A Critique of Black Box Economics

A central debate in the philosophy of science concerns the purpose of scientific theories: should they aim to represent the world as accurately as possible or is it sufficient for them to produce useful predictions? Milton Friedman’s influential essay The Methodology of Positive Economics argues for the latter, suggesting that the primary goal of a theory is to generate accurate predictions, regardless of whether its assumptions reflect reality. This instrumentalist-empiricist view has had a profound impact on economics, encouraging the development of models that may not be true in their assumptions, but are judged primarily on their predictive success.

However, this approach risks reducing economics to a mere “black box” tool for prediction, devoid of genuine insight into the underlying mechanisms that drive economic phenomena. From the perspective of scientific realism, which holds that a theory should aim to represent the world as accurately as possible, Friedman’s instrumentalism-empiricism neglects the deeper task of understanding the economy’s workings.

Friedman’s approach can be understood as an application of instrumentalism, a view in the philosophy of science that focuses on the utility of theories rather than their truth. According to Friedman, the assumptions of an economic theory need not be realistic; what matters is whether the theory yields accurate predictions. In his view, the ideal model of a “perfect” billiard player—who makes shots as if they were a mathematical genius, calculating angles and velocities with precision—is valid as long as it predicts the trajectory of the billiard balls accurately. Similarly, an economic theory need not reflect the actual decision-making processes of individuals or firms as long as it predicts their behavior well enough to be useful for policy or analysis. This emphasis on prediction has shaped much of contemporary economics, leading to the proliferation of abstract models that aim for predictive accuracy without concern for the realism of their assumptions.

However, a purely instrumentalist approach to economic theory comes with significant risks. By prioritizing prediction over understanding, instrumentalism can turn economics into a “black box” that provides useful predictions without explaining why those predictions hold. A black box refers to a system whose internal workings are opaque, and whose outputs can be observed, but not fully understood. In this context, economic models may provide policymakers and analysts with predictions about inflation, unemployment, or market trends, but they offer little insight into the causal mechanisms that generate those outcomes. As a result, instrumentalist models may lead to successful predictions in the short-term, but they may also mislead economists about the true nature of economic systems and the dynamics at play.

A wholly instrumentalist model of history may claim that any country that has a national flower, for example, is destined to be invaded. If one examines various countries with national flowers, such as Japan (cherry blossom), France (lily), and the United Kingdom (rose), one might notice that these nations have all experienced invasions at various points in history. From this observation, one could absurdly conclude that having a national flower somehow invites foreign powers to invade. For instance, Japan faced invasions from Mongol forces in the 13th century, while France has a long history of invasions throughout the Middle Ages and beyond. This reasoning, however, overlooks the myriad of geopolitical factors that contribute to a country being invaded, such as military alliances, economic resources, and territorial disputes.

Predictive frameworks, unmoored from real causal mechanisms, have no means of establishing causality. To establish causality, the following are required: (1) one event precedes the other; (2) there is a substantial relationship between the two events; and, (3) there are no confounding variables. While the first two can be established, the third cannot. One can never demonstrate that they have accounted for all confounding variables.

Hence, unless one invokes causal mechanisms, combined with heuristics that approximate truth, such as Occam’s razor, establishing causality is impossible, such that economists could only observe correlations and precedence. Under such a narrow framing, the rejection of the flower theory proposed above can only occur on the grounds that it is insufficiently predictive, not that there is no reasonable causal mechanism one can establish between flowers and invasion. Yet, if this is the case, then all that can be said as it pertains to correlation and causation is that there are stronger and weaker correlations between a lagged variable and a dependent variable. Under this framework, the flag theory can only be dismissed if one could demonstrate a theory with a stronger correlation, but in approaching the problem in this manner, one is rejecting causality.

Without a firm understanding of the underlying causes, economists may struggle to adjust their models when conditions change or unexpected events occur, leading to policy missteps or failures in prediction when those models are applied to new situations. A striking example of this failure is the 2008 financial crisis, in which instrumentalist models—derived from historical market data—failed to predict the collapse because they were overfitted to the preceding period of stability, overlooking the deeper systemic risks, such as unsustainable debt and financial interconnectivity. Those focused on the underlying causal relationships, as opposed to making predictions, were able to counterintuitively, successfully predict the crisis, such as the many Austrian economists who provided accurate warnings of the impending collapse.

From the perspective of scientific realism, this instrumentalist approach is deeply flawed. Scientific realism holds that the goal of a theory is not merely to make predictions, but to represent the world as accurately as possible. In this view, a theory is valuable not only because it predicts future events, but because it explains why those events occur, shedding light on the causal structures and relationships that underpin observable phenomena. In the natural sciences, for instance, a successful theory of planetary motion not only predicts the movements of celestial bodies but also explains them in terms of gravitational forces and physical laws. The predictive power of such a theory is derived from its accurate representation of the real-world mechanisms that govern planetary motion. By contrast, an economic model that accurately predicts inflation while relying on patently false assumptions about human behavior lacks explanatory power and fails to provide a deeper understanding of the economy.

Theories that accurately reflect the underlying reality of the phenomena they study are more likely to be durable and adaptable over time. In contrast, models that prioritize prediction over understanding may work well in specific contexts but break down when applied to new situations. In the natural sciences, the history of scientific progress has often involved the refinement of theories to better represent the underlying reality. Newton’s laws of motion were refined by Einstein’s theory of relativity, which provided a more accurate representation of the universe at large scales and in conditions of high gravity. Similarly, in economics, a theory that accurately represents the decision-making processes of individuals and firms, the dynamics of markets, and the role of institutions is more likely to provide robust predictions across a range of different contexts than one that relies on unrealistic assumptions simply because it works in a narrow set of cases.

Finally, Friedman’s instrumentalism assumes that predictive success is the ultimate test of a theory’s validity, however, in complex systems like the economy, predictions can sometimes succeed for the wrong reasons. A model might produce accurate predictions in the short-term due to chance, or because it captures superficial regularities in the data, without actually representing the deeper structures that cause those regularities. In such cases, the model’s predictions may fail when the underlying conditions change. Hence, it is hardly remarkable that economists have a notoriously poor track record when it comes to making predictions. Deirdre McCloskey pointedly remarked, “the industry of making economic predictions…earns only normal returns.”

By contrast, a theory that accurately represents the underlying reality of the causal mechanism of the economy is more likely to produce robust predictions that hold up even when the environment changes. Just as understanding the genetic basis of evolution allows biologists to make more reliable predictions about the development of species, understanding the real structures of the economy allows economists to make predictions that are grounded in the actual dynamics of markets, institutions, and human behavior.

While Friedman’s instrumentalist approach has been influential in shaping the development of economic theory, it risks reducing economics to a mere “black box” for generating predictions without offering real insight into the workings of the economy. From the perspective of scientific realism, the goal of a scientific theory should be to represent the world as accurately as possible, not just to produce useful predictions. In economics, this means striving to develop theories that accurately reflect the real mechanisms driving economic phenomena, rather than relying on assumptions on empirical data that yield predictions but fail to provide genuine understanding. By aiming for accuracy in representation, economists can produce more robust, durable theories that not only predict future events but also explain why those events occur, offering deeper insights into the nature of economic systems.

Originally Posted at https://mises.org/

zerohedge
Economics , News , Politics , Science
November 22, 2024
2 views

Utility Companies Are Not On Our Side

Authored by Linnea Leuken & H. Sterlin Burnett via RealClearPolitics,

When electric power was a novel idea and just beginning to be adopted in urban centers, the industry had a Wild West feel to it as multiple companies strung wires, opened power plants, and sold electricity on an unregulated market. Competition was fierce, but state and local governments concluded that the inefficiencies and redundancies endangered the public and imposed higher costs.

So states set up service territories with monopolistic or oligopolistic service providers, who were entrusted with providing reliable power and sufficient reserve for peak periods in return for being guaranteed a profit on rates proposed by the utilities but approved or set by newly established state public utility commissions (PUCs). These commissions were charged with ensuring public utilities served the general public universally within their territory, providing reliable service at reasonable rates.

Much has changed since then. Politicians began to supplant engineers to decide, based on self-interested calculations, what types of power should be favored and disfavored, and what types of appliances and modes of transportation Americans could use. As the 21st century dawned, a new consideration entered the picture: Climate change.

Under the banner of combatting global warming, utilities were at first encouraged and then coerced into adopting plans and policies aimed at achieving net zero emissions of carbon dioxide. The aim of providing reliable, affordable power – the rationale for the electric utilities’ monopolies in the first place – was supplanted by a controversial and partisan political goal. Initially, as states began to push renewable energy mandates, utilities fought back, arguing that prematurely closing reliable power plants, primarily coal-fueled, would increase energy costs, compromise grid reliability, and leave them with millions of dollars in stranded assets.

Politicians addressed those concerns with subsidies and tax credits for renewable power. In addition, they passed on the costs of the expanded grid to ratepayers and taxpayers. Effectively, elected officials and the PUCs, with a wink and a nod, indemnified utilities for power supply failures, allowing utilities to claim that aging grid infrastructure and climate change were to blame for failures rather than the increased percentage of intermittent power added to the grid at their direction.

Today, utilities have enthusiastically embraced the push for renewable (but less reliable) resources, primarily wind and solar. PUCs guarantee a high rate of return for all new power source (wind, solar, and battery) installations, which has resulted in the construction of ever more and bigger wind, solar, and battery facilities. The costlier, the more profitable – regardless of their compromised ability to provide reliable power or the cost impact on residential, commercial, and industrial ratepayers.

A new report from The Heartland Institute demonstrates the significant financial incentives from government and financiers for utilities to turn away from affordable energy sources like natural gas and coal, and even nuclear, and instead aggressively pursue wind and solar in particular. All of this is done in the name of pursuing net zero emissions, which every single major utility company in the country boasts about on their corporate reports and websites. Reliability and affordability come secondary to the decarbonization agenda.

Dominion Energy is a good example, as they are one of the most aggressive movers on climate-focused policy. Dominion CEO Robert Blue speaks excitedly about government-forced transitions to a wind- and solar-dominated grid in interviews. During one interview with a renewable energy podcast, he said:

“[S]ometimes the government needs to focus on outcomes. We’re trying to address a climate crisis, and we are going to need to move quickly to do that.” In the same interview, he expressed enthusiasm about federal policy that would achieve a government-directed transition.

And why wouldn’t he? Dominion, like most utilities, is granted government tax credits and guarantees on returns for investing in large, expensive projects like offshore wind, the most expensive source of electric power. The bigger the project, the bigger the profit with guaranteed returns.

Also, onshore wind companies have received special “take limits” from the Fish and Wildlife Service to kill protected bald eagles and golden eagles, while prosecuting oil companies if birds are injured or killed on their sites.

Net zero policies are not the environmental panacea that climate change activists proclaim. Industrial-scale wind and solar use substantially more land than conventional energy resources, disrupting ecosystems and destroying wildlife habitats in the process.

And despite recent technological advances, wind and solar are still not dispatchable resources, meaning they cannot provide consistent power at all times needed. Refuting claims made by environmentalists and utilities that wind and solar are the cheapest sources of electric power, costs have risen steeply as the use of wind and solar has increased. Customers of Duke Energy in Kentucky, for example, are paying 78% higher rates in the wake of coal-fired plant closings.

Politicians and utilities are pushing for even more electrification for appliances and vehicles despite the fact that Federal Energy Regulatory Commission officials have repeatedly warned in recent years that adding more demand for electric power while replacing reliable power sources with intermittent renewables is destabilizing the power system.

It appears that the utilities prioritize short-term profits over grid reliability or keeping costs reasonable – and the government officials who are supposed to keep them in check are only encouraging them. It doesn’t need to be this way. The U.S. grid was not always this way. Only in recent years, with the obsessive pursuit of net zero, have rolling black and brownouts become so common.

Today, utility companies are sending lobbyists to conservative policymakers in order to convince them that the utilities have our best interests in mind. Their track record tells another story. Meanwhile, Americans have less reliable electricity at higher costs.

Linnea Lueken (llueken@heartland.org, X: @LinneaLueken) is a research fellow with the Arthur B. Robinson Center on Climate and Environmental Policy at The Heartland Institute.

Tyler Durden
Fri, 11/22/2024 – 06:30