Ian Fletcher and Marc Fasteau write:
* The US has been losing the international competition for high-value industries and the good jobs, wealth, tax revenues, and national defense capabilities they provide. From 1998 to 2010, 6 million US manufacturing jobs disappeared. Many – 3.5 million between 1991 and 2019 alone – are estimated to have been lost due to imports. Real wages for nonsupervisory workers have stagnated for 40 years in part because of such job losses. Consumers have benefited from the imports, but not enough to outweigh the lost industries and jobs.
* Key military components now come from abroad, some from China and other adversaries, leaving the US exposed to supply cutoffs, sabotage, and spyware. The COVID-19 pandemic revealed the vulnerability of America’s medical and other important supply chains.
* Until very recently, industrial policy has been dismissed in the US as a recipe for ill-advised, inefficient interventions in free markets, both domestically and abroad. Domestically, it has been associated with failing industries, such as steel in the 1980s, lobbying for bailouts.15 It has been associated with companies, such as Lockheed in the 1970s and Chrysler in the 1980s, that appeared to fail the test of market competition and needed government help to survive.16 And it has been associated with purported boondoggles such as synthetic fuels and the breeder reactor.
Abroad, industrial policy has been associated with governments propping up failing state-owned companies while mismanaging them. It has been blamed for commercially unviable lunges for technological sophistication such as the Anglo-French Concorde and the European computer industry. It has been blamed for expensive, failed attempts to transplant modern industry to developing nations.
But systematic, proactive industrial policy is in fact the norm for the rich, technologically advanced nations America competes with, especially in East Asia and Continental Europe.
* Over the past 15 years, industrial policy has been slowly filtering back into America’s policy space. Although not so characterized at the time, the 2008–10 federal rescue of the auto industry was a classic act of industrial policy. It was a success, though more needs to be done if the industry is to remain healthy.
In 2021 and 2022, Biden proposed and Congress enacted the Bipartisan Infrastructure Act (BIA), the CHIPS and Science Act (CHIPS), and the Inflation Reduction Act (IRA). These ambitious new programs, combined with their explicitly pro–industrial policy rationales, were a big step forward.
* In a significant departure from past practice, the BIA, the IRA, and CHIPS explicitly endorsed and embodied a number of core industrial policy principles. First, that economic and technological leadership and secure supply chains in civilian industries, not just defense industries, are critical to national security. Second, that making things, not just inventing them, is required for prosperity. Third, that large-scale government investment is needed to foster commercialization of new technologies and support US manufacturing in high technology and other economically important industries.
* Mainstream US economists have opposed industrial policy for decades…. Paul Krugman: “Economists have been extremely negative about the idea of industrial policy even in principle. The general presumption of most economic theory is that the best industrial policy is to let the market work – that decentralized incentives of the marketplace will push resources to the places with the highest expected return, and that no second-guessing of market decisions is necessary or desirable.”
* A major problem with mainstream economics, once one gets past recitations of free-market theology and looks at the actual scholarship, is that it is only good at understanding things that are well suited to mathematical modeling, such as price setting in freely competitive markets.37 It has weak insight into organization, management, corporate strategy, R&D, engineering, workforce development, the origins of technology, product design, policymaking by governments, and the strategies nations use to compete. None of
these reduce to equations well, but they are central to how modern firms, governments, and thus economies function. Even some fields that do mathematize well, such as finance and international trade, are prone to oversimplifications that lead to false conclusions.
ChatGPT says:
Mainstream (neoclassical) economics is strongest where behavior can be reduced to clean, tractable equations — especially competitive markets, supply/demand, and price theory.
The weaknesses it’s pointing to are real:
Organizational behavior, management, and innovation: These are often studied outside mainstream econ (in business schools, sociology, and management science) because they involve messy human, institutional, and historical factors that resist tidy mathematical modeling.
Technology and product development: Economists tend to model innovation abstractly (endogenous growth theory, R&D spillovers), but the real processes are complex and case-specific.
Government policy and national strategy: Political economy tackles these, but often at a high level; actual decision-making and strategic maneuvering are better captured by political science, history, or strategic studies.
Finance and trade oversimplifications: Classic models can make unrealistic assumptions — perfect markets, rational expectations, frictionless capital flows — that sometimes lead to predictions or policies that fail in practice.
Here are some concrete examples where mainstream economics’ math-friendly bias led to blind spots or bad calls:
1. The 2008 financial crisis
Many macro and finance models (especially DSGE and efficient markets theory) assumed away systemic risk and the possibility of cascading failures in interconnected financial institutions.
Complex mortgage derivatives and shadow banking were treated as if markets would price them correctly, ignoring institutional incentives and human error.
Result: Widespread failure to predict or even recognize the crisis building.
2. Industrial policy and East Asian growth
Standard trade theory (comparative advantage, free trade) predicted countries like Japan, South Korea, and Taiwan should specialize according to their factor endowments (e.g., low-cost labor).
In reality, these nations used heavy-handed government intervention, protectionism, and coordinated industrial strategy to leapfrog into advanced manufacturing.
Many economists only recognized the validity of these strategies decades later.
3. Innovation and technology adoption
Models often treat innovation as a smooth, continuous variable (“R&D spending leads to productivity gains”) instead of a lumpy, path-dependent, organizationally complex process.
This misses why some breakthrough technologies stall (e.g., nuclear power) while others take off (e.g., semiconductors), despite similar initial economics.
4. Labor markets and skill formation
Standard labor economics assumes workers are easily reallocated when industries decline (“just retrain and move”), ignoring the deep role of local institutions, social networks, and employer-specific skills.
This helped produce overly optimistic predictions about how fast communities could recover from globalization shocks (e.g., the “China shock” in U.S. manufacturing).
5. Privatization and deregulation in developing countries
The Washington Consensus models assumed competitive markets would emerge after privatization and trade liberalization.
In many cases (e.g., Russia in the 1990s), what emerged were oligopolies and kleptocracies, because the institutional prerequisites for competition were absent.
Here’s how economics’ preference for elegant, tractable math pushed it toward those blind spots:
1. Preference for equilibrium models
Economists favor models where markets clear and forces balance. These are mathematically tidy but assume away instability, feedback loops, and crises.
This bias meant pre-2008 models often ignored how leverage and panic can drive systems far from equilibrium.
2. Homogenizing assumptions
To keep the math solvable, models often treat firms, workers, or countries as “representative agents” with identical behaviors.
This erases real-world diversity in capabilities, strategies, and institutions — exactly what matters in technology adoption, industrial policy, and workforce development.
3. Aversion to qualitative complexity
Fields like management, engineering, and political decision-making involve factors that don’t translate neatly into equations — power struggles, corporate culture, bureaucratic politics.
Because these are harder to formalize, mainstream economics often sidestepped them entirely, leaving them to other disciplines.
4. Over-reliance on rational optimization
Rational-actor models are mathematically elegant and analytically tractable, but they miss human error, bounded rationality, and institutional inertia.
This leads to overly optimistic predictions about smooth adjustment to shocks (e.g., labor retraining, market liberalization).
5. The “as if” defense
Milton Friedman popularized the idea that unrealistic assumptions don’t matter if predictions work. This licensed models that ignored institutional detail and historical context in favor of abstract optimization, even when predictions failed badly in new contexts.
6. Data fit over process understanding
Statistical models can be fine-tuned to match past data (e.g., asset pricing models before 2008), but that’s not the same as understanding the underlying mechanisms.
Without institutional and organizational insight, good fit can hide fragility.
Grok says:
The core of the critique—that mainstream economics offers “weak insight” into organization, management, corporate strategy, R&D, engineering, workforce development, origins of technology, product design, policymaking, and national competition strategies—is substantiated by widespread scholarly criticism. These areas involve dynamic, institutional, and behavioral factors that resist reduction to clean equations due to uncertainty, path dependency, power relations, and social embeddedness.
Organization, Management, and Corporate Strategy: Mainstream models often rely on methodological individualism, reducing complex social structures to individual utility maximization, which ignores how institutions shape behavior.
For example, transaction cost economics (a semi-mainstream extension via new institutional economics) attempts to model organizations via contracts and hierarchies, but critics argue it still oversimplifies real-world management dynamics like culture, leadership, and adaptive strategies.
Heterodox economists, such as those in evolutionary economics, contend that firms aren’t just profit-maximizing black boxes but evolve through routines and learning that defy static equations.
R&D, Engineering, Origins of Technology, and Product Design: Mainstream endogenous growth models (e.g., Paul Romer’s work) incorporate R&D as a driver of innovation through knowledge spillovers and increasing returns, but they abstract away from the messy processes of invention, engineering challenges, and design iteration.
Critics note that these models treat technology as an exogenous “black box” or simplify it into production functions, failing to capture historical contingencies or the role of public institutions in breakthroughs (e.g., government-funded basic research).
Engineering and product design, in particular, involve interdisciplinary factors like user feedback and prototyping that don’t lend themselves to equilibrium math.
Workforce Development and Policymaking: Human resource management (HRM) and workforce skills are poorly modeled in mainstream economics, which often treats labor as a homogeneous input in production functions.
Institutional economists argue for a focus on training systems, unions, and social norms, which mainstream approaches undervalue.
Policymaking by governments and national competition strategies (e.g., industrial policy) are critiqued for being sidelined in favor of market-centric views, as seen in debates over “new economic geography” models that struggle with geography-specific institutions and strategies.
The quote’s source itself, from a book on U.S. industrial policy, emphasizes how these elements are central to economic function but resist equations.
These limitations stem from an overreliance on mathematics, which prioritizes deductive rigor over empirical realism. Critics like Deirdre McCloskey and Tony Lawson argue that math enforces unrealistic assumptions (e.g., perfect rationality) and can’t handle open systems where variables aren’t isolable, leading to “scientism” that obscures real-world understanding.
This was evident in the 2008 financial crisis, where models failed to predict systemic risks due to ignored complexities.
Gemini says:
The human element, political factors, and social dynamics that drive things like corporate strategy, government policymaking, and technological innovation are often not easily reducible to a set of equations. For example, a model might predict a certain outcome based on price and demand, but it may not account for a company’s brand loyalty, a CEO’s risk tolerance, or a government’s political motivations.
The statement accurately lists several areas where mainstream economics is often criticized for having weak insight. These include:
Organization and Management: These fields are heavily influenced by psychology, sociology, and leadership, which are not easily captured by standard economic models. The efficiency of a team, for instance, isn’t just about individual incentives; it’s also about communication, trust, and shared purpose.
Corporate Strategy and R&D: A firm’s strategy often involves a mix of long-term vision, risk-taking, and unique competitive advantages that are hard to quantify. Similarly, R&D is an unpredictable process of discovery, not a predictable input-output function.
Government Policymaking: This involves balancing competing interests, political pressures, and social goals that go beyond pure economic efficiency. The “optimal” economic policy might be politically unfeasible or socially unacceptable.
Give me some other sacred incantations that have a similar problem — they sound great in theory, but they suck in reality.
ChatGPT says:
Here’s a short list of “sacred” institutions or values that sound glorious in the abstract but are messy, compromised, or outright dysfunctional in practice:
1. Democracy
Theory: Power resides in the people, leaders are accountable, policies reflect the public will.
Reality: Low voter knowledge, capture by organized interests, minority rule through gerrymandering or electoral quirks. Almost no organizations outside government actually use it to run themselves.
2. Meritocracy
Theory: The most talented and hardworking rise to the top.
Reality: Access to opportunity is heavily shaped by inherited wealth, social networks, and institutional biases; “merit” is often just a veneer over privilege.
3. Free speech
Theory: A robust marketplace of ideas ensures truth wins out.
Reality: Speech is shaped by attention economies, misinformation, and unequal platforms — the loudest or richest often drown out the best arguments.
4. Rule of law
Theory: Everyone is equal before the law; rules apply predictably and impartially.
Reality: Enforcement is selective, resources buy better legal outcomes, and informal power often trumps formal rules.
5. Capitalism / Free markets
Theory: Competition drives efficiency, innovation, and consumer choice.
Reality: Markets tend toward concentration, regulatory capture, and rent-seeking; many key sectors operate as oligopolies or monopolies.
6. Education as a great equalizer
Theory: Schooling levels the playing field and allows social mobility.
Reality: School quality is tied to neighborhood wealth; cultural capital and connections matter as much as degrees.
7. Scientific objectivity
Theory: Science is self-correcting, unbiased, and purely evidence-based.
Reality: Funding incentives, publication bias, and political pressure shape what gets studied, published, and believed.
8. International law
Theory: A framework for resolving disputes peacefully and protecting rights across borders.
Reality: Lacks enforcement against powerful states; compliance is optional for those with enough leverage.
9. Human rights
Theory: Universal, inalienable rights apply to all.
Reality: Enforcement is selective; often invoked to justify interventions while ignored when inconvenient.
10. Diversity & Inclusion
Praise: Diverse perspectives strengthen institutions, innovation, and fairness.
Reality: Often reduced to surface-level representation without changing underlying power structures; can become tokenistic or PR-driven.
11. . International Cooperation / Multilateralism
Praise: Global problems require collective solutions, from climate change to human rights.
Reality: Agreements are often symbolic, unenforceable, or undermined by national self-interest; powerful states ignore rules when convenient.
Here’s the emotional and rhetorical function of liberal pieties — why they stay powerful despite their flaws:
1. Democracy
Emotional function: Gives people a sense of dignity and agency, even if their actual influence is small.
Rhetorical function: Frames opponents as “anti-democratic” if they question institutional arrangements; legitimizes the existing political system.
2. Diversity & Inclusion
Emotional function: Signals moral virtue and fairness; lets individuals and institutions feel progressive without tackling deeper structural problems.
Rhetorical function: Acts as a shield against criticism — “we can’t be unjust, look at our diverse team.”
3. Free Speech
Emotional function: Evokes the romance of fearless truth-telling and the belief that society is open-minded.
Rhetorical function: Used selectively — to defend speech one supports, or to critique the suppression of aligned voices while ignoring suppression of opponents.
4. Education as the Great Equalizer
Emotional function: Offers hope that social mobility is achievable; reassures people that inequality can be fixed without wealth redistribution.
Rhetorical function: Justifies inaction on deeper economic inequality by pointing to education reform as the cure-all.
5. International Cooperation / Multilateralism
Emotional function: Fosters a sense of shared humanity and global solidarity.
Rhetorical function: Allows governments to appear constructive while avoiding costly commitments; criticism of non-cooperation is framed as parochial or reactionary.
6. Human Rights
Emotional function: Appeals to a universal moral standard, providing a sense of moral high ground.
Rhetorical function: Tool of soft power — can justify interventions, sanctions, or pressure against rivals while ignoring allies’ abuses.
7. Scientific Consensus
Emotional function: Provides comfort in uncertainty — there’s an expert-backed “right answer.”
Rhetorical function: Used to shut down debate by framing dissent as anti-science, even when the consensus is partial or evolving.
Here’s the same kind of breakdown for conservative pieties — the values often praised as sacred in conservative rhetoric but messy or contradictory in practice.
1. Free Markets
Emotional function: Promises self-reliance, opportunity, and prosperity through hard work.
Rhetorical function: Used to resist regulation and justify policies favoring business interests; “interference” is framed as anti-freedom, even when markets are already heavily shaped by law and lobbying.
2. Law and Order
Emotional function: Provides a sense of safety, stability, and moral clarity.
Rhetorical function: Can be used to justify harsh policing and punitive laws, often applied unevenly; sidesteps systemic causes of crime.
3. Family Values
Emotional function: Evokes stability, tradition, and belonging.
Rhetorical function: Used to oppose social changes (e.g., same-sex marriage, reproductive rights) while ignoring family breakdown driven by economic factors.
4. Patriotism / National Pride
Emotional function: Gives identity and meaning through shared history and symbols.
Rhetorical function: Criticism of national policy or history can be framed as unpatriotic; used to unify in-group and marginalize dissenters.
5. Religious Faith
Emotional function: Offers moral certainty, purpose, and community.
Rhetorical function: Invoked to sanctify political goals; critics can be cast as anti-religion or immoral, regardless of the policy details.
6. Small Government
Emotional function: Conjures the ideal of individual freedom and personal responsibility.
Rhetorical function: Used to oppose social spending while making exceptions for defense, subsidies, or enforcement that benefit favored groups.
7. Constitutional Originalism
Emotional function: Appeals to timeless wisdom and stability in governance.
Rhetorical function: Selectively applied — historic interpretations are embraced when they match current policy preferences, ignored when inconvenient.
Ian Fletcher and Marc Fasteau write:
* Businesses necessarily make money by exploiting market imperfections because where markets are perfect, competition squeezes profits toward zero. Everybody wants perfect markets when they are the buyer but imperfect markets when they are the seller.
* Industrial policy is government interventions in the economy based on the following propositions.
1. Economic activities differ in value.
2. A laissez faire policy will not maximize a nation’s capture of the most valuable economic activities.
3. Government interventions can enable the capture of more.What America Needs:
1. Expansion of domestic programs designed to support manufacturing, especially in the creation and commercialization of innovation.
2. Controls on international capital flows to drive the dollar down to a value that produces balanced trade – that is, an average of surpluses and deficits close to zero.
3. Tariffs (and occasionally quotas and related policies) to protect specific industries of high economic value, especially in advanced manufacturing.
4. Tariffs (and ditto) to protect industries important for military reasons, for public health, or because they are strategic chokepoints for the whole economy, such as semiconductors.
5. Policies to deny economic and geopolitical adversaries key technologies developed by the US and its allies.