City Region Food Systems - Part IIIA – Scale and Production Strategy

My analysis of this derives from my thinking over the last twenty years as well as engagement in a broad range of food system localization efforts. Early in the noughts I gave a conference plenary talk and made the following statement:

“I’d like to live in a food system in which I know where a significant percentage of my food comes from, not necessarily all of it … I’d like to know that the production, processing, distribution, and waste were done in an environmentally sensitive manner. I’d like to know that the democratic principles upon which this nation (U.S.) was founded are made stronger and not weakened through consolidation and monopolization. I’d like to know that the farmers who grow our food are honored as heroes and not marginalized as commodity producers. I would like to know that every person and consumer working in the food system has the opportunity to reach their potential and is not limited by less than living-wage jobs, poor nutrition, and substandard education. I would like a food system in which food is a right and working honestly is a responsibility.”

That still resonates with me and is the starting point for much of my thinking.  It is also at odds with the notion that the only way to ‘feed the world’ is by large scale, conventional, commodity-driven agriculture.  It is also at odds with the notion that we can continue consuming an average U.S. diet that is so at odds with eating patterns that are both healthier for people and the environment. 

Yet there are many who think our current production and eating patterns are not just good, but necessary.  Jason Lusk, in his recent book The Food Police: A Well Fed Manifesto About the Politics of Your Plate captures the zeitgeist of this position quite well.  He fundamentally argues that those in the supply chain understand the problems in the food system and the rest should just get out of the way.  As he says, “… We seem to have forgotten the achievements of our efficient and safe food system.  It is not a perfect system but it is one that has led to the greatest prosperity ever witnessed in human history and has allowed us to feed the world… Let’s get out of their way. “[1] Below I will explore three key fallacies in this argument.

The latest U.S. Dietary Guidelines Advisory Committee decided to incorporate ‘environmental sustainability’ into the 2015 Dietary Guidelines and included a category of ‘sustainability’ for public comments. Any reading of the public comments reveals two things about various animal commodity groups positioning on this topic.  First, agricultural production industries generally don’t think that dietary guidance and sustainability should be linked.  As stated by the North American Meat Institute,  “to recommend a ‘sustainable dietary pattern’ when the very definition of sustainable is still at issue does a disservice to the importance of affordability, nutrient adequacy, safety and cultural relevance to the foods consumed by U.S. population” (12/30/14 written testimony).  In other words, nothing should be done in this regard because we don’t have precise agreement on definitions.  Second, they typically concern themselves with environmental indicator improvements (e.g. carbon footprint) within their industry over time (I’ll call this intra-industry concerns).  As stated in written testimony from the Beef Checkoff Program, “beef’s overall sustainability has improved 5% in six years and the overall environmental and social fingerprint of the beef industry has been reduced by 7% in this time.”  In none of the testimony from the beef industry is there any recognition of the relative environmental footprint of beef compared to other parts of the food plate or other meats.  Other testimony from commodity group representatives similarly tends to look at change over time within their own industry and ignores comparisons with other, substitutable foods.  In other words, they don’t concern themselves with the relationship of their particular commodity relative to the overall food plate – it is apparent from reading the comments that any individual commodity group is incapable of looking outside itself and considering the ‘common good’ as opposed to its ‘commodity good’.  For more on this please see blogs by myself and Sam Lee-Gammage.  In addition, as I write this blog the U.S. House of Representatives has included a policy rider in the USDA funding bill that prevents the inclusion of sustainability guidelines in the U.S. Dietary Guidelines when released later this year.

So fallacy number 1 is that producers or producer-representative groups will consider the impact of their industry in relationship to other possible food choices to their potential detriment. It is not in the interest of the beef industry, for example, to argue that beef either per pound or per calorie has a relatively high environmental footprint (measured as GHG emissions or land use).  Does this mean that everyone should quit eating beef? Not necessarily.  Does this mean that the beef industry should continually increase production as population increases to maintain an average U.S. per capita annual intake of about 28 kgs?  No.  I think there is a role for ruminant meats both in the landscape of America and on the food plate of American households (if they so choose).  However, it does mean that if we are truly to feed America and ‘feed the world’ now and in the future that the nature of the food plate, the level of consumption, needs to evolve more quickly and more transformationally. 

Table 2 illustrates what would happen if U.S. beef consumption were kept constant (28 kgs/person) as population grows as well as what would happen if production (assuming for the sake of argument that none is imported) were kept constant during the same period.  Keeping consumption constant would be equivalent to about 2 and 9 million additional beef cattle raised annually (2020 & 2050 respectively) while keeping production constant would result in a 20% consumption decrease by 2050.   Land and water use data would drive us to the conclusion that if the U.S. is to truly participate in ‘feeding the world’ then we should decrease consumption, on average, more than this.

Fallacy number two is the disparagement of other production strategies other than ‘conventional’ (organic production and pasture-based animal production in particular).  The primary argument is they are less productive than the current, dominant paradigm.  It is true - at this moment organic production systems often do not yield at the same levels as conventional systems. This is the wrong way to approach the question.  We should ask - is the playing field of comparison level? If it were would we have the same analysis?  Lusk and others have argued that research, especially land grant university research, is important to farm productivity and yield (see for e.g. Huffman and Evenson).  If so then there is a major research gap- for e.g., the resources expended on researching cropping systems as agroecosystems, the internalization of nutrient management, lowest-input strategies of production and organic systems specifically is small by comparison to optimization of conventional systems with a range of external inputs. Such a disproportionate comparison should not be the basis of critique in the absence of context.  Rather, it should be the basis for shifting research resources in a more cropping systems oriented direction that develops generalizable and regionally-specific strategies for lowest-input and resource conserving production strategies.  It is certainly not a basis for critical rejection.

On the animal side let’s consider dairy production[2].  Whether production is calculated in pounds of milk per cow (typical consideration) or pounds of milk per acre (maybe more relevant in a period when land and water will probably be the limiting factors) there is no question that a grain-based system is more productive than a grass-based system.  However, consider that U.S. land-grant researchers (and others) have been working for decades to maximize production per cow in a grain-based system.  Everything from animal genetics and animal dietary needs to corn and soybean yield maximization has been researched.  It is only more recently that U.S. land-grant universities have put research funds into pasture-based systems in any sort of concerted fashion. Even now these are scattered and typically less well funded with less scientist-years committed to the research. However, research programs in places like Michigan, North Carolina, and Wisconsin are exploring topics such as intensive pasture management, pasture plant mixes and cow genetics to great effect.  They will never reach the milk levels per cow but may well reach the milk levels per acre of grain-based systems. 

In an emerging era of constraints on land for production maybe this should be the denominator in our calculations.  What if county fairs gave ribbons and awards for the farmers with the most productivity per total acres of inputs?  So fallacy number two is, in part, extremely dependent on research inputs and, in part, on farmer experiences – it is dependent on a critical mass of participants engaging in these systems.  A more level playing field at least on the U.S. land-grant research side might tell a different story. 

Also, it would be instructive to bring other indicators of success into the picture.  Environmental costs that are largely externalized from market costs at this point could be included as indicators of value – such as minimization/elimination of soil erosion, storage of atmospheric carbon, reduction of surface water run-off, better soil water holding capacity and increased aquifer recharge, reduced nutrient flow to surface waters and reduced pesticide use or contamination of groundwater.  For e.g. in 1997 about 1.9 pounds of pesticide were applied to an acre of soybeans, 2.9 pounds per acre for corn, and 0.08 pounds per acre of pasture (the figures are for Michigan from the national pesticide database).  If we were to look at production per unit of pesticide applied or per unit of total external inputs, for example, pasture-based systems would probably look strikingly ‘better’ right now.  Hence, from a standpoint of minimizing external inputs even in the absence of comparable levels of system optimization research, it may well be that pasture-based ruminant systems perform better – depending on the comparison criteria.

Another way to look at this may be from productivity over time.  Recognizing that climate change implies greater weather variability it might be useful to look for more predictable production systems to supply food in a century of greater weather unpredictability.  The long-term farming systems trial at the Rodale Institute is illustrative.  Over a twenty-year period the organic system was somewhat less productive (yield/acre) than the conventional system in wet years and somewhat more productive in dry years.  On average the organic cropping system demonstrated greater reliability.  “Organic crop management techniques will be a valuable resource in an era of climatic variability, providing soil and crop characteristics that can better buffer environmental extremes.”[3] Wouldn’t it be better to develop production systems that are more predictable rather than less, especially in periods of greater weather uncertainty and water challenges?

Let’s consider organic vs. conventional vegetable production.  Every U.S. land grant university has research programs to aid their state’s farming population.  Resources are largely devoted to maximizing crop harvest in conventional production.  A large part of this research portfolio is in the realm of basic plant physiology and biochemistry – and somewhat system neutral.  While the U.S. Land Grant Universities are getting better at funding organic production research there is still relatively little research in production systems and optimization of organic vegetable production.  The most recent meta-analysis of organic vs. conventional systems illustrates organic to be about 25-30% lower in vegetable yields.  What would happen with concerted research in organic pest and disease control in more humid environments?  How about strategies to localize the acquisition of plant nutrients– for e.g. human waste recycling?  If research has been important to agricultural development then it is also true that we won’t know the relative merits of different types of production systems unless resources are put into research.  It may also be true that in a less meat-intensive diet the reduced land requirements allow greater tradeoffs between crop yield, planetary boundary indicators, and a variety of social attributes.  One thing is for sure – making these types of comparisons at this point AND using them as proof that organic can’t compete on a production per acre basis may be a useful political argument but is a false and unscientific analysis. 

My gut feeling is that in the near term we will find the most sustainable and resilient fruit and vegetable production systems in higher humid environments (for e.g. east of the Mississippi River in the U.S.) will be hybrids of the organic and conventional systems we see today.  Milk production levels in a number of organic dairies indicate that they probably use about 25-40% grain overall- they are hybrid systems with a heavy reliance on pasture and hay but with grain used as well. 

The third fallacy in many arguments from the ‘large and conventional farms should supply the food’ is that scale and efficiency are, in essence, infinitely linked and that a farm must be very large to be efficient.  The corollary to this is that production exists where it is because of the comparative advantage of their location due to climate or other attributes.   

A mantra in most arenas of agricultural production is the notion of scale economy – that is per unit product cost decreases as the production scale increases.  We typically don’t stop and think that this can’t be an infinite relationship – at some point the curve must at best asymptote.  Mike Duffy of Iowa State University published interesting research in which he identified peak efficiency (cost/bushel) in soybean production at 160-325 hectares and in corn production at 325-490 hectares – at larger scales production efficiency leveled off or even declined somewhat.  Efficiency of scale is not infinite and probably peaks at scales lower than imagined.

Location of production is often wrapped up in this efficiency argument under the heading of ‘comparative advantage’. We might agree that many small farms are inefficient (production per unit labor and/or cost per unit production) compared to many large farms.  This does not in and of itself mean that smaller farms can’t approach the efficiency of large farms with access to the appropriate tools nor does it imply that only size and comparative advantage locations are all that matter.

There are a number of farmers in my home area raising a diverse array of vegetables.  As I watch their farms I ask myself – will they be able to do this at a scale providing a good livelihood with retirement when they are 60? My response (in my head, to myself, and as someone who passed 60 a while back) is ‘probably no’.  Why not?   By and large most of them are heavily dependent on hand labor for key production processes - especially harvesting of crops like greens or carrots.  By and large the equipment manufacturers and land grant bioengineering departments have ignored the opportunities in equipment for very small and smaller-scale producers.  There are some in the walk-behind tractor category but very limited options -beyond weed tillage - in the full-size tractor category.  This is especially true for harvesting.  Of course these farms are less efficient compared to those using equipment like multi-row mechanical harvesters for salad greens in California.  This presents an opportunity for entrepreneurs and inventors to develop reasonable-cost equipment for the small-scale producer.  Production on 0.5-1 hectare scale might include a combination of small tractor implements for tillage and walk-behind implements for weeding and some harvesting (root crops and greens for e.g.).  It would then scale such that farmers could imagine, and develop a business plan for, producing 6-8 hectares of produce at reasonable cost and with reasonable inputs of labor – hence a system that becomes more efficient while enabling the incorporation of a range of environmental attributes and at a scale appropriate for whatever communities they are supplying.  The case study of Wisconsin’s Harmony Valley Farm, cultivating 32 hectares, is illustrative.   It might also be true, at least I think it’s worth exploring, that production at this scale will be more amenable to renewable energy adoption for equipment and hence become much more environmentally efficient.

Why is this important?  For today’s small scale farmers to grow, to produce food more efficiently and thus be able to offer it at a price that is broadly available to consumers while making a decent living they need to be at a reasonable scale.  At very small scales there are tradeoffs that must be made – if sufficient income is to come from very small acreage to support a household the unit prices will necessarily be high and that means direct sales to relatively high end consumers (i.e. those with sufficient disposable income to pay a premium and/or restaurants with a similar clientele).  At this scale it is hard, if not impossible to simultaneously generate a livable income AND provide food to consumers across the economic spectrum.  The livability half of this is buttressed by recent research from the USDA indicating that farms less than 4 hectares must gross about $100,000 per year to be viable and profitable - very difficult on a 1 hectare farm without a significant infrastructure for year round production (in my area of 42^o north and the center of North America at least) and a strong customer base of higher end markets. In Michigan we have some fine young farmers developing extensive unheated hoop house infrastructure and selling year-round.  Yet, the harvest of most crops still occurs by hand- greens harvested by scissors and carrots by fork for example – so that scaling means continuous increases in labor.  Beyond hobby farming, scale of production has to be above a certain threshold to be as efficient as needed if the farmer both wants to provide a livelihood for the farm household and have the potential to supply food across a range of consumers. 

What about the location of that production?  Why does it matter if it all comes from a single national production center? 

I think there is good reason to argue we should encourage a distribution of production across our landscape and our nation.  Why?  Concentrating production in a narrow range of locations looks to some like the right thing to do. The implication is that this will always be the case or that we will have sufficient time to ‘move’ should these locations become untenable.  I think it just as likely that current centers of production will do everything possible, at whatever environmental cost, to maintain themselves functionally as they currently exist – there is a lot of vested infrastructure and political/economic power attached to these current locales. Right now we have a variety of production concentration centers.  Vegetables in California (over half of the U.S. domestic production); hogs in Iowa, North Carolina and Minnesota (about 52% of the U.S. production); while Georgia, Arkansas, Alabama, North Carolina and Mississippi produce nearly 60% of the U.S. broilers.   Egg production and dairy tend to be more distributed across the country.  Given that no one can provide a perfect prediction of the future the question is – can these centers of production continue into the future under a variety of scenarios?  And if they can’t do we have the political, economic, and social will to make changes in anticipation?  If we have the will is it possible to take land-based, human skill-based professions and move them to other parts of the country?  Do the human skills exist with enough depth to start in various city regions?  One case scenario is California’s extensive vegetable production.  In the summer of 2014 California experienced a record drought; 2015 looks like more of the same as the measurement of snowpack in the mountains (a predictor of spring/summer runoff for fresh water availability) revealed that there was none – zero snowpack.  Mandatory water restrictions are now in place – including some agricultural users. California is also relying on underground aquifer water more this year than last – increasingly reducing aquifer stores and laying the basis for even greater shortages over the next few years.  There is also a constant battle for water with the growing population – a population projected to grow from 39 million (2015) to 41 million (2020) and then 50 million (2050).  Projections over a longer time period are more dire - climate models indicate that California may lose up to 70% of its snow pack by 2050.  This means less water storage in the mountains as snow and thus less run-off when irrigation water is most needed. 

While, in a national and global context, we need every bit of California’s production now and moving forward, it is reasonable to predict California farmers will not be able to maintain their historic production levels in the face of climate change, water stresses, and population growth.  Hence, I would argue it is a smart hedging-our-bets scenario to consciously encourage fruit and vegetable production over a broader swath of the U.S. landscape.  And soon.  For me ‘local’ and ‘regional’ is, in part, a strategy for moving towards this distributed national production system.  Having these consciously focused on city regions means helping to ensure food and nutrition security for a concentrated population.  It recognizes that a resilient food system is one with a broader range of diversity – including location.

This means other functions and attributes of the food system need developing to take advantage of expanded production.  We currently see developments on three fronts that help food from these emerging ‘local’ farms get to market – the expansion of direct market sales via farmers markets, farmstands, subscription and CSA farms; the inclusion by existing distribution systems (e.g. large scale distributors and large retail chains) of products more locally produced; and the emergence of new distribution systems in the form of food hubs. 

It seems to me that these three inadequacies in contemporary critiques of smaller scale and alternative food system strategies: the inability of vested interests to critique at their own expense, the unscientific critique of alternative systems, and the shortsightedness of perceived scale and location superiority provides a strong rationale for asking how it could be done different in a city region perspective.  Does this point to a scale and production diversity strategy for city regions? What might be a strategy for hedging our bets about future uncertainty – embracing opportunities now that also provides more options in an uncertain future? What would increase resiliency?  What would help us in our ability to meet these challenges and adapt while maintaining the same basic core function of providing a healthy diet for all Americans (and global citizens) and allow for sustainable exports?  For a human-ecological system to be resilient in part means that at all levels of organization we preserve and enhance the natural resource base.

The answer by some to these, and other, questions is to argue for a landscape of small-scale organic farms providing the population’s food.  In the next essay I will take up the question of whether small-scale can feed the world, or at least the United States, and then offer some general thoughts on a compromise strategy for moving forward. 

[1] J.Lusk The Food Police: A Well-Fed Manifesto About the Politics of Your Plate. p. 190

[3] Quoted from: D.W. Letter, R. Seidel and W. Liebhardt (2003). The performance of organic and conventional cropping systems in an extreme climate year. American Journal of Alternative Agriculture, 18, pp 146-154 doi:10.1079/AJAA20034