Open The Echo Chamber

Chris Albon copied me on a retweet today from World Concern that said:

A beautiful sight: things growing in #Somalia. This is what’s possible in the #HornofAfrica. twitpic.com/7c8y24

For those not inclined to click the link, it went to this picture:

I have mixed feelings about this tweet and this picture.  On one hand, it expresses what I am sure is genuine relief from an organization that is concerned with the well-being of people living in the Horn of Africa.  On the other hand, the phrase “this is what is possible” suggests that this does not usually happen . . . except, of course, now we are in the Dayr, the October to December rainy season.  Though the Dayr is the shortest rainy season in this part of the world, wet fields and new growth do in fact usually happen right about now.  Further, the phrase “things growing in Somalia” suggests that nothing was growing before.  This was not the case – things have been growing, even in famine-struck parts of southern Somalia.  Not enough has been growing in some places, and this shortage has been compounded by all sorts of political challenges that have created a widespread problem.  Finally, there is a bit of tone to this – as if we are out of the woods in the Horn.  Well, maybe – but it will be months until a real harvest comes in, and much longer than that before accountable governance and functioning markets return, so we have a ways to go.  And given that this famine was not caused by drought (the drought exacerbated other underlying factors), the fact that we are having trouble addressing those underlying factors means the next drought (and there will be another one relatively soon) may create a very similar set of circumstances and challenges.

In summary, I believe in hope.  That is why I call myself an optimist.  But at the same time, we have to be careful about conflating hope with triumph . . . which is why I call myself a hopelessly realistic optimist.

So, it seems I have been challenged/called out/what-have-you by the folks at Imagine There Is No . . . over what I would do (as opposed to critique) about development.  At least I think that is what is going on, given that I received this tweet from them:

@edwardrcarr what would You do with 1 Billion $ for #development? bit.ly/rQrUOd #The.1.Bill.$.Question

In general, I think this is a fair question.  Critique is nice, but at the end of the day I strive to build something from my critiques.  As I tell my grad students, I can train a monkey to take something apart – there isn’t much talent to that.  On the other hand, rebuilding something from whatever you just dismantled actually requires talent.  I admit to being a bit concerned about calling what I build “better”, mostly because such judgments gloss over the fact that any development intervention produces winners and losers, and therefore even a “better” intervention will probably not be better for someone.  I prefer to think about doing things differently, with an eye toward resolving some of the issues that I critique.

So, I will endeavor to answer – but first I must point out that asking someone what s/he would do for development with $1 billion is a very naive question.  I appreciate its spirit, but there isn’t much point to laying down a challenge that has little alignment with how the world works.  I think this is worth pointing out in light of the post on Imagine There Is No . . ., as they seem to be tweaking Bill Easterly for not having a good answer to their question.  However, for anyone who has ever worked for a development agency, the question “on what would you spend a billion dollars” comes off as a gotcha question because it is sort of nonsensical.  While the question might be phrased to make us think about an ideal world, those of us engaged in the doing of development who take its critique and rethinking seriously immediately start thinking about the sorts of things that would have to happen to make spending $1 billion possible and practical.  Those problems are legion . . . and pretty much any answer you give to the question is open to a lot of critique, either from a practical standpoint (great idea that is totally impractical) or from the critique side (and idea that is just replicating existing problems).  When caught in a no-win situation, the best option is not to answer at all.  Sure, we should imagine a perfect world (after all, according to A World Of Difference, I am “something of a radical thinker”), but we do not work in that world – and people live in the Global South right now, so anything we do necessarily must engage with the imperfections of the now even as we try to transcend them.

Given all of this, I offer the following important caveats to my answer:

1) I am presuming that I will receive this money as individual and not as part of any existing organization, as organizations have structures, mandates and histories that greatly shape what they can do.

2) I am presuming that I have my own organization, and that it already has sufficient staff to program $1 billion dollars – so a lot of contracting officers and lawyers are in place.  Spending money is a lot harder than you’d think.

3) I am presuming that I answer only to myself and the folks in the Global South.  Monitoring and evaluation are some of the biggest constraints on how we do development today.  As I said in my talk at SAIS a little while ago, it is all well and good to argue that development merely catalyzes change in complex systems, which makes its outcomes inherently unpredictable.  It is entirely another to program against that understanding – if the possible outcomes of a given intervention are hard to predict, how do you know which indicators to choose?  How can you build an evaluation system that allows you to capture unintended positive and negative outcomes as the project matures without looking like you are fudging the numbers?  This sounds like constrained thinking, but it is reality for anyone working in a big donor agency, and for all of the folks who implement the work of those agencies.

4) I am presuming there are enough qualified staff out there willing to quit what they are doing and come work for this project . . . and I am going to need a hell of a lot of staff.

5) I am presuming that I am expected to accomplish something in the relatively short term – i.e. 3-5 years, as well as trigger transformative changes in the Global South over the long haul.  If you don’t produce some results relatively soon, people will bail out on you.

All of these, except for 5), are giant caveats that basically divorce the question and its answer from reality.  I just need to point that out.  Because of these caveats, my answer here cannot be interpreted as a critique of my current employer, or indeed any other development organization – an answer that would also serve as a critique of those institutions would have to engage with their realities, blowing out a lot of my caveats above . . . sorry, but that’s reality, and it is really important to acknowledge the limits of any answer to such a loaded question.

So, here goes.  If I had $1 billion, I would spend it 1) figuring out what people really do to manage the challenges they face day-to-day, 2) identifying which of these activities are most effective at addressing those challenges and why, 3) evaluating whether any of these activities can be brought to scale or introduced to new places, and 4) bringing these ideas to scale.

Basically, I would spend $1 billion dollars on the argument “the new big idea is no more big ideas.”

Why would I do this, and do it this way?  Well, I believe that in a general way those of us working in development have very poor information about what is actually happening in the Global South, in the places where the challenges to human well-being are most acute.  We have a lot of assumptions about what is happening and why, but these are very often wrong.  I wrote a whole book making this point – rather convincingly, if some of the reviews are to be believed.  Because we don’t know what is happening, and our assumptions are wide of the mark, a lot of the interventions we design and implement are irrelevant (at best) or inappropriate (at worst) to the intended beneficiaries.  Basically, the claim (a la Sachs and the Millennium Villages Project) that there are proven development interventions is crap.  If we had known, proven interventions WE WOULD BE USING THEM.  To assume otherwise is to basically slander the bulk of people working on development as either insufficiently motivated (if we weren’t so damn lazy, and we really cared about poor people, we could fix all of the problems in the world with these proven interventions) or to argue that there simply needs to be more money spent on these interventions to fix everything (except in many cases there is little evidence that funding is the principal cause of project failure).  Of course, this is exactly what Sachs argues when asking for more support for the MVP, or when he is attacking anyone who dares critique the project.

The only way to really know what is happening is to get out there and talk to people.  When you do, what you find is that the folks we classify as the “global poor” are hardly helpless.  They are remarkably capable people who make livings under very difficult circumstances with very little resource and limited fallback options.  They know their environments, their economy, and their society far better than anyone from the outside ever will.  They are, in short, remarkable resources that should be treated as treasured repositories of human knowledge, not as a bunch of children who can’t work things out for themselves.  $1 billion would get us a lot of people in a lot of places doing a lot of learning . . . and this sort of thing can be programmed to run over 6 months to a year to run fieldwork, do some data analysis, and start producing tailored understandings of what works and why in different places . . . which then makes it relatively easy to start identifying opportunities for scale-up.  Actually, the scale-up could be done really easily, and could be very responsive to local needs, if we would just set up a means of letting communities speak to one another in a free and open manner – a network that let people in the Global South ask each other questions, and offer their answers and solutions, to one another.  Members of this project from the Global North, from the Universities and from development organizations, could work with communities to convey the lessons the project has gleaned from various activities in various places to help transfer ideas and technology in a manner that facilitates their productive introduction in new contexts.  So I suppose I would have to carve part of the $1 billion off for that network, but it would come in under the scale-up component of my project.  Eventually, I suspect this sort of network would also become a means of learning about what is happening in the Global South as well . . .

With any luck at all, by year 3 we would see the cross-fertilization of all kinds of locally-appropriate ideas and technology happening around the world and the establishment of a nascent network that could build on this momentum to yield even more information about what people are already doing, and what challenges they really face.  We would have started a process that has immediate impacts, but can work in tandem with the generational timescales of social change that are necessary to bring about major changes in any place.  We would have started a process that likely could not be stopped.  How it would play out is anyone’s guess . . . but it would sure look different than whatever we are doing now.

Yep, no sooner do I post on failure and how we account for it and learn from it, then I come upon a big fail of my own.  That I can learn from. Irony, anyone?

As many of you know, I have been working in Ghana since 1997.  I’ve spent some 20 months there, though it has been a while since I was last on the ground (I need to change that) – basically, the last meaningful research trip I took was in the summer of 2006.  That work, along with the fieldwork that came before it, was so rich that I am still working through what it all means – and it has led me down the path of a book about why development doesn’t work as we expect, and now a (much more academic) complete rethinking of the livelihoods framework that many in development use to assess how people make a living.

One of my big findings (at least according to some of my more senior colleagues) is that inequality and (depending on how you look at it) injustice are not accidental products of “bad information” or “false consciousness” in livelihoods strategies, but integral parts of how people make a living (article to this effect here, with related work here and here, as well as a long discussion in Delivering Development).  One constraint specific to the livelihoods in the villages in which I have been working is the need to balance the material needs of the household with the social requirement that men make more money than their wives.  I have rich empirical data demonstrating this to be true, and illustrating how it plays out in agricultural practice (which makes up about 65% of most household incomes).

In other words, I know damn well that men get very itchy about anything that allows women to become more productive, as this calls one of the two goals of existing livelihoods strategies into question.  Granted, I figured this out for the first time around 2007, and have only very recently (i.e. articles in review) been able to get at this systematically, but still, I knew this.

And I completely overlooked it when trying to implement the one village improvement project with which I have been involved.  Yep, I totally failed to apply my own lessons to myself.

What happened?  Well, to put it simply, I had some money available after the 2006 fieldwork for a village improvement project, which I wanted the residents of Dominase and Ponkrum to identify and, to the extent possible, design for themselves.  We had several community meetings that meandered (as they do) and generally seemed to reflect the dominant voices of men.  However, at the end of one of these meetings, one of my extraordinarily talented Ghanaian colleagues from the University of Cape Coast had the experience and the awareness to quietly wander off to a group of women and chat with them.  I noticed this but did not say anything.  A few minutes later, he strolled by, and as he did he said to me “we need to build a nursery.”  Kofi had managed to elicit the womens’ childcare needs, which were much more practical and actionable than any other plans we had heard.  At the next community meeting we raised this, and nobody objected – we just got into wrangling over details.  I left at the end of the field season, confident we could get this nursery built and staffed.

Five years later, nothing has happened.  They formed the earth blocks, but nobody cleared the agreed-upon area for the nursery.  It was never a question of money, and my colleagues at the University of Cape Coast checked in regularly.  Each time, they left with promises that something would get going, and nothing ever did.  I don’t fault the UCC team – the community needed to mobilize some labor so they would have buy-in for the project, and would take responsibility for the long-term maintenance of the structure. This is on the community – they just never built it.

And it wasn’t until yesterday, when talking about this with a colleague, that I suddenly realized why – childcare would lessen one demand on women that limits their agricultural productivity and incomes.  Thus, with a nursery in place women’s incomes would surely rise . . . and men have no interest in that, as this is not the sort of intervention that would drive a parallel increase in their own incomes.  I have very robust data that demonstrates that men move to control any increase in their wives incomes that might threaten the social order of the household, even if that decreases overall household income and access to food.

So why, oh why, did I ever think that men would allow this nursery to be built?  Of course they wouldn’t.

I can excuse myself between 2006-2008 for missing this, as I was still working through what was going on in these livelihoods.  But for the last three years I knew about this fundamental component of livelihoods, and how robust this aspect of livelihoods decision-making really is, even under conditions of change such as road construction.  I have been looking at how others misinterpret livelihoods and design/implement bad interventions for years, all the while doing that very thing myself.

Healer, heal thyself.

I am a big fan of the idea of admitting failure and trying to learn from it.  I like ambitious projects with potentially huge payoffs, but a lot of risk of failure – they’re just much more interesting than going at things incrementally.  Besides, if you are going to fail, why not fail spectacularly?  As I tell my grad students, if you are going to ride it all the way to the ground, you might as well dig a big hole when you get there.  At least people will notice the hole, and try to figure out what the hell you were up to . . . of course, I am an academic (with tenure), so I have a pretty big cushion to land on these days.

All that said, I wonder about the utility of these admitting failure efforts that I see coming from groups like Engineers without Borders.  I had the good fortune to catch up with Tom Murphy (or, as the twitterati know him, @viewfromthecave) the other day while he was here in DC, and we started talking about learning from failure.  In the course of our conversation, we came around to two key problems.  First, really admitting failure requires reframing the public image of development as an inherently do-no-harm effort, where just doing something is better than nothing.  Second, given this first problem, when we really start talking about what failure means, even in the most constructive of settings, we will call the entire development enterprise into question. How do we avoid throwing the baby out with the bathwater?

We have long allowed ourselves and our donor constituencies to believe that development work should never have bad outcomes – there is a pervasive belief (under challenge right now, at least by some) that, at worst, a failed project will not change anything – that is what development failure means. Of course, this is simply untrue – development efforts can make things much, much worse for people if they are poorly framed, designed, and implemented – a point I try to make in Delivering Development.  This has a lot to do with the very imagery of a helpless and oppressed global poor the aid world relies upon to raise funds.  When people see someone in a situation that difficult, they assume things could not get worse.  There is no discussion of what is working in the lives of the poor, and therefore the public has little sense that there are fragile things in peoples’ lives and livelihoods that should be protected as we bring new programs and projects to ground. As a result, development takes on the image of a low-risk enterprise in which social protection and “do no harm” safeguards are superfluous, as the worst we could do is leave people as they were.

Up against that worldview, admitting failure seems just fine – “hey, we didn’t really move the needle with that project, but we’ll figure out what we did wrong and try again” sounds much better than “we are incredibly sorry for utterly devastating the physical basis of your livelihoods and forcing many of you to abandon your farms because we ignored your existing land management practices.”  Unfortunately, admitting failure means a lot of the latter, and I am not at all convinced that anyone has the stomach to really wade into that.

This issue has to be combined with a concern for the scale of failure.  It is all well and good to admit failure, even ugly failures, at the project level – stuff happens.  A failed project can usually be traced to concrete causes that can then be addressed and remedied.  But how can a bilateral aid agency, or even a multilateral agency, do the same for its programs?  It is one thing for such huge organizations to talk about the failure of individual projects, and learn from them, but how can we talk about learning from entire programs that don’t live up to expectations without attracting serious challenges to the aid budget that end up wrecking even successful programs, or preventing the scale-up of things that we know work? Put another way, how can we create an environment where learning from our activities is truly possible, and balance that environment with the political reality of aid agencies and NGOs that answer to (different) constituencies that expect only good things to happen?

This framing of global poverty, and the persistent need to justify aid budgets, puts everyone involved with development on a terrible tightrope – at least for those of us interested in evidence-based programming and policy.  Just saying that admitting failure is good does not begin to get us to a world in which we can see that as more than a slogan.  We will have to unwind decades of public relations and fundraising practice, and back out of some very long-standing and pervasive views of global poverty, before we have any real hope of bringing real learning to the fore of development practice.

Or, we could just give everyone tenure . . .

Ah, that familiar refrain – a mix of love and derision provoked by the vagaries of life in my favorite West African country: the power cuts out randomly in the midst of a big soccer match, “Oh, Ghana!” The new road washes out because of inadequate culverts? “Oh, Ghana!” And now, the country’s economy grows 34% in the second quarter of 2011 – expanding the GDP by 3.4 percent in that quarter alone (h/t to Andy Sumner for pointing this out to me)?

Wait, isn’t that good news?

Well, on its face, yes – this surge in growth suggests there is a lot more money at play in Ghana, and that will hopefully result in new and better jobs, greater revenues for the state, and eventually better services for the population.  But there are two big caveats that really, really worry me here.

1. The growth was driven mostly by growth in the mining and quarrying sector – of which oil has about a 2/3 share. So the economy has grown, but it is still commodity-dependent.  Admittedly, they now have oil on top of cocoa and gold, but these don’t exactly track independently of one another.  Building your whole economy on three commodities is not a path to a stable, sustainable future.
2. Ghana does not seem to have a plan to spend all of this new revenue in a manner that will trigger the virtuous process I was describing above.  Without a plan, the possibility of misuse and redirection of funds into private accounts rises dramatically (h/t to Mark Weston).

Oh, Ghana!

Even the oddly good news – agricultural (economic) growth seems to be matching the growth of mining and quarrying – isn’t really that good.  At first glance, this news seems to suggest that ag production is increasing, or that more of that production is getting to market before spoiling, trends that would benefit much of the Ghanaian population.  Maybe not, though – Ghana’s light-crop cocoa crop doubled over the same period last year, suggesting this increase is largely pegged to cocoa.  Worse, a big chunk of this improvement is tied to good weather, which is difficult to gamble on year-to-year.

Oh, Ghana!

I was on a panel at the Organic Trade Association‘s research series at the Natural Products Expo East in Baltimore last Friday, discussing the issue of organic farming and the need to feed the world.  As I heard over and over from proponents of organic agriculture, the argument “you can’t feed the world on organic” is something thrown at them all the time.  As I argued, though, this is a production-based argument: that is, organic farming often has somewhat lower levels of productivity than industrial farming (though there are several cases where this does not seem to hold, and a number of confounding factors that make it entirely possible that the productivity difference is actually quite small).  Well, that would be a relevant argument if we were already using our food resources carefully.  Except we aren’t.  Consider:

• We still produce more than enough food globally to feed everyone a very healthy number of calories, and probably enough that those calories could be accompanied by adequate nutrients.  The current problems of food insecurity are primarily about distribution, not production.
• Anywhere between 20% and 40% of all food grown globally spoils before it reaches market.  The figures are lower for grains (which tend to travel well) and much higher for vegetables.
• In the US, we throw away roughly 30% of all food we purchase.
• Consider those two numbers together: In the US, we probably lose a lot less of the crop between farm and purchase at market, but then throw 30% of it away.  In other places, the food that reaches the table is nearly completely eaten, but we could lose up to 40% of that food before it reaches market.  In other words, no matter where you go on Earth, there is a hell of a lot of waste in the food system.
• Finally, consider that 33% of all farmland is used for animal feed, one of the less efficient ways of getting calories out of the environment.  It is unclear to me if this 33% includes biofuel crops, but in any case biofuels would only add a few percentage points to this at most.

In short, we have distribution problems and an astonishing amount of waste in our food systems, but it seems that a lot of the food security debate in policy circles is driven by production arguments.  Enhancing production is not a low hanging fruit.  Enhancing production is often used as an excuse for ignoring local knowledge and capacity in favor of reworking entire agroecological systems (which usually ends badly).  Those of us working in development would be well-served to consider all the ways we might address hunger, including waste and distribution, rather than focus myopically on one cause for what might be a phantom problem.  Welcome to another central theme of Delivering Development: misunderstanding/misidentifying the development challenge, and then trying to solve the wrong thing.

One caveat: there are places in the world in absolute production crises – that is, they lack market access to facilitate the movement of needed food, and their agricultural systems are no longer resilient in the face of current challenges.  In these places, waste may be less of an issue, and distribution solutions may be years in the future (good infrastructure and markets require good governance, which is no easy fix), and therefore the application of new agricultural technologies might become the low hanging fruit solution for the time being, until the other challenges can be met. It’s about finding the right tool for the job (and knowing exactly what the job is, too).

Over at the Guardian, Damian Carrington has a blog post arguing that “Food is the ultimate security need.”  He bases this argument on a map produced by risk analysts Maplecroft, which sounds quite rigorous:

The Maplecroft index [represented on the map], reviewed last year by the World Food Programme, uses 12 types of data to derive a measure of food risk that is based on the UN FAO’s concept. That covers the availability, access and stability of food supplies, as well as the nutritional and health status of populations.

I’m going to leave aside the question of whether we can or should be linking food security to conflict – Marc Bellemare is covering this issue in his research and has a nice short post up that you should be reading.  He also has a link to a longer technical paper where he interrogates this relationship…I am still wading through it, as it involves a somewhat frightening amount of math, but if you are statistically inclined, check it out.

Instead, I would like to quickly raise some questions about this index and the map that results. First, the construction of the index itself is opaque (I assume because it is seen as a proprietary product), so I have no idea what is actually in there.  Given the character of the map, though, it looks like it was constructed from national-level data.  If it was, it is not particularly useful – food insecurity is not only about the amount of food, but access to that food and entitlement to get access to the food, and these are things that tend to be determined locally.  You cannot aggregate entitlement at the national level and get a meaningful understanding of food insecurity – and certainly not actionable information.

Further, you can’t aggregate food markets or prices at the national level and get anything meaningful with regard to food security – let’s compare Maplecroft’s map with FEWS-NETs maps for the immediate future (August-September 2011):

First Maplecroft:

Now FEWS-NET:

While FEWS-NET does not have global coverage, compare their maps to those of Maplecroft and you see two things: One, FEWS is clearly working at a much finer geographic scale, because they have on-the-ground information about actual markets and access, as well as a deep understanding of climate and livelihoods through which to contextualize their grounded data.  This is what it takes to represent variable vulnerability within a country.  The variability you see on their map illustrates my point about the problems of national-level statistics – clearly food insecurity is a regional-to-local problem in every country, even Somalia.  Two, FEWS is not projecting major risk in the same places as Maplecroft, whose map has painted most of equatorial and dryland Africa as problematic at best.  Now, FEWS-NET’s medium-term projections (October-December 2011):

Again, no real resemblance to the Maplecroft map.

Now, you can argue that the Maplecroft map is aimed at a different goal than the FEWS-NET maps, as Maplecroft is trying to create a risk-assessment picture of food security in the region.  However, Maplecroft’s timescale is unclear (does it cover the next 6 months? 1 year? 5 years?), and its data is so over-aggregated as to be non-actionable.  You can’t build policy or programs from this, and I would argue that you can’t really assess the risk of food insecurity from the map or the underlying index either.  FEWS-NET’s maps are what actionable information looks like . . .

I appreciate the point Carrington is trying to make on his blog – food security is a really important issue.  But if we are to address the challenges of hunger and conflict, we need to build our understanding of the connection between them from meaningful data . . . and probably work from the outstanding material already available via FEWS-NET and others.

OK, a last thought on the development initiatives and markets thread: let’s leave the predictive markets thing aside for the moment, and get to what I think is a more serious question for development initiatives – do we use all the information we might to evaluate the likely impact of our programs?  I think a lot of folks misread the intent of my initial post – I was NOT suggesting we bet on mortality rates and other direct measures of project effectiveness.  That is something I could see as an academic exercise, but is way too morbid for my tastes, even in that setting.

But everyone who lunged in that direction seemed to miss the point that any major development initiative will, if it succeeds, have radiating impacts through different markets.  That is, a successful food security initiative will change harvest sizes of different crops, thereby influencing commodities markets.  A successful public health intervention might increase the size of the workforce, or its efficiency.  And so on.  My simple thought was that any fund investor worth his/her salt should be examining these initiatives and their expected outcomes to decide 1) if the initiative worked, what markets might be affected, how and when and 2) do they think the initiative will actually work.

If there is no movement around these initiatives, it seems to me that these two factors might be important – at the first step in this decision-making, investors might decide that in the event of a successful intervention, the markets affected might not be accessible or profitable, or the timeframe of any movement in the market might be so long as to make immediate response unnecessary.  Thus, we would see no market response to the announcement of an intervention.  At that point, it doesn’t matter if the intervention will work or not – that assessment never comes into the picture.

However, in at least some cases, I have to think that there are initiatives out there (in a world of rising food prices, I am a bit fixated on food security at the moment) that would affect significant markets, and not only at a national scale (where markets might be illiquid or otherwise inaccessible).  Take the case of cocoa and Cote d’Ivoire this past winter: the civil conflict in CIV cut off a significant amount of global supply, and futures markets got skittish over the further constriction of trade, driving cocoa prices upward.  This is a niche crop, heavily produced by only a few countries, but the price movement could have meant big dollars for a fund that correctly anticipated this trend.  Surely there are (or will be) food security initiatives that could similarly affect the overall supplies of and access to particular (perhaps niche) crops for entire regions, or even shift global availability/perception enough to shift commodities prices in much larger, more transparent markets in the short term. Don’t fixate on national markets for these initiatives – what about really big development movers that could affect global supplies of grain in an era where all the slack has been taken out of various global grain markets?  You can’t tell me that everyone at these trading desks is simply ignoring the food security world . . . surely they are at least assessing through step 1) above.  So if there is no market response to these initiatives, either the timeframe of movement is too distant to warrant interest, or the traders simply don’t think these initiatives will succeed enough to significantly influence the markets in which they trade.  Perhaps the price of oil and its impact on transport is much, much more important than increasing harvest size when it comes to shaping food commodities prices . . . in which case, it would probably be good for those designing food security initiatives to know this at the outset and address it in project design (for example by thinking about transportation issues as integral to the initiative).

Of course, there is option 3): traders have no idea what sorts of initiatives are out there, and are operating in ignorance of these potential large drivers.  This is entirely possible, but a bit hard to believe . . .

In a comment on my earlier post critiquing the recent ENSO and conflict piece that appeared in nature Nature , Joe pointed out that my argument that the authors of the piece did not understand livelihoods was not necessarily clear to the reader.  I think this is completely fair – I am buried in livelihoods . . . it is a concept at the core of what I have researched for the past 14 years, and therefore what may seem obvious to me is not so obvious to everyone else.

First, to clarify: I think the top-line issue I was shorthanding in my response to Solomon was the causal framework: it is totally unclear to me how they think environmental change is translated into conflict.  It is possible that they had no explicit notion of how this connection is made, but I think that would create an enormous set of problems for the study as it would make it impossible to know what variables to control for in the study (to some extent, I think this is a problem with the study anyway).  However, the study, and Solomon’s response, led me to believe that they did have a very basic framing of this connection, where weather impacts livelihoods which impacts behavior.  In this apparent framing, it seems to me that they treated livelihoods as a straightforward set of activities – and the impact of weather on those activities could be easily and generally understood, and the human outcomes of those impacts could also be easily and generally understood.  If this is true, it is a serious misunderstanding of livelihoods.

There is a lot of stuff I could say about livelihoods – my current intellectual project involves rethinking how we understand livelihoods, because I think current analytical frameworks cannot really engage with actual livelihoods decision-making on the ground.  As a result, a lot of our understandings of what people do, and why they do it, are wide of the mark, and the interventions we design to improve/augment/replace existing means of making a living in particular places are often misguided and prone to “surprise” outcomes.

First, a quick definition of livelihoods as they are treated in the contemporary literature: “the capabilities, assets (stores, resources, claims and access) and activities required for a means of living” (Chambers and Conway, 1992:7).  As Brent McCusker and I have argued:

this definition of livelihoods moves past income toward a more holistic consideration of the manner in which a person obtains a living. In practice, this definition has resulted in a number of approaches to livelihoods that focus closely on access to various types of assets drawn upon by individuals to make a living. These approaches tend to categorize these assets as one of five types of capital: natural, physical, human, financial and social. Land comes under natural capital, “the natural resource base (land, water, trees) that yields products utilized by human populations for their survival,” though an improved field might come under the heading of physical capital, which generally includes “assets brought into existence by economic production processes.”

My problem with the livelihoods approach that dominates the literature, and subtly undergirds the Nature piece I was critiquing, is not the broad definition of livelihoods.  Instead, the problem lies in the subtle assumption of this approach that, in its focus on the requirements for a means of living, concentrates on material circumstances and outcomes as a metric for the success and viability of particular livelihoods.  As I have demonstrated repeatedly (for example here, and in my book Delivering Development), livelihoods are double-edged: they are aimed at both meeting certain material requirements of life and maintaining the privileges of the powerful.  Above certain very, very low thresholds, the social goals of livelihoods actually trump the material goals.  Therefore, if we want to understand livelihoods decisions and outcomes, we must understand the social context at least as well as we do the material conditions in a particular place.  Using generalized assumptions about human motivations to explain responses to livelihoods shifts will smooth over really significant differences in decision-making, and therefore obscure any possible causal connection between things like environmental change and the incidence of conflict – material maximization/deprivation is only part of the story of human motivations, and a relatively small part at that.

How does this all relate to the Nature piece and my criticism? While the authors never specified the means by which this would happen in the piece, only offering general speculation in their response to my criticism, I found Solomon’s response to my blog post really telling:

The study is trying to understand whether choosing to engage in conflict is a “livelihood decision” that individuals in modern societies select more often when El Nino events occur. Our findings tells us that for some reason, people’s willingness to engage in organized violence changes when the global climate changes. One hypothesis is that perhaps “predation” (i.e. the forceful extraction of property from others) is a form of “adaptation” to climate changes.

It is possible that Solomon’s reference to conflict as a livelihoods decision was simply echoing the terms of my criticism.  However, both the article and his response seems to reflect an implicit framing of the environment-to-conflict connection as somehow passing through livelihoods in a straightforward manner.  Because the authors never actually unpack how the environment impacts livelihoods, and in turn how those impacts are translated into human impacts, they become guilty of the same issue that plagues nearly everyone using the livelihoods framework these days: they implicitly embrace an over-generalized framing of livelihoods decisions that relies too heavily on a relatively minor driver of decision-making (material conditions), and completely ignores the dominant factors that shape the character of particular activities and therefore result in particular outcomes for the well-being of those living under that strategy.  I am sure that predation does occur.  I am also absolutely certain that this is not a general response – it does not happen very often (plenty of empirical studies show other behaviors).  It is not interesting to know that it occurs – we already know that.  What is interesting and important is why it occurs.  Going for “story time” explanations of complex behavior does not contribute to our understanding of human behavior, or the impact of climate change on human well-being.

I am working on a reframing of livelihoods that elevates the social component to its proper place in livelihoods decision-making (in review at the Journal of Development Studies).  The thinking behind this reframing is intensely theoretical and really, really academic (for a taste of what I mean, see this piece I wrote with Brent).  My goal in the forthcoming piece is to take this really esoteric theory and turn it into an approach that can be understood and employed widely.  With any luck it will be accepted and published relatively soon . . . I will put up a pre-print as soon as I am able.  But even with this reframing, we are going to have to work really hard at understanding when large-scale studies such as the one I have been critiquing are appropriate for furthering our understanding of things we really need to know, when they merely illustrate what we already know, and when they present really problematic findings with a misleading level of certainty.

So, how do we fix the way we think about development to address the challenges of global environmental change?  Well, there are myriad answers, but in this post I propose two – we have to find ways of evaluating the impact of our current projects such that those lessons are applicable to other projects that are implemented in different places and at various points in the future . . . and we have to better evaluate just where things will be in the future as we think about the desired outcomes of development interventions.

To achieve the first of these two is relatively easy, at least conceptually: we need to fully link up the RCT4D crowd with the qualitative research/social theory crowd.  We need teams of people that can bring the randomista obsession with sampling frames and serious statistical tools – in other words, a deep appreciation for rigor in data collection – and connect it to the qualitative social theoretical emphasis on understanding causality by interrogating underlying social process – in other words, a deep appreciation for rigor in data interpretation.  Such teams work to cover the weaknesses of their members, and could bring us new and very exciting insights into development interventions and social process.

Of course, everyone says we need mixed methodologies in development (and a lot of other fields of inquiry), but we rarely see projects that take this on in a serious way.  In part, this is because very few people are trained in mixed methods – they are either very good at qualitative methods and interpretation, or very good at sampling and quantitative data analysis.  Typically, when a team gets together with these different skills, one set of skills or the other predominates (in policy circles, quant wins every time).  To see truly mixed methodologies, this cannot happen – as soon as one trumps the other, the value of the mixing declines precipitously.

For example, you need qualitative researchers to frame the initial RCT – an RCT framed around implicit, unacknowledged assumptions about society is unlikely to “work” – or to capture the various ways in which an intervention works.  At the same time, the randomista skill of setting up a sampling frame and obtaining meaningful large-scale data sets requires attention to how one frames the question, and where the RCT is to be run . . . which impose important constraints on the otherwise unfettered framings of social process coming from the qualitative side, framings that might not really be testable in a manner that can be widely understood by the policy community.  Then you need to loop back to the qualitative folks to interpret the results of the initial RCT – to move past whether or not something worked to the consideration of the various ways in which it did and did not work, and a careful consideration of WHY it worked.  Finally, these interpretations can be framed and tested by the qualitative members of the team, starting an iterative interpretive process that blends qualitative and quantitative analysis and interpretation to rigorously deepen our understanding of how development works (or does not work).

The process I have just described will require teams of grownups with enough self-confidence to accept criticism and to revise their ideas and interpretations in the face of evidence of varying sorts.  As soon as one side of this mixed method team starts denigrating the other, or the concerns of one side start trumping those of the other, the value of this mixing drops off – qualitative team members become fig leaves for “story time” analyses, or quantitative researchers become fig leaves for weak sampling strategies or overreaching interpretations of the data.  This can be done, but it will require team leaders with special skill sets – with experience in both worlds, and respect for both types of research.  There are not many of these around, but they are around.

Where are these people now?  Well, interestingly the answer to this question leads me to the second answer for how development might better answer the challenges of global environmental change: development needs to better link itself with the global environmental change community.  Despite titles that might suggest otherwise (UNEP’s Fourth Global Environment Outlook was titled Environment for Development), there is relatively little interplay between these communities right now.  Sure, development folks say the right things about sustainability and climate change these days, but they are rarely engaging the community that has been addressing these and many other challenges for decades.  At the same time, the global environmental change community has a weak connection to development, making their claims about the future human impacts of things like climate change often wildly inaccurate, as they assume current conditions will persist into the future (or they assume equally unrealistic improvements in future human conditions).

Development needs to hang out with the scenario builders of the global environmental change community to better understand the world we are trying to influence twenty years hence – the spot to which we are delivering the pass, to take up a metaphor from an earlier post on this topic.  We need to get with the biophysical scientists who can tell us about the challenges and opportunities the expect to see two or more decades hence.  And we need to find the various teams that are already integrating biophysical scientists and social scientists to address these challenges – the leaders already have to speak quant and qual, science and humanities, to succeed at their current jobs.  The members of these teams have already started to learn to respect their colleagues’ skills, and to better explain what they know to colleagues who may not come at the world with the same framings, data or interpretations.  They are not perfect, by any stretch (I voice some of my concerns in Delivering Development), but they are great models to go on.

Meanwhile, several of my colleagues and I are working on training a new generation of interdisciplinary scholars with this skill set.  All of my current Ph.D. students have taken courses in qualitative methods, and have conducted qualitative fieldwork . . . but they also have taken courses on statistics and biogeographic modeling.  They will not be statisticians or modelers, but now they know what those tools can and cannot do – and therefore how they can engage with them.  The first of this crew are finishing their degrees soon . . . the future is now.  And that gives me reason to be realistically optimistic about things . . .