Chapter 18: Public Economy
Democracy is “the worst form of government except for all the others,” as Winston Churchill famously observed. This chapter applies economic reasoning to the political process — examining why many people don’t vote, how small interest groups can dominate policy, and why majority rule can sometimes fail to produce logical outcomes.
Table of Contents
1. Voter Participation and Costs of Elections
1.1 Why Don’t People Vote?
U.S. voter turnout:
- Presidential elections: 55–65% of voting-age citizens
- Congressional/local elections (no presidential race): often less than 50%
International comparison:
- Germany, Spain, France: 75–80% turnout
- Some countries mandate voting: Australia, Belgium, Italy, Greece, Turkey, Singapore, most Latin American nations
1.2 Rational Ignorance
Rational ignorance: The theory that people will not vote — or invest time in learning about issues — if the costs of becoming informed and voting are too high, or if they feel their single vote will not be decisive in the election outcome.
Anthony Downs (An Economic Theory of Democracy, 1957): “It seems probable that for a great many citizens in a democracy, rational behavior excludes any investment whatever in political information per se… he realizes that his vote has almost no chance of influencing the outcome.”
B. F. Skinner (Walden Two, 1948): “The chance that one man’s vote will decide the issue in a national election… is less than the chance that he will be killed on his way to the polls.”
1.3 Who Votes?
Research shows that people who are more “connected” to society vote more:
| More Likely to Vote | Less Likely to Vote |
|---|---|
| Married people | Single people |
| Employed | Unemployed |
| Long-term neighborhood residents | Newcomers |
| Know their neighbors | Socially isolated |
| Higher income & education | Lower income & education |
Policy bias: Since voters tend to be married, employed, and well-educated with middle-class or higher incomes, politicians are likely to focus on their interests — potentially neglecting the needs of the unemployed, those experiencing poverty, and the less educated.
1.4 The Cost of Elections
2020 U.S. election spending:
- Total spending (president, Congress, state/local): $14.4 billion — more than double the 2016 level
- U.S. GDP in 2020: ~$21 trillion → election spending = about 1/15th of 1% of GDP
- Total government spending in 2020: ~$8.8 trillion → election spending was less than 2/10 of 1% of what officials would decide how to spend
Perspective: Americans spend ~$2 billion/year on toothpaste and ~$7 billion on hair care. Procter & Gamble alone spent almost $5 billion on advertising in 2020.
Key legal landmarks:
- Buckley v. Valeo (1976): Supreme Court ruled that the First Amendment protects campaign spending as a form of speech — government cannot forbid spending above voluntary limits
- Bipartisan Campaign Reform Act (2002): Strengthened disclosure rules but limited certain political donations and election-period advertising
- Citizens United v. FEC (2010): In a 5–4 decision, the Court ruled that spending limits on corporate contributions to political action committees were unconstitutional → led to a sharp increase in election spending
2. Special Interest Politics
2.1 How Small Groups Win Big
Special interest groups: Groups that are small in number relative to the nation but well organized and focused on a specific issue, exerting a disproportionate effect on political outcomes.
Lobbyists: Professionals who study legislation in detail, suggest alternative wording, contribute to campaigns, and may offer legislators high-paying jobs after they leave office.
Environmental regulation example:
An air pollution rule costs 10 large companies $8 million each (total: $80 million). The social benefit is $10 per person × 330 million people ≈ $3.3 trillion.
- Each company has $8 million at stake → fierce lobbying incentive
- Each citizen gains only $10 → not worth the effort to organize
- Result: The well-organized minority can defeat policy that benefits the majority
Carbon tax example (Tobaccoville):
A town of 10,000 people. A carbon tax would benefit each resident by $300 (total: $3 million). It would cost two factories $1 million each (total: $2 million).
- Factory owners: easy to coordinate, stand to lose $1M each → strong lobbying incentive
- 10,000 townspeople: costly and difficult to coordinate for just $300 each
- Despite total benefits exceeding total costs by $1 million, society is unlikely to achieve the optimal outcome
2.1.1 The Lobbying Cost-Benefit Formula
Why do small groups dominate? We can formalize it:
\[\text{Net incentive to lobby} = \text{Stake per member} - \text{Cost of organizing per member}\]For a group of $n$ members sharing total benefit (or cost) $B$:
\[\text{Stake per member} = \frac{B}{n}\]Worked Example: Steel Tariff Lobbying
Pro-tariff side: 5 steel companies share $200 million in tariff benefits.
- Stake per firm: $\frac{$200M}{5} = $40M$
- Cost to lobby (hire lobbyists, campaign contributions): ~$2M per firm
- Net incentive: $38M per firm → Absolutely worth it
Anti-tariff side: 250 million consumers share $400 million in costs.
- Cost per person: $\frac{$400M}{250M} = $1.60$
- Cost to organize politically: at least $5 per person (time, effort)
- Net incentive: $-$3.40$ per person → Rationally not worth opposing
Result: Despite total consumer losses ($400M) exceeding producer gains ($200M), the tariff passes — a deadweight loss of $200M caused by political asymmetry.
Concentrated Benefits vs. Dispersed Costs
2.2 Real-World Examples
| Case | Special Interest | Cost to Public |
|---|---|---|
| Chinese tire tariffs (2009) | United Steelworkers union (5,000 lost jobs) | Tariffs raised from 4% to 39%; cost consumers $1.11 billion/year; saved 1,200 jobs but cost 3,700 jobs elsewhere; $926,500 cost per job saved |
| Medicare prescription drugs (2003) | AARP and elderly voters (76% of ages 65–74 voted vs. 51% of 18–24) | Benefit cost $40 billion in 2006, projected $121 billion by 2016 |
| U.S. sugar quotas | Domestic sugar lobby (since 1789) | U.S. prices 40% higher than EU; cookie/candy makers must use 85% domestic sugar |
| ACA (2010) | HCAN (labor unions, community groups) spent $60 million | Offset corporate lobbying to pass new insurance regulations |
2.3 Regulatory Capture
Regulatory capture: When the industries being regulated exert considerable influence over the regulators, shaping laws and regulations to benefit themselves rather than the public.
2.4 Identifiable Winners, Anonymous Losers
Democratic systems have a bias toward policies with clearly identifiable beneficiaries and widely dispersed, hard-to-identify costs:
| Policy | Identifiable Winners | Anonymous Losers |
|---|---|---|
| Rent control | Current renters | Would-be renters who can’t find units |
| Import restrictions | Domestic firms and their workers | Consumers paying higher prices |
| Tax breaks for specific industries | Those industries | All other taxpayers |
2.5 Pork-Barrel Spending
Pork-barrel spending: Legislation that benefits mainly a single political district, with costs spread across the entire country.
Senator Robert C. Byrd of West Virginia (served 1958–2010) directed a steady stream of federal funds to his state. At least 16 structures in West Virginia were named after him — including highways, dams, institutes, courthouses, health centers, academic centers, libraries, and office complexes.
Scale of pork-barrel spending:
| Year | Earmarks | Cost |
|---|---|---|
| FY 2017 | 163 | $6.8 billion |
| FY 2021 | 285 | $16.8 billion |
| Change | +75% | +147% |
(Source: Citizens Against Government Waste, Congressional Pig Book)
2.6 Logrolling
Logrolling: A situation in which groups of legislators agree to vote for a package of otherwise unrelated laws that they individually favor — e.g., “I’ll vote for your bridge if you vote for my hospital.”
If 51% of legislators come together, they can pass a bill that includes a project for each of their districts — even if no single project would pass on its own. This is one mechanism through which pork-barrel spending proliferates.
3. Flaws in the Democratic System of Government
3.1 The Median Voter Theory
Median voter theory: Politicians will try to match policies to the preferences of the median voter — the voter in the middle of the political spectrum. This predicts “middle of the road” policy outcomes.
Reality check: The sugar lobby case shows that the minority, not the median, can dominate policy when the minority is well-organized. The median voter theory assumes equal political engagement, which rarely holds.
3.2 The Voting Cycle
Voting cycle: A situation in which, with three or more choices: A is preferred by a majority over B, B is preferred by a majority over C, and C is preferred by a majority over A. It becomes impossible to identify what the majority truly prefers.
The New Year’s dinner problem:
| Ortega Family | Schmidt Family | Alexander Family | |
|---|---|---|---|
| 1st choice | Turkey | Roast beef | Lasagna |
| 2nd choice | Roast beef | Lasagna | Turkey |
| 3rd choice | Lasagna | Turkey | Roast beef |
Pairwise voting results:
- Turkey vs. Beef → Turkey wins 2–1 (Ortega + Alexander)
- Beef vs. Lasagna → Beef wins 2–1 (Ortega + Schmidt)
- Lasagna vs. Turkey → Lasagna wins 2–1 (Schmidt + Alexander)
This forms a cycle: Turkey > Beef > Lasagna > Turkey… The outcome depends on the order choices are presented, not on majority preference.
Voting Cycle Diagram
Condorcet Paradox (1785): The Marquis de Condorcet proved that majority rule can produce intransitive (circular) preferences even when every individual voter has perfectly transitive preferences. This is not a flaw of the voters — it’s a mathematical impossibility of aggregating preferences.
Arrow’s Impossibility Theorem (1951): Kenneth Arrow proved more generally that no voting system can simultaneously satisfy all of: unanimity, independence of irrelevant alternatives, and non-dictatorship. This earned him the Nobel Prize in Economics.
Real-world implications: Imagine voting on defense spending vs. healthcare spending vs. a tax cut — a voting cycle could emerge, making the outcome dependent on who controls the agenda.
3.3 Where Is Government’s Self-Correcting Mechanism?
| Private Firms | Government Agencies |
|---|---|
| Sell products in competitive markets | Funded by tax revenue, not sales |
| Fail → go bankrupt | Rarely face existential threats |
| Customers can switch to competitors | Citizens cannot choose alternative agencies |
| Constant pressure for efficiency | Weaker pressure to improve |
Implication: When government provides goods or services directly, it may do so with less efficiency than private firms — except when government agencies compete directly with private providers (e.g., garbage collection contracted to competing firms).
3.4 A Balanced View: Three Simultaneous Truths
As F. Scott Fitzgerald wrote: “The true test of a first-rate mind is the ability to hold two contradictory ideas at the same time.” Economics requires holding three:
- Markets are extraordinarily useful — allocating scarce resources through prices, competition, and voluntary exchange
- Markets sometimes produce unwanted results — monopoly, pollution, poverty, inequality, discrimination, insurance failures, financial bubbles
- Government can help but is also imperfect — subject to rational ignorance, special interest capture, pork-barrel politics, voting cycles, and lack of market discipline
Joan Robinson: “Economic theory, in itself, preaches no doctrines and cannot establish any universally valid laws. It is a method of ordering ideas and formulating questions.”
Economics is neither conservative nor liberal. It provides the tools to analyze policy — the conclusions depend on facing the actual strengths and weaknesses of both markets and government.
4. Key Takeaways
- Rational ignorance predicts that most voters won’t invest time in learning about issues because their single vote is extremely unlikely to affect outcomes
- Voter turnout in the U.S. (55–65% in presidential elections) is lower than most developed nations
- Special interest groups succeed because they are small, organized, and each member has a large stake — while the diffuse majority has little individual incentive to oppose them
- Policies often have identifiable winners (who lobby) and anonymous losers (who don’t even realize they bear costs)
- Pork-barrel spending concentrates benefits in single districts while spreading costs nationally; logrolling multiplies such projects
- Voting cycles make majority rule logically impossible when three or more options exist
- Government agencies lack the self-correcting mechanism of market competition (no bankruptcy, no alternative providers)
- Good economic policy requires recognizing the strengths and shortcomings of both markets and government
5. Practice Questions
Q1. Explain the theory of rational ignorance. Why does it predict low voter turnout?
Answer
Rational ignorance holds that the costs of becoming informed about political issues and voting (time, effort) exceed the expected benefits, because a single vote is extremely unlikely to change the election outcome. A utility-maximizing person therefore rationally chooses to remain uninformed and may not vote at all. This helps explain why only 55–65% of Americans vote in presidential elections.Q2. U.S. election spending reached $14.4 billion in 2020. Is this too much? Frame your answer in economic terms.
Answer
In absolute terms, $14.4 billion sounds enormous. But relative to the ~$21 trillion U.S. economy, it's about 1/15th of 1% — less than what Americans spend on toothpaste and hair care combined. These elections determine how $8.8 trillion in government spending is allocated, so the campaign spending represents less than 2/10 of 1% of what's at stake. Whether this is "too much" depends on one's perspective, but the economic context suggests it's relatively small.Q3. A pollution regulation would cost 5 factories $10 million each but benefit 200 million consumers by $5 each. Will it likely pass? Why or why not?
Answer
Total costs: $50 million. Total benefits: $1 billion. Despite benefits far exceeding costs, the regulation may not pass. Each factory has $10 million at stake and strong incentive to lobby against it. Each consumer gains only $5 — far too little to motivate political action. The well-organized minority (5 factories) will likely outlobby the diffuse, unorganized majority.Q4. The Chinese tire tariff cost consumers $1.11 billion annually to save 1,200 jobs. Was this efficient? Explain.
Answer
No. The cost per job saved was $926,500 — far exceeding the ~$40,000 average annual wage of tire workers. Moreover, the tariff actually cost 3,700 jobs in other sectors (as consumers reduced spending elsewhere), resulting in a net loss of 2,500 jobs. The tariff caused a net decline in U.S. social surplus.Q5. What is regulatory capture, and how does it relate to special interest politics?
Answer
Regulatory capture occurs when the industries being regulated exert significant influence over the regulatory agencies — studying every word of legislation, lobbying legislators, contributing to campaigns, and sometimes hiring regulators after they leave government. This means the regulated industry effectively shapes the rules meant to restrain it, a special case of special interest politics.Q6. Three legislators represent districts A, B, and C. Each wants a $50 million project for their district. Individually, each project would fail a vote. Explain how logrolling enables all three.
Answer
Through logrolling, the three legislators agree to vote as a bloc for a single bill containing all three projects ($150 million total). Since they form a majority (3 out of, say, 5 legislators), the combined bill passes even though none of the individual projects had majority support. Each legislator gets their district's project, but the total cost ($150 million) is spread across all districts — including those that receive no benefit.Q7. Construct a voting cycle with three voters and three pizza toppings.
Answer
| | Voter 1 | Voter 2 | Voter 3 | |---|---|---|---| | 1st choice | Pepperoni | Mushroom | Olive | | 2nd choice | Mushroom | Olive | Pepperoni | | 3rd choice | Olive | Pepperoni | Mushroom | Pairwise: Pepperoni beats Mushroom 2–1 (Voters 1, 3); Mushroom beats Olive 2–1 (Voters 1, 2); but Olive beats Pepperoni 2–1 (Voters 2, 3). This creates a cycle: Pepperoni > Mushroom > Olive > Pepperoni. No option has a true majority preference.Q8. Why is the median voter theory sometimes wrong in practice?
Answer
The median voter theory assumes all voters participate equally and politicians reflect the median preference. In reality: (1) many voters don't vote (rational ignorance), (2) well-organized special interests can outlobby the majority, (3) campaign contributions give disproportionate influence to certain groups, and (4) voters who do participate aren't representative of the full population. The U.S. sugar quota (minority lobby dominating since 1789 despite costing all consumers) is a direct counterexample.Q9. Why doesn’t government have the same self-correcting mechanism as private markets?
Answer
Private firms face competitive pressure: if they produce unwanted products or are inefficient, customers switch to competitors and the firm goes bankrupt. Government agencies receive tax revenue regardless of performance and face no competing providers. Citizens cannot "buy" education or defense from an alternative government agency. While elected officials can reassign leadership, the pressure for efficiency is much weaker than the threat of bankruptcy.Q10. Give an example of a policy with identifiable winners and anonymous losers.
Answer
Import restrictions on foreign steel: the identifiable winners are domestic steel companies and their workers, who benefit from reduced competition. The anonymous losers are every consumer and manufacturer who buys steel — they pay higher prices but may not realize the cause. Construction companies, auto manufacturers, and appliance makers all face higher costs that are passed to their customers, but no individual consumer attributes their higher car or appliance price to the steel tariff.Q11. An election has one Pepsi-party candidate and four Coca-Cola-party candidates. Despite the population overwhelmingly preferring Coca-Cola, Pepsi wins. Explain.
Answer
The four Coca-Cola candidates split the Coca-Cola vote among themselves, while the single Pepsi candidate receives all Pepsi votes. Even if 80% prefer Coca-Cola, those votes might split 20% to each of four candidates, while Pepsi gets the remaining 20% — winning as the plurality in a "first past the post" system. This demonstrates how vote-splitting in multi-candidate races can produce outcomes that don't reflect majority preferences.Q12. “Economic theory preaches no doctrines.” Explain this quote from Joan Robinson and its relevance to public economy.
Answer
Robinson meant that economics provides analytical tools — supply and demand, cost-benefit analysis, incentive structures — without dictating political ideology. The same economic framework can reveal both market failures (justifying government intervention) and government failures (justifying market-based solutions). In public economy, this means we should neither idealize free markets nor assume government always improves outcomes. Good policy requires realistic assessment of both.Q13. A proposed regulation would impose $5 million in costs on each of 8 chemical plants but provide $2 in health benefits to each of 300 million citizens. Calculate: (a) total costs, (b) total benefits, (c) net social benefit, (d) the lobbying asymmetry ratio (stake per firm ÷ stake per citizen). Will it pass?
Answer
(a) Total costs: $5M × 8 = **$40 million** (b) Total benefits: $2 × 300M = **$600 million** (c) Net social benefit: $600M − $40M = **$560 million** (hugely beneficial) (d) Lobbying asymmetry ratio: $5,000,000 ÷ $2 = **2,500,000 : 1** Each firm has 2.5 million times the incentive per member to lobby compared to each citizen's incentive to support the regulation. Despite the massive net social benefit, this regulation will likely **fail** due to special interest lobbying — a textbook case of concentrated costs vs. dispersed benefits.Q14. A city council votes on three proposals — a new park (P), a library (L), and a stadium (S). Five council members have these preferences:
| Member | 1st | 2nd | 3rd |
|---|---|---|---|
| A | P | L | S |
| B | P | S | L |
| C | L | S | P |
| D | S | L | P |
| E | S | P | L |
(a) Is there a Condorcet winner? Test all pairwise matchups. (b) Would logrolling change the outcome?
Answer
(a) Pairwise matchups: - **P vs. L:** P preferred by A, B, E (3); L preferred by C, D (2) → **P wins 3–2** - **P vs. S:** P preferred by A, B (2); S preferred by C, D, E (3) → **S wins 3–2** - **L vs. S:** L preferred by A, C (2); S preferred by B, D, E (3) → **S wins 3–2** S beats both P and L. P beats L. The ranking is **S > P > L**. **Yes, there IS a Condorcet winner: Stadium (S)** — it beats every other option in pairwise comparison. No voting cycle exists here. (b) Even with a Condorcet winner, logrolling could change the outcome. Members A and C could form a coalition: "I'll vote for your library if you vote for my park," bundling P and L into a single bill. But since S already has majority support, logrolling is unlikely to override the S preference unless combined with side payments or unrelated issues.Q15. In 2010, the U.S. sugar program kept domestic sugar prices at roughly $0.34/lb while the world price was $0.12/lb. The U.S. consumes about 22 billion pounds of sugar annually, and there are approximately 4,500 domestic sugar farms. Calculate: (a) annual cost to consumers, (b) annual benefit per sugar farm (assuming all benefit goes to farms), (c) annual cost per U.S. household (130 million households). Does this explain why the program survives?
Answer
(a) Price premium: $0.34 − $0.12 = $0.22/lb Annual cost to consumers: $0.22 × 22B = **$4.84 billion** (b) Benefit per farm: $4.84B ÷ 4,500 = **$1,075,556 per farm** (~$1.08 million) (c) Cost per household: $4.84B ÷ 130M = **$37.23 per household per year** This perfectly illustrates the special interest asymmetry: - Each sugar farm gains **$1.08 million** → enormous incentive to lobby - Each household loses **$37** → barely noticeable, not worth opposing politically - Lobbying asymmetry: $1,075,556 ÷ $37 ≈ **29,000 : 1** This is why the sugar program has survived since 1789 — over 230 years of concentrated benefits defeating dispersed costs.6. Glossary
| Term | Definition |
|---|---|
| Rational ignorance | Theory that people won’t invest in political information because their single vote is extremely unlikely to change outcomes |
| Special interest groups | Small, well-organized groups that exert disproportionate political influence on specific issues |
| Lobbyists | Professionals who study and influence legislation on behalf of interest groups |
| Regulatory capture | When regulated industries exercise significant influence over the agencies meant to regulate them |
| Pork-barrel spending | Legislation whose benefits are concentrated on a single political district while costs are spread nationally |
| Logrolling | Legislators agreeing to vote for each other’s unrelated bills to ensure mutual passage |
| Median voter theory | Prediction that politicians match policies to the median voter’s preferences |
| Voting cycle | Situation where majority preferences are circular (A > B > C > A), making it impossible to identify a majority-preferred option |
| Tariff | Tax on imported goods |