Between the relative and the absolute falls the shadow.
By Mark W. McElroy and Jo Van Engelen
As a distinct school of thought in the field of corporate sustainability management (CSM), the context-based approach gives rise to its own style of metrics that should be used in measuring and reporting sustainability performance. We call these context-based metrics, or CBMs. More familiar to practitioners in sustainability management, however, are so-called relative and absolute metrics. It is important to understand which of these two categories of metrics CBMs fall into, if any.
Starting with absolute metrics, these are perhaps the simplest form of measurement used in sustainability reporting, although they do have their problems. First, here’s a definition of absolute metrics we can use: Absolute metrics express operational performance in terms of what overall levels of performance are in specific areas of interest (e.g., water use) for an organization as a whole—as in the total volume of water consumed by an organization during a specific period of time.
Although simple in form, issues abound in the use of absolute metrics when measurements are taken for different periods of time with the intent of comparing performance on, say, an inter-annual basis. This is because organizations rarely remain unchanged over time. The makeup and size of an organization, for example, will routinely vary from one year to the next, thereby undermining attempts to compare performance on an apples-to-apples basis. Indeed, an organization that doubles its carbon emissions in a year in which it has also doubled in size should hardly be seen as somehow having worsened its performance compared to the year before, since for all intents and purposes, it is not the same organization anymore.
Water and Pencil Sharpeners
It is for these reasons that absolute measures, in order to be meaningful, must be normalized in a way that adjusts for changes in organizational makeup and size. One good way of doing so is to tie performance to a common denominator of some kind, such as the number of employees in a firm, thereby giving rise to a per capita unit of measurement. A per capita metric is particularly appropriate in this case because organizations, per se, are composed of individuals working together to achieve a common goal. If it is organizational performance that we are concerned with, the normalized unit of measurement we choose should be one that is as descriptive of the organization as possible. By contrast, organizational performance expressed in terms of, say, office square footage (e.g., water consumed per square foot of occupied office space) makes about as much sense as water consumed per pencil sharpener. Neither constitutes a measure of the organization, per se, and instead focuses on some other thing that, at best, may be loosely correlated with the size of the organization, but which is not in any way a direct measure or descriptor of it.
Thus, absolute metrics, intended to support inter-annual comparisons of performance, but which do not adjust for changes in the ways discussed above are dubious and should be avoided. More important, even those that are properly constructed should not be regarded as sustainability metrics, per se. Why not? Because they typically fail to take context-based standards of performance (for impacts on vital capital resources and stakeholder well-being) explicitly into account.
Relative metrics are similar to normalized absolute metrics except that they, too, do not result in measures that describe the sustainability of organizations. Instead, they most often provide measurements of efficiency. Here, it should be clear that there is a difference between sustainability and efficiency—the two are fundamentally different. For that very reason, a relative metric for water use in an organization, for example, need not be tied to the constraint earlier noted for absolute metrics (i.e., that the variable chosen for normalization purposes should be descriptive of the organization or of the activities being assessed). In the case of relative (or efficiency) metrics, no such constraint exists. Instead, the variable chosen for a relative metric must be descriptive of the non-organizational thing of interest, such as water consumption per unit of production or output, or water consumption per hour of employee time, or water consumption per unit of revenue.
Here, then, is a definition we can use for relative metrics: Relative metrics express operational performance in terms of how performance in one area (e.g., water use) correlates to performance in another area (e.g., revenue or total production)—as in water consumed per unit of revenue or water consumed per unit of production during a specific period of time.
Relative metrics are also sometimes referred to as intensity metrics, precisely because of their efficiency orientation. Once again, water provides us with an easy to understand example. Imagine a metric that reports water consumed per unit of production. This is an efficiency metric, pure and simple, or what some might refer to as a water intensity metric; it expresses the intensity of water use per product produced. The same thing can be done for greenhouse gas emissions, solid waste emissions, energy consumption, and material throughput, as well.
It should also be clear that an organization could have both an absolute and a relative metric in play at the same time for the same issue, such as water use. In that case, the absolute metric would be water used by a company or manufacturing plant in a year, and the relative metric could be water used per unit of production in the same year. And just as we can say that absolute metrics are not sustainability metrics, per se, so we can say that relative metrics are not sustainability metrics, either. To understand the efficiency of an organization’s water use is not to know what the sustainability of its use is. Even reduced or optimal levels of water use can be unsustainable—without more information (i.e., context) to go by, one simply cannot tell.
Back to the Absolute
This raises the question of what absolute measures really are, and whether or not they might be a kind of efficiency metric—or, if not, perhaps some other thing. In point of fact, we think it safe and valid to say that absolute metrics are merely a special type of relative metric since they express impacts in a normalized way, albeit at the level of whole systems. Instead of expressing impacts per underlying unit of production or what have you, they express impacts per unit of overlying organization, as in water consumed by a company (a legal entity) from one year to the next. Thus, for some purposes, the fact that organizations change in makeup and size from one year to the next is of no consequence. Imagine growth rates, themselves, for example. A metric of that kind is already predicated on the assumption that organizations change in size every year; its very purpose is to measure by how much. So the fact that organizations change in size from one year to the next would in no way prevent a metric of that kind from being used.
The motivation behind most sustainability measures, however, is very different. Once again, take the case of water. Imagine a world in which one year there were only ten companies in business and one source of water. Imagine, as well, that each company used exactly one-tenth of the available renewable supply that year, and that the sustainability of their use was measured in absolute terms. Now imagine that the following year each company spins off a new division and transfers half of its employees to run it. In year number two, then, there would be twenty companies, not ten. Should each of the original ten companies in the second year be graded sustainable if their overall use of water resources remains unchanged? How could this be justified, since the original ten companies would be half the size of what they were, and the newer ten companies would have no water to use at all—not without dipping into non-renewable stocks, that is? All of these changes and contextual considerations, however, are hidden from view when using absolute metrics. And that is exactly what makes them so ill-suited for purposes of measuring sustainability.
A better solution in the imaginary case above would be to use a relative metric that measures organizational performance, not on an enterprise-wide or legal entity level, but on a normalized per capita or per employee level. That way, the sustainability of a company’s use of water resources could be expressed in terms of water use per employee (or per employee hour worked), and then be measured, reported, and compared in those terms from one year to the next. Thus, despite the fact that a company might be half the size in year two of what it was in year one, its water use per employee hour worked might very well be the same.
To sum up what we have said thus far, we can say that the distinction between absolute and relative metrics is a fuzzy one, and that absolute metrics are merely a special case of relative metrics that operate at the level of the enterprise as a whole, or sub-groups within it. Thus, the thing being measured never really changes at a logical level from one period to the next—but does in a physical sense—which for some purposes poses no problems at all. Acme Widgets is Acme Widgets, no matter how many employees it has this year versus last.
From Relativity to Context
For sustainability measures, however, variability in make-up, size and other factors matters greatly, as the example above makes clear. To help cope with such variability, other forms of relative metrics can be used, such as the ones already discussed—water consumption per employee, water consumption per unit of production, water consumption per unit of revenue or what have you. But this, too, has its problems.
If all we measured in the fictitious case above, for example, was water consumption per employee, the size and sufficiency of the shared water resource from which each company was taking its supplies would never be known, much less factored into the assessment. Thus, relative metrics, no matter what their statistical advantages over absolute measures might be, suffer – in all of their forms – from a fatal lack of context. It is precisely such context, however, that is so critical to the measurement of sustainability performance, notwithstanding its irrelevance to efficiency. This, of course, leads us to context-based metrics.CBMs measure and report sustainability performance in a literal or genuine sense. They do so by explicitly taking the demand for and supply of valuable goods and services from vital capitals explicitly into account. Using water, once again, as an example, a CBM would include a measure of an organization’s proportionate share of available renewable supplies, given its size in terms of headcount, contribution to GDP, or other factors that can be used in making such allocations. Actual water consumption can then be measured against the allocated level, resulting in a numerical score that is either less than or equal to the allocation (sustainable), or greater than the allocation (unsustainable).
Here, then, is a definition we can use for CBMs: Context-based metrics express organizational performance in terms of impacts on vital capitals, relative to norms or standards for what such impacts ought to be (for specific periods of time) in order to ensure stakeholder well-being (e.g., total water consumed per employee per year compared to a fair or equitable allocation of available renewable supplies).
In a very important sense, then, context-based metrics are nothing but relative metrics with factual and normative context added (e.g., how much renewable water is there? And how much of it should an individual organization or facility be entitled to use?). The relative aspect of such metrics makes it possible to perform inter-annual analyses in order to resolve the problem of changes in organizational size, make-up, etc. that always come up; while the context element makes it possible to take the availability, quality, and sufficiency of relevant vital capitals explicitly into account. The result is a powerful tool for measuring the sustainability performance of organizations in the real world, including all of the variations in capital resource conditions, human populations and company-specific stakeholder groups that must be fully accounted for if measurements are to be meaningful.
In our own work with clients, we have constructed CBMs in the form of numerical quotients, with denominators representing standards or norms of performance for impacts on capitals, and numerators representing actual impacts on the same capitals. For environmental areas of impact, scores of less than or equal to 1.0 signify sustainable performance, while scores of greater than 1.0 signify unsustainable performance. This is because the means of natural capital is limited and we must therefore live within it.
In the case of non-environmental impacts, by contrast, the logic of scoring reverses, since unlike natural capital, the other capitals associated with social sustainability (human, social and constructed) are human-made (i.e., are anthropogenic, or anthro capitals) and must therefore be created and maintained at certain minimum levels. Social sustainability performance, then, is largely a measure of capital production, whereas environmental sustainability, by contrast, is a measure of
Notwithstanding the important distinction we have made between sustainability and efficiency metrics, most sustainability measurement and reporting practices today fall into the efficiency category (i.e., they are context-free, relative metrics), while still clinging to sustainability terminology. That said, some important exceptions exist. Two formal (i.e., codified) exceptions, in particular, are worth noting. One is the Ecological Footprint Method (see the work of William Rees and Mathis and a second is the Social Footprint Method (see our article from the July/August 2008 issue of Corporate Social Responsibility and Environmental Management). Each of these context-based methods is currently being used to varying degrees, and will presumably—or hopefully—become more widely used as the shortcomings of context-free thinking become better understood. This will serve as the underpinning of our next piece, exploring the very contours of sustainability as a concept.
Mark W. McElroy, Ph.D. is the founder and executive director at the Center for Sustainable Organizations in Thetford Center, Vermont. Jo van Engelen, Ph.D. is a full professor at the University of Groningen in The Netherlands, and also serves as chair of Integrated Sustainable Solutions at Delft University of Technology.