Economics and the Realm of Second Order Effects
The social cost of reductionism
“It counts the destruction of the redwood and the loss of our natural wonder in chaotic sprawl. It counts napalm and counts nuclear warheads and armored cars for the police to fight the riots in our cities. It counts Whitman's rifle and Speck's knife, and the television programs which glorify violence in order to sell toys to our children. 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.”
– Robert F. Kennedy
Before undergraduate economics majors are freed to sample electives at the edges of their discipline, they must pay fealty through standardized prerequisites to learn the foundational principles on which all subsequent coursework depends. The mathematical models taught in these core classes like Micro and Macroeconomic Theory comprise the foremost intellectual explanations of how wealth is generated and value exchanged amongst societies. As such, they are often the scientific justifications behind public and private policy decisions made by governments, think tanks, corporations, and any other social institution looking for recommendations on how to maximize peaceful prosperity.
Every model sorts relevant economic factors into either endogenous and exogenous variables. Endogenous variables are what enable social scientists to ask interesting questions, as they can be changed or determined by other variables within the model. For example, the researcher can adjust variable X to measure its impact on variable Y, and vice versa, in order to maximize or minimize the target output of the model. In contrast, exogenous variables are pre-set assumptions imposed upon the model from the outside. They are one-directional, in that they can influence endogenous variables but cannot vary within the model themselves. How these variables are defined, as well as the overall structure of the model, reveals a lot about what the theorist deems important.
One of these especially important mathematical anchors, the Solow Growth Model, models long-run economic growth as measured by change in Gross Domestic Product (GDP) per quarter or per year. The Solow Model is primarily concerned with the interactions of only two endogenous variables in its explanation of economic growth: labor output and capital stock. Although rightfully recognized as key to long-run growth in the real world, additional considerations like population growth, human capital, the investment rate, and technology are exogenous to the model. The Solow Theory of Economic Growth and similar neoclassical models tend to dominate Macroeconomic Theory textbooks and academic institutions.
In my own textbook, at the end of the chapter on economic growth there was a tiny blurb titled “Creative Destruction” outing a theory that broke with the Solow orthodoxy dominating the prior pages. Its text failed to fill a whole page yet described growth as an evolutionary process progressing from one “state” of political economy to the next, where the relationship between labor and capital within states are important but secondary to the political, technological, and psychological factors determining institutional transitions between states.
The writings of Joseph Schumpeter, the creator of this “Theory of Creative Destruction”, reflect an expansionist perspective standing in marked contrast to the reductionism of its more prevalent neoclassical cousins. Rather than isolating the factors dictating growth into a few exogenous variables and a simple linear equation, Schumpeter sought to broaden the scope of his lens to include the non-linearity of technological diffusion and human behavior. Through the Popperian perspective of knowledge expansion, one could say that Schumpeter took bigger, more imaginative swings and aimed at deeper truths. If this is so, why is the Solow Model so dominant?
Ultimately, models are tools, and the major distinction between Solow and Schumpeter is that Schumpeter’s theory is harder to use. Many neoclassical economists concede there are nuanced explanations for why many exogenous variables, encompassing everything from religion and cultural beliefs to natural resource endowments and geography, play a role in dictating long-run economic growth. However, it’s impossible to distill all this social and biophysical complexity into any model, nevermind one that can provide actionable recommendations; practicality necessitates a reductionist approach that can cram qualitative richness into clean, quantifiable relationships capable of answering questions like: “what is the output per worker and quantity of capital stock that maximizes GDP?”. By embracing an expansionist approach, Schumpeter concedes the ability to provide a straightforward answer to a politician or CEO asking for scientific recommendations.
Actionable theories of economic growth are immensely important given the role GDP, and other neoclassical measures of economic health such as the unemployment rate and the Consumer Price Index (CPI), play in determining the outcome of political elections. Since winning re-election during a recession is remarkably difficult, formulating strategies to avoid slowdowns and boost these key metrics within a term limit is of the utmost importance. It is even better if these strategies are defensible; reductionism also benefits from a moat of blamelessness because so much of an outcome can be determined by factors “outside the scope of the model” due to “unforeseen circumstances not within our control”.
Tweaking policy levers according to recommendations from the Solow Model and its ilk often produce the intended outcomes for highly visible target metrics but incubate a breeding ground of clandestine second and third order effects in the exogenous realm of the model. In the long run, these effects may ultimately cannibalize the targeted positive outcomes (like GDP growth) as well.
Reducing complexity makes it easier to justify action but creates a perilous illusion of comprehension. Any act of quantification, by definition, is an explicit assumption about what is worthy of being measured and what is not; setting the mathematical structure of a model and defining exogenous and endogenous variables refines this complexity further still. The distinction between “pure” and “applied” science, with the latter carrying the expectation of producing tools, can create a trade-off between usability and proximity to the truth.
In my lifetime, neoclassical economic policy and global specialization has (mostly) reliably delivered growing GDP and tangible consumer benefits to Americans. However, the US has also seen exacerbated wealth inequality, drastic increases in the cost of higher education and healthcare, a decline of artisan industry and erosion of the domestic manufacturing base, and a population suffering from obesity, opioid addiction and mental health problems. It feels as if focusing on one narrow definition of prosperity prescribed by reductionist economic models has severely affected two exogenous inputs, human capital and technological progress, and perhaps created derivative effects in many other critical, interrelated variables outside the scope of neoclassical economics entirely. It has taken exogenous shocks like a pandemic and energy crisis to expose the fragility of the current system and truly threaten quarterly GDP figures.
Popper might say that economists and policymakers have neglected the falsifiable and imaginative theories striking farther into the unknown – the theories, like Schumpeter’s, that push knowledge forward and approach truth. In practice, the “truths” of neoclassical economics have not encompassed enough of the broad spectrum of human activity to safely use their recommendations without undue risks outside the scope of their models. In reducing the richness of technological progress, energy and environmental processes, and the irrationality of human behavior and culture into a few black boxes, we have missed the forest for the trees.
Robert F. Kennedy presciently and eloquently said as much in 1968.