Econometrician Charles Manski summarized the logic of empirical inference by the relationship: Assumptions + Data = Conclusions. In the world of big data, the importance of domain knowledge often gets neglected. However, in reality, the assumptions develop the perspective for evaluating the data. Assumptions, according to Manski, also includes theory along with the axioms. When conclusions depend significantly on the assumptions, how does the uncertainty and subjectivity associated with these assumptions affect the conclusions raised? This article will look into the intricacies of assumptions in the empirical inference process and examine the case of GDP from that lens.

Former US Secretary of Defence Donald Rumsfeld said, “there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say, we know there are some things we do not know. But there are also unknown unknowns— the ones we don’t know we don’t know. And if one looks throughout the history of our country and other free countries, it is the latter category that tends to be the difficult ones.” Here, he distinguishes between known knowns, known unknowns and unknown unknowns. This distinction is critical to have cognizance about the cogency of the inference made based on assumptions that are not wholistic. Because of unknown unknowns, the assumptions that include the theory cannot be claimed to be complete of all the necessary knowledge and axioms.

To give an example, Newton’s laws of motion were developed in classical mechanics. However, classical mechanics was based on assumptions that were not complete. Those assumptions began to be challenged when the dual nature of an object, both as a particle and a wave, was discovered. Till then, this phenomenon was an unknown unknown and therefore, that made the theory deficient. However, no one would have predicted the nature of unknown unknowns till this discovery was made. Similarly, the complete character of unknown unknowns to the present academic understanding cannot be predicted with certainty. What we as the human race know at present is minute as compared to the infinite possibility of knowledge that we don’t know. Perhaps that should also give us humility and a sense of the magnitude of our existence.

One of the core problems in econometrics is suitable feature extraction. To find the true causal relation and thus the dynamics of a system, one needs first to detect the variables or features that constitute the system. The system might range from simple to complex organizations. Simple systems might be a picture that is comprised of pixels. The complex system can be as complicated as the human body. The challenge of unknown unknowns increases with indulgence with more complex systems.

For example, for finding the critical variables for the evaluation of health in the human body, the framework, which is an assumption, matters a lot. A great example would be the difference in perception of Allopathy and the Ayurvedic system of health. According to Dr. Ivan Vladić,

While allopathic medicine sees the illness as an attack on the body, and the treatment as a fight against it, Ayurveda approaches the disease as an imbalance in the body, and the treatment as bringing the body back into balance.”

Ayurveda views human health from the balance of tridosha- vata, pitta and kapha, whereas Allopathy considers the biochemistry and physiology of the human body. Therefore, the essential variables for primary aspects such as fever vary for Ayurveda and Allopathy since the view of the human body is different for the systems. Since the crucial variables are different, the causal relationships are also different, and the cause’s attribution is also dissimilar. While the cause for fever in the Allopathy system may be a bacterial, viral or parasitic infection, the cause for fever in Ayurveda could be as Vataja, Pittaja, Kaphaja, Vata-Pittaja, Vata-Kaphaja, Tridoshaj (intrinsic fevers), Agantuja (extrinsic fever), Krodhaja (anger induced), Kamaja (related to sex), Vishama jwar (malaria), Shokaja (anxiety induced), and Bhuta-vistha (of unknown origin).

Even when the variables are the same, there remains scope for researchers to make assumptions about the sample selection, feature connections, construction of hypothesis, operationalization of variables, analytical process to employ, the saliency of different parts of inference, presentation of results etc. This considerable degree of freedom available to the researchers makes the possibility of inferences diverse. Recent research has shown that with the same data, contradictory inferences were made by different research groups. Schweinsberg et al. compiled a list of recent studies where different or contradictory inferences were formed using the same data. Charles Manski coined the term duelling certitude for such happenings in his famous book ‘Public Policy in an Uncertain World’.

When such occurrences are common in academic circles, even with a well-defined data set, it should be a much more regular occurrence with ordinary people without well-defined data set in their day-to-day lives. The frameworks that ordinary people have developed are so vast that there are as many mental versions of reality as many people are there. This subjectivity in the interpretation of domain and data has led to differences culminating in rival theories and schools of thought. The most prominent example of this could be the political science spectrum of left and right. People have corroborated different theories with various researches on real-world data. The differences start from the objective each school of thought is optimizing. For example, to show better economic growth, research supporting capitalism shows that the economic pie has grown, whereas researches supporting communism show the reduction in inequality. They, in reality, have strong reasons to believe and claim that the school of thought they endorse is better. However, the assumptions are so abstruse that one gets confused as to where to start unravelling the conflict.

This subjectivity in the assumed assumptions makes a fertile ground for the perpetuation of propaganda. Keeping the data as it is, one can presuppose a conclusion and work out the assumptions needed to reach the desired conclusion. Charles Manski describes this as advocacy. Certain think tanks and media houses reap the benefit of this fuzziness and propagate propaganda. One needs a highly objective mindset and firm honesty to communicate the truth with sincerity. Charles Manski, in his paper, showcases how Milton Friedman, a very famous economist, conflates science and advocacy in an example.

One place where fudging has a significant impact happens with the calculation of GDP. There is a lot of ambiguity associated with GDP as a measure. GDP is the calculation of all market transactions, including goods and services that are traded for money. However, there are serious limitations associated with employing GDP as a proxy for a country’s wellbeing. The challenge of quantifying wellbeing has been acknowledged in all social sciences domains. However, the use of GDP as a proxy for wellbeing has raised the challenge raised by Goodhart. Goodhart said, “When a measure becomes a target, it ceases to be a good measure.” GDP captures just one of the means for the overall wellbeing of the citizens. However, focussing solely on GDP have instead created challenges for wellbeing.

According to McCulla and Smith,  GDP answers questions such as “how fast is the economy growing,” “what is the pattern of spending on goods and services,” “what percent of the increase in production is due to inflation,” and “how much of the income produced is being used for consumption as opposed to investment or savings”. However, according to Costanza et al., “many important economic activities are entirely excluded from GDP measurements, such as volunteer work, social capital formation within healthy family units, the costs of crime and an increasing prison population, and the depletion of natural resources.” Therefore, they say that “GDP is a measure of economic activity, not economic wellbeing.”

Robert F Kennedy beautifully expressed this. He said, “Yet the Gross National Product does not allow for the health of our children, the quality of their education or the joy of their play. It does not include the beauty of our poetry or the strength of our marriages, the intelligence of our public debate or the integrity of our public officials. It measures neither our wit nor our courage, neither our wisdom nor our learning, neither our compassion nor our devotion to our country, it measures everything, in short, except that which makes life worthwhile. And it can tell us everything about America except why we are proud that we are Americans.”

Not just GDP is not an apt tool to capture wellbeing, it is not also a competent tool to capture material prosperity.  According to Indic tradition, material wealth has been related to the balance of the panchmahabhutas- earth, water, fire, air, and ether. A proper balance of these panchmahabhutas is essential for the progress of the human race in subtle life objectives. Costanza et al. also expressed this essentially same concept a bit differently. They explain that all the components- natural, social, and human capital influence the economy and that the quantity and quality of such capital are affected by net investment from the economy. By measuring only marketed economic activity, GDP neglects alterations in the natural, social, and human components of community capital on which the community relies for continued sustenance and welfare.

Dependence on GDP as a proxy for wellbeing also results in depletion of material wealth. Costanza et al. explain that GDP, amongst all other natural degradations, motivates forest depletion, as lumber provided by cutting forest is valued more when we calculate contribution to GDP. GDP calculation does not integrate, and neither does it have well-defined techniques to include services such as biodiversity habitat, vulnerability to flooding from natural conditions such as storms, water filtration that improves its quality, the segregation of carbon dioxide and manufacture of oxygen. Only services that are part of the market economy are counted.

Therefore, a lot of brainstorming needs to happen regarding the ideation of prosperity and wellbeing. A holistic framework to view life will help in getting the right perspective to growth. Such a framework will help form comprehensive objectives, thorough measurement techniques, sound feedback mechanisms and a more agreeable approach to growth. Such comprehensive objective will produce sustainable growth models, measurement techniques will provide apt variables to regulate, feedback mechanisms will help maintain a dynamic balance, and a more agreeable approach will unite positive efforts. Bharat already has a strong foundation to contribute such frameworks to the world. She already has concepts such as Panchmahabuta, Panchkosha and Purushartha and interaction between them established that can be modified to the modern needs. However, thorough effort needs to be put in moulding these Indic frameworks to satisfy the modern needs in a modern communication method to grow together into a healthier, prosperous and joyful world.

 

References

Schweinsberg, M., Feldman, M., Staub, N., van den Akker, O. R., van Aert, R. C., Van Assen, M. A., … & Schulte-Mecklenbeck, M. (2021). Same data, different conclusions: Radical dispersion in empirical results when independent analysts operationalize and test the same hypothesis. Organizational Behavior and Human Decision Processes.ument.

Costanza, R., Hart, M., Talberth, J., & Posner, S. (2009). Beyond GDP: The need for new measures of progress. The pardee papers.

McCulla, S. H., & Smith, S. (2007). Measuring the Economy: A primer on GDP and the National Income and Product Accounts. Bureau of Economic Analysis, US Department of Commerce.

Manski, C. F. (2013). Public policy in an uncertain world. Harvard University Press.

Manski, C. F. (2020). The lure of incredible certitude. Economics & Philosophy, 36(2), 216-245.