📈 economics

Why Efficient Reforms Fail: Endogenous Game Transformation under Status Quo Bias and Social Preferences

This paper demonstrates that under status quo bias and social preferences, "hard" institutional reforms that remove inferior options are often more effective than "soft" price incentives, which merely shift behavior continuously without eliminating the persistence of suboptimal actions.

Original authors: Madjid Eshaghi Gordji, Mohammadali Berahman, Hasti Eshaghi

Published 2026-05-22

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Madjid Eshaghi Gordji, Mohammadali Berahman, Hasti Eshaghi

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a group of people trying to decide which path to take to get to a destination. One path is the old, familiar road they've always used (the Status Quo). The other path is a newer, faster, and better road (the Superior Alternative).

Even though everyone knows the new road is better, the group keeps getting stuck on the old road. Why? This paper uses a mix of game theory and psychology to explain why "gentle nudges" often fail to get people to switch, while "hard bans" work much better.

Here is the breakdown of the paper's main ideas using simple analogies:

1. The Problem: The "Lazy Switch"

Imagine you are used to driving a gas car. Switching to an electric car is better for the planet and saves money in the long run. But there are Switching Costs: you have to sell your old car, learn new habits, find charging stations, and deal with the hassle of change.

The paper argues that people aren't perfect robots who instantly calculate the best move. They are Boundedly Rational. They make decisions with a bit of "noise" or hesitation. Even if the new road is slightly better, the fear of the hassle (the switching cost) keeps them on the old road.

2. The Two Tools: "The Price Tag" vs. "The Roadblock"

The government (or a leader) wants to get everyone to switch to the new road. They have two main tools:

Tool A: The Price Tag (Soft Reform)
This is like putting a tax on gas or giving a fine for using the old road. It makes the old road more expensive.
- The Paper's Finding: No matter how high you set the tax, as long as the old road still exists, some people will still choose it. Why? Because people are hesitant. Even if the tax is huge, a few people will still say, "I'll pay the fine because I don't want to deal with the hassle of switching." The tax shifts behavior, but it never completely wipes out the old choice.
- Analogy: It's like putting a $1,000 fee on eating broccoli. Some people will stop eating it, but a few stubborn people will still pay the fee because they really like broccoli.
Tool B: The Roadblock (Hard Reform)
This is like deleting the old road entirely. You ban the gas car or remove the "infinite scroll" button from a phone app.
- The Paper's Finding: This is the only way to guarantee the old choice disappears. If the option is physically removed from the menu, the probability of choosing it drops to zero.
- Analogy: If you lock the door to the broccoli section of the grocery store, nobody can buy it, no matter how much they want to.

The Big Lesson: You can make the old way so expensive that almost nobody chooses it, but you can never make it exactly zero unless you simply delete the option. A "Roadblock" is qualitatively different and more powerful than a "Price Tag."

3. The "Spiteful Neighbor" Problem

The paper also looks at what happens when people care about each other (or hate each other). Imagine a group of friends voting on whether to switch to the new road.

The Scenario: Everyone agrees the new road is better for everyone. It's a win-win.
The Twist: One person in the group is "spiteful." They don't care about their own gain; they just hate the idea of the others winning.
The Result: If the group requires Unanimity (everyone must agree), that one spiteful person can block the entire reform. They will say "No" just to spite the others, even though it hurts them too.
The Fix: If the group uses Majority Voting instead, the spiteful person can be outvoted, and the reform can happen.

4. Real-World Examples the Paper Uses

The authors apply these ideas to two specific, real-world situations:

Climate Change (Gas vs. Electric):
- Soft Approach: A Carbon Tax (making gas expensive). The paper says this will reduce gas use, but it will never reach zero because the option still exists.
- Hard Approach: A Ban on new gas cars (like the EU's 2035 plan). This deletes the option, forcing the switch to zero.
- Conclusion: To fully stop fossil fuels, you often need the "Roadblock" (ban), not just the "Price Tag" (tax).
Social Media (Addictive Features):
- Soft Approach: Fines for platforms that use "infinite scroll" or "autoplay." The platforms might pay the fine and keep the feature because it makes them money.
- Hard Approach: A law that deletes the feature entirely.
- Conclusion: Fines won't stop the feature completely; only deleting it will. Also, if one board member hates safety regulations because they love the ad revenue, they can block the change if the company requires unanimous agreement.

Summary

The paper proves mathematically that gentle nudges (taxes/fines) have limits. They can change behavior gradually, but they cannot force a complete stop as long as the bad option is still available. To truly eliminate an inferior habit or technology, you often need hard constraints (bans/deletion) that remove the option from the table entirely. Additionally, if the decision-making process requires everyone to agree, a single grumpy or spiteful person can ruin a good plan for everyone.

Technical Summary: Why Efficient Reforms Fail

Problem Statement
The paper addresses the persistent failure of societies to transition from inferior institutions, technologies, or social norms to widely recognized superior alternatives. While standard explanations cite coordination failure, path dependence, and strategic uncertainty, this work focuses on a complementary dimension: the strategic choice of the game itself. The authors investigate why "efficient" reforms often fail when agents exhibit status-quo bias (inertia due to switching costs) and bounded rationality (noisy decision-making). Specifically, the paper asks: under what conditions do "soft" price interventions (taxes/subsidies) succeed versus "hard" feasibility interventions (deletion/bans), and how do antagonistic social preferences at the meta-level (rule-changing level) obstruct Pareto-improving reforms?

Methodology
The authors develop a formal framework integrating three core elements:

Bounded Rationality with Status-Quo Bias: The model utilizes a Quantal Response Equilibrium with Status-Quo Bias (QRE-SB). Agents choose actions based on a logit choice rule where the probability of selecting an action depends on its expected payoff minus a switching cost ( $\kappa$ ) if the action differs from the default status quo. The precision parameter $\beta$ governs the sensitivity to payoff differences.
Endogenous Game Transformations: Agents (or a policymaker) can choose "meta-actions" to transform the underlying game. The model distinguishes between:
- Price-only transformations ( $T^P$ ): Modifying payoffs via taxes or subsidies while keeping all actions feasible.
- Feasibility transformations ( $T^D, T^A, T^R$ ): Deleting, adding, or replacing actions, thereby altering the feasible set.
Meta-Level Equilibrium Concepts: The paper defines two equilibrium concepts for the "meta-game" (the game of choosing rules):
- Meta-Nash Equilibrium: Standard Nash equilibrium where players maximize their own expected payoffs in the transformed game.
- Hyper-Meta-Nash Equilibrium: Extends the concept to include social preferences (friendship or enmity/antagonism), where players maximize a weighted sum of their own payoff and the payoffs of others.

The analysis focuses on a symmetric binary coordination game where one equilibrium is Pareto-superior but the status quo is locked in by switching costs.

Key Contributions and Results
The paper derives several comparative statics and theoretical results regarding the efficacy of reforms:

Comparative Statics of Persistence: In a QRE-SB framework, the probability of choosing the status quo increases with the switching cost ( $\kappa$ ) and decreases with taxes imposed on the status quo action.
Qualitative Non-Equivalence of Interventions (Theorem 4.2 & Proposition 4.6): Under logit QRE-SB with finite precision ( $\beta < \infty$ $β < \infty$ ), no finite tax can replicate the effect of deletion.
- Price-only interventions: Because logit probabilities are strictly positive for all feasible actions, a tax can only shift behavior continuously; the status quo action retains a non-zero probability of being chosen regardless of how high the tax is.
- Deletion interventions: Removing an action from the feasible set forces its selection probability to exactly zero.
Approximation vs. Replication (Theorem 4.3): While arbitrarily high taxes can approximate the effect of deletion (driving the probability of the status quo arbitrarily close to zero), they can never achieve it exactly.
Blocking by Antagonistic Preferences (Theorem 4.4): Under a unanimity rule for reform, a single sufficiently spiteful player (one with strong negative weights on others' payoffs) can block a reform that is individually beneficial for every player. The condition for blocking is that the spiteful player's loss in utility from others' gains outweighs their own gain.
Welfare Thresholds: The paper establishes a cost threshold determining when deletion dominates taxation, providing a formal rationale for when hard constraints are superior to soft incentives.

Significance and Applications
The authors claim the framework provides a formal rationale for why "hard" feasibility restrictions often dominate "soft" price incentives in the presence of inertia. The paper explicitly applies these findings to two pressing policy domains:

Climate Transition: The model explains why carbon taxes (price-only) may be insufficient to achieve complete decarbonization compared to fossil-fuel phase-outs (deletion). Even high taxes leave a non-zero probability of fossil fuel use, whereas bans eliminate the option entirely. It also suggests that unanimity rules in international agreements may be vulnerable to blocking by spiteful incumbents (e.g., oil companies), whereas majority voting might overcome this.
Platform Regulation: The framework justifies the deletion of addictive features (e.g., infinite scroll) over merely fining them. Since no finite fine can reduce the probability of using a harmful feature to zero, regulatory bans are necessary to fully eliminate the behavior.

Conclusion
The paper concludes that under bounded rationality and status-quo bias, the distinction between changing payoffs and changing the set of available actions is fundamental. While soft incentives shift behavior continuously, hard constraints can eliminate undesirable actions entirely. Furthermore, the design of voting rules (unanimity vs. majority) is critical when social preferences (specifically antagonism) are present, as they can prevent Pareto-improving reforms even when all players would individually benefit. The authors note that future work will extend the model to dynamic learning, network interactions, and empirical calibration.