: “why do we always say that developing country leaders have mixed motives at best whereas the motives of donors [and other aid actors] are always considered impeccable?” We should find a way to ensure that these aid “physicians” first heal themselves.
The final course of action is to try to expand and adapt the concepts and models used in our work. This effort (of which this blog is one small part) is still very much a work-in-progress, but the growing interest among researchers and practitioners should give us some small cause for hope. After all, the key to paradigm shifts in science – and in other fields – is not just logical argument and experimental proof. In the words of Thomas Kuhn:
as in political revolutions, so in paradigm choice—there is no standard higher than the assent of the relevant community.”
Innovation is getting a lot of attention at the moment in development and humanitarian work. Many, including myself, see this as long overdue. But, according to an article in this weeks Economist, this attention may be misplaced. The piece makes a strong argument for the importance of imitation in business, and its advantages over innovation. In this post I want to take a look at these arguments for imitation. I also want to see what complex systems research tells us about the limits and possibilities of such an approach.
I: The Virtues of Copying?
Innovation is essential. Countless speeches, articles and books attest to its central importance – in economic growth, business success, and organisational effectiveness. As a result, imitation is a “heretical idea”. But the uncomfortable truth, according to the piece in this weeks Economist at any rate, is that in the real world, firms that copy others are more successful.
Examples include:
The iPod, the iPhone and the iPad were not the first of their kind: “Apple imitated others’ products but made them far more appealing.”
Pharmaceutical firms can be divided into inventors and imitators, and some inventors have joined the copycats, selling generic drugs
Supermarket own-label products copy well-known brands, making for a multi-billion dollar product category
High street fashion firms consistently copy innovations from the catwalk.
Such imitators often proved to be the winners in business:
copying is not only far commoner than innovation in business… but a surer route to growth and profits…. studies show that imitators do at least as well and often better from any new product than innovators do. Followers have lower research-and-development costs, and less risk of failure because the product has already been market-tested…”
So why isn’t imitation lauded? The key issue is that the incentives for copying are weak – whether legal, individual, organisational or cultural. Set aside the obvious issue around patents, and we learn that “praise and promotion do not go to employees who borrow from other firms.” A study of how new product development firms go about their work found that none had a formal or informal policy for responding to other firms’ innovations, making them slow to learn from others successes. And there may well be cultural factors at play here, although the piece made what for me was a rather lazy and out-dated comparison: US firms tend to be obsessed with innovation, whereas Asian firms are far better at legal imitation.
II: Imitation or Exploring Adjacent Possibilities?
I finished the piece feeling that there was more to this issue than the author acknowledged. At best, the analysis was incomplete, at worst, very simplistic.
First, many of the successful “copying” efforts highlighted in the article highlights were far from easy or straightforward. The process of imitation involves numerous adaptations and innovations – some of them significant and certainly not cheap.
Take the iPhone, which is held up as a kind of archetype of copying. Clearly it wasn’t the first smartphone, but compare it to what was around at the time and it is clear that Apple weren’t simply copying and pasting ideas. The idea that Apple somehow saved on the R&D costs of the iPhone because of the advances made in prior products seems risible.
To borrow an idea from evolutionary biology, I would argue that the key to the most of the successful imitations that the article presents is really that many of the so-called copycat firms explored the “adjacent possibilities” – the diverse and emergent possibilities that spring up around a new idea, product or process.
The core of the idea is that people arrive at the best new ideas when they combine prior, adjacent, ideas in new ways. Most combinations fail; a few succeed spectacularly. So, copying might succeed – but there are all kinds of examples where it doesn’t.
III: Innovation in the App Ecosystem
Second, and perhaps even more importantly – you need innovators to copy from in the first place. No innovators means nothing to copy which means nothing to sell. To really get to grips with the ‘imitation vs innovation’ debate, the Economist piece needed to pay more attention to the overall system of firms, consumers and products – and the dynamics and interactions in that system. Fortunately, recent and excellent work by two researchers at University College London addresses exactly this issue in the context of mobile applications, and presents some very pertinent conclusions.
The “app economy” of producers and users that has sprung up around the applications that run on mobile phones and devices is nothing short of startling. It has been referred to by industry experts as “one of the biggest economic and technological phenomena today”.
Soo Ling Lim and Peter Bentley of UCL were interested in the dynamics of the app economy, in particular wanted to understanding the role of innovation.
Lim and Bentley – one is a prize-winning software engineer, the other a successful “app entrepreneur” – started with the insight that will be familiar to Aid on the Edge readers: that the app economy was best seen as a “co-evolving system of apps, developers, and users [who] form complex relationships, filling niches, competing and cooperating, similar to species in a biological ecosystem.”
Apple releases very little data on its stores and so the researchers decided that the best way to get around this would be to build an ecosystem model to simulate the dynamics they observed. They programmed agents in their model – appropriately named AppEco – to mimic the behaviour of developers and consumers.
Developers build and upload apps to the app store; while consumers browse the store and download the apps. They also programmed apps – passive artefacts in the ecosystem which are the key means by which the agents interact.
They then programmed their developers with different characteristics. They identified five broad types of developers in the real app economy, and built these characteristics into their model agents. They are innovators, optimisers, milkers and copycats, and flexibles. Although specific to the app economy, there are clear parallels in most other sectors and contexts.
Innovators are those developers who come up with groundbreaking apps – like AroundMe or TuneIn Radio. In te model, innovators were developers who produced different apps in a variety of categories – including social networking, business, utilities, and productivity.
Optimisers take a hit formula, such as the Angry Birds franchise, and try to adapt and continually improve it. In the model, these developers were the learners – they took their own best app and made variations on it.
Milkers have one specific idea and use it repeatedly. They might, for example, create apps for each of hundreds of town maps, rather than building one app that can call up many maps. In the model, they used their most recent idea and varied it repeatedly.
Copycats build knock-offs of top-selling apps – see Angry Chickens or Angry Dogs – and work by appealing to or confusing users who end up buying the facsimiles.
Flexibles begin with one of the strategies above, but change their strategy based on the strategy of the top developers.
Lim and Bentley then calibrated the model to match the behaviour of a real app store, in this case, the Apple iOS App Store, which is the oldest and best established store. They used three years worth of publicly available data from the store and primed the model until it closely resembled the behaviour of the real store.
The next stage was to run some “what if” experiments. The specific questions that inspired these experiments have direct relevance to the Economist piece. For example, with so many developers trying out different strategies to increase their downloads, Lim and Bentley wanted to know if an innovative developer would receive more downloads compared to a copycat developer.
At the start of each simulation of the App economy, all five categories of developers contributed an equal number of apps, but different constraints were placed on the system. A whole range of different scenarios then evolved, including the following:
If the proportion of apps from each group was kept constant, copycats made the most money initially.
Over time, however, the overall ecosystem suffered from a lack of novel products. Dissatisfied users moved onto better platforms
Copycats rely on good apps created by other strategies; it is extremely difficult for an ecosystem to support a large proportion of copycats. (This result mirrors the app stores in the real world – copycat developers regularly appear and take advantage of the success of others, but nevertheless their strategy remains in the minority.)
When consumer choices dictated which apps were successful, it was the optimisers who sold the most apps, followed by innovators, milkers and finally the copycats.
The general conclusions in the authors’ summary paper seem clear:
In a complex ecosystem no strategy can be a guaranteed winner, but our results indicate that some strategies should be chosen more frequently than others. Innovators produce diverse apps, but they are hit or miss – some apps will be popular, some will not. Milkers may dwell on average or bad apps as they churn out new variations of the same idea. Optimisers produce diverse apps and tailor their development towards users’ needs. Finally, Copycats may seem like the best strategy to guarantee downloads in an app ecosystem, but the strategy can only work when there are enough other strategies to copy from. In addition, this strategy can only exist in a minority, otherwise app diversity will decrease (many duplicated apps result in a scarcity of some features desired by users) and the fitness of the ecosystem will suffer.” (emphasis added)
IV: Conclusions: Mix it Up
Both of these ideas – “adjacent possibilities” and the ecosystem model – suggest that the Economist article downplayed the difficulties and limitations of imitation. While it briefly acknowledges Schumpeter’s concerns about imitation dominating industry, the overall tone is bullish, suggesting that: “copying is here to stay; businesses may as well get good at it.”
The reality, however, is that copying is seldom a straightforward matter, and can take as much creativity and resources as innovation.
And there should be serious concerns about the overall health of a ‘ecosystem’ dominated by copycats. Copying is a strategy that can only work for a minority of players, and then only for limited periods.
The challenges of achieving global public policy consensus aside, new research is highlighting a range of other pressing concerns that need urgent attention.
Last week saw the launch of the summary of the IPCC special report on ‘Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation’ (SREX). This was the result of a two-and-a-half year long global collaboration between 220 scientists (full disclosure: I was one of the 220 & wrote sections of chapter 6 on managing climate risks at a national level).
One of the key messages of the IPCC report is that existing risk management and adaptation measures need to be improved dramatically. Many countries were found to be poorly adapted to current extremes and risks – let alone those projected for the future.
As recent Stockholm Resilience Centre research shows, this is more than just a technical issue. In fact, the study suggests the time is ripe for a serious rethink of the way resilience and adaptation measures are being designed and implemented.
Resilience is popular. Over the last year, I have being doing an increasing amount work on this ‘big new idea’ in international affairs – for DFID, ODI and others.
While it’s clear that there is a lot of value in this area for development and humanitarian efforts, there are some conceptual and operational challenges that need to be addressed. One widespread issue that I’ve noticed is that while aid agencies are embracing resilience, they are also tending to put the underlying theoretical framework of complex systems to one side as ‘too complicated’.
The study from experts at the Stockholm Resilience Centre shows that such conceptual simplifications of resilience can have considerable downsides – in the extreme, they can lead to interventions that actually diminish resilience.
There are now hundreds (if not thousands) of major public sector initiatives that have been developed in response to climate change, in high, middle and low income countries alike. Adaptation strategies include adjusting economic activities, changing land and energy use practices, and reforms to the design and implementation of infrastructure.
The study authors evaluated nine adaptation policy responses to assess how much they were affecting the resilience of various social-ecological systems. The findings were sobering: ‘Out of the nine initiatives analysed, only three had elements that could enhance resilience as much as reducing it. The other six had effects that predominantly reduced the resilience of a system.’
The reason? Many of the policy approaches to climate risks focused too much on short-term benefits and sought simple technological fixes to problems that were more complex. Such responses, designed with a focus on one single risk factor, can inadvertently undermine the capacity to address other stresses. As the authors put it:
There is growing evidence that current policy approaches… fail to significantly address multiple and interacting factors which affect system resilience and the needs of vulnerable populations.
Such over-simplistic efforts ‘create bizarre distortions in public policy’ precisely because climate vulnerabilities are created through multiple stresses, and not single factors. The problems went beyond how risks were defined – issues of governance, feedback and learning were also identified as critical. As the authors put it:
[In those] situations in which system stresses were defined as narrow, technical problems with short-term horizons… governance structures were top-down, did little to link actors at different scales, masked system feedbacks, and did not provide incentive or structure to promote learning…. In contrast, in the two examples where the issue was framed in a broader manner, policy implementation tended to enhance characteristics that supported the ability to manage resilience, including flexibility and learning.
Is there any explanation for this widespread focus on single risk factors? There are numerous reasons cited in the study. These include:
a desire for readily observable metrics
existing political structures and incentives
entrenched institutional cultures, and
long histories of dealing with social and ecological problems in narrow and limited ways
All these factors have been identified as systemic problems in international aid agencies, both on this blog and elsewhere. Indeed, some of the most troubling manifestations of the push for simplification were to be found in developing country case studies. For example, fisheries management in Uganda and drought responses in Kenya both highlighted the importance of local sources of resilience based knowledge of local ecosystems and social networks. But in both cases, the local sources of potential resilience were diminished by actors and forces operating at wider level.
Given this important new evidence, we are left with what seems like an obvious choice. To paraphrase the study authors, do we want efficient and effective adaptation measures, narrowly and technologically defined? Or do we want strategies that are more open-ended and innovative and seek to build resilience by understanding and strengthening local capacities?
The answer may seem obvious, but as global climate policy debates have repeatedly highlighted, in this realm the obvious choices are often the hardest agree upon. Politics and special interests clearly play a major role, and can all too often inhibit the space for evidence-based considerations.
One would hope that the adaptation issue is less entrenched than the battle that continues to be being waged around mitigation. At the very least, policy makers and practitioners alike need to become more aware of, and work with the key finding of the study – namely that:
dealing with specific risks without full accounting of the nature of system resilience leads to responses that can potentially undermine long–term resilience…”
Ricardo Hausmann of Harvard and Cesar Hidalgo of MIT (whose work I have blogged about previously here) have just published the deeply impressive Atlas of Economic Complexity. It is built around an innovative, network-based methodology for understanding economies and their potential for growth. It represents perhaps the most systematic and in-depth application of the ideas and methods of complexity research to issues of development to date. Readers can download the Atlas and experiment with a powerful interactive visualiser here.
Following an interview with Cesar Hidalgo last week, this extended post explores the implications of this important new work.
I. What is the premise of the Atlas?
The basic idea underpinning the Atlas of Economic Complexity is straightforward. As Hausmann notes:
The fundamental proposition… is that the wealth of nations is driven by productive knowledge. Individuals are limited in the things they can effectively know and use in production so the only way a society can hold more knowledge is by distributing different chunks of knowledge to different people. To use the knowledge, these chunks need to be re-aggregated by connecting people through organizations and markets. The complex web of products and markets is the other side of the coin of the accumulating productive knowledge. [emphasis added]
The secret to modernity is that we collectively use large volumes of knowledge, while each one of us holds only a few bits of it. Society functions because its members form webs that allow them to specialize and share their knowledge with others.”
At the heart of the Atlas is the attempt to measure the amount of productive knowledge that each country holds by applying network analysis techniques to this complex web.
Much standard development – and economic – thinking doesn’t engage very well with the idea of webs and networks. As Hidalgo told me, such ideas run counter to much standard thinking, which seeks to identify differences between individuals and groups based on their inherent qualities – demographic criteria and suchlike. Experts then puzzle over why, for example, communities with the same criteria, or countries with very similar starting points end up with very different development pathways and social and wealth outcomes. It turns out that in many cases, their relationships and networks prove to be a key differentiating factor. If the data is available, it is possible to develop very precise and rigorous analysis of these differences.
II. How does the Atlas work?
So how does the Atlas make these ideas relevant to development economics? Well, for starters, it acknowledges that accumulating productive knowledge is difficult: “For the most part, it is not available in books or on the Internet. It is embedded in brains and human networks. It is tacit and hard to transmit and acquire. It comes from years of experience more than from years of schooling. Productive knowledge, therefore, cannot be learned easily like a song or a poem. It requires structural changes. Just like learning a language requires changes in the structure of the brain, developing a new industry requires changes in the patterns of interaction inside an organization or society.”
As readers will be well aware, the social accumulation of productive knowledge has not been universal: “The enormous income gaps between rich and poor nations are an expression of the vast differences in productive knowledge amassed by different nations.”
These differences are expressed in the diversity and sophistication of the things that each nation makes. In order to put knowledge into productive use, societies need to reassemble these distributed products through teams, organisations and markets. These issues are explored in detail in the Atlas, through the concept of the ‘product space’. This is a map which captures the products made by different countries in terms of their knowledge requirements. This maps provide a way of understanding how productive knowledge is accumulated.
As Hidalgo said in interview:
Knowledge doesn’t add up like capital. There is a lot of redundancy in knowledge. Some countries may have diverse knowledge but small populations. Product space is an expression of different kinds of knowledge – and its much better than other indicators.
Cesar’s TED talk from August 2010 gives more information about this idea and how it works.
The underlying notion of this analysis is that the complexity of an economy is related to the range of useful knowledge embedded in it:
For a complex society to exist, and to sustain itself, people who know about design, marketing, finance, technology, human resource management, operations and trade law must be able to interact and combine their knowledge to make products. These same products cannot be made in societies that are missing parts of this capability set. Economic complexity, therefore, is expressed in the composition of a country’s productive output and reflects the structures that emerge to hold and combine knowledge… Increased economic complexity is necessary for a society to be able to hold and use a larger amount of productive knowledge, and we can measure it from the mix of products that countries are able to make.”
III: What does it all mean?
So what does this give us in practical terms? As a starter, representing such a huge amount of data – covering 128 countries, 99% of world trade, 97% of the world GDP and 95% of the world population – in visual form is in itself a remarkable feat. As people like Hans Rosling have powerfully demonstrated, innovations in how we visualise data can yield tremendous new insights and ideas.
Here’s an example of a product space diagram, this one for the United States. To learn more about the diagrams and how to interpret them, I would strongly recommend having a play with the visualiser, then scanning the report, then having another play.
Hausmann, Hidalgo and their team have also developed an Index of Economic Complexity to represent their data systematically. This Index tells us about the richness of the product space of a given country, and by extension, is one useful indicator of the potential to grow. It can also be used to compare economic complexity across countries, as shown in this chart showing the ranking of different countries (from 1 to 128, highest is most red).
The authors acknowledge that these ideas are not always easy to grasp, and provide a useful thought-experiment to help readers get their heads around the implications of the Index.
Think of a particular country and consider a random product. Now, ask yourself the following question: If this country cannot make this product, in how many other countries can this product be made? If the answer is many countries, then this country probably does not have a complex economy. On the other hand, if few other countries are able to make a product that this country cannot make, this would suggest that this is a complex economy.
So for example, Japan and Germany are the two countries with the highest levels of economic complexity and if a good cannot be produced there, the list of other potential countries is likely to be very short. Conversely if a product cannot be made in Mauritania or Sudan, the list of other potential countries is likely to be a long one.
One useful way of understanding the benefits of the Atlas is to think about what the analysis adds to some of the key questions in growth economics. One of the classic comparisons made in the growth literature is between African and East Asian countries – which were at comparable levels of development in the 1950s-1970s, but which are now literally worlds apart.
Hausmann and Hidalgo give their take on this by comparing the Economic Complexity Index for Ghana and Thailand. The lessons are resonant for aid agencies. Both countries had similar levels of schooling in 1970, and Ghana expanded education more vigorously than Thailand in the subsequent 40 years, supported of course by external assistance and policy recommendations.
Despite this, “Ghana’s economic complexity and income stagnated as it remained an exporter of cocoa, aluminium, fish and forest products. By contrast, between 1970 and 1985 Thailand underwent a massive increase in economic complexity, equivalent to a change of one standard deviation in the Economic Complexity Index. This caused a sustained economic boom in Thailand after 1985. As a consequence, the level of income per capita between Ghana and Thailand has since diverged dramatically.”
The Economic Complexity Index has been shown to be a better predictor of economic growth than a number of other existing development indicators. For example, as reported in the Economist last week, it outstrips the WEF index of competitiveness by a factor of 10 in terms of the accuracy of its predictions. It also outperforms the World Governance Indicators and the standard variable used to measure human capital as predictors of growth.
There are many other rich and varied insights from the work which cannot be covered in detail here. There is also tremendous potential to build on and extend this data and analysis in the future. One of the areas I have been working on recently is on resilience, both as a means of reducing the impact of future crises and disasters, and as a means of securing development gains. This issue is understandably at the forefront of many policymakers’ minds at the moment. The network analysis underpinning the Atlas could be used as a very useful comparative indicator of economic resilience, comparing the sustainability of growth in different countries, and help us think through growth scenarios which might enhance or diminish resilience.
There may also be scope to use this kind of thinking to bring more rigour and realism to problems of industrial reform. Take for example the ubiquitous issue of how we move to low-carbon industrial strategies. It would need more data and analysis, but the product space is clearly a powerful way to start to think about the key issues in a systematic and data driven fashion. There are numerous climate change benchmarks out there but none – as far as I know – employ the kind of network analysis used in the Atlas, and so a key aspect of how industrial economies work is missed out. By understanding better the carbon reliance of a particular countries product space, it is possible to think through the implications – the likely successes and failures – of existing adaptation policies.
IV: In conclusion
Perhaps the most important contribution of the Atlas is the analytical rigour that it brings to the complex and dynamic nature of economic growth, and the ability it gives us to ask new and challenging questions more precisely. Cesar summed it up for me as follows:
What we really want to do is to inspire a new kind of conversation. Our traditional approach to economics has retained measures developed in 1930s and 1940s to deal with the situations and crises we faced back then. We think there should be a new breed of measures – that bring much more precision and resolution, and that mean we don’t continue to build our analysis on the over-simplification of a complex system.
Let’s hope we see more of this way of thinking in development debates. While there will inevitably be a degree of resistance from the old guard, it seems to me that the underlying premise of the report is something that no one could disagree with:
Ultimately, this Atlas views economic development as a social learning process, but one that is rife with pitfalls and dangers. Countries accumulate productive knowledge by developing the capacity to make a larger variety of products of increasing complexity. This process involves trial and error. It is a risky journey in search of the possible.”
Such lessons clearly need to play a much more central role in development policy and practice. Haussman, Hidalgo and their team have done us a real service with this work.
Despite increased prominence and funding of global health initiatives, attempts to scale up health services in developing countries are failing, with serious implications for achieving the Millennium Development Goals. A new paper argues that a key first step is to get a more realistic understanding of health systems, using the lens of complex adaptive systems.
Much ongoing work in development and humanitarian aid is based on the idea of ‘scaling up’ effective solutions. Healthcare is one of the areas where this idea has played a central role – from WHO’s Health for All in the 1960s to UNICEF’s child healthcare programmes, from rolling out HIV-AIDS, malaria and TB treatments to the package of interventions delivered to achieve the Millennium Development Goal on health.
However, despite the fact there are many cost-effective solutions to health problems faced in developing countries, many agencies are still frustrated in their attempts to deliver them at scale. This may be because of a widepread failure to understand the nature of health systems.
complex systems made up of networks of many heterogeneous components that interact non-linearly. While pathways of change can be shaped by governance and are influenced by path dependencies, they are not entirely controllable or predictable; there will always be uncertainties and unintended consequences and new ‘emergent’ interactions and behaviours.’
If we accept this eminently sensible description, then it is little wonder that scaling up efforts continue to be frustrated. The paper by Ligia Paina and David H Peters, published in Health Policy and Planning in August, argues that there is a drastic need for a shift in thinking:
…from the current models around scaling up health services, which revolve around linear, predictable processes, to models that embrace uncertainty, non-linear processes, the uniqueness of local context and emergent characteristics.”
Their argument is supported by the fact that existing assumptions about the nature and hoped-for successes of scaling up have led to a lot of disappointments. Moreover, these efforts ‘offer little insight on how to scale up effective interventions in the future.’
The paper explores 5 concepts of complexity science, illustrated below.
All of these ideas carry relevant lessons for the design, planning, implementation and evaluation of health policy and programmes. As the authors conclude:
The implications include paying more attention to local context, incentives and institutions, as well as anticipating certain types of unintended consequences that can undermine scaling up efforts, and developing and implementing programmes that engage key actors through transparent use of data for ongoing problem-solving and adaptation.”
The authors close with a proposal that future efforts to scale up should adapt and apply complex systems models and methodologies which have been used in other fields but which remain underused in public health. These include network scinece approaches, modelling techniques, and tools to better understand systems dynamics.
The potential benefits are clearly stated:
This can help policy makers, planners, implementers and researchers to explore different and innovative approaches for reaching populations in need with effective, equitable and efficient health services.”
These are all fascinating developments, and suggest that health may be a key area where the ideas from complexity science can prove of tangible value for development and humanitarian work.
Interested readers can hear an podcast about the article here, with David Peters talking about his ideas and experiences (and me saying a few words about the history of complexity science and the relevance for health efforts.) David has also blogged about it here.
The latest issue of American Scientist features some superb reflections by Robert L Dorit on the limitations of reductionist thinking in the biological sciences. They have clear parallels for social sciences and, by extension, for social policy. Selected extracts are below.
Despite Descartes’ contention that we could not distinguish a well-made automaton of an ape from an actual ape (“were there such machines exactly resembling organs and outward form an ape or any other irrational animal, we could have no means of knowing that they were in any respect of a different nature from these animals”), the relations of parts to wholes in living systems is entirely different from that in machines—and most unclocklike. If anything, living systems consistently violate all of the criteria for reducibility. The number of elements that compose any living system—an ecosystem, an organism, an organ or a cell—is enormous. In living systems, the specific identities of these component parts matter. Unlike chemistry, for instance, in which an electron in a lithium atom is identical to an electron in a gold atom, all proteins in a cell are not equivalent or interchangeable. Each protein is the result of its own evolutionary trajectory. We understand and exploit their similarities, but their differences matter to us just as much. Perhaps most importantly, the relations between the components of living systems are complex, context-dependent and weak. In mechanical machines, the conversation taking place between the parts involves clear and unambiguous interactions. These interactions result in simple causes and effects: They are instructions barked down a simple chain of command.
In living systems, by contrast, virtually every interesting bit of biological machinery is embedded in a very large web of weak interactions. And this network of interactions gives rise to a discussion among the parts that is less like a chain of command and more like a complex court intrigue: ambiguous whispers against a noisy and distracting background.
As a result, the same interaction between a regulatory protein and a segment of DNA can lead to different (and sometimes opposite) outcomes depending on which other proteins are present in the vicinity. The firing of a neuron can act to amplify the signal coming from other neurons or act instead to suppress it, based solely on the network in which the neuron is embedded. The disappearance of a single species can stabilize an ecosystem or send it spinning into chaos, depending on (you guessed it) the network of interactions that surrounds that species. This extensive and subtle connectivity, which gives meaning to the behavior of the underlying components, turns out to be a consistent feature of living systems.
The recurrent evolution of these networks of weak interactions suggests that they may allow biological systems to incorporate information from the environment while also maintaining stability in the face of constant perturbation. This general feature of living systems also has clear methodological consequences for modern biology. Once this gossamer web is taken apart in search of the smallest components we can study, the process of putting it back together bears no resemblance at all to reconstructing a clock. Thus we find ourselves, early in the 21st century, with extraordinarily detailed descriptions of the components of many biological systems. But reconstructing those systems is proving to be a monumental and consistently surprising enterprise.
The promise of reductionism rested on the belief that an intelligent dissection of complex phenomena would not only yield progress, but would eventually reduce any problem to its component parts. Complexity, we naively hoped, was simply a by-product of incomplete understanding, an illusion that would fall away once the parts were fully understood. But this is the dirty little secret of contemporary biology: Despite our reductionist successes, the central conceptual problems of biology have not yielded to study. We have revealed the elegant workings of neurons in exquisite detail, but the material understanding of consciousness remains elusive. We have sequenced human genomes in their entirety, but the process that leads from a genome to an organism is still poorly understood. We have captured the intricacies of photosynthesis, and yet the consequences of rising carbon-dioxide levels for the future of the rain forests remain frustratingly hazy. We are, in short, the king’s horses and the king’s men: We stare at the pieces, knowing what Humpty should look like, but unable to put him together again.
How does one deal with such complexity? In living systems, Dorit argues, the component parts are always embedded in an intricate web of interactions. Such complexity appears at multiple levels of organisation as shown in the three illustrations below: a network of interactions among 1,700 different human proteins (top), a food web in a Puerto Rican rain forest (middle) and a neuronal network in a mouse brain (bottom).
Dorit continues:
the days when we could have blind faith in the power of reductionist deconstruction are over. Humpty lies in fragments. Fortunately for us, a new approach is taking shape to replace the seductive appeal of reductionism. We biologists may have bought a little too much of what Descartes had to sell, but as the limits of naive reductionism become more obvious, additional methods emerge for understanding the complexity of life. This new, interactionist perspective on living systems, with its emphasis on the interplay of parts, has benefited from new computational tools and experimental approaches. Whole new subfields in the life sciences, as well as productive interactions among existing disciplines, have emerged. Systems biologists, complexity theorists and newly minted biologists now attend as carefully to the ways in which parts come together as they do to the parts themselves. In the process, features of living systems that we once carelessly overlooked (or destroyed) in our haste to deconstruct now snap into focus. We are, for instance, beginning to understand that modularity and redundancy are inherent features of all levels of biological organization. These features characterize systems that are simultaneously resilient and capable of evolving.
