Fragile Harvest

by Misha Lepetic

We never fully grasp the import of any true statement until we have
a clear notion of what the opposite untrue statement would be.

~William James

I recently went to a food security panel held by Columbia University’s Earth Institute. There was, as is customary with the Earth Institute, both much to celebrate – such as a nuanced understanding of climate change and its responses – and bemoan – such as the casual dismissal of organic agriculture and a whole-hearted endorsement of GMO crops. While the subject was the very real phenomenon of food insecurity of the developing world, I became curious about how our perceptions of food security in the developed world – and what constitutes a desirable food system – inform our views of the developing world. One small detail from the afternoon, concerning a specific kind of fragility, was especially striking, and forms a convenient basis for the following critique of certain institutional worldviews.

How quickly can a system crumble under pressure? The idea of fragility in food systems can be characterized in many ways, such as crop vulnerability to weather shocks, or falling yields due to environmental degradation or ever-more resistant pests. Fragility can also be more formally defined as the way in which a system is – oftentimes endogenously – vulnerable to disruption or outright breakdown, as defined by Charles Perrow’s important work on complex technological systems. But it was economic fragility that was the focus of the following chart, shared by the Earth Institute’s Jessica Fanzo, from FAO’s “The State of Food Insecurity in the World 2011” (p14):


Although this is only the lowest quintile of the population for each country, it makes clear the extent to which financial fragility is a determinant of food security. Any increase in food prices requires a significant additional portion of a family’s income in these countries, if they are to maintain the same level of caloric intake, let alone nutrition. More frequently, families are not able to spend more money on food, and must employ other strategies to make ends meet: fewer meals; less caloric or nutritional value in each meal; the reallocation of meals away from members who are not income earners; taking children out of schools when a family must choose between education and food (i.e., as a result of school fees); the preferencing of employment for children over education, etc.

This kind of anxiety is inherently difficult for people in the U.S. to envision. Just how foreign is the concept of food security is to us? Simply put, the United States’ citizens spend less of their income per capita on food than any other nation. Setting aside the arguments that have been amply made elsewhere about the fact that we as end-consumers do not pay for the “true cost” of food, there are three things worth pointing out: how little we pay; the consistency with which we pay so little; and the further invariability of how we spend our food money.

The first point is illustrated by the USDA’s research that, for every year this century, our food costs as a proportion of income have come in under 10% of total income. Not only that, but during this time that number has consistently ranged from 9.5% to 9.9%. In fact, from a superficial point of view, neither of the two recessions during the past 11 years, nor stagnation of real income, have moved the needle very much. This is altogether remarkable. Moreover, we have seen a steady decline since the post-War peak of 23.5% in 1947, so this is a trend that has decades of socio-economic and technological momentum behind it.

Third, and most interestingly, is how we choose to eat. In 2011, eating at home took up 5.7% of our income, while eating out constituted 4.1%. In fact, the same USDA surveys inform us that during the last decade or so, we have consistently averaged a 60-40 split between eating at home and eating out. This too is remarkable, since the received wisdom dictates that eating out is one of the first activities attenuated by a recession. Perhaps we cut out more big-ticket meals for cheaper but more frequent ones, but the ratio essentially stays the same.

These three indicators imply that the system for growing, processing and distributing food in the U.S. has been supremely successful, if only because it has smoothed out any shocks that would suddenly raise prices for consumers (or depress prices for producers). This is especially impressive when one considers that price spikes have consequences, such as the Arab Spring.

Obviously, the Earth Institute, like anyone else, would love to see the poor of Tajikistan or Ghana progress towards a world where their food expenses are both a smaller fraction of their income, and much more resistant to disruption (we can hold off on getting them to Applebee’s for a while yet). However, is reproducing the food security that we seemingly enjoy in this country at all viable – and is our own food security even all that realistic?


To understand this, we ought to look beyond price stability, which is really only symptomatic of the underlying system, and examine the infrastructure that makes this stability possible. Our food systems are in fact optimized and stretched to the breaking point. Recall Charles Perrow, who characterized complex systems by analyzing how “tightly coupled” they were, among other traits. To paraphrase,

The sub-components of a tightly coupled system have prompt and major impacts on each other. If what happens in one part has little impact on another part, or if everything happens slowly (in particular, slowly on the scale of human thinking times), the system is not described as “tightly coupled.” Tight coupling also raises the odds that operator intervention will make things worse, since the true nature of the problem may well not be understood correctly.

This is an excellent characterization of our current food system. As a simple example, consider the trajectory of a salmonella outbreak, where the ground meat from a single tainted animal can then be intermingled with hundreds of others. The consolidation and verticalization of our food system amplifies the ease with which tainted food is spread. Moreover, the difficulty of tracing tainted food back through the supply chain leads to even greater volumes of food being recalled. It should also be noted that the frequency of food recalls has been rising for the last decade.

Another example of tight coupling in our food system is the just-in-time nature of our distribution systems. Most of our cities have only a few days’ worth of groceries available in stores; a complex distribution network is constantly replenishing the needs of an ever-more urbanizing nation, boutique items such as urban agriculture notwithstanding. Disruptions such as Hurricane Sandy, which brought a one-two punch of taking down both transportation and electrical systems – that is, both delivery and refrigeration – highlight the thin ice on which we skate when we put a premium on “freshness” to the detriment of resilience.


Even more compelling is our reliance on monocultures that increasingly bend our food production system towards a state of brittleness. Nowhere is this more in evidence than in our excessively enthusiastic use of GMO crops, which I will discuss below. But suffice to say that the years of R&D necessary to come up with drought-resistant strains of cereal crops put all our eggs in a basket, and expose our staples to the catastrophic emergence of so-called “superweeds” and “superpests.” In this way we are engineering our food system to rapidly converge onto a linear system that is, by definition, bottlenecked by a series of single points of failure.

There are other factors that indirectly support and ensure the success of food price stability, some of which are under differing degrees of pressure, and some of which are thriving. These include an extensive highway and railroad system; cheap fossil fuels that undergird our transportation system and fertilizer production; and a rich set of subsidies that have been enjoyed by farmers for decades. Add to this steady technological progress, including GMOs; advanced financial markets that smooth price fluctuations via futures contracts; and the communications and IT infrastructures that nearly erase market information costs, and we can see that there is a tremendous amount of context that goes into ensuring food price stability.

Clearly, the extraordinary food price stability we see in the United States is really a red herring. More ominously, if my speculation about systemic brittleness are borne out, this stability is also a lagging indicator – not dissimilar to the stock price of a company whose sales are still robust, but whose R&D pipeline has been exhausted. It is difficult to conceive that poor countries can approach greater levels of food price stability through the aping of only certain parts of this model, and yet that is precisely the panacea being championed by the Earth Institute.


Recall the first chart, which showed the portion of income spent on food by the poorest population quintile by income, versus the food price stability I have just discussed. The implied macro-level thinking here inspires us to compare the two and state, “Yes, these are the outcomes we want to achieve.” But the outcomes we choose to privilege will determine the direction of our efforts. Let’s choose carefully.

One of the things I like about the Earth Institute is its emphasis on the importance of data, and the use of that data to drive policy. However, data is intrinsically backward-looking, and the evidence-driven attitude can become fundamentally conservative to a fault. For example, if your primary outcome is to successfully feed lots of people, it is logical to favor GMO crops (as we do in the United States). If you are only measuring increases in productivity, ceteris paribus, you won’t necessarily be calculating the probability of a pest evolving the ability to do an end-run around your sprays, thereby setting up the risk of some genetic Maginot Line.

But we don’t need to look to possible future consequences to find trouble. It is not so much the productivity of GMOs that is at issue, but their dependence on a restrictive, indeed almost monopolistic, intellectual property regime. The below chart should indicate the extent to which a single company controls the technology that has ensured, on the supply side, cheap food enjoyed by millions of Americans (and their livestock).

Monsanto-share-gm_h328To drive home the intensity of Monsanto’s monopoly, it should be mentioned that while this 2004 chart represents only GMO-planted acreage, currently 90% of US soybean planting is GMO. Monsanto also flexes its intellectual property rights: as of 2010, the company has “filed 136 lawsuits against farmers for alleged violations of its Technology Agreement and/or its patents on genetically engineered seeds…in 27 states… Sums awarded to Monsanto in 70 recorded judgments against farmers totaled $23,345,820.99.” Thus food price stability, in the form of GMO crops, is provided  courtesy of a monopolist unhesitant to sue its own customers into bankruptcy. Despite all the attendant criticisms, please contrast this with the original open-source ethos of Norman Borlaug and the hybrids that precipitated the Green Revolution of the 1970s, and it is easy to conclude that we are generating fewer options for ourselves, not more.

Beyond its endorsement of GMOs, another troubling aspect of Earth Institute thinking came up during the Q&A, when the panel took my own question. I wanted to know about the impact of long-term leases on a country’s agricultural productivity, a relatively recent trend. Following the 2008 food price spike, rich countries with a dearth of agricultural land have sought to secure basic food supplies. By 2011, South Korea, Saudi Arabia and others have pioneered 80 million hectares’ worth of deals, and Africa, due in no small part to its weak governance, has become the favored site for these so-called “land grabs.” What are the implications of this?

Earlier, one of the panelists had quite correctly framed the central question of food security as, “How well can the market, on its own, ensure food security, and what is the right way for governments to intervene?” The phenomenon of land grabs implies that the market is, in fact, handling the issue of food security – by fundamentally transforming the relationships between the market participants themselves. In the past, governments collaborated with development agencies and other NGOs to improve crop strains. Producers sold on the open market, and prices were set in a fairly transparent fashion. Futures markets – first developed by Japanese rice traders in the 18th century– smoothed commodity prices.


Of course, this is not a free market by any stretch – subsidies provided to both farmers and consumers continue to distort prices today. Governments set export quotas that pay more attention to a nation’s balance of payments, or make deals with other governments for their natural resources. But the leasing trend has entirely rearranged these relationships, leading to vast territories that are virtually extrajudicial in nature; do not generate promised employment; and certainly do not contribute to a country’s GDP, coffers or technological progress – let alone its own food security. Even the Economist isn’t optimistic about this state of affairs.

Thus I was surprised at the rosy response that “we have been trying to get the private sector interested in agricultural partnerships for a long time, and now we have it.” Why yes, yes you do. Except that the private sector has no interest in you, Earth Institute, excepting perhaps the free IP they can cadge from you along the way. Governments that sign away their land are also no longer interested, or even empowered, to collaborate with development agencies. This is because the lessee brings in all the manpower, seed and infrastructure needed to grow their crops. In this sense, the old models of development no longer apply, and the Earth Institute and its peers may find themselves in the unenviable position of having to rethink their place in the development ecosystem.

In the meantime, this is resource extraction writ large, and anyone looking for real consequences need look no further than Madagascar:

When Madagascan President Marc Ravalomanana was ousted in March 2009, a major factor in his new unpopularity was the deal he had made with South Korea’s Daewoo corporation, effectively handing over 1 million hectares of Madagascar’s best agricultural land. The firm planned to use a South African workforce to grow 5 million metric tons of corn a year in addition to cultivating 120,000 hectares of palm oil, mainly for export to land-poor Korea to ensure its food security. With two-thirds of the Madagascan population living below the poverty line of less than US$ 0.64 a day and more than 500,000 relying on food aid, the deal provoked outrage from the international community. The 99-year land lease was quickly repealed by incoming Madagascan President Andry Rajoelina.

However, the really compelling bit will happen when the two threads I cite above become entwined – that is, when companies like Monsanto begin selling GMO technologies for planting millions of hectares that virtually no longer belong to a country. Indeed, will host governments have any say when GMO seeds are used to fuel the monocultures that naturally accompany such economies of scale? Will Monsanto, et al. enforce the integrity of their intellectual property against other farmers of the host countries with equal gusto as against farmers in the United States, for example when GMO seed spills across the boundaries of leased land, as it inevitably will? And, to return to our first chart, will these arrangements have any effect on the ability of the most vulnerable, in the poorest countries, to better afford the food they need? Outcomes are indeed something that ought to be chosen carefully.

There is a lot to like about the Earth Institute. A sprawling enterprise whose mandate is to cultivate basic and applied research to inform public policy and generate awareness about the great issues of our time must be on to something. And yet, emerging from Lerner Hall on a chilly November day, I wanted to caution it and other development institutions playing by yesterday’s rules: careful what you wish for, you might just get it.