Chapter 24: Interest Groups and Lobbying — The First Amendment Right That Distorts Democracy

"Congress shall make no law... abridging... the right of the people peaceably to assemble, and to petition the Government for a redress of grievances." — First Amendment, U.S. Constitution

"I see in the near future a crisis approaching that unnerves me and causes me to tremble for the safety of my country... corporations have been enthroned and an era of corruption in high places will follow." — Letter attributed to Abraham Lincoln, 1864 (disputed authenticity, but a real fear of the era)

"If you wanted to design a constitutional democracy in which a small number of well-funded and well-organized groups could dominate the policy process — and you wanted that domination to be perfectly legal and protected by the Constitution itself — you would have a hard time improving on what we have." — Lee Drutman, The Business of America Is Lobbying (2015)

Learning objectives

By the end of this chapter, you will be able to:

1. Explain the First Amendment foundation for lobbying and the constitutional implications for any reform proposal.
2. Describe the empirical scale of the federal lobbying industry — registered lobbyists, total spending, and the "shadow lobbying" workforce.
3. Distinguish among the activities interest groups pursue: direct lobbying, issue advocacy, electoral activity, research, coalition building, and litigation.
4. Identify the major categories of interest groups (economic, labor, professional, single-issue, civil rights, public-interest, foreign, intergovernmental, corporate) and recognize examples across the political spectrum.
5. Analyze the revolving door — including its empirical patterns, the corruption critique, and the expertise defense — without adopting either as the only honest read.
6. Interpret the Kalla and Broockman (2016) field experiment showing access asymmetry between donors and non-donor constituents.
7. Compare the iron-triangle and policy-network models of how interest groups, agencies, and committees interact, and identify when each model fits.
8. Steel-man both the case for stronger lobbying regulation and the First Amendment defense of the lobbying industry.

24.1 The right to petition: where lobbying begins

Open the Bill of Rights. Read the First Amendment slowly. After "the freedom of speech, or of the press," the sentence keeps going: "or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances."

That last clause — the petition clause — is doing more constitutional work than most Americans realize. It is the textual hook on which the modern lobbying industry hangs. When a pharmaceutical company hires a former senator to argue against drug-price negotiation, when the Sierra Club organizes a delegation to meet with a House committee on wilderness protection, when a steelworkers' union sends representatives to the Department of Commerce on tariff policy — all of that activity is, in the most literal constitutional sense, the people petitioning the government for a redress of grievances.

This matters because it sets a high bar for any reform proposal. You cannot simply "ban lobbying." You cannot easily restrict who is allowed to talk to a member of Congress, or what they are allowed to say, or how much they are allowed to spend organizing other people to say it. The Supreme Court has said, repeatedly, that the petition clause and the speech clause together protect a wide range of activities that look — to a frustrated reformer — like nothing more than rich and well-organized interests buying influence.

The Court has also said, at the same time, that disclosure requirements are constitutional, that anti-corruption rules are constitutional, and that process limits (like requiring lobbyists to register, or requiring contributions to be reported) survive First Amendment scrutiny. So the constitutional space for reform is real. It is just narrower than its loudest advocates often pretend.

This chapter takes that constitutional starting point seriously. It also takes the empirical reality seriously: organized interests, especially well-funded business and trade-association interests, dominate the federal policy process in ways that distort the representative function the Founders designed Congress to perform. Both things are true. Holding both at once is the analytical work of this chapter.

24.2 What interest groups are and what they do

An interest group is an organization that seeks to influence public policy without itself running candidates for office. That last clause is what distinguishes interest groups from political parties. Parties run candidates and seek to staff government with their own people. Interest groups stand apart from the electoral apparatus and try to move policy in their direction by acting on whoever is in office.

The activities interest groups pursue fall into roughly six categories.

Direct lobbying. Meetings, phone calls, emails, and letters — between paid representatives of an organization and members of Congress, congressional staff, executive-branch officials, and agency rulemakers. This is the activity that "lobbying" most narrowly refers to, and it is the activity covered by the Lobbying Disclosure Act (1995) and the Honest Leadership and Open Government Act (2007). We will return to disclosure rules in §24.5.

Issue advocacy. Public campaigns to shape opinion. Television and digital ads, op-eds, social-media campaigns, mass mailings, billboards. The Affordable Care Act fight (2009-2010) saw an estimated $200+ million in issue advocacy on each side. Climate-policy fights, gun-policy fights, and tax-policy fights all generate similar volumes. Issue advocacy reaches voters; voters then pressure (or fail to pressure) their representatives.

Electoral activity. Endorsements, scorecard ratings (the NRA's letter grades, the AFL-CIO's voting record, the Heritage Action scorecard), PAC contributions to candidates, super PAC independent expenditures, voter-mobilization programs. Many large interest groups have a 501(c)(3) charity arm (no electoral activity), a 501(c)(4) advocacy arm (some electoral activity), and a separate PAC and/or super PAC. The legal walls between these entities are real but porous, and the entities cooperate more closely than the doctrinal separation suggests.

Research and analysis. Think-tank reports, white papers, academic studies, legal briefs (including amicus briefs in litigation), and economic models. The Congressional Budget Office is the official scorekeeper for federal legislation, but it operates inside an ecosystem of outside research that frames issues, supplies counter-models, and provides intellectual ammunition for each side. The Brookings Institution, the American Enterprise Institute, the Heritage Foundation, the Center for American Progress, the Cato Institute, the Center on Budget and Policy Priorities, and dozens of others compete in this space. So do industry-funded research shops and academic-affiliated centers funded by foundations of varying ideological commitment.

Coalition building. No single interest group, however large, can pass major legislation by itself. Interest groups assemble coalitions — sometimes ideologically homogeneous, sometimes surprisingly diverse. The Reagan-era coalition for the 1986 tax reform combined business groups, anti-poverty advocates (who liked the expanded earned-income tax credit), and reformers who wanted base-broadening. The 2017 Tax Cuts and Jobs Act assembled a different coalition. The 2010 Affordable Care Act assembled yet another, including, importantly, a hard-fought neutralization of pharmaceutical-industry opposition through a deal that protected drug-company interests. We will return to that deal in case study 1.

Litigation. Strategic lawsuits to advance policy goals through the courts. The NAACP Legal Defense Fund's litigation strategy from the 1930s through Brown v. Board (1954) is the prototype: pick the cases carefully, control the legal theory, build precedent step by step. The American Civil Liberties Union, the Federalist Society's network of affiliated lawyers, the Pacific Legal Foundation, the Center for Reproductive Rights, the Alliance Defending Freedom, the Institute for Justice, and many others all run organized litigation programs. Increasingly, the strategic lawsuit is the most efficient lever for moving policy when the legislative branch is gridlocked.

The "lobbying industry," in the narrow sense, is the first activity. The "interest group ecosystem," in the broad sense, is all six.

24.3 The empirical scale

How big is this thing?

Registered federal lobbyists. As of recent years, between roughly 12,000 and 13,000 individuals are registered under the Lobbying Disclosure Act. The number peaked around 14,800 in 2007-2008 and has gradually declined. (Source: OpenSecrets and the Senate Office of Public Records, both of which compile from quarterly LD-2 filings.)

Federal lobbying spending. Total reported federal lobbying spending has grown from approximately $1.6 billion in 2000 to around $3.5 billion in 2010 to roughly $4.5 billion per year as of 2024-2025 (in nominal dollars; the inflation-adjusted growth is smaller but still real). Note the puzzle: spending has risen while the number of registered lobbyists has fallen. That means spending per registered lobbyist has risen substantially, which raises an obvious question — what is that money paying for if not registered lobbying activity?

The shadow lobbying workforce. Lee Drutman, James Thurber, and others who have studied this question estimate that the people whose actual work involves influencing federal government policy — but who do not register as lobbyists — number somewhere between 50,000 and 100,000. They are former officials operating just below the registration threshold (which generally requires 20% of one's professional time on lobbying activities for a single client over a quarter), strategic communications specialists, "policy advisors" and "senior counselors" at law and consulting firms, executive-branch outreach professionals at trade associations, and former staffers running grass-tops mobilization shops. The work is largely the same as registered lobbying. The reporting requirements are not.

This shadow lobbying problem is not a fringe concern of reformers. It is acknowledged in academic studies, in the Project on Government Oversight's analyses, in Congressional Research Service reports, and in occasional reform bills introduced from both parties. Closing the gap is technically and politically harder than it sounds — see §24.10 — but the gap itself is real.

The biggest spenders. As of the most recent quarterly filings, the top organizational spenders on federal lobbying are typically the U.S. Chamber of Commerce, the National Association of Realtors, the Pharmaceutical Research and Manufacturers of America (PhRMA), Blue Cross Blue Shield, Meta, Amazon, Alphabet (Google), the American Hospital Association, and a rotating set of defense contractors and energy companies. (Live data: open the OpenSecrets "Top Spenders" page for the most recent quarter.)

By industry. Aggregating across companies and trade associations:

(Figures are approximate; consult OpenSecrets industry tables for current data. Some industries' totals include trade associations whose membership crosses category lines.)

Pharmaceutical/health is consistently the single largest industry, by a substantial margin. Tech has grown the fastest over the past decade — Meta and Google were near-zero lobbying presences in 2005 and are now top-ten spenders. Defense is reliably substantial but not dominant; the popular imagination of an "all-powerful" defense lobby overstates the case. Energy lobbying is split between fossil-fuel interests, renewables, utilities, and nuclear, and these subgroups often lobby against each other.

The geography of lobbying. Most federal lobbying happens within a few blocks of the U.S. Capitol. K Street, the corridor running west from Mt. Vernon Square through Foggy Bottom, has been synonymous with the federal lobbying industry since the 1970s, although today's lobbying firms are scattered across downtown Washington, Georgetown, and northern Virginia. The major law firms that house substantial lobbying practices — Akin Gump, Brownstein Hyatt Farber Schreck, Holland & Knight, BGR Group, Covington & Burling, Hogan Lovells, Squire Patton Boggs — are based in Washington and have offices that ring the Capitol within a 15-minute drive. The geographic concentration matters because it produces a thick professional culture: lobbyists know each other, know the staff, know the members, attend the same fundraisers, send their children to the same schools, and operate within a tight social network that transcends individual employer relationships.

Outside Washington, state-capital lobbying mirrors the federal pattern at smaller scale. Sacramento, Albany, Springfield, Tallahassee, and the other large state capitals each support a lobbying ecosystem of several hundred to several thousand registered lobbyists. State-level lobbying is often more concentrated by industry (the gambling industry in Nevada, the energy industry in Texas, the pharmaceutical industry in New Jersey, the financial-services industry in New York and Connecticut) than federal lobbying, and state-level revolving-door dynamics can be even more aggressive. Federal officials retiring to Washington-area lobbying firms is the most-discussed pattern; state legislators retiring to in-capital state-level lobbying practice is the much larger, less-discussed pattern in absolute headcount.

The international dimension. American-based lobbying for international clients has grown substantially. Foreign governments hire registered FARA-listed agents in Washington at increasing rates; foreign multinational corporations operate Washington offices; international NGOs (Amnesty International, Human Rights Watch, the International Crisis Group) have substantial Washington presences. The line between "American interest group" and "foreign interest group" can blur — a Mexican-government-funded research organization in Washington, a U.S. corporation lobbying for trade-policy positions favorable to its overseas operations, and an American Jewish organization lobbying on Israeli-policy issues are all engaged in legitimate, constitutionally protected activity, but the international entanglements create disclosure questions that the FARA regime addresses incompletely.

Lobbying inequality. The most important fact about the lobbying ecosystem is the imbalance among the categories of organizations doing the lobbying. The political scientists Kay Lehman Schlozman, Sidney Verba, and Henry Brady, in their landmark study The Unheavenly Chorus (2012), found that business and trade-association interests account for roughly 70-80% of all federal lobbying activity, while public-interest groups, labor unions, and consumer organizations together account for a small fraction. They quoted E.E. Schattschneider's famous 1960 line: "The flaw in the pluralist heaven is that the heavenly chorus sings with a strong upper-class accent." Six decades later, the chorus is louder, the accent has not changed, and the empirical gap between business and other organized interests is, if anything, wider.

This is not a partisan claim. The business-lobbying tilt advantages legislators of both parties who serve business-favorable constituencies, and disadvantages legislators of both parties who attempt to legislate against entrenched business interests. The pharmaceutical pricing fight, which we examine in case study 1, has had Republican and Democratic skeptics of negotiation and Republican and Democratic supporters of negotiation — the alignment runs more along electoral-competitive-incumbency lines than partisan ones.

24.4 Types of interest groups, with examples

Interest groups are wildly diverse. Here is a working taxonomy with examples drawn — deliberately — from across the political spectrum.

Economic / business. The U.S. Chamber of Commerce, the National Association of Manufacturers (NAM), the National Federation of Independent Business (NFIB), and hundreds of industry trade associations (American Petroleum Institute, the American Bankers Association, the Edison Electric Institute, the Recording Industry Association of America). Trade associations represent firms collectively; individual large corporations also lobby for themselves. The largest single category by spending.

Labor. The AFL-CIO is the umbrella federation; it includes affiliated unions covering most of the unionized private and public workforce. Major unions: the United Auto Workers (UAW), the Service Employees International Union (SEIU), the International Brotherhood of Teamsters, the National Education Association (NEA), the American Federation of State, County, and Municipal Employees (AFSCME), the American Federation of Teachers (AFT), the Communications Workers of America (CWA), the United Food and Commercial Workers (UFCW). The Industrial Workers of the World (IWW) and other independent labor organizations operate outside the AFL-CIO. Labor lobbying expenditures are large in absolute terms but a small fraction of business lobbying. Membership is at historic lows (about 10% of the workforce, down from a peak of 35% in the mid-1950s) but the public-sector share of unionized workers is much higher than the private-sector share.

Professional associations. The American Medical Association (AMA), the American Bar Association (ABA), the American Institute of Certified Public Accountants (AICPA), the American Dental Association, the American Psychological Association, the American Hospital Association. Each lobbies for the interests of its members on issues of licensure, scope of practice, reimbursement rates, and regulatory authority. Often very effective on issues affecting the profession; less visible than political-celebrity lobbying.

Single-issue groups. Organizations focused on one policy area. Examples across the spectrum:

• Gun policy. The National Rifle Association (NRA), Gun Owners of America (further-right of NRA), the National Shooting Sports Foundation; on the other side, Everytown for Gun Safety, Brady (formerly the Brady Campaign to Prevent Gun Violence), Giffords.
• Abortion policy. Susan B. Anthony Pro-Life America, Concerned Women for America, the National Right to Life Committee; on the other side, Planned Parenthood Action Fund, NARAL Pro-Choice America (renamed Reproductive Freedom for All in 2023), the Center for Reproductive Rights.
• Environmental. The Sierra Club, the League of Conservation Voters, the Natural Resources Defense Council (NRDC), the Environmental Defense Fund (EDF); on the other side, the American Energy Alliance, the Heartland Institute, and various industry-funded research and advocacy operations.
• Aging. AARP (Association for the Advancement of Retired Persons) is the largest membership organization in the country; ~38 million members. Lobbies extensively on Social Security, Medicare, and aging-related issues.
• Immigration. Federation for American Immigration Reform (FAIR), NumbersUSA, Center for Immigration Studies on the restrictionist side; United We Dream, the National Immigration Law Center, the Migration Policy Institute on the pro-immigration side.

Civil rights and civil liberties. The American Civil Liberties Union (ACLU), the National Association for the Advancement of Colored People (NAACP) and its Legal Defense Fund (LDF), the Anti-Defamation League (ADL), the Asian American Legal Defense and Education Fund (AALDEF), the Mexican American Legal Defense and Educational Fund (MALDEF), the National LGBTQ Task Force, the Lambda Legal Defense and Education Fund, the Becket Fund for Religious Liberty, the Alliance Defending Freedom, the First Liberty Institute, the Religious Freedom Institute. Civil-liberties and religious-liberty organizations sometimes ally across the left-right divide on free-speech and free-exercise issues.

Public-interest and good-government. Common Cause, RepresentUs, the Campaign Legal Center, the Brennan Center for Justice, Public Citizen (left-coded); Heritage Action, the Federalist Society, the Heritage Foundation, the Cato Institute (right-coded, with the Cato Institute libertarian rather than conservative on many issues); the League of Women Voters and the Committee for a Responsible Federal Budget (often centrist or non-partisan). The "public interest" label is itself contested — most groups in this category believe they represent the public interest as they conceive it; their opponents disagree.

Foreign-government and foreign-interest lobbying. Governed primarily by the Foreign Agents Registration Act (FARA) of 1938, which requires anyone acting as an agent of a foreign principal in a political capacity to register and disclose. FARA enforcement was minimal for decades and tightened substantially after 2017. Significant U.S. lobbying expenditures by foreign governments include Saudi Arabia, the United Arab Emirates, Qatar, Israel, Taiwan, and others. The disclosure requirements are stricter than for domestic lobbying, but enforcement remains uneven.

State and local government lobbying. State governments lobby the federal government — through the National Governors Association, the National Conference of State Legislatures (NCSL), the U.S. Conference of Mayors (USCM), the National Association of Counties, and individual state and city offices in Washington. Municipal lobbying is a $100+M annual industry by itself.

Corporate lobbying. Individual companies hiring representation directly. Boeing on aerospace, ExxonMobil on energy, Pfizer on health, Lockheed Martin on defense, Walmart on labor and trade, Microsoft on technology and antitrust. Direct corporate lobbying is on top of trade-association lobbying — a company can be a member of the Chamber of Commerce, of an industry-specific trade association, and also have its own lobbyists. The same applies to individual companies' PAC contributions and super PAC support.

The diversity of this list matters. When commentators say "lobbying" they often mean "the kind of lobbying I do not like." The reality is that every organized interest in American politics is lobbying. The Sierra Club lobbies. The NRA lobbies. The ACLU lobbies. The U.S. Chamber lobbies. Your AARP membership card pays for someone to lobby on Social Security. Your union dues pay for someone to lobby on labor law. The reform conversation is not "lobbying versus no lobbying." It is "which kinds of lobbying are advantaged and disadvantaged by the current rules."

24.4.1 The collective-action problem and Mancur Olson

To understand why business and trade-association interests dominate lobbying — and why public-interest groups struggle — return to a foundational text: Mancur Olson's The Logic of Collective Action (1965). Olson asked a simple question with surprising answer. Why do small groups with concentrated interests organize easily, while large groups with diffuse interests struggle to organize at all?

The answer, Olson showed, is that organizing is costly and the benefits of successful policy advocacy are often non-excludable. If the dairy industry succeeds in lobbying for a milk-price-support program, every dairy producer benefits, regardless of whether they contributed to the lobbying effort. So why would any individual producer pay to lobby? The rational answer: don't pay, and let your competitors pay; you get the benefit either way. This is the free-rider problem, and it makes large-group organizing structurally difficult.

Small groups with high per-member stakes solve the problem because each member's contribution makes a noticeable difference. The handful of large U.S. sugar producers, each with millions or billions of dollars at stake in sugar-tariff policy, can easily fund lobbying — they would lose more in a single bad policy outcome than the lobbying costs them. Their consumers, whose individual stakes in sugar prices are perhaps $50 per year, cannot organize at all. The economics make the lobbying inevitable on one side and impossible on the other.

This is why business interests — concentrated, high-stakes, small in numbers — dominate lobbying. And why diffuse public interests — millions of consumers, taxpayers, or beneficiaries each with small individual stakes — usually do not lobby effectively. The asymmetry is not a moral failing of the public-interest community. It is a structural consequence of how collective-action incentives operate.

Olson's framework also explains why "public-interest" groups that do exist usually depend on selective incentives to overcome the free-rider problem. The Sierra Club offers magazines, hiking trips, and membership identity. AARP offers prescription-drug discounts and travel benefits. Labor unions offer tangible workplace representation. The "interest group" function is bundled with a "membership benefits" function that gives individuals a reason to pay dues regardless of the public-policy work. Without the selective incentive, the public-policy work would not get funded.

The Olson framework is indispensable for understanding the lobbying ecosystem. It also has reform implications. If diffuse public interests cannot organize on their own, the system either depends on policy entrepreneurs (think tanks, advocacy organizations funded by foundations rather than members) to advocate on behalf of unorganized publics, or it accepts that those publics will be systematically under-represented relative to organized concentrated interests. Both options are operative. Neither closes the gap.

24.5 Disclosure: what the rules require

Federal lobbying is governed by three main disclosure regimes:

The Lobbying Disclosure Act of 1995 (LDA). Requires anyone meeting the statutory definition of "lobbyist" — broadly, anyone paid to spend a significant fraction of professional time on lobbying contacts with covered government officials — to register with the Senate Office of Public Records and the House Clerk, and to file quarterly reports (the LD-2) listing clients, issues lobbied, expenditures, and which chambers or agencies were contacted. The LDA replaced an earlier, weaker 1946 statute that had been read narrowly by the Supreme Court in United States v. Harriss (1954) to apply only to direct contacts about pending legislation.

The Honest Leadership and Open Government Act of 2007 (HLOGA). Strengthened the LDA after the Jack Abramoff scandal of the mid-2000s. Required quarterly (rather than semi-annual) reporting, added separate disclosure of campaign contributions and political-committee activity, required certification that registrants understood the gift and travel rules, and lengthened cooling-off periods (see §24.7).

The Foreign Agents Registration Act of 1938 (FARA). As described above, governs lobbying on behalf of foreign principals. Stricter than the LDA in some respects (requires disclosure of substantive communications, not just registration), looser in others (the threshold for what counts as "political activity" is contested). Enforcement was sporadic for decades and substantially tightened in the late 2010s after several high-profile prosecutions including those connected to the Mueller investigation.

The substantial gap between rules and practice. A few important holes in the current disclosure regime:

• The 20% threshold. Under the LDA, you only have to register if you spend at least 20% of your time on lobbying activities for a single client over a quarter. People who spend 19% — or who structure their work across multiple clients to keep any single one below the threshold — do not have to register. This is the principal driver of the shadow-lobbying workforce.
• The "lobbying contact" definition. Only direct communications with "covered officials" count. Strategic advice, PR, opposition research, and grassroots-mobilization work do not require registration even if they are commissioned by a lobbying client and aimed at a legislative outcome.
• The "consultant" loophole. Senior advisors to lobbying clients can describe themselves as "consultants" or "strategic counselors" and avoid the definition.
• Limited enforcement. The Government Accountability Office reviews compliance annually; the Department of Justice can prosecute violations. Prosecutions are extremely rare. Civil penalties for non-compliance are small and inconsistently applied.

The result: the lobbying disclosure regime captures the formal core of the industry but undercounts the periphery. Reform proposals — which we discuss in §24.10 — focus heavily on closing these gaps.

24.6 Iron triangles and policy networks

How do interest groups, agencies, and congressional committees actually work together to make policy? Two competing models, both with empirical support.

The iron triangle. The classic framework. Three corners:

1. A bureau or agency with regulatory authority over a sector.
2. A congressional committee or subcommittee with jurisdiction over that agency and its budget.
3. The interest group(s) representing the regulated industry or constituency.

Each corner serves the other two. The agency provides favorable rules and enforcement; the committee provides favorable budgets and oversight; the interest group provides political support, campaign contributions, and information. The arrangement is mutually reinforcing — and durable, because the people who would most want to disrupt it (general taxpayers, broad public interests) do not pay close enough attention to overcome the focused attention of the three corners.

Iron triangles are most common in stable, low-salience policy areas: agricultural commodity programs, certain veterans' services, particular defense procurement lines, sub-issues within transportation policy. In these areas the same handful of committees, agencies, and interest groups have been doing business with each other for decades, and elections rarely change the dynamics.

Policy networks (or "issue networks"). The political scientist Hugh Heclo, in a 1978 essay, argued that the iron-triangle model was outdated. Many policy areas, he said, do not have a stable triangle but a much larger and more fluid network: dozens of interest groups (often on multiple sides of the issue), think tanks, academic experts, journalists, congressional staff, executive-branch officials, state and local actors, foreign governments, and an interested attentive public. Membership in the network shifts as new entrants appear and old ones lose interest. Conflict within the network is normal, not anomalous.

Policy networks fit high-salience, partisan-contested policy areas: climate policy, abortion policy, gun policy, healthcare reform, immigration, electoral reform. In these areas no three-cornered triangle could possibly capture what is happening. The cast of characters is too large and too contentious.

When does each model fit?

The two models are not mutually exclusive; both are happening simultaneously across the federal government. Healthcare policy at the high level is a chaotic policy network; healthcare reimbursement schedules for specific procedures are something closer to an iron triangle. Defense at the strategic level is networked and contested; the procurement of a particular subcomponent of an aircraft carrier is iron-triangular.

The reform implication: iron triangles are stable because the participants like them. Networks are unstable because nobody has stable control. Reformers who want to break iron triangles need to raise salience — to drag the issue into the network. Reformers who want to organize a network into a coalition for change need to do the opposite — to build durable partnerships in a chaotic environment.

24.7 The revolving door

The phenomenon is named for the literal architecture of office-building entrances, but the metaphor extends well beyond that. People go from government service to private-sector lobbying, sometimes back to government, sometimes back to lobbying again. The pattern is decades old; the institutional scale of it is more recent.

A note on the K Street Project, the Tom DeLay era, and partisan revolving doors. From roughly 1995 through the mid-2000s, House Majority Leader Tom DeLay (R-TX) and the Republican leadership pursued what came to be called the K Street Project: a coordinated effort to ensure that lobbying firms hired Republican rather than Democratic former officials, on the theory that Republican leadership in Congress should be matched by Republican relationships in the lobbying industry. Firms that hired Democrats were threatened with loss of access to the Republican leadership; firms that hired Republicans were rewarded with introductions and consideration. The project collapsed under the weight of the Jack Abramoff scandal — Abramoff, a Republican lobbyist, pled guilty in 2006 to fraud and corruption charges that implicated dozens of officials and led directly to the 2007 HLOGA reforms.

Critics noted at the time that the K Street Project differed from ordinary partisan revolving-door behavior only in its degree of explicitness. Both parties had always tilted their post-government lobbying placements toward firms with friendly partisan affiliations. The K Street Project made the tilt operational and rewarded enforcement. The principle applies in reverse: when Democrats control a chamber, lobbying firms hire more Democratic former staff; when Republicans control it, they hire more Republican former staff. Lobbying firms are pragmatic — they want access wherever the votes are.

This is a feature of the revolving door that even careful reformers sometimes underweight. The door is not corrupt for one party and clean for the other. The structural pattern is bipartisan and durable, regardless of which party holds the majority at any given moment.

The empirical pattern. Studies by the Center for Responsive Politics (now part of OpenSecrets), Public Citizen, and academic researchers consistently find:

• Approximately 50% of former members of Congress who leave office (and are not deceased or disqualified) end up registered as lobbyists or working in lobbying-adjacent positions within five years.
• Approximately 70% of senior congressional staff (chiefs of staff, legislative directors, committee staff directors) end up in lobbying or government-relations roles within five years.
• Senior executive-branch political appointees follow a similar pattern — at the Departments of Defense, Treasury, Health and Human Services, Energy, and Justice especially, the path from senior official to industry consultant or trade-association executive is well-traveled.

Methodology matters here: the exact numbers vary depending on how broadly one defines "lobbying-adjacent" (does corporate government-relations count? does trade-association leadership? does a law firm partnership focused on regulatory work?). The basic pattern, however, is robust across methods.

Cooling-off periods. Federal law requires:

• Senators: 2-year cooling-off period before they can lobby their former colleagues (set by HLOGA 2007).
• House members: 1-year cooling-off period.
• Senior staff (very senior — paid above a threshold): 1-year cooling-off period for direct lobbying contacts with their former office.
• Cabinet secretaries and senior executive-branch officials: 1- to 2-year limits depending on category.

The cooling-off periods restrict only "direct lobbying contacts" with the former office or chamber. They do not prevent the former official from being employed by a lobbying firm, advising on strategy, supervising other lobbyists, attending fundraisers, or lobbying any other office. So a former senator from Ohio can immediately lobby California senators, and after two years can lobby anyone. A former Senate Banking Committee staffer can immediately lobby House members on banking issues.

Steel-manning the corruption critique. The reform argument runs roughly as follows: when a sitting member of Congress or a senior staffer knows that her future income will likely come from one of the industries her current committee oversees, and when the lucrative industry job is a much better-paid version of her current public-service work, the alignment of interests is structural. She does not have to accept any quid-pro-quo to subtly tilt her work toward the industry's perspective. The bias is built into her professional incentive structure. Multiply this across hundreds of officials and thousands of staff, and the cumulative effect on policy is substantial — even if no individual case looks corrupt and no laws are broken.

This argument has been made carefully by Lawrence Lessig (Republic, Lost, 2011), Lee Drutman (The Business of America Is Lobbying, 2015), and many others. It does not require the assumption that anyone is bribed. It only requires that the structure of post-government employment opportunities shapes the in-government work, on average and on the margin, in directions that favor the industries with the most attractive post-government jobs.

Steel-manning the expertise defense. The counter-argument runs like this: lobbying is, ideally, the activity of bringing useful information to elected officials and their staff who do not have the expertise or the bandwidth to know everything they need to know. The single most useful kind of lobbyist is someone who has worked in the policy area, knows the institutional history, understands the technical details, and can communicate effectively with current decision-makers. That person, by definition, is a former insider. The revolving door is a feature of the recruitment system that produces effective lobbyists. Without it, the lobbying industry would consist of people who learned the policy from outside — and would be measurably worse at the information-provision part of their job. The result would not be "less lobbying"; it would be "less informed lobbying," which would not obviously be better.

This argument has been made by political scientists studying the lobbying industry (including David Vogel, Beth Leech, and James Thurber), by trade associations defending the practice, and by some former officials who become lobbyists and write candid memoirs about it. It does not deny the corruption critique; it argues that the structural alignment of interests is the price of having a competent lobbying industry, and that the alternative is informational worse.

Both arguments are partially right. The revolving door produces both expertise and structural bias. The honest empirical question is whether the expertise gains exceed the structural-bias costs, and that question is not currently answerable from data alone. Reasonable people, looking at the same evidence, come down on opposite sides.

24.8 Money, access, and the Kalla and Broockman experiment

A key empirical question: do wealthy donors and well-funded organizations get more legislative access than ordinary constituents? The Powell 2014 observational study found yes; donors got more time. But observational studies can confound — perhaps donors are simply more sophisticated, or follow up more aggressively.

The Kalla and Broockman 2016 randomized field experiment settled the question for one important measure of access: meetings.

In partnership with a national progressive advocacy organization, the researchers randomly assigned 191 congressional offices to receive meeting requests that revealed (in the treatment condition) or did not reveal (in the control condition) that the requestors were donors to the member's campaign. The meeting requests were otherwise identical. The researchers then tracked which offices granted meetings, and at what level (was it the member herself, a senior staffer, a junior staffer, or no one?).

The result: revealed donors were granted meetings at substantially higher rates and at substantially higher levels of seniority than identical non-donor constituents. The point estimate suggested that revealing donor status increased the probability of a meeting with a senior staffer or member by a factor of 3-4, depending on the office.

Read that carefully. This is not a study of vote-buying. It is not a study of corruption. It is a study of access. Donors get more legislative access than non-donors, holding everything else constant.

That access matters. The legislative process runs on relationships, on attention, on the ability to put your case in front of someone who can act on it. The reformers who say "money buys access" are not making a normative claim; they are stating an empirically demonstrated fact. The disagreement is about what to do about it — whether the access asymmetry constitutes a corruption of representation, or whether it is the inevitable byproduct of a participatory democracy in which some citizens choose to engage more than others.

The Kalla and Broockman result, combined with the Schlozman/Verba/Brady work on the upper-class accent of organized interests, paints a consistent picture. The lobbying ecosystem advantages those who have already organized, who have already donated, and who already have the institutional infrastructure to be heard. That advantage compounds over time. It is built into the system as currently constituted.

24.9 The Powell Memo and the institutional rise of conservative legal infrastructure

In August 1971, Lewis F. Powell Jr. — then a Virginia corporate lawyer, later (October 1971) nominated by Richard Nixon to the Supreme Court — wrote a confidential memorandum for the U.S. Chamber of Commerce. The memo, titled "Attack on American Free Enterprise System," argued that American business was being marginalized in the political and intellectual battle over economic policy, and that business needed to invest seriously and systematically in long-term institutional infrastructure to defend free-market principles.

The memo recommended: - Funding scholars, fellowships, and research institutes friendly to free-enterprise principles. - Building a presence on college campuses. - Pushing back in the press through op-eds, letters, and media campaigns. - Building legal infrastructure to bring cases that would shape constitutional and regulatory law. - Engaging at the regulatory-agency level, not just the legislative one. - Coordinating among business, rather than letting individual companies be picked off one at a time.

The memo was leaked and published in 1972, shortly after Powell's confirmation. Whether or not the memo itself caused what followed — the historiography is contested — the next two decades saw exactly what the memo recommended. The Heritage Foundation was founded in 1973. The Cato Institute in 1977. The Manhattan Institute in 1978. The Federalist Society in 1982. The American Enterprise Institute (1938) was modernized and dramatically scaled up. State-level free-enterprise legal foundations multiplied. By the 1990s, conservative legal and policy infrastructure rivaled or exceeded the longer-established progressive infrastructure on most issue areas.

The Powell Memo is often cited as the founding document of the modern conservative legal and policy network. Whether one views that network as a healthy expression of First Amendment-protected organizing or as a structural distortion of policy depends on one's prior politics. What is not in dispute: the network was built deliberately, and it works. Mirror-image investments by progressive donors and foundations, particularly accelerating after 2004, have built a partly comparable progressive infrastructure (Center for American Progress was founded in 2003, the Democracy Alliance in 2005). The institutional landscape is now bilateral. The asymmetry that worried Powell in 1971 is not the asymmetry of 2026, though new asymmetries persist.

24.10 Reform proposals

Reform proposals fall into several families. We steel-man each and note the constitutional and practical difficulties.

Stronger disclosure (closing the shadow-lobbying gap). Lower the LDA registration threshold from 20% to 10% (or eliminate the threshold entirely); broaden the definition of "lobbying contact" to include strategic advice and grassroots-mobilization work commissioned by a lobbying client; require disclosure of meeting subjects and substantive arguments, not just contact occurrence. The case: closes the gap between rules and practice; addresses the most-cited failure of the current regime; survives First Amendment scrutiny easily because disclosure-only regulations are constitutionally robust. The challenge: enforcement requires resources the GAO and DOJ do not currently have; broadening the definition risks capturing constitutionally protected advocacy that should not be regulated; the line between "lobbying" and "First-Amendment-protected speech" is genuinely fuzzy in a way that makes drafting hard.

Longer cooling-off periods. Extend the senator/representative cooling-off period to 5 years (some proposals say lifetime); apply similar periods to senior executive-branch appointees; expand the "covered" period to include not just direct contacts but firm partnerships and strategic-advisory roles. The case: addresses the structural-incentive problem more directly than disclosure; reduces the financial pull of post-government industry employment on in-government decisions. The challenge: lifetime bans likely violate First Amendment associational rights; longer bans reduce the incentive for talented people to enter government in the first place (the "talent drain" objection); cooling-off periods are easy to evade by hiring the spouse, the chief of staff, or operating through proxies.

Lobbying tax. Impose a per-contact or percentage-of-spending tax on lobbying activity; use the revenue to fund congressional staff and analytical resources, reducing member dependence on lobbyist-supplied information. (Variants of this proposal have been introduced as bills repeatedly, never enacted.) The case: reduces the volume of lobbying without restricting any speech; provides resources for the under-staffed congressional research function. The challenge: probably constitutional, but a tax that is high enough to materially reduce activity raises serious First Amendment concerns under cases like Murdock v. Pennsylvania (1943); does not address the asymmetry between well-funded and less-well-funded interests (a flat tax actually makes the asymmetry worse).

Public financing of campaigns. If candidates do not depend on donors for campaign funds, donor leverage diminishes. Variants include the small-dollar matching system used in New York City, the "democracy vouchers" used in Seattle, and full public financing as in some European systems. (We treat campaign finance in depth in Ch 34.) The case: addresses the donor-access asymmetry directly by reducing candidates' financial dependence on donors. The challenge: post-Citizens United, public financing systems must be voluntary and cannot prevent outside spending; designing a system that adequately funds modern campaigns without creating perverse incentives is harder than it sounds; constitutional questions remain about how restrictive a public-financing system can be.

Banning corporate lobbying. The Sanders 2020 campaign proposed prohibiting registered lobbyists from lobbying for corporations (some variants would ban for-profit lobbying altogether). The case: addresses the upper-class-accent problem by removing the largest single category of well-funded lobbying. The challenge: directly violates the First Amendment on a straightforward reading of NAACP v. Button (1963), Citizens United v. FEC (2010), and the general First-Amendment-protects-petition principle. A ban specifically on corporate lobbying — while permitting union, NGO, and individual lobbying — is even more constitutionally vulnerable because it is a content/identity-based restriction.

Strengthening congressional staff. Increase the budgets of congressional committees and member offices so that members and staff have the analytical capacity to evaluate policy without depending on lobbyist-supplied research and analysis. The Congressional Research Service, the Congressional Budget Office, and the Government Accountability Office have all seen real-dollar budget compressions over decades; restoring and expanding these resources reduces members' dependence on outside expertise. The case: addresses the demand side of the lobbying market rather than the supply side; constitutionally unproblematic; bipartisan-aligned reformers have called for this. The challenge: not zero-cost; doesn't on its own change the asymmetry of organized interests; institutional momentum favors the status quo.

A reform agenda that actually addresses the empirical problems would likely combine several of these — stronger disclosure, longer cooling-off periods, a public-financing option for campaigns, and stronger congressional analytical resources — rather than relying on any one alone.

24.11 The defense of the lobbying industry

Before this chapter ends, the defense deserves its full hearing.

The First Amendment grounds. Lobbying is constitutionally protected speech and petition. The right to organize, to associate, to fund organizations that represent your views, to hire skilled representation to articulate your views to government, and to engage in collective political activity is at the core of what the First Amendment protects. The Supreme Court has affirmed this repeatedly, in cases ranging from NAACP v. Alabama (1958) — where the Court protected the NAACP's membership lists from state subpoena — to NAACP v. Button (1963) — where the Court protected the NAACP's litigation-as-speech activities — to Citizens United v. FEC (2010) — where the Court extended protection to corporate and association political spending. A reform proposal that meaningfully restricts lobbying activity must navigate this constitutional doctrine. Many cannot.

Information provision. Modern policy is technical. A junior congressional staffer working on, say, regulatory standards for a new class of medical devices does not have time to become an expert in regulatory affairs, FDA process, biomechanical engineering, and clinical-trial design. The lobbyists for the medical-device industry, the patient-advocacy organizations, and the FDA-watchdog reform groups do have that expertise. They bring it to the table. The information they provide is not always accurate or balanced — that's why it comes from multiple competing sources — but on average and in aggregate, it improves the quality of the policy that gets made. Studies of congressional staff (Kingdon, Agendas, Alternatives, and Public Policies; Drutman; others) consistently find that staff rely on lobbyist-supplied information as a major input, and that they consider this normal and useful, not corrupting.

Coalition building. Many legislative coalitions cross ideological lines in ways that are visible only to people who track interest-group activity. The 2018 First Step Act on criminal-justice reform combined libertarian, evangelical, civil-rights, and prosecutorial-reform organizations. The 2017 sex-trafficking legislation (FOSTA-SESTA) combined feminist organizations and conservative-Christian organizations. The 2010 Affordable Care Act ultimately succeeded in part because the pharmaceutical industry's deal with the administration removed it as an opponent (a fact celebrated by some of the law's supporters and lamented by others — this is the case-study-1 question). The infrastructure of American politics is built on the ability of organized interests to build temporary coalitions across persistent partisan divides. Without the lobbying industry, that infrastructure does not exist.

Expertise. Some of the most useful organizations on Capitol Hill are technical-expert organizations — the American Mathematical Society on STEM funding, the American Society of Civil Engineers on infrastructure, the Society for Industrial and Organizational Psychology on workplace policy. These organizations are technically lobbyists. They bring genuine, hard-won expertise. The cost of restricting their activity would be paid in worse policy.

The "imagine the alternative" test. What would a federal policy process without organized interest groups look like? It would not look like a populist citizen-driven democracy. It would look like a system in which only the people who had the personal time, expertise, and access to advocate for themselves got heard — which is, by definition, a system that strongly advantages the wealthy and well-connected. Organized interest groups, including unions, civil-rights organizations, advocacy groups for the poor, the disabled, the elderly, and the marginalized, are how people without individual political capital pool their voices into something that can be heard. Reformers should remember: the people most disadvantaged by a sweeping anti-lobbying reform are the people whose individual voices are weakest.

The defense, in short: lobbying is not a problem to be solved. It is a constitutionally protected activity that, on the whole, is better than its alternatives, even though specific abuses warrant specific reforms.

24.11.1 The reform proposals as a package: what would actually change?

It is worth pausing on a question reformers often skip: if a comprehensive lobbying-reform package were enacted tomorrow — say, the package recommended in §24.10 — what would actually change? The honest answer is, less than reformers hope and more than defenders fear.

Stronger disclosure would shrink the shadow-lobbying workforce by perhaps 20-40%, force an additional 5,000 to 20,000 people into formal registration, and produce a noticeably more complete picture of who is influencing whom. It would not change what the dominant interests are or how much they spend; it would only make the influence more visible. That visibility matters — sunlight remains a useful disinfectant — but it is not transformative.

Longer cooling-off periods would reduce the most flagrant cases (the senior committee staff member who leaves on Friday and represents a lobbying client on Monday). It would not change the fundamental career trajectory of senior officials, who would continue to land at law firms, trade associations, and corporate-affairs offices with five-year-delayed direct-lobbying privileges. The structural-incentive problem would diminish but not disappear.

A lobbying tax of, say, 10% on registered lobbying spending would generate roughly $400-500 million per year and fund a substantially expanded congressional support apparatus. The tax would be passed through to clients (who would lobby slightly less, or pay slightly more for the same lobbying); the volume of activity would diminish modestly. The bigger effect would be on the demand side — congressional staff and analysts would have alternative information sources, reducing dependence on lobbyist-supplied research.

Public financing of campaigns, even at the optional matching-funds level, would dramatically reduce the donor-dependence of candidates who opted in. It would not affect outside spending (super PACs and dark-money groups), which would remain the dominant force in expensive races post-Citizens United. The access asymmetry that Kalla and Broockman documented would shrink for participating candidates and remain unchanged for non-participating ones.

Banning corporate lobbying outright would not survive constitutional review. Variants that survive review (heavy disclosure, taxation, restrictions on revolving-door employees of corporate lobbyists) would have effects similar to the disclosure and tax measures above.

Strengthened congressional staff would have, perhaps, the most underrated long-term effect. A Congressional Budget Office with twice its current analytical capacity, a Government Accountability Office with restored staffing, and committee staffs at 1980 levels would change the power dynamics of who informs whom. Lobbyists would still lobby. Members would just have alternative sources of expertise to balance the lobbyist-supplied frame.

The combined effect of the full package: a federal lobbying ecosystem that is more visible, marginally smaller, less revolving-door-driven, less donor-leveraged for participating candidates, and more competitively informed at the staff level. A meaningful improvement. Not a transformation. The underlying constitutional architecture — the First Amendment right to organize and petition — and the underlying social architecture — that organized interests dominate where unorganized interests do not — would persist.

This is the realistic scope of plausible reform. Anyone who promises more is selling something. Anyone who claims that nothing meaningful can be done is also selling something — typically a defense of the status quo.

24.11.2 The Federalist Society as a case in non-traditional interest-group activity

A useful illustration of how the lobbying-and-influence ecosystem extends beyond the traditional lobbying frame: the Federalist Society for Law and Public Policy Studies, founded in 1982 at Yale Law School and the University of Chicago Law School by a group of conservative and libertarian law students.

The Federalist Society is, technically, not a lobbying organization. It does not hire registered lobbyists to make legislative contacts. Its tax status (501(c)(3), debate-and-education) prohibits it from engaging in lobbying as a substantial activity. Its public-facing operation is a network of campus chapters, lawyer chapters in cities across the country, debate panels, and a flagship annual convention.

What it does do is build a community of conservative and libertarian lawyers, identify them, develop their thinking through panels and publications, and — through a separately-incorporated practice — recommend lawyers for federal judicial nominations. The organization's role in shaping the conservative judicial pipeline since the Reagan administration, and especially during the Trump administrations, has been decisive. Of the six conservative justices on the current Roberts Court, all have direct or strong indirect ties to the organization. Hundreds of federal trial and appellate judges have been associated with it.

This is influence on policy at the deepest level — the shaping of the legal and constitutional architecture within which all other policy operates. It runs through none of the traditional lobbying channels. It would not be touched by any of the lobbying-reform proposals discussed above. It is constitutionally robust on every doctrine the Court has articulated.

A mirror-image progressive infrastructure exists, though it was built later and is less institutionally consolidated. The American Constitution Society (founded 2001), the People for the American Way (founded 1981), and the various progressive judicial-nomination-focused organizations operate in the same space. The combined picture: judicial influence is itself an interest-group activity, conducted through organizations that do not appear on any LD-2 quarterly report, governed by no FARA-style disclosure regime, and fully protected by the First Amendment.

The takeaway is not that the Federalist Society or its progressive analogues are doing anything wrong — they are doing exactly what the First Amendment protects, and reasonable observers across the political spectrum can agree that the activity is legitimate. The takeaway is that "lobbying" as conventionally measured captures a slice of the influence ecosystem. The most consequential influence often happens outside that slice. Any honest analysis of how policy gets made has to look beyond the LD-2 filings.

24.12 Holding the tensions

Two truths exist in tension throughout this chapter, and the chapter does not resolve them.

The empirical reality is that the federal lobbying ecosystem distorts the representative function of Congress. Business and trade-association interests dominate. Donors get more access. The revolving door creates structural alignments between officials and industries. The shadow-lobbying workforce is large. Public-interest, labor, consumer, and civil-rights organizations are systematically outspent by 5-to-1 or worse. The accent in the heavenly chorus, six decades after Schattschneider, remains upper-class.

The constitutional reality is that the First Amendment protects the right to petition, to associate, to speak, to organize, and to fund organizations that represent one's views. Most of what looks like the problem is constitutionally protected. Reform proposals that ignore the First Amendment will not survive judicial review. Reform proposals that work within the First Amendment — disclosure, cooling-off periods, public-financing options, strengthened congressional capacity — are real but partial.

A democracy that is honest with itself about both realities can make incremental progress on the parts of the system that are most clearly broken — the shadow-lobbying gap, the under-resourced congressional staff, the gameable cooling-off periods — while accepting that it cannot eliminate the underlying asymmetry of organized interests. That asymmetry is the price of a free society in which people are allowed to organize. The work of citizenship is to organize on the side of broader interests, not to ban the organizing.

The next chapter examines social movements and protest — the form of political activity that, when interest-group politics fails to produce change for a marginalized group, becomes the alternative. The two together — organized interests inside the system and organized movements outside it — are the primary engines through which Americans translate concern into policy. Both are essential. Both are constitutionally protected. Neither is sufficient by itself.

Up next: - Case Study 1: The pharmaceutical industry and Medicare drug pricing (1965-2026). - Case Study 2: The NRA's institutional power and recent decline. - Exercises, Quiz, Key Takeaways, Further Reading.