Yes, because what we measure affects what we do, argue Stiglitz, Sen & Fitoussi in the 2009 report of the Commission on the Measurement of Economic Performance and Social Progress Starting from the basic assumption that the quality of policies adopted in a certain sphere of activity depends on the appropriateness and the explanatory power of the measurements available assessing the respective field, the Commission’s report points out that inasmuch as some of our current metrics are flawed, so too may be the inferences we draw.
With that said, Stiglitz, Sen ; Fitoussi express the ambitious aim to provide an impetus for on-going research into the development of better metrics that will allow a better assessment of economic performance and social progress. By doing so, they academics hope to pave the way for advancing social progress further and detecting ways how to better sustain it over time. With this in mind, Stiglitz, Sen ; Fitoussi develop a broad set of recommendations aimed at the identification of new and more relevant indicators of social progress as well as at the evaluation of the feasibility of alternative measurement tools.
In the authors’ eyes, statistical concepts are usually not wrong as such, but they are often wrongly used. Therefore, in addition to better metrics, a better understanding of the appropriate use of each measure is needed, too. The authors make a clear distinction between the assessment of current well-being and the assessment of the levels of well-being over time – summarized in the term “sustainability”.
Before they define ways of how the sophisticated use of tools from social science could advance social progress, meaning an increase of the well-being of each of us, the authors stress the significance of better measures of economic performance given the fast growing complexity of our globalized economy. In this context, Stiglitz, Sen & Fitoussi highlight the fact that obviously most of our current metrics, first of all the GDP, have serious flaws.
The GDP for example fails in projecting quality improvements, which leads to significant flaws in the measurement of overall productivity. The academics emphasize that with the growing share of services and of complex products in proportion to the total output, such shortcomings in measurement become even more striking. Moreover – because it is mainly focusing on expenditures – the GDP systematically distorts the measurement of government-provided individual services and therefore fails to integrate the government’s true contribution to an economy.
Given such conclusions, Stiglitz, Sen ; Fitoussi go one step further: Most of the current metrics do not just fail – although they may not be wrong as such, but wrongly used – in explaining their “core area”, namely the measurement of economic production; but they are often incorrectly brought in as alleged yardsticks for societal well-being, too. However, since these metrics are first and foremost designed for measuring economic growth they hardly picture social progress in an appropriate form.
On these grounds the authors argue that by being aware of these circumstances one might easily realizes that since we have much more sophisticated metrics to measure the economic progress, we tend to use them as “stopgaps” to fill the void of available and commonly accepted metrics of societal well-being. However, the authors claim that measuring well-being – a complicated multi-dimensional construct – requires both considering its objective as well as its subjective dimensions.
Evidently, economic metrics such as the GDP fail miserably by tackling this task. In view of these findings, the academics postulate the necessity of a shift of emphasis from measuring economic production to measuring people’s well-being. More precisely, the development of a statistical system that complements measures of market activity by measures centered on people’s well-being and by measures that capture sustainability.
Furthermore, such a system should be able to document the diversity of peoples’ experiences and the linkages across various dimensions of people’s life. In order to transform the current approach to statistics towards such a system, the authors propose several recommendations, including advices such as evaluating material well-being by looking at income and consumption rather than production, emphasizing the household perspective, considering income and consumption jointly with wealth, or giving more prominence to the distribution of income, consumption and wealth.
Likewise, Stiglitz, Sen ; Fitoussi detect room for improvement for areas such as measuring sustainability in terms of social well-being as well as with a view to environmental pressures. Pushing forward these claims, Stiglitz, Sen ; Fitoussi stress the great significance of social science in the search of new – and with a view to an intended increase of people’s well-being – more adequate patterns of thinking about the interplay of economic growth and social progress1. B. Evaluation Stiglitz, Sen & Fitoussi demonstrate in a well-structured manner new ways of dealing with statistical metrics.
In doing so, they plausibly urge the required caution and critical attitude when dealing with statistical projections. By addressing its recommendations to an accurately defined group of stakeholders as well as by outlining its own history of origins and unveiling its underlying agenda, the report holds his promise and constitutes a solid basis for further reflections. On the one hand the authors show the limits of indiscriminate “one-fits-all approaches” towards measuring economic as well as social progress.
On the other hand, they emphasize the huge potential of social science on the path to a general increase of living standards through a better assessment of the status quo – provided it is implemented in an extensive and well-considered manner. The authors convincingly call attention to the high complexity of the development of sound metrics and their appropriate use, and – by doing so – confront the reader with the fragility of the perceived reliability of widely used statistical metrics.
The report shows well, how tempting it can be to “overstretch” certain metrics by – falsely – apprehending them as adequate measures for explaining a certain phenomenon, which is actually not the case. Taken together, the report clearly identifies pending issues in the context of the present state of statistical information about the economy and the society, and guides the way for further research on the subject matter.
Nevertheless, one could argue that concerns about the adequacy of current measures of economic performance, particularly doubts regarding the appropriateness of the GDP as a main performance benchmark, are nothing really new. The same applies for the insight that our current measures might be useless for measuring societal well-being or – even less – sustainability in its economic, environmental and social dimensions. However, in the authors’ defend one must acknowledge that their intention is certainly not to reinvent the wheel and to completely overthrow the main pillars of the current measurement systems.
The report basically presents a broad set of recommendations in which ways the current system could be complemented as well as adapted by implementing carefully targeted adjustments. Yet, given the complexity and the multifacetedness of the subject matter the academics – correctly – admit that the present report rather must be seen as a starting point for further discussion than as a “final document” in itself. However, aiming at tackling the subject matter in its entirety as well as speaking to a multitude of stakeholders, the report tends to risk a loss of focus and depth in its claims.
This applies especially, when it comes to the practicability of its suggestions. Furthermore, facing a variety of recommendations, the reader may be confused how to assign priorities or where to start transforming the current system. C. Recommendation Although the practical implementation of the numerous recommendations in the report may be questionable, the Commission’s report approaches profound questions of welfare economics and gives venue for a discussion of societal values.
With this end in view, Stiglitz, Sen & Fitoussi explicitly address themselves to a broad range of stakeholders, including political leaders, policy makers, the academic community, statisticians and users of statistics, as well as civil society organizations and – more generally – the public at large. Being aware of the relevance of their deliberations for this wide audience, the authors renounce (as far as possible) the use of a technical language, ensuring the comprehensibility of their arguments. However, while reading the present report one should be sensible for some words of warning, too.
Although the idea of a better world through the ingenious use of social science seems to be likewise reasonable as tempting, one should not be blinded by the “technocratic illusion” painting a state free of ideology, only amenable to scientific research and the faculty of reason. To bring this “doctrine” into question, it suffices to have a closer look at the basic assumption of the report: “(… ) what we measure shapes what we collectively strive to pursue – and what we pursue determines what we measure (… )”. Consider such world purely determined by what we are able to prove statistically.
Taking up the allegory drawn in the report: imagine the “pilots” would be equipped with better, more sophisticated compasses. Would that solve the problem at heart? I am in doubt about it. A perfectly reliable compass may be a very useful tool if you know well the direction of your journey. However, it does not tell you whether to head north or south. Even though, we base such a decision – equipped with a broad set of powerful metrics – on a precise analysis of past experiences, there is no way to entirely exclude value judgments from the decision making process.
One can even go one step further: Since we are far from having a perfectly reliable compass in reality – even if we would be able to implement all the recommendations given in the report – the proportion of normative judgments in policy-making is even more evident. Therefore, although the solution presented in the present report may be convincing in theory, when it comes to reality universal principles of actions – ways to judge actions without knowing their consequences – become equally relevant.
Given that insight, one could ask if it would not be more reasonable to invest more efforts in the (feasible) elaboration of such universal principles shaping our actions in any situation, than wasting our endeavors in the pursuit of ever better and more powerful metrics – a race conceivably doomed to fail? Characters: 11’136 1 By conducting this line of argument, the authors are in line with Layard’s reasoning presented in the course of the Lionel Robbins Memorial Lectures from March 2003.