Measuring Sustainability

This article in Beef Magazine by Bryan Weech notes the growing interest in beef sustainability but he also emphasizes the challenge for producers in knowing what sustainability means or how to measure it.  He writes:

Often these events either focus on comparing the sustainability of beef industry of 1970 versus today or discuss the developing definition of sustainable beef, which seems to be coalescing on beef that is environmentally friendly, socially responsible, and economically viable.

That is a broad sentence that nicely sets the boundaries of sustainability. However, it leaves most of us wondering what it means from a management perspective and how exactly is environmentally friendly, socially responsible, and economically viable beef accomplished from a day-to-day, on-the-ground management perspective?

He then provides a nice table listing 14 different approaches advocated by different parties to promote sustainability including approaches such as "reduced consumption/vegan", organic, and grass fed.  The only approach he lists as having a potential meaningful effect on sustainability is a Verified or Certified Sustainable program.  However, few details were mentioned as to what exactly such a program would entail.  

I'll throw an idea out here.  I've been arguing for the past several years that productivity growth is an often forgotten cornerstone of sustainability.  Enhanced productivity means being able to produce more (or the same amount of) beef using fewer resources.  That sounds like a pretty important part of being sustainable.  The challenge is, however, that the way productivity growth is often measured is in aggregate terms over long periods of time, such as the comparison to the 1970s that Weech mentioned above.  It is also the sort of comparison I used in this New York Times piece.  But, to put such measures of productivity into practice (or to form a basis of a certification program), one would have to measure the productivity and efficiency of individual farms over time and as compared to each other.

Fortunately, economists have a well established set of tools to conduct precisely these sorts of calculations.   This suite of methods go by a variety of names such as technical efficiency analysis, frontier analysis, data envelopment analysis, and more.  The basic idea is take data on a set of farms and figure out which ones are on the "frontier" of the production function or cost function.  That is, for a given level of inputs, which farm produces the highest level of output?  Once the farms along this frontier are determined, then one can measure the distance of every other farm from the frontier to establish measures of relative efficiency.  It is also possible to use multiple farms over time to see whether the frontier is shifting over time or to measure how far a farm was from the frontier in, say, 1980 compared to, say, 2017.  This sort of approach has been widely used to compare relative efficiencies of organic and conventional dairy farms, Guatemalan corn farmers,  hog farms in Hawaii, US agricultural cooperatives, and much much more.  There have been so many studies conducted on these topics that there are even meta analyses attempting to quantitatively summarize the vast literature and identify which types of farms are more or less likely to be efficient.  

Just to give an example, consider this 1997 study by Allen Featherstone, Michael Langemeier, and Mohammad Iset that compared 192 Kansas cow-calf operations in the year 1992.  Here is one of their main findings:

Technical efficiency ranged from 0.37 to 1, with an average measure of 0.78 (table 2). Thus, the output of the farms potentially could be increased by roughly 22% if each farm were purely technical y efficient (i.e., if each farm operated on the production frontier). Forty-nine out of the 195 farms were technically efficient.

So, there were 195-49 = 146 cow calf operations that were not as sustainable as they could be in the sense that they could have produced more output for the particular amount of resources (or inputs) they were using as compared to other farms in the sample.  

This sort of analysis also allows one to easily visualize how farms compare to each other in terms of efficiency.  Here is a figure from the paper which shows scale efficiency of all 192 operations (each farm is represented as an asterisk in the figure).  The solid dark line is the frontier showing operations with the lowest costs for a given amount of revenue.  

The authors write:

The results generally indicate that a greater proportion of overall inefficiency was due to farms producing above the cost frontier than to farms being of an inefficient scale. The individual analyses of the farms showed that 62 farms were operating in the region of increasing returns to scale, one farm was producing at constant returns to scale, and 132 farms were operating in the region of decreasing returns to scale.

I'm not claiming that this sort of efficiency analysis is a perfect measure of sustainability or that it is capable of capturing all the dimensions of sustainability consumers might find important.  However, it strikes me a productive way to help move forward the debate on how one might attempt to start quantifying an important aspect of sustainable production.