Chapter 27: Economic Policy — Taxes, Spending, Regulation, and the Debate Over Government's Economic Role

"Whenever there is in any country, uncultivated lands and unemployed poor, it is clear that the laws of property have been so far extended as to violate natural right." — Thomas Jefferson, letter to James Madison, 1785

"There is no such thing as a free lunch." — Milton Friedman, There's No Such Thing as a Free Lunch, 1975

Why this chapter is here

Most chapters in this book have been about how American government is structured: the Constitution, the institutions, the processes by which power is allocated and contested. This chapter — and the ones that follow in Part IV — turn from structure to substance. What does the federal government do, in any given policy area? Why does it do it that way, and not some other way? And — most importantly for the design of this book — what are the strongest arguments on each side of the central disagreements?

Economic policy is the right place to start Part IV for two reasons. First, it touches everything: tax rates fund the spending priorities that Chapter 28 covers (healthcare and the safety net), the regulations that Chapter 30 covers (environment and energy), and the immigration and trade decisions in Chapters 31 and 32. The federal government is, before it is anything else, a fiscal entity that moves about $7 trillion a year. Second, the disagreement about how active government should be in the economy is the cleanest, most articulated, longest-running ideological split in American politics. If the book is going to demonstrate its commitment to charitable presentation of competing positions — to steel-manning, in the phrase from the Balance Guide — economic policy is the place where the discipline shows.

This chapter builds on Chapter 16 (the federal budget) but inverts its emphasis. Chapter 16 walked through the mechanics — appropriations, mandatory vs. discretionary, the deficit as fiscal arithmetic. This chapter is about choices. How progressive should the income tax be? Should capital gains be taxed at the same rate as wages? When does antitrust enforcement protect consumers, and when does it chill investment? Is industrial policy a new tool or a return to a pre-1980 norm? These are not technical questions. They are questions about what kind of country the United States is supposed to be.

A note before we begin. This chapter uses two terms — "progressive" and "conservative" — that are imperfect labels in 2026. Many self-identified progressives are skeptical of large parts of the Democratic Party. Many self-identified conservatives are skeptical of large parts of the Republican Party. The terms are still useful as shorthand for two broad clusters of beliefs about the proper role of government in the economy, and the chapter will use them in that broad sense, while noting where the labels strain.

1. The role-of-government debate

1.1 Why this is the central question

Every other question in economic policy depends on a logically prior question: what is government's job? Different answers produce radically different policy menus.

If you think the role of government in the economy is to protect property rights, enforce contracts, and maintain stable money — and to do as little else as possible because every additional intervention introduces distortions, rent-seeking, and reduced incentives for productive activity — your tax policy will be flat or near-flat, your regulatory state will be small, and you will be skeptical of "industrial policy" in any form.

If you think the role of government is to steward broad-based growth, redistribute opportunity, regulate against externalities and market failures, and provide public goods that markets cannot supply efficiently — your tax policy will be more progressive, your regulatory state will be more expansive, and you will see industrial policy as a normal tool of economic management.

Most American voters and most American economists are clustered between these extremes. But the extremes structure the conversation. So we take both seriously, presenting each in the form its strongest advocates would recognize.

1.2 The conservative / classical-liberal / free-market case

The intellectual tradition begins with Adam Smith (The Wealth of Nations, 1776), runs through nineteenth-century classical liberals like John Stuart Mill (in his more market-friendly moments), and is reinvigorated in the twentieth century by Friedrich Hayek (The Road to Serfdom, 1944; The Constitution of Liberty, 1960) and Milton Friedman (Capitalism and Freedom, 1962; Free to Choose, 1980, with Rose Friedman). Modern academic standard-bearers include Glenn Hubbard (Columbia, former chair of George W. Bush's Council of Economic Advisers), Edward Glaeser (Harvard), and Tyler Cowen (George Mason).

The strongest version of the case has four pillars.

Pillar 1: Markets aggregate dispersed information that no central planner can aggregate. This is Hayek's "knowledge problem" (Hayek 1945, "The Use of Knowledge in Society"). The price system is not a clumsy approximation of central planning; it is a mechanism that solves a coordination problem central planning cannot solve. When the price of copper rises, copper users economize, copper producers expand, substitutes come on the market — without any single mind needing to know why copper got scarce. Government interventions that override prices (price controls, large subsidies, restrictive licensing) destroy this information.

Pillar 2: The incentive structure of government is not the incentive structure of markets. Public-choice economics — James Buchanan, Gordon Tullock, The Calculus of Consent (1962) — observes that politicians and bureaucrats respond to political incentives, not necessarily public-interest incentives. A regulation that benefits one organized industry at the expense of unorganized consumers will pass, even if it is net negative, because the industry shows up at the rule-making and the consumers don't. The case is not that government workers are bad people. It is that the institutional incentive structure produces predictable failures — rent-seeking, in the technical term.

Pillar 3: Growth compounds; redistribution is one-shot. A 2-percentage-point increase in long-run growth, sustained for thirty years, produces vastly more wealth at every income level than any plausible redistribution. Tyler Cowen makes this case explicitly in Stubborn Attachments (2018): the moral case for growth is not that the wealthy deserve to be wealthier but that everyone, including the poor, gets richer when the economy grows. Conservatives argue that policies which slow growth — high marginal tax rates, restrictive regulation, large welfare states that reduce labor-force participation — purchase short-run redistribution at long-run cost.

Pillar 4: Property rights and the rule of law are the deep precondition for prosperity. This is the development-economics literature: Daron Acemoglu and James Robinson, Why Nations Fail (2012); Hernando de Soto, The Mystery of Capital (2000). Countries with secure property rights, contract enforcement, and constraints on government predation become rich. Countries without them stay poor. The first job of economic policy, in this view, is not to choose between growth and redistribution but to maintain the institutional conditions under which any economic activity is possible.

The conservative position is not "government should do nothing." Hayek explicitly endorsed a social safety net. Friedman proposed the negative income tax (the intellectual ancestor of the EITC) as a more efficient anti-poverty tool than the existing welfare bureaucracy. The position is that government should do less than the post-1933 American state typically does, and that the burden of proof on each new intervention should be on the intervention's advocates.

1.3 The progressive / Keynesian / social-democratic case

The intellectual tradition begins with the Progressive Era (1890s–1920s) and the New Deal, runs through John Maynard Keynes (The General Theory, 1936), and is reinvigorated by Joseph Stiglitz (Globalization and Its Discontents, 2002; People, Power, and Profits, 2019), Paul Krugman, Mariana Mazzucato (The Entrepreneurial State, 2013), Heather Boushey (chair of Biden's CEA, Unbound, 2019), Emmanuel Saez and Gabriel Zucman, and Dani Rodrik.

The strongest version of the case has four pillars.

Pillar 1: Markets fail in systematic, predictable, well-documented ways. Externalities (pollution, congestion, antibiotic resistance), public goods (basic research, national defense, lighthouse services), information asymmetries (used cars, health insurance, financial products), monopoly and concentrated economic power, coordination failures, and macroeconomic instability — these are not exceptional cases. They are pervasive, and they call for policy responses that markets cannot supply on their own. Stiglitz's career-defining work on information asymmetries (Nobel 2001) established that even small information failures can produce large welfare losses.

Pillar 2: Public investment generates private returns. Mazzucato's The Entrepreneurial State documents that almost every component of the iPhone — touch screen, GPS, internet, voice recognition — traces to publicly funded research (DARPA, NIH, NSF). The interstate highway system, the land-grant universities, the GI Bill, the polio vaccine, the human genome project, the COVID-19 mRNA platform: each began as a public-sector investment with returns far exceeding its costs. The "private sector creates and the government redistributes" framing of the conservative case understates how much of what private actors monetize was first invented at public expense.

Pillar 3: Inequality is not a side effect; it is corrosive to growth and democracy. Joseph Stiglitz's The Price of Inequality (2012) argues that high inequality reduces aggregate demand (the rich save more), reduces investment in human capital (poor children get less education), and concentrates political power in ways that distort policy. Heather Boushey's Unbound makes a similar case empirically. The conservative pillar 3 — growth compounds — is conceded; the progressive response is that the distribution of growth matters too, and that an economy where the top 10% capture most of the gains from growth is not delivering on the promise of broad-based prosperity.

Pillar 4: The post-1980 deregulatory experiment had specific, identifiable costs. The Savings and Loan crisis, the dot-com crash, the 2008 financial crisis, opioid epidemic, persistent wage stagnation in much of the country, the disappearance of pension protections, manufactured-home lending abuses — these were not random events but predictable consequences of allowing financial-sector innovation, market concentration, and regulatory rollback to outrun their oversight. Progressives argue that the answer is not to abolish markets but to reembed them in regulatory and democratic structures that align private incentives with public outcomes.

The progressive position is not "government should run the economy." Stiglitz explicitly endorses markets where they work. Krugman has spent decades defending free trade against protectionist arguments from both left and right. The position is that government should do more, in specific ways, than the post-1980 American policy consensus has done, and that the burden of proof on the absence of government action — when externalities, market failures, or distributional consequences are obvious — should be on those defending the absence.

1.4 The flank positions

The mainstream conservative and mainstream progressive positions, presented above, are within a few standard deviations of where most American politicians and most American economists actually are. Two flank positions sit further out and shape the conversation.

The libertarian / market-fundamentalist position holds that even mainstream conservatism concedes too much to the regulatory and welfare state. The Cato Institute, the Mercatus Center (George Mason), Don Boudreaux's Café Hayek blog, and economists like Tyler Cowen (in some moods) and David Henderson articulate this position. Core commitments: end the federal Department of Education, privatize most of what the federal government currently does directly, abolish the FDA's pre-market approval requirement (allow drugs to enter the market with disclosure rather than approval), liberalize immigration substantially, end agricultural subsidies, end most occupational licensing, and tax consumption rather than income. The position has limited electoral traction but disproportionate influence in policy think-tank work.

The progressive / social-democratic position holds that even mainstream progressivism concedes too much to market institutions that have failed. Jacobin magazine, the Democratic Socialists of America (DSA), economists associated with the Roosevelt Institute and the Levy Economics Institute, and elected officials including Bernie Sanders and (some of) the House "Squad" articulate this position. Core commitments: Medicare for All replacing private health insurance, free public college, a federal jobs guarantee, large-scale public housing investment, sectoral collective bargaining, much higher marginal tax rates on top incomes (50%+ on income above some threshold), wealth taxes on the very rich, and aggressive antitrust to break up concentrated platforms and banks. The position has more electoral traction than the libertarian flank but is a minority within the Democratic coalition.

The flank positions matter for two reasons. First, they sometimes win — the 2017 Tax Cuts and Jobs Act drew on libertarian and Reaganite economic theory; the 2022 Inflation Reduction Act drew on progressive industrial-policy theory. Second, they shape the Overton window: ideas that would once have been beyond consideration become the new mainstream when one of the flanks pulls hard enough.

The chapter will not advocate for any of these positions. It will present each — and the disagreements among them — fairly.

2. Tax policy

The federal government raised about $4.92 trillion in fiscal year 2024 (CBO, Monthly Budget Review, FY 2024 final). The composition: 49% individual income tax, 36% payroll taxes (Social Security and Medicare), 11% corporate income tax, and the remainder from estate, excise, and customs duties. Within these big categories sit some of the most consequential and most-contested choices in American policy.

2.1 The income tax

The federal income tax has progressive marginal rates: as taxable income rises, additional dollars are taxed at higher rates. As of tax year 2024 (for returns filed in early 2025), the seven brackets for a single filer are 10%, 12%, 22%, 24%, 32%, 35%, and 37%. The 37% top rate kicks in at about $609,350 of taxable income for single filers and about $731,200 for joint filers (IRS Rev. Proc. 2023-34). The structure is marginal: a filer with $700,000 in taxable income does not pay 37% on the whole amount but 37% only on the slice above the bracket threshold.

A common misunderstanding is worth addressing here, because students bring it up every semester: most Americans do not understand that "moving up a tax bracket" never causes a worker to take home less in absolute terms. The marginal rate applies only to the next dollar earned. Surveys consistently find that about a third of voters believe otherwise (Tax Foundation polling, 2024).

Above the rate structure, the actual amount any household pays depends heavily on two further choices: the standard deduction and the itemized deduction alternative. The 2017 Tax Cuts and Jobs Act (TCJA) approximately doubled the standard deduction (to $14,600 for single filers and $29,200 for joint filers in 2024) while capping or eliminating several itemized deductions. The result: the share of filers itemizing dropped from about 30% in 2017 to about 10% in 2018 (IRS Statistics of Income, 2018–2022 series). Most filers now take the standard deduction.

The TCJA's individual provisions are scheduled to sunset at the end of 2025. Unless Congress acts — and as of this writing, the post-2024-election Congress is debating whether to extend, modify, or let some sunset — the lower marginal rates, larger standard deduction, and SALT cap revert to pre-2017 law. This sunset provision, sometimes critiqued as a budgetary maneuver to fit the legislation within reconciliation rules, has become a forcing function for major tax legislation. Case Study 1 of this chapter examines the TCJA's record in detail.

2.2 Capital gains

Income from selling appreciated assets (stocks, real estate, art) is taxed at lower rates than ordinary wage income — historically by a substantial margin. The current long-term capital-gains rate (for assets held more than a year) is 0%, 15%, or 20% depending on income, plus the 3.8% Net Investment Income Tax for higher earners. Compare these to the top ordinary-income rate of 37%.

This is a contested choice. The conservative defense rests on three arguments: (1) corporate profits are already taxed at the corporate level, so taxing the same profits again as capital gains at the shareholder level constitutes double taxation; (2) lower rates on long-term gains reward patient investment, reducing the bias toward short-term churning; (3) much of nominal capital gains is inflation, and taxing the inflation portion at the same rate as real wage income overstates the real income being taxed.

The progressive critique is also threefold: (1) a tax-rate gap between wages (taxed at up to 37%) and capital gains (taxed at up to 23.8% with NIIT) systematically favors households whose income comes from owning assets over households whose income comes from working; (2) the corporate-tax double-taxation argument assumes corporate taxes are actually paid at statutory rates, which they often are not (effective rates are much lower); (3) the inflation argument can be addressed directly through indexation, without lowering the rate. Stiglitz, Saez, and Zucman have argued at length that the capital-gains preference is the single largest source of tax-driven inequality in the federal code.

The empirical record is genuinely mixed. Studies do not show a clean relationship between capital-gains rates and aggregate investment (the conservative "lower rates spur investment" claim has weaker empirical support than political rhetoric suggests). But studies also do not show that capital-gains preferences mostly benefit the very wealthy in ways unrelated to investment — a substantial share of long-term gains accrues to middle-income retirees selling decades-held assets. This is one of the cases where the strongest claims of both sides are stronger than the data supports.

2.3 The estate tax

The federal estate tax applies to transfers at death above an exemption threshold. As of 2024 the threshold is about $13.6 million per individual ($27.2 million per couple), indexed to inflation, and the rate above the threshold is 40%. The TCJA roughly doubled the exemption from its pre-2017 level; this doubling sunsets after 2025 absent Congressional action.

About 99.7% of decedents owe no federal estate tax (Joint Committee on Taxation, 2023). The total revenue raised — about $30 billion in FY 2023 — is small as a share of the federal budget.

Despite the small fiscal footprint, the estate tax is one of the most politically charged taxes. The conservative position frames it as the "death tax": a moral wrong because it taxes the same wealth twice (during the lifetime that built it and again at death) and threatens to force liquidation of family farms, family businesses, and other illiquid holdings. The progressive position frames it as a tax on dynastic wealth: among the few mechanisms in the U.S. tax code that prevent intergenerational concentration of fortunes, and among the few that get to wealth at all (most of the system taxes income, not wealth).

The "family farms" argument deserves attention because it has been politically powerful and is empirically contested. Tax Policy Center analyses (most recently 2023) estimate that about 80 small farm or business estates owe any federal estate tax in a given year, and the number actually forced to liquidate to pay the tax is far smaller. The American Farm Bureau Federation contests this — they argue the headline number underestimates indirect effects on farm planning, succession, and credit. The honest summary: the estate tax does affect a small number of family farms, but the political prominence of the family-farm framing exceeds its quantitative importance. Both observations are true.

2.4 Corporate income tax

The federal corporate tax rate is 21%, lowered from 35% by the TCJA in 2017. The 21% is the statutory rate; the effective rate that companies actually pay is generally lower because of credits, deductions, and accelerated depreciation. Joint Committee on Taxation analyses put the effective rate for large profitable corporations in the 13–17% range over recent years, with substantial variation across industries.

The TCJA also introduced new international tax rules. GILTI (Global Intangible Low-Taxed Income) imposes a minimum tax on certain foreign earnings of U.S. multinationals. The 2022 Inflation Reduction Act added a 15% domestic minimum tax (the Corporate Alternative Minimum Tax, or CAMT) on financial-statement income of corporations with $1 billion+ in average annual income. And separately, the OECD Pillar Two framework — which the Biden Administration negotiated and which the Trump 2.0 administration has signaled it may withdraw from — would impose a 15% global minimum on multinational corporate income across signatory jurisdictions.

The conservative case for the 21% rate and the broader 2017 corporate-tax restructuring: prior to 2017, the U.S. statutory rate of 35% was the highest in the OECD, creating incentives for corporate inversions (where a U.S. firm "moves" its tax domicile abroad through merger), profit-shifting to low-tax jurisdictions, and reduced domestic investment. The 2017 reform brought the U.S. into the middle of the OECD pack and was followed by some real wage growth and increased domestic capital expenditure (though the size and durability of these effects is contested).

The progressive case for higher corporate taxes (or at least for closing more loopholes): corporate profitability hit record highs in the late 2010s and early 2020s while corporate tax revenue as a share of GDP fell. The corporate-tax burden was among the largest TCJA giveaways and went disproportionately to shareholders (Mertens and Ravn 2013 / CBO retrospective 2022). In the progressive view, the post-2017 corporate-tax structure is a transfer from the government to capital owners that has produced limited offsetting growth.

Like the capital-gains debate, this one features stronger rhetoric than the data supports on both sides. The 35% rate was unsustainable for the reasons conservatives stated; the 21% rate has not produced the growth surge its advocates predicted. The middle-ground position — that the corporate rate should be in the low-to-mid 20s with closed loopholes — is what most academic public-finance economists, regardless of party, actually recommend.

2.5 Tax expenditures

A tax expenditure is a deduction, exclusion, credit, or preferential rate that reduces tax liability for some category of activity. From a budget-accounting perspective, tax expenditures are economically equivalent to direct spending: the government foregoes revenue it would otherwise collect, just as it could send the same dollar amount as a spending check. The Joint Committee on Taxation publishes a tax-expenditures list annually.

Tax expenditures totaled about $1.7 trillion in FY 2024 (JCT, 2024) — a number larger than the entire non-defense discretionary budget. The largest items:

• Exclusion of employer-paid health insurance (~$300 billion). Employees are not taxed on the value of employer-provided health coverage. This is the single largest tax expenditure. It is also the policy choice that produced the U.S. employer-based health insurance system after WWII.
• Capital-gains and dividend preferential rates (~$200 billion).
• Retirement plan contributions (~$200 billion). 401(k), IRA, and similar accounts allow contributions to be excluded from current-year income.
• Mortgage interest deduction (~$30 billion, down from much higher levels pre-TCJA, which capped the deduction).
• Charitable contribution deduction (~$60 billion).
• Earned Income Tax Credit (EITC) and Child Tax Credit (CTC) (~$80 billion combined). These are refundable credits — they pay out even to filers with no tax liability — and so function more like direct anti-poverty spending than ordinary tax preferences.
• State and local tax (SALT) deduction, capped at $10,000 by the TCJA. The cap is set to expire with the rest of the TCJA individual provisions at the end of 2025.

The size of the tax-expenditure system is one of the strongest practical arguments for tax simplification, and one of the points of unusual cross-ideological agreement. Both the conservative and progressive critiques converge on the observation that the tax code is being used to subsidize an enormous range of specific activities, often with regressive distributional consequences (the mortgage-interest deduction, before the TCJA cap, mostly benefited high-income homeowners). Where the agreement breaks down is on which expenditures to cut.

3. Spending policy

3.1 The mandatory-discretionary split

Chapter 16 covered the budget mechanics. Recap: mandatory spending (Social Security, Medicare, Medicaid, federal employee retirement, SNAP, refundable tax credits, interest on the debt) flows automatically based on eligibility rules and benefit formulas set in law; it requires no annual appropriation. Discretionary spending (defense, education, transportation, science, foreign aid, environmental enforcement, etc.) flows only when Congress passes the annual appropriations bills.

In FY 2024, mandatory spending was about 64% of federal outlays. Discretionary was about 26%. Net interest on the debt was about 13% (this last figure has been climbing rapidly since 2022 as interest rates rose; CBO projects it will exceed defense spending by 2026 if current rates persist).

The political consequence is profound: the most consequential drivers of the federal budget — Social Security, Medicare, Medicaid — are not on the table in most years. They run on autopilot. Congress fights over the discretionary 26%. This is why "balanced budget" proposals that exempt Social Security, Medicare, Medicaid, and defense (the typical political package) cannot mathematically balance the budget — there is too little non-protected spending.

3.2 Mandatory programs and reform debates

Both parties have positions on entitlement reform; neither party in office actually moves on it. The structural problem: the programs are genuinely popular, the people receiving benefits vote, and the demographic math (more retirees per worker) means that without reform, scheduled benefits cannot be paid in full beyond about the mid-2030s for Social Security and the late 2020s/early 2030s for the Medicare Hospital Insurance Trust Fund (Social Security Trustees Report 2024; Medicare Trustees Report 2024).

The conservative reform menu: raise the retirement age (slowly, with adjustment for life expectancy), means-test benefits more aggressively at the top, allow younger workers to direct part of their payroll-tax contribution into private retirement accounts, change cost-of-living adjustment formulas to slow benefit growth, premium-support models for Medicare. Paul Ryan's Path to Prosperity (2011 and successors) is the textbook articulation.

The progressive reform menu: raise or eliminate the cap on Social Security payroll taxes (currently $168,600 in 2024); raise the payroll tax rate; expand Medicare eligibility (lower the age, broaden services); transition Medicaid toward a federal floor that is uniform across states; add a public option to the ACA marketplaces that competes with private insurers. Social Security 2100, sponsored by John Larson (D-CT), is the textbook articulation.

The political reality: substantial reform of either kind requires bipartisan cover. The 1983 Social Security reforms (Greenspan Commission, signed by Reagan) are the model. No comparable bipartisan effort has succeeded since. Each cycle brings the math closer to the cliff.

3.3 Discretionary spending: defense and non-defense

Discretionary outlays in FY 2024 were about $1.7 trillion. Roughly 50% was defense; 50% was non-defense (CBO classifications). Within non-defense, the large categories are veterans' benefits, education, transportation, housing assistance, and basic research.

The defense-discretionary breakdown gets political each cycle, because the post-2011 Budget Control Act framework — and the various continuing resolutions and "deals" that have managed around it — has tied defense-spending caps to non-defense-spending caps. The result has been recurring conflicts where conservatives want higher defense and lower non-defense, while progressives want the opposite, and the resulting deal raises both (or, as in 2025-2026, neither, while continuing-resolution mechanics keep the government open).

3.4 Industrial policy

For perhaps three decades after about 1980, "industrial policy" was a term of derision in mainstream American economic discourse. The bipartisan consensus, articulated by economists across the ideological spectrum, was that government attempts to "pick winners" in specific sectors of the economy generally produced worse outcomes than letting markets allocate capital. This consensus was not absolute — defense procurement, agricultural support, and some research subsidies always existed — but the rhetoric was strongly against active sectoral targeting.

That consensus has substantially eroded. The shift began with the Trump administration's 2018 tariffs on steel and aluminum (Section 232) and on Chinese imports (Section 301), which were industrial policy by another name. It accelerated during COVID-19 (vaccine development through Operation Warp Speed; semiconductor shortages exposing supply-chain dependencies; renewed concern about pharmaceutical-API manufacturing). It crystallized in the Biden administration's three signature laws:

• The CHIPS and Science Act (2022) — about $52 billion in subsidies for U.S. semiconductor manufacturing, plus research investment. This passed with substantial bipartisan support (64 senators).
• The Inflation Reduction Act (2022) — climate, healthcare, and tax legislation; the climate provisions (about $370 billion in tax credits over ten years) constitute a major industrial-policy push toward domestic clean-energy manufacturing.
• The Bipartisan Infrastructure Law (2021) — about $1.2 trillion authorized, $550 billion in new spending, on physical infrastructure.

The Trump 2.0 administration has continued or expanded several elements of this push (tariffs, particularly on China; Buy American provisions; some industrial subsidies) while reversing others (rollback of clean-energy tax credits is in active legislative debate as of 2026). Whatever the outcome of these specific provisions, the broader pattern is striking: the United States now openly pursues industrial policy as a normal tool of economic management for the first time since the 1970s.

The intellectual case for industrial policy has likewise been rebuilt. Mariana Mazzucato's work, mentioned above, is the most-cited academic argument. Dani Rodrik's writing on "smart" industrial policy is influential among practitioners. Even market-oriented economists who would not have endorsed industrial policy in the 2000s (Tyler Cowen, for example) have written more sympathetically about it for national-security-relevant sectors since 2018.

The conservative critique remains: industrial policy is vulnerable to capture, picks winners that markets would not pick for good reasons, and tends to produce permanent constituencies that prevent later course correction. The progressive critique of current industrial policy: the IRA's tax-credit structure delivers most of the benefit to large corporations, with limited assurance that the production it subsidizes would not have happened anyway. The honest summary: industrial policy is back, both parties have at least partially endorsed it, the specific design choices are contested across multiple dimensions, and the empirical record is too short to evaluate.

4. Regulatory policy

The regulatory state was the subject of Chapter 11. This section is narrower: the substantive economic content of major regulatory traditions, and the ideological argument about each.

4.1 What economic regulation does

Federal economic regulation operates in roughly six domains:

• Financial regulation. Banking (Federal Reserve, OCC, FDIC); securities (SEC); commodities (CFTC); consumer financial products (CFPB). Post-2008 Dodd-Frank significantly expanded the regulatory perimeter; the 2017 Trump administration regulatory rollback partially trimmed it.
• Environmental regulation. EPA, primarily — Clean Air Act, Clean Water Act, NEPA review; FERC for energy infrastructure. Discussed at length in Chapter 30.
• Labor regulation. NLRB (collective bargaining); OSHA (workplace safety); Wage and Hour Division (Fair Labor Standards Act enforcement).
• Consumer protection. FTC (advertising, antitrust, unfair practices); CPSC (product safety); FDA (drugs, food, devices).
• Antitrust. DOJ Antitrust Division; FTC Bureau of Competition. Treated separately below because it is the most ideologically interesting.
• Sector-specific regulation. FCC (communications), DOT (transportation), FAA (aviation), HUD (housing finance), and many more.

4.2 Cost-benefit analysis

Since the Carter administration, and reinforced by Reagan's 1981 Executive Order 12291 and Clinton's 1993 Executive Order 12866 (which remains in force, modified), federal agencies have been required to conduct cost-benefit analyses for major rules. The Office of Information and Regulatory Affairs (OIRA), within OMB, oversees this review.

Cost-benefit analysis is one of the rare procedural innovations that has bipartisan elite support. Cass Sunstein, who ran OIRA under Obama, and Susan Dudley, who ran it under George W. Bush, agree on most of the methodology even when they disagree on individual rules.

The progressive critique of cost-benefit analysis: it systematically undervalues benefits that are diffuse or hard to monetize (clean air, biodiversity, distributional fairness) while overvaluing costs that are concentrated and easy to monetize (compliance burden on a specific industry).

The conservative critique: cost-benefit analysis as practiced often understates compliance costs and uses inflated assumptions for benefits (the "value of statistical life" multiplier, in particular).

Both critiques have force. The methodology has improved over time and remains imperfect.

4.3 Banking and securities

The post-2008 Dodd-Frank Wall Street Reform and Consumer Protection Act (2010) is the largest piece of financial regulatory legislation since the New Deal. Key provisions: the Volcker Rule (limiting proprietary trading by federally insured banks); the Consumer Financial Protection Bureau (CFPB); resolution authority for failing systemically important financial institutions; central clearing for most derivatives; stress testing for large banks.

The 2018 Economic Growth, Regulatory Relief, and Consumer Protection Act partially rolled back Dodd-Frank for mid-sized banks (raising the systemically-important threshold from $50 billion to $250 billion in assets) — a rollback that critics argue contributed to the 2023 Silicon Valley Bank collapse, while defenders argue the SVB failure was idiosyncratic to that bank's interest-rate-risk management.

This is a clean example of where the data is contested but the structural argument is clear. The progressive position: financial deregulation correlates historically with crisis (Glass-Steagall repeal 1999 / 2008 crisis; S&L deregulation 1980s / S&L crisis), and post-2008 reforms should be preserved or extended. The conservative position: Dodd-Frank imposed disproportionate costs on smaller banks and may have contributed to industry concentration, which was the opposite of the legislation's goal.

4.4 Environmental, labor, consumer protection

These traditions are covered more fully in subsequent chapters:

• Environmental regulation (NEPA, Clean Air Act, Clean Water Act, Endangered Species Act): Chapter 30.
• Labor regulation (NLRA, OSHA, FLSA): partially Chapter 28.
• Consumer protection (FDA, CPSC, FTC consumer-protection authority): partially Chapter 28.

For the purposes of this chapter, note that each tradition has been the subject of partial deregulation under Republican administrations (Reagan, George W. Bush, Trump 1, Trump 2) and partial expansion under Democratic administrations (Carter, Clinton, Obama, Biden), with the long-run trend a slow accretion of regulatory scope punctuated by partial rollbacks.

5. Antitrust: the Chicago School and the neo-Brandeisians

Antitrust law has been one of the most important intellectual battlefields in U.S. economic policy since the 1970s. The current debate is the closest thing to an open ideological war in academic and policy economics.

5.1 The Chicago School consensus

For most of American antitrust history before the 1970s, courts applied a structural standard: large market share, especially when paired with conduct that excluded competitors, could be challenged on its own. The 1911 breakup of Standard Oil and the 1980s breakup of AT&T are exemplars.

In the 1970s and 1980s, scholars associated with the University of Chicago Law School and Department of Economics — Robert Bork (The Antitrust Paradox, 1978), Richard Posner, George Stigler, Frank Easterbrook, William Landes — successfully argued that this structural approach had become incoherent. Their alternative: antitrust should focus on consumer welfare, narrowly defined, and should intervene only when conduct produces measurable harm to consumers (typically, higher prices). Mere bigness is not bad; firms get big, in this view, by being efficient, and harming them harms consumers.

By the 1990s, federal courts had largely adopted the Chicago School framework. The 2001 Microsoft case (United States v. Microsoft) was prosecuted under it, with mixed outcomes. The 2018 attempt to block the AT&T / Time Warner merger (vertical integration) failed under it. By the late 2010s, the Chicago consensus was so well-established that critics on both left and right began arguing it had ossified into a presumption against intervention.

5.2 The neo-Brandeisian challenge

Beginning around 2017, a group of younger scholars and activists began arguing that the Chicago consumer-welfare standard was incomplete. The intellectual leaders include Lina Khan (whose 2017 Yale Law Journal note "Amazon's Antitrust Paradox" is the canonical text), Tim Wu (Columbia, The Curse of Bigness, 2018), Barry Lynn (Open Markets Institute), and Zephyr Teachout (Fordham, Break 'Em Up, 2020). The label "neo-Brandeisian" refers to Justice Louis Brandeis, who in the early twentieth century argued that concentrated economic power was a threat to political democracy as well as to consumer welfare.

The neo-Brandeisian arguments are several:

• The consumer-welfare standard, narrowly defined as "lower prices," misses important harms: reduced innovation, suppressed wages, political-economic concentration, weakened bargaining power of suppliers and workers.
• Modern platforms (Amazon, Google, Apple, Meta, Microsoft) often offer "free" services, so price-based analysis fails.
• Vertical integration (Amazon owning third-party sellers' platform AND competing with them as a seller; Google owning search AND advertising tools) creates conflicts that the Chicago framework systematically discounts.
• The Chicago presumption against intervention has become a self-fulfilling prophecy: by not enforcing, the government has allowed concentration that now requires intervention to undo.

Khan was nominated by Biden and confirmed as FTC chair in 2021. From 2021 to 2024, the FTC under her leadership pursued an aggressive enforcement agenda: a challenge to the Microsoft / Activision Blizzard merger (lost on the merits, twice); the Amazon antitrust case (filed September 2023; ongoing); the Meta case (ongoing); challenges to non-compete clauses; updates to merger guidelines (2023, jointly with DOJ).

5.3 The conservative critique of neo-Brandeisianism

The strongest version of the conservative critique:

• The Chicago framework is not arbitrary; it is the result of decades of empirical and theoretical work that established the limits of structural antitrust as actually administered.
• Neo-Brandeisian theories of harm — "innovation harm," "labor-market concentration," "political-economic power" — are real concerns but not amenable to legal enforcement under existing statutes. Expansion would require Congress to act, not the FTC.
• Aggressive enforcement creates regulatory uncertainty that itself chills investment, particularly in M&A activity that may be procompetitive (acquirer brings management talent, acquirer integrates the target into a more efficient supply chain, etc.).
• Several of the FTC's high-profile cases have been losses on the legal merits, suggesting the theories are outpacing the law.

The 2024 FTC win-loss record on its merger challenges was poor by historical standards (Microsoft/Activision: lost. Tapestry/Capri: won, but the parties terminated. Other cases mixed.). Trump 2.0's FTC has signaled a return toward Chicago-influenced standards, though the specifics depend on the chair (Andrew Ferguson, confirmed 2025).

5.4 The progressive defense

The strongest version of the progressive defense:

• The Chicago consumer-welfare standard was never the only legitimate reading of the antitrust statutes; the Sherman Act and Clayton Act reach broader conduct than narrow consumer-welfare doctrine recognizes.
• Industrial concentration has measurably increased across most U.S. sectors since 2000 (Grullon, Larkin, and Michaely 2019; Council of Economic Advisers Issue Brief 2016 and 2023). Whatever the proper legal response, the underlying empirical claim is well documented.
• Cases sometimes lose on the merits as part of building precedent for future cases. The Standard Oil and AT&T precedents were not won on the first try.
• Political-economic concerns about concentrated power are legitimate even when not fully captured by current statutes; legislative reform (such as the Open App Markets Act, repeatedly proposed) could fill gaps. Aggressive enforcement under existing statutes is one of the legitimate ways to surface these gaps.

Case Study 2 of this chapter examines the 2021–2024 FTC experiment in detail. Both critiques and defenses have purchase. The chapter does not adjudicate.

6. Macroeconomic policy

6.1 Fiscal policy

Fiscal policy refers to the use of government spending and taxation to influence aggregate economic output. The classic Keynesian prescription: in a recession, expand fiscal policy (increase spending, cut taxes, run a larger deficit); in an overheating economy, contract it (raise taxes, cut spending, run smaller deficit or surplus).

In practice, U.S. fiscal policy is poorly suited for countercyclical use. The legislative process is too slow; by the time Congress responds to a recession, the recession is often ending. The political process is asymmetric: it is much easier to expand than to contract, so deficits accumulate. The 2008–2009 American Recovery and Reinvestment Act (Obama) and the 2020–2021 CARES Act / American Rescue Plan (Trump and Biden) are examples of large fiscal interventions that did appear to help; their critics (on growth grounds and on inflation grounds, respectively) have substantial points to make as well.

The 2022 inflation surge illustrates the fiscal-policy challenge. Most academic analyses (Federal Reserve Bank of San Francisco, 2022; Bernanke and Blanchard, 2023) attribute about a third of the inflation to demand-side factors (including fiscal stimulus exceeding the size of the output gap), with the remainder to supply-side factors (energy, supply chains, post-COVID labor reallocation). The exact decomposition is contested.

6.2 Monetary policy and the Federal Reserve

Monetary policy — the management of money supply and interest rates — is the responsibility of the Federal Reserve. The Fed is technically independent within government: its Board of Governors is appointed by the President with Senate confirmation, but its operational decisions are not subject to executive override.

The case for Fed independence: monetary policy operates on a timeline (effects show up 12–24 months after policy changes) that does not match the political cycle. Politicians have systematic biases toward easier money (lower rates, more accommodation) than economic conditions justify, because easier money is popular in the short run. Cross-country empirical work (Alesina and Summers 1993; subsequent literature) found that more independent central banks produced lower inflation without sacrificing growth.

The case against unlimited Fed independence: monetary policy has substantial distributional consequences (interest rates affect borrowers and savers differently; quantitative easing inflates asset prices held disproportionately by the wealthy), and decisions of this magnitude should answer to democratic processes. Some progressive economists argue for a more democratically accountable central bank; some conservative economists argue the Fed has overstepped its mandate.

Recent dynamics: the Fed engaged in unprecedented quantitative easing (QE) after the 2008 financial crisis, dramatically expanding its balance sheet from about $900 billion to about $4.5 trillion. After a brief tapering, COVID prompted further expansion, peaking at about $9 trillion. The 2022 inflation surge prompted the Fed to begin a rapid rate-hiking cycle, raising the federal funds rate from near zero to a 5.25–5.50% target by mid-2023. As of 2024–2025, the Fed has begun a measured rate-cutting cycle as inflation has moderated. The Trump 2.0 administration has been publicly critical of the Fed Chair (Jerome Powell) and the question of Fed independence has become more politically salient than at any time since the early 1980s.

6.3 Trade policy

Trade policy — tariffs, free-trade agreements, the WTO — was, for the period from roughly 1990 to 2016, an area of unusual bipartisan elite consensus. The Washington Consensus — open trade, capital mobility, financial-sector liberalization, exchange-rate flexibility — was the official position of the U.S. government, the IMF, and the World Bank. NAFTA (1994), the WTO (1995), and China's accession to the WTO (2001) were the high-water marks.

That consensus has substantially eroded — first on the populist right, then on the populist left, and increasingly across both party establishments. Several causes:

• The "China shock" literature (Autor, Dorn, Hanson 2013 and successors) documented that the post-2001 surge in Chinese imports caused larger and more persistent local labor-market dislocations in the U.S. than trade theory had predicted.
• National-security concerns about supply-chain dependencies, especially in semiconductors, pharmaceuticals, rare earths, and defense-relevant manufacturing, have moved up the policy agenda.
• The political-economic case — that the gains from trade, while real in aggregate, were unevenly distributed across regions, occupations, and demographic groups in ways that the existing safety net did not adequately compensate — has won broader assent.

The Trump 1 administration imposed Section 301 tariffs on a broad range of Chinese goods beginning in 2018. The Biden administration retained most of these tariffs and added some (electric vehicles, batteries, solar, semiconductors). The Trump 2 administration has expanded tariffs further, with broader reach across countries and sectors. Whatever one's view of whether tariffs are good policy, the empirical fact is that U.S. trade policy is now substantially more restrictive than at any time in the post-WWII period, and that this shift has bipartisan acquiescence even where it lacks bipartisan endorsement.

The empirical record on the Trump 1 tariffs: peer-reviewed studies (Amiti, Redding, and Weinstein 2019; Fajgelbaum et al. 2020) estimated that the burden of the tariffs fell substantially on U.S. consumers and downstream producers, with modest welfare costs in aggregate but concentrated impacts in specific industries. A subsequent 2024 analysis by some of the same authors found that the political-economy effects (the bargaining and signaling value vis-à-vis China) were positive enough to be worth examining seriously, even granting the welfare costs. The honest summary: tariffs imposed real costs on Americans, the Trump 1 China tariffs produced little measurable change in Chinese behavior, and yet the political logic for continuing them remained strong enough that both the Biden and Trump 2 administrations retained or expanded them. This is exactly the kind of policy question where empirical and political-economic considerations point in different directions.

7. Inequality

7.1 The empirical landscape

Income inequality in the United States, by most measures, increased substantially between 1980 and roughly 2010, was relatively stable from about 2010 to 2020, and has shown mixed signals since. The most-cited measures:

• The Gini coefficient — a summary statistic of distributional inequality, ranging from 0 (perfect equality) to 1 (perfect concentration). U.S. household-income Gini was about 0.40 in 1980 and about 0.49 in 2020 (Census Bureau, P-60 series). Comparable measures for peer democracies place the U.S. above all other OECD countries except Mexico and Chile.
• The top 1% income share (Piketty, Saez, Zucman) — the share of total income going to the top 1% of households. The U.S. figure was about 10% in 1980 and rose to about 19% by the mid-2010s, by Piketty-Saez methods.
• The top 1% wealth share — the analogous measure for wealth (assets minus liabilities) — is even more concentrated, with the top 1% holding about 30% of household wealth (Federal Reserve Survey of Consumer Finances, 2022).

These figures, however, are subject to significant measurement disputes. The 2024 paper by Auten and Splinter ("Income Inequality in the United States: Using Tax Data to Measure Long-term Trends," Journal of Political Economy) re-examined the Piketty-Saez series with adjusted assumptions about tax-unit definitions, transfer income, and unreported income, and concluded that top-1% income share rose by perhaps half as much as Piketty-Saez found — from about 9% to about 12%, rather than to 19%. The Auten-Splinter analysis has not displaced the Piketty-Saez series in academic discourse, but it is now part of the debate. The honest summary: income inequality has clearly risen, but by exactly how much depends on measurement choices that economists genuinely disagree about.

7.2 Mobility

A separate question from static inequality is intergenerational mobility: how strongly does a child's economic position predict their adult economic position? The classic finding (Chetty, Hendren, Kline, Saez 2014 and subsequent updates from the Opportunity Insights team) is that intergenerational mobility in the U.S. is lower than in most peer democracies (Canada, Denmark, much of Europe). The U.S. is also unequal across geography: a child born to low-income parents in some U.S. metropolitan areas has substantially better prospects than a child born to similar parents elsewhere.

The mobility data has produced a degree of bipartisan agreement that the static inequality data has not. Both progressives and conservatives can read the mobility findings and conclude that the U.S. is failing on a value (equal opportunity, broad-based prosperity) that both sides endorse. Where disagreement returns: what to do about it.

7.3 Causes

The academic literature identifies multiple causes of rising inequality, none individually decisive:

• Skill-biased technical change. Technology has raised the wage premium for high-skill workers and depressed wages for routine labor (Autor 2014; Goldin and Katz 2008).
• Globalization. Trade and offshoring put downward pressure on wages for workers in import-competing industries (the China-shock literature).
• Decline of unions. Unionization in the private sector fell from about 35% in the 1950s to about 6% in 2024. Union wage premiums and the within-firm wage compression unions produced have fallen with them.
• Tax-policy changes. Top marginal rates have fallen substantially since the 1970s (the top federal income-tax rate was 70% in 1980, 28% by 1988, 39.6% in 2000, 37% currently). Capital-gains rates have likewise fallen.
• Educational sorting. Returns to college have risen; gaps between college-educated and non-college-educated workers have widened.
• Geographic sorting. High-wage industries cluster in high-cost metros; the gap between wages in high-productivity metros (Bay Area, NYC, Boston) and elsewhere has grown.
• Family-structure change. Marriage and family-formation patterns have diverged sharply by education and income (Murray 2012; Putnam 2015; Cherlin, in some moods).

Different ideological perspectives weight these causes differently. The progressive emphasis tends to fall on tax-policy changes, declining union power, and declining real minimum wages. The conservative emphasis tends to fall on skill-biased technical change, family-structure change, and the unintended consequences of welfare-state policies that reduce labor-force participation. The empirical literature does not pick a single decisive cause; the magnitudes are contested.

7.4 Policy responses

The progressive policy menu emphasizes redistributive taxation (higher rates on top incomes, wealth taxes, expanded EITC and CTC, paid leave), labor-market interventions (higher minimum wages, sectoral bargaining, expanded NLRB protections), social investments (universal pre-K, expanded Pell grants, public-option health insurance), and direct public provision of housing and childcare.

The conservative policy menu emphasizes growth-oriented tax policy (lower marginal rates, broader base), supply-side reforms in housing and licensing (zoning reform, occupational-license relaxation), pro-family policies (expanded child tax credits, marriage-neutral safety nets), and human-capital investment (school choice, vocational education, work requirements that maintain attachment to the labor force).

There is more overlap in this list than political rhetoric suggests. Both parties endorse some version of the EITC. Both parties have endorsed some form of child tax credit. Both parties endorse zoning reform in some venues (the YIMBY movement has cross-party endorsement). The disagreements are about magnitude and design as much as about direction.

8. Contested questions

8.1 Minimum wage

The federal minimum wage is $7.25 per hour, unchanged since 2009. Many states and cities have higher minimums; California's is $16.50 (2025), Washington's is $16.66, New York City's is $16.50. Twenty states have higher minimums than the federal floor; thirty maintain the federal floor.

The empirical literature on minimum wages has evolved substantially. Through about the early 1990s, the textbook prediction — minimum wages above the market-clearing wage cause unemployment — was the consensus, supported by time-series evidence and basic labor-market theory.

The 1994 Card and Krueger study of New Jersey's minimum-wage increase (using Pennsylvania as a control) found no measurable employment effect from a moderate increase. This produced two decades of academic controversy and methodological refinement.

The current state of evidence (Cengiz, Dube, Lindner, Zipperer 2019; Manning 2021; Neumark and Wascher in updated analyses; Congressional Budget Office 2021 minimum-wage assessment): moderate increases in minimum wages — say, raising the federal floor from $7.25 to $10 or $12 — have small employment effects in most labor markets, with some redistribution from employers and consumers to low-wage workers. Large increases — particularly in low-wage regions where the new floor binds for a high share of workers — have larger employment effects, with the magnitude depending on local conditions.

This evolution illustrates how empirical disagreement should be discussed in a textbook. The 1980s consensus was not wrong about the direction of effects; it was wrong about the magnitude. Modern monopsony-influenced labor models (where employers have wage-setting power because workers face frictions in switching jobs) explain why moderate increases can be welfare-enhancing while large increases are not. The honest summary: the federal minimum wage is low enough that an increase to $10–$12 would have small employment effects and modestly redistribute income; an increase to $20–$25 in a low-wage region would have larger and more uncertain effects.

8.2 Healthcare (preview)

Healthcare policy is the subject of Chapter 28. Brief preview: the Affordable Care Act (2010) is the major recent statute; about 92% of Americans now have health coverage, the highest rate in U.S. history; expanded Medicaid eligibility in states that took the option is the largest single source of post-ACA coverage gains; per-capita health spending in the U.S. remains roughly twice that of peer democracies, with mixed health outcomes.

8.3 Climate (preview)

Climate and energy policy is the subject of Chapter 30. The IRA's $370B clean-energy package, the Trump 2 partial rollback, and the underlying scientific and economic landscape are treated in detail there.

8.4 Immigration (preview)

Immigration policy is the subject of Chapter 31. Economic effects: most academic studies find that immigration produces modest aggregate gains, with concentrated effects on workers in the most-affected segments of the labor market and modest fiscal effects depending on the immigrant population.

8.5 Trade and tariffs (recap)

Section 6.3 above covered trade. Brief recap of the empirical landscape: aggregate gains from trade are substantial; distributional effects are concentrated; tariffs on intermediate goods raise prices and reduce competitiveness for downstream U.S. producers; political-economy considerations about supply-chain resilience and bargaining leverage have moved policy in a more restrictive direction across both parties.

9. Where this chapter leaves us

A reader who finishes this chapter should be able to do four things:

1. Articulate the strongest version of both the progressive and conservative cases for the role of government in the economy. Not the talk-radio versions; not the worst-of-Twitter versions. The Stiglitz version and the Hayek version. The Mazzucato version and the Cowen version.
2. Read a tax bill or a regulatory rule and identify the specific design choices it embeds. Marginal rates, deduction structure, expense classification, sunset provisions, cost-benefit analysis, regulatory perimeter — these are the components of any actual policy.
3. Distinguish empirical from normative claims. "The 2017 corporate tax cut delivered most of its benefits to shareholders" is empirical, with a literature behind it. "The 2017 corporate tax cut was bad policy" is normative, depending on a value judgment about the relative weights placed on growth, fairness, and revenue.
4. Hold the disagreement honestly. This is the discipline the textbook tries to model: that two intelligent, informed people can read the same data and reach different conclusions, because they weight different values differently. The job of an educated citizen is not to win every argument, but to know which arguments are worth having.

The next chapter turns from economic policy in general to the specific contested space of healthcare. Many of the same patterns — mainstream-conservative-vs-mainstream-progressive disagreement, with libertarian and social-democratic flanks; intricate institutional history; substantial empirical literature with genuine residual uncertainty — recur there.

Your district: economic policy applied locally

Your representative voted on the 2017 TCJA, on the 2021 Bipartisan Infrastructure Law, on the 2022 IRA, and on the 2022 CHIPS Act. Pull their record on these four votes. (GovTrack and Congress.gov list every roll call.) Compare to the median voter in your district on each of these issues, using whatever public polling is available (Cook Political Report's House Vote Tracker and FiveThirtyEight's archive can help, though some 2026 issues may be undercovered).

Two questions to take to discussion: (1) Do the votes match what the median voter in your district would have wanted? (2) When the votes don't match, what is the representative's account of why? (Their public statements, press releases, and town-hall remarks are usually accessible.)

This is a Democracy Audit task. The goal is not to find that your representative is bad; the goal is to learn what the relationship between representative and district actually looks like in your case.