Post
JA EN

Why Rules Only Ever Increase — The Psychology Behind How One Incident Creates a Hundred Regulations

Posted
By
21 min read
Why Rules Only Ever Increase — The Psychology Behind How One Incident Creates a Hundred Regulations
  • Target audience: Business professionals, managers, and executives
  • Reading time: 20 min

Overview

“Do we really need this rule?” — Everyone has thought this at least once. Yet the question is rarely voiced aloud, and the rules keep accumulating. The U.S. Code of Federal Regulations (CFR) ballooned from 22,877 pages in 1960 to 188,343 pages in 2021 — more than an eightfold increase1. The EU’s body of law exceeds 660,000 pages2. At American universities, administrative staff have grown at ten times the rate of faculty3.

This is not a coincidence. Rules keep growing because of cognitive biases and structural incentives that create nearly irresistible psychological mechanisms. This article examines six such mechanisms — backed by empirical data — to explain why rules are a one-way street.

1. Loss Aversion — How “Just in Case” Distorts Rational Judgment

1.1 Losses Hurt Twice as Much as Gains Feel Good

Kahneman and Tversky’s Prospect Theory (1979) and subsequent research uncovered a fundamental asymmetry in human decision-making4:

The pain of losing is felt roughly 2 to 2.25 times more intensely than the pleasure of an equivalent gain.

Finding $100 on the sidewalk feels good. Losing $100 feels devastating. This is loss aversion.

1.2 Applying This to Organizational Decision-Making

Map this asymmetry onto how organizations design rules:

 Add the ruleDon’t add the rule
Problem occurs“We had safeguards in place” (small loss)“Why didn’t you do something?” ( large loss )
No problem occursOngoing cost burden, but invisibleNothing happens (zero gain)

The critical insight is that the cost of adding a rule — reduced productivity, lower engagement, wasted time — is difficult to see. Meanwhile, the cost of a compliance failure — media coverage, stock price drops, lawsuits, career damage for the person responsible — is highly visible and attributed to specific individuals.

The result: “add the rule just in case” is always the individually rational choice, even when it is collectively irrational for the organization.

1.3 The Endowment Effect — Once a Rule Exists, It Stays

Thaler’s (1980) endowment effect compounds the problem5. People assign greater value to things they already possess than to things they could acquire.

The same applies to rules. Once implemented, a rule becomes organizational “property,” and abolishing it feels like losing something — even when it produces no measurable benefit.

2. Availability Cascades — How One Incident Distorts Reality

2.1 Self-Reinforcing Risk Perception

Kuran and Sunstein (1999) formalized the availability cascade, a self-reinforcing mechanism that amplifies risk perception far beyond its objective level6:

flowchart TB
    A["A compliance violation<br>is discovered"] --> B["Media and internal<br>channels repeat the story"]
    B --> C["Perceived risk rises<br>far above actual risk"]
    C --> D["Not having countermeasures<br>becomes a risk in itself"]
    D --> E["Regulations adopted that are<br>disproportionate to actual risk"]
    E --> F["Existence of regulations<br>reconfirms the severity of risk"]
    F --> C

    style A stroke:#cc0000,stroke-width:3px
    style E stroke:#cc9900,stroke-width:3px

What makes this cycle so powerful is the dual-driver structure of informational and reputational cascades:

  • Informational cascade: “Other people say it’s dangerous, so it must be.”
  • Reputational cascade: “If I say the risk is overblown, I’ll be seen as reckless — better to stay quiet and go along.”

In other words, the incentive to conform is stronger than the incentive to speak the truth.

2.2 “Availability Entrepreneurs” — Who Profits from Rule Proliferation

Kuran and Sunstein coined the term “availability entrepreneurs” for people who deliberately exploit — or accelerate — these cascades6.

They use incidents and scandals as raw material to advance their own agendas:

Availability entrepreneurMotivationBehavior
Compliance departmentExpand headcount and budgetArgue that “more controls are needed”
External consultantsWin consulting contractsEmphasize severity of risks, propose solutions
Legal departmentMinimize litigation exposureDesign rules using the most conservative interpretation
Risk-averse managersAvoid problems on their watchReduce subordinate discretion, add approval layers

This is not a question of bad faith. It is a question of structural incentives. As long as people within an organization stand to gain personally from adding rules, rules will keep growing.

2.3 Case Study: The Alar Scare (1989)

One of the central examples in Kuran and Sunstein’s paper is the Alar scare in the United States6.

In 1989, the NRDC (Natural Resources Defense Council) published a report on the cancer risk posed by Alar, a growth regulator used on apples. The media ran with the story, triggering a nationwide boycott of apples. The state of Washington alone suffered an estimated $100 million or more in losses.

Yet the report’s risk estimates diverged significantly from the scientific consensus. The manufacturer withdrew Alar voluntarily under consumer panic before the EPA could even complete its formal review.

One report leads to media amplification leads to consumer panic leads to disproportionate response — the exact same pattern that drives rule proliferation inside organizations.

3. Status Quo Bias — The Deck Is Always Stacked Against Removing Rules

3.1 The Overwhelming Power of the Default

Samuelson and Zeckhauser (1988) demonstrated status quo bias: people prefer the current state of affairs to an irrational degree7.

Both their experiments and real-world data — including Harvard faculty choices among health insurance and retirement plans — showed that people stick with clearly inferior options simply because those options are the status quo.

Apply this to rules:

  • Keeping a rule (status quo) = the default. No action required.
  • Abolishing a rule (changing the status quo) = an active decision. The proposer bears all responsibility.

This asymmetry is decisive. The person who adds a rule may be praised for “protecting the organization.” The person who removes one risks being blamed for “taking unnecessary risks.”

3.2 Regret Avoidance Blocks Abolition

A powerful driver behind status quo bias is regret avoidance7:

  • Keep the rule and no problem occurs: “See, the rule was protecting us.”
  • Keep the rule and a problem occurs: “The rule wasn’t enough” — and more rules are added.
  • Abolish the rule and no problem occurs: Nobody notices (zero gain).
  • Abolish the rule and a problem occurs: “Removing that rule caused this” — the individual is blamed.

Of the four scenarios, abolishing a rule is rewarded in none. And in one scenario, it results in severe career damage. For any rational individual, there is zero incentive to propose eliminating a rule.

4. Organizational Moral Panic — How One Incident Constrains Everyone

4.1 The Five Conditions of Moral Panic

Sociologist Cohen (1972) introduced the concept, and Goode and Ben-Yehuda (1994) systematized moral panic theory, which explains how societies react disproportionately to perceived threats89.

Goode and Ben-Yehuda’s five conditions map neatly onto compliance incidents within organizations:

ConditionSocietal levelHow it manifests inside organizations
1. ConcernWidespread belief that a behavior or group poses a threat“Could the same thing happen here?” anxiety spreads
2. HostilityIncreased hostility toward “folk devils”Blame directed at the offending department or individual; criticism of “employees who don’t take compliance seriously”
3. ConsensusBroad agreement that the threat is real“Not taking action would be negligent” becomes the prevailing view
4. DisproportionalityResponse exceeds objective harmA single violation triggers company-wide rule changes and mandatory training
5. VolatilityPanic rises rapidly and then subsidesIntense reaction right after the scandal, interest fades within months — but the rules remain

Condition 5 — volatility — is the key. The panic passes, but the rules created during the panic are never rescinded. Regulations crafted at the peak of emotional intensity become permanent fixtures of the institution.

4.2 “Moral Entrepreneurs” Meet “Availability Entrepreneurs”

Cohen’s (1972) “moral entrepreneurs” and Kuran and Sunstein’s (1999) “availability entrepreneurs” are, within organizations, effectively the same people68.

They exploit the “window” that a scandal opens to push through rules and departmental expansions that would never have been approved under normal circumstances. In the immediate aftermath of an incident, the prevailing atmosphere is “if not now, when?” — and any discussion of costs or side effects is suppressed.

Recognizing this structure is essential. The issue is not whether the motives of those proposing new rules are pure. The issue is that the timing of the proposal — during a panic — prevents sober evaluation.

5. Parkinson’s Law and the Expansion of Administrative Functions

5.1 “Work Expands to Fill the Time Available”

In 1957, British historian C.N. Parkinson articulated Parkinson’s Law — an observation that reads as satire but contains a profound truth about organizational growth10.

Parkinson identified two core mechanisms:

  1. The law of multiplication of subordinates: Managers want to hire subordinates, not rivals.
  2. The law of multiplication of work: Administrators create work for each other.

Klimek et al. (2009) tested this quantitatively11. When the Medical University of Vienna became independent in 2004, its administrative staff grew from 15 to 100 — while academic staff remained essentially unchanged. When a new organization is created, the administrative layer balloons.

5.2 Administrative Bloat in U.S. Universities

The most striking data comes from American higher education:

MetricPeriodChange
New administrative and professional hires1987-2012517,636 added (87 per business day)3
Growth rate of administrators1993-200960% increase ( 10 times the rate of full-time faculty)3
Administrators per 100 students1993-200739% increase (professors: 18%)3
Administrative spending per student1993-200761% increase3

At Yale University, the number of administrators and managers ( over 5,460 ) exceeds the number of undergraduates ( under 5,000 )3.

Why does this happen? Administrative departments create rules. Those rules require monitoring, which requires a new department. That department generates reports, which require a process to manage, which requires additional staff. Rules create positions, and positions create rules — a self-perpetuating cycle.

6. The Ratchet Effect — Crises Make Rules Irreversible

6.1 The Peacock-Wiseman Hypothesis

Peacock and Wiseman (1961) analyzed UK fiscal data and discovered that government spending increases in a staircase pattern12:

Peacetime -> Crisis (war) -> Spending surges -> Crisis ends -> Spending never returns to pre-crisis levels

This “displacement effect” occurs because the public’s threshold for acceptable burden shifts upward during a crisis — and the new baseline persists long after the crisis is over.

6.2 Higgs and the Ratchet Effect

Higgs (1987) extended this analysis and formalized it as the ratchet effect13.

Crises expand government authority, regulation, and bureaucracy. Post-crisis contraction is always partial, leaving behind institutional residues. More importantly, the crisis triggers an ideological shift in which new forms of intervention come to be accepted as “normal.”

flowchart TB
    A["Peacetime<br>rule level"] -->|"Crisis occurs"| B["Rules<br>surge"]
    B -->|"Crisis ends"| C["Some rules<br>are rolled back"]
    C --> D["New 'peacetime'<br>level<br>(higher than before)"]
    D -->|"Next crisis"| E["Rules surge<br>even further"]
    E -->|"Crisis ends"| F["Some rules<br>are rolled back"]
    F --> G["Even higher<br>'peacetime' level"]

    style B stroke:#cc0000,stroke-width:3px
    style E stroke:#cc0000,stroke-width:3px
    style D stroke:#cc9900,stroke-width:3px
    style G stroke:#cc9900,stroke-width:3px

6.3 The Corporate Ratchet in Practice

Though originally a theory about government, the same pattern is observable inside corporations.

Enron (2001) -> The Sarbanes-Oxley Act (SOX, 2002)

In response to the accounting fraud at Enron and WorldCom, SOX imposed stringent internal controls on all publicly traded U.S. companies. Academic research estimated compliance costs at roughly $19 billion per year across 1,428 firms, with profitability declines lasting up to four years after implementation14. A GAO (2025) report confirmed that the cost burden falls disproportionately on smaller companies15.

More than two decades after SOX was enacted, its core requirements have barely been relaxed. The goal of preventing corporate fraud is legitimate, but not every company commits Enron-level fraud — yet every company bears the same compliance cost. This is the ratchet in action.

September 11 (2001) -> The TSA (Transportation Security Administration)

The TSA, created in the wake of 9/11, has seen its annual budget grow steadily from its founding to approximately $11.8 billion in FY2025. Yet as security expert Bruce Schneier famously dubbed it, much of what the TSA does amounts to “security theater”16:

  • GAO (2010): The $900 million behavior-detection program (SPOT) identified zero terrorists while at least 16 individuals with known ties to terrorism passed through undetected.
  • GAO (2013): “No evidence” that behavioral indicators can identify aviation security risks.
  • Red-team testing: The TSA failed to detect 95% of smuggled weapons and explosives.

Against an annual budget of $11.8 billion, this performance is plainly disproportionate. Yet any political proposal to cut the TSA’s budget is virtually impossible — because “reducing counterterrorism” triggers loss aversion. The TSA is itself a textbook ratchet effect.

7. The Numbers Behind the One-Way Street

The psychological mechanisms above are not just theoretical. Here is what they have produced in practice.

7.1 The Expansion of the U.S. Code of Federal Regulations

YearCFR pagesNotes
196022,877Pre-Kennedy era
197571,224Tripled
2008157,9747x the 1960 level
2021188,343Over 8x

From Truman to Biden, every single president left office with more CFR pages than when they entered — including presidents who campaigned on deregulation1.

Coffey, McLaughlin, and Peretto (2020) estimated that cumulative regulation since 1980 has depressed U.S. GDP growth by approximately 0.8% per year. Had regulatory volume been held at its 1980 level, the U.S. economy would have been roughly 25% larger — about $4 trillion — by 201217.

7.2 The EU’s Regulatory Mass

The EU’s body of law (the acquis communautaire) has reached 666,879 pages since its founding in 1957, with over 170,000 pages of currently active legislation. In the past decade alone, more than 100,000 pages were added2.

7.3 Japan’s Internal Control Costs

Japan exhibits the same pattern. The Kobe Steel quality data falsification scandal (2017) affected manufacturers worldwide — Toyota, Nissan, Honda, Boeing — and sent shockwaves through Japanese corporate governance18. In response, Japan’s Corporate Governance Code (introduced 2015, revised 2018 and 2021) was progressively tightened, increasing compliance costs for listed companies.

Ironically, recent discussions in Japan have begun to focus on simplifying the Governance Code — recognition that the cost of formally maintaining requirements that have already been internalized is becoming counterproductive18.

8. Rules Can Be Reduced — Counterexamples and Prescriptions

8.1 British Columbia’s Success

The Canadian province of British Columbia (BC) introduced a “two-out-for-every-one-in” policy in 2001 and achieved dramatic results19:

  • By 2004, 37% of regulatory requirements had been eliminated (exceeding the one-third target)
  • Cumulative reduction to date: 42.8%
  • Currently operating under a “net-zero increase” policy
  • The Canadian federal government adopted this model, enacting it as the Red Tape Reduction Act

8.2 The UK’s One-In-One-Out

The UK introduced One-In-One-Out (OIOO) in 2011, escalating to One-In-Two-Out (OITO) in 2012 and One-In-Three-Out in 201620.

Between 2010 and 2015, departments exceeded their targets and removed 963 million pounds more in business burden than they introduced. However, a National Audit Office (NAO) analysis found that 46% of regulatory decisions were excluded from the OIOO/OITO calculations, and critics argue that the actual burden on business may have increased. The fact that even a “rule-reducing mechanism” develops loopholes speaks to how deeply these biases are entrenched.

8.3 Denmark’s Trust Reform

In 2013, Copenhagen launched a trust reform (Tillidsreform), which subsequently spread to Sweden (2014) and Norway (2021)21. The approach replaces excessive control with trust: in home care services, for example, staff were granted greater professional autonomy.

8.4 Common Success Factors

Three conditions emerge from these counterexamples:

ConditionBCUKDenmark
1. Quantitative targets2-for-1 rule1-in-X-out ruleLimited (qualitative)
2. Clear commitment from the topPremier-ledPrime Minister-ledMayor / municipal leader-led
3. InstitutionalizationEnacted into lawCabinet decision -> department KPIsLabor-management agreements

Exhortations to “reduce rules” are insufficient. What is needed are rules for reducing rules — structural mechanisms like sunset clauses and one-in-X-out policies.

Precisely because the psychological mechanisms push so strongly toward accumulation, the countervailing force must come from deliberate institutional design.

Conclusion — Six Biases Create a One-Way Ratchet

Rules keep growing not because of laziness or malice, but because of the structure of human cognition itself.

flowchart TB
    INCIDENT["Incident or scandal<br>occurs"] --> LA["1. Loss aversion<br>The pain of loss feels<br>2x the cost of prevention"]
    INCIDENT --> AC["2. Availability cascade<br>Risk perception<br>self-amplifies"]

    LA --> RULE["Rules are added"]
    AC --> RULE

    RULE --> EE["3. Endowment effect<br>Rules become<br>organizational 'property'"]
    RULE --> SQ["4. Status quo bias<br>Only the person proposing<br>abolition bears risk"]

    EE --> STAY["Rules are maintained"]
    SQ --> STAY

    STAY --> PK["5. Parkinson's law<br>Administrative functions expand<br>and reproduce rules"]
    STAY --> RT["6. Ratchet effect<br>The next crisis<br>raises the floor again"]

    PK --> INCIDENT
    RT --> INCIDENT

    style INCIDENT stroke:#cc0000,stroke-width:3px
    style RULE stroke:#cc9900,stroke-width:3px
    style STAY stroke:#cc6600,stroke-width:3px
MechanismHow it worksOutcome
1. Loss aversionThe cost of a violation looks greater than the cost of a ruleAdding a rule is always the “safe” choice
2. Availability cascadeRisk perception self-amplifiesExcessive countermeasures appear “necessary”
3. Endowment effectAbolishing an existing rule feels like a lossEven useless rules cannot be let go
4. Status quo biasOnly the proposer of abolition bears riskNobody proposes removing rules
5. Parkinson’s lawAdministrative functions self-replicateThe number of rule-makers keeps growing
6. Ratchet effectCrises irreversibly raise the baselineThe old level is never restored

All six mechanisms operate simultaneously, all in the same direction. There are six engines driving rules upward and no brakes pulling them back. It is no accident that the U.S. CFR grew eightfold in sixty years, or that corporate compliance costs keep rising. It is an inevitability.

That is precisely why jurisdictions like British Columbia and the UK have shown that the only viable prescription is to deliberately embed rule-reducing mechanisms into institutional design. Telling people to “be aware of their biases” is futile. To counteract a bias, you must engineer a structure that pushes in the opposite direction.

“Do we really need this rule?” — the goal is to guarantee that this question is asked not through individual courage, but through organizational mechanisms. That is the only way to stop the one-way ratchet.

For more on related themes, see these articles:

References

References are listed by citation number as they appear in the text.

Additional references (not cited by number in the text)

  1. Code of Federal Regulations Statistics - George Washington University Regulatory Studies Center / Pacific Legal Foundation analysis. CFR page counts over time: from 22,877 pages in 1960 to 188,343 in 2021, an eightfold increase. [Reliability: High] ↩︎ ↩︎2

  2. The Acquis Communautaire: Just How Big Is It? - AALEP. The EU’s total legislative volume stands at 666,879 pages, with over 170,000 pages currently in force. [Reliability: Medium-High] ↩︎ ↩︎2

  3. Administrative Bloat in US Universities - Goldwater Institute (2010) / AEI analysis / Bloomberg / Department of Education data. Administrative growth was 10x the rate of full-time faculty. [Reliability: Medium-High] ↩︎ ↩︎2 ↩︎3 ↩︎4 ↩︎5 ↩︎6

  4. Prospect Theory: An Analysis of Decision under Risk - Kahneman, D. & Tversky, A. (1979). Econometrica, 47(2), 263-291. / Advances in Prospect Theory: Cumulative Representation of Uncertainty - Tversky, A. & Kahneman, D. (1992). Journal of Risk and Uncertainty, 5(4), 297-323. The original Prospect Theory paper and the follow-up study quantifying the loss aversion coefficient at approximately 2.25. [Reliability: High] ↩︎

  5. Toward a Positive Theory of Consumer Choice - Thaler, R. (1980). Journal of Economic Behavior & Organization, 1(1), 39-60. Introduced the endowment effect. [Reliability: High] ↩︎

  6. Availability Cascades and Risk Regulation - Kuran, T. & Sunstein, C.R. (1999). Stanford Law Review, 51(4), 683-768. Introduced the availability cascade concept and the term “availability entrepreneurs.” Includes case analysis of the Alar scare. [Reliability: High] ↩︎ ↩︎2 ↩︎3 ↩︎4

  7. Status Quo Bias in Decision Making - Samuelson, W. & Zeckhauser, R.J. (1988). Journal of Risk and Uncertainty, 1(1), 7-59. Empirical demonstration of status quo bias. [Reliability: High] ↩︎ ↩︎2

  8. Folk Devils and Moral Panics: The Creation of the Mods and Rockers - Cohen, S. (1972). London: MacGibbon & Kee. The foundational text on moral panic theory. [Reliability: High] ↩︎ ↩︎2

  9. Moral Panics: The Social Construction of Deviance - Goode, E. & Ben-Yehuda, N. (1994). Wiley-Blackwell. Systematized the five conditions of moral panic: concern, hostility, consensus, disproportionality, and volatility. [Reliability: High] ↩︎

  10. Parkinson’s Law, and Other Studies in Administration - Parkinson, C.N. (1957). Boston: Houghton Mifflin. Analysis of bureaucratic expansion mechanisms. [Reliability: Medium-High] ↩︎

  11. Parkinson’s Law Quantified: Three Investigations on Bureaucratic Inefficiency - Klimek, P., Hanel, R., & Thurner, S. (2009). Journal of Statistical Mechanics: Theory and Experiment, P03008. Quantitative validation of Parkinson’s Law, including the case of the Medical University of Vienna where administrative staff grew from 15 to 100. [Reliability: High] ↩︎

  12. The Growth of Public Expenditure in the United Kingdom - Peacock, A.T. & Wiseman, J. (1961). Princeton University Press. Identified the “displacement effect” in government spending: post-crisis expenditure levels never return to pre-crisis norms. [Reliability: High] ↩︎

  13. Crisis and Leviathan: Critical Episodes in the Growth of American Government - Higgs, R. (1987). Oxford University Press. Formalized the ratchet effect: crises irreversibly expand government authority. [Reliability: High] ↩︎

  14. How Costly is the Sarbanes Oxley Act? Evidence on the Effects of the Act on Corporate Profitability - Ahmed, A.S., McAnally, M.L., Rasmussen, S.J. & Weaver, C.D. (2010). Journal of Corporate Finance, 16(3), 352-369. Estimated SOX compliance costs at approximately $19 billion per year across 1,428 firms, with profitability declines lasting up to four years. [Reliability: High] ↩︎

  15. Sarbanes-Oxley Act: Compliance Costs Are Higher for Larger Companies but More Burdensome for Smaller Ones - GAO (2025). Confirmed that SOX compliance costs fall disproportionately on smaller companies. [Reliability: High] ↩︎

  16. TSA Security Theater - Schneier, B. (2003). Beyond Fear: Thinking Sensibly About Security in an Uncertain World. / GAO: TSA Should Limit Future Funding for Behavior Detection Activities (2013). / GAO: Covert Testing (2015). The SPOT program detected zero terrorists. Red-team testing found a 95% failure rate for detecting weapons. [Reliability: Medium-High] ↩︎

  17. The Cumulative Cost of Regulations - Coffey, B., McLaughlin, P.A., & Peretto, P. (2020). Review of Economic Dynamics, 38, 1-21. Estimated that cumulative regulation has depressed GDP growth by approximately 0.8% per year, meaning the U.S. economy could have been roughly 25% ($4 trillion) larger by 2012. [Reliability: High] ↩︎

  18. Corporate quality scandals in Japan - Corporate misconduct in Japan (2025) / The Diplomat - Trust Deficit (2025). The Kobe Steel data falsification scandal and subsequent tightening of Japan’s Corporate Governance Code. [Reliability: Medium-High] ↩︎ ↩︎2

  19. Lessons from the British Columbia Model of Regulatory Reform - Jones, L., Mercatus Center / BC Government. BC’s 2-for-1 policy achieved a cumulative 42.8% reduction in regulatory requirements and served as the model for Canada’s federal Red Tape Reduction Act. [Reliability: Medium-High] ↩︎

  20. One-In-One-Out / One-In-Two-Out - UK Government. Between 2010 and 2015, departments exceeded targets, removing 963 million pounds more in business burden than they introduced. [Reliability: High] ↩︎

  21. On the diffusion and implementation of trust-based management in Scandinavia - Bentzen, T.O. et al. (2024). International Journal of Public Sector Management, 37(1). Analysis of how Denmark’s trust reform spread to other Scandinavian countries. [Reliability: High] ↩︎

Psychology
cognitive bias psychology management decision-making organizational behavior
This post is licensed under CC BY 4.0 by the author.