Unintended Consequences: How Good Intentions Lead to Bad Outcomes

Whether you are governing a country, managing a company, or simply trying to raise your children, when you hold power, you have an instinctive urge: I will reward what I want; I will prohibit what I don’t want; I will approve what I worry about; I will evaluate what I value.

If you think this way, you are tacitly treating society as a linear machine, believing that as long as the remote control is in the hands of good people — and you are that good person — everything will be fine.

Little do you know that good intentions do not necessarily lead to good results. Reality often turns out contrary to expectations, and good intentions frequently pave the way to bad outcomes.

Good Intentions, Unintended Outcomes: Real-World Cases #

In 1994, to protect poor tenants, the San Francisco city government implemented rent control. As a result, landlords calculated that the rents were no longer profitable, so they simply sold their properties or converted them to other uses. Consequently, the supply of rental housing on the market decreased by about 15%, and rents increased by 5%, making it even harder for the poor to find housing [1].

In 1973, to protect endangered animals, the United States passed the Endangered Species Act. As a result, some landowners, fearing development restrictions if endangered animals were found on their land, simply preemptively cut down trees that could serve as habitats [2].

In 2014, to reduce white pollution, California banned supermarkets from providing traditional free thin plastic bags. As a result, retailers replaced them with thicker, “reusable” plastic bags that came with a charge — but consumers still used them as single-use bags and discarded them after one use. By 2022, plastic bag waste in landfills across the state did not decrease but surged, skyrocketing from 157,000 tons to 231,000 tons [3].

In 2009, to stimulate car consumption, the U.S. launched the “Cash for Clunkers” program. The short-term result was indeed an increase in new car sales — but subsequent research found that most purchases merely brought forward decisions that would have been made a few months later: as soon as the subsidies stopped, sales plummeted back to their original levels [4].

…There are simply too many similar examples. In the words of sociologist Robert K. Merton, these are the “Unanticipated Consequences” of purposive social action [5].

The Deep Root Cause of Unintended Consequences: People Respond to Incentives #

Why do unintended consequences occur? We cannot simply lament, “You have your plans, and the world has its own.” One of the most important insights in economics is that “people respond to incentives” — and pay particular attention to this:

People respond to incentives, not intentions.

If you believe that because you have good intentions, people should strive to realize those intentions under your call, then you are oversimplifying society. No matter how much power you wield, or how freely you can introduce any policy or establish any new rule, what you change is not the outcome itself, but the situation.

People will react to this new situation, but not necessarily in the way you expect.

There was a kindergarten where parents were always late picking up their children, which bothered the administration (intention). So, they set a rule (policy): a fine for being late (incentive situation). The parents’ reaction to this new situation was to be even more late.

Why? Before the fine policy, picking up children on time was a moral obligation, and being late would cause guilt. With the fine, being late became a service with a clear price tag — I pay a bit more, buying an extra half-hour of the teacher’s time, and any feeling of guilt vanished.

This was a classic experiment conducted by two economists in several kindergartens in Israel [6]. Later, the administration recognized the policy’s failure and removed the fine, but the lateness rate never dropped back down. Once that layer of moral obligation was squeezed out by price, it couldn’t be restored.

You might think:

Policy → Behavior Change → Desired Outcome.

The real world, however, is:

Policy → Change in Incentives + Change in Expectations + Distortion in Implementation → People’s Re-optimization of Behavior → System Response → Unintended Consequences.

The key point is that people will re-optimize their behavior based on your new incentive situation.

Above, it’s about intentions; below, it’s about price.

Above, it’s about values; below, it’s about profit statements.

Above, it’s about direction; below, it’s about constraints.

This is not about “the public having low awareness.” This is the fundamental physics of social systems. In the words of Nobel laureate economist Robert Lucas: You cannot use behavioral relationships observed under old regimes to predict the effects of new policies, because new policies themselves change people’s expectations and behavior — people call this the “Lucas critique” [7].

With power, you can choose your policies, but you cannot choose the consequences of your policies.

Alienation in Policy Transmission: When Intentions Meet Evaluation #

Let’s look at how good intentions, in more serious contexts, can gradually turn into their opposite during transmission down the hierarchy.

Take Wang Anshi’s “Green Sprouts Loan Program” during his reforms.

The original intention sounded excellent. Traditionally, farmers facing lean seasons could only borrow from private usurers at annual interest rates as high as 70%, which was extremely difficult. Wang Anshi’s Green Sprouts Loan Program offered a low interest rate of 40%, which would save farmers from the pit of usury and increase national revenue. Wouldn’t that be killing two birds with one stone?

This policy had issues at the design level from the start. According to Mr. Luo Zhenyu’s analysis in the program “Journey of Civilization” [8], high private interest rates have their reasons. Farmers’ repayment capacity is low, making lending to them a very high-risk business; many debts are simply unrecoverable. If interest rates were low, this business would not be viable. How could the government offer low interest rates when it didn’t understand local farmers any better than local gentry? Under normal market logic, this business would surely result in losses.

Perhaps Wang Anshi didn’t consider this, or perhaps he did — either way, Wang Anshi’s confidence stemmed from the fact that this was a government-run business.

To ensure the policy’s implementation, Wang Anshi set an evaluation metric for local officials: the amount of Green Sprouts loans disbursed would count as a performance achievement.

So, what do you think local officials would do? Some farmers borrowed Green Sprouts loans but couldn’t repay them, asking for a one-year deferral — could you agree to that? Some farmers said their harvest was good and they didn’t need to borrow Green Sprouts money — could you allow that? If you failed to implement the Green Sprouts Loan Program effectively, not only would your evaluation be unsatisfactory with no hope of promotion, but you would also be resisting reform!

The magistrate next door invented a method: he forcibly allocated loan quotas to every household, then had local gentry and clerks manage a portion, not caring how they got the money, just demanding, “I’ll give you this much money in the first year, and you must return it to me in the second year, plus 40% interest!” And they succeeded. Because they knew how to effectively leverage the government’s machinery of power.

Will you learn from this? Officials do not respond to the intention of reforming the country to enrich the people and strengthen the nation; officials respond to incentives.

American political scientist Michael Lipsky has a theory called “street-level bureaucracy” [9], stating that what truly brings policies to fruition is not the documents from superiors, but the frontline implementers — grassroots civil servants, police officers, social workers, teachers, and so on — who operate on the streets. They all have their own resource limitations, performance pressures, discretionary power, and self-preservation motives. Therefore, what a policy truly looks like is not determined by how it’s written in documents, but by how frontline personnel execute it.

The top talks about grand goals: environmental protection, fairness, safety, reducing burdens. At the middle level, these goals are compressed into metrics; at the grassroots level, metrics are translated into actions. Transmitted down layer by layer, the original public goals often transform into organizational self-preservation goals.

Superiors want performance achievements, local authorities need to report completion, officials want to avoid liability, and the common people want to survive. No one feels they are doing evil. And so, with a perfectly good initial intention and a group of ordinary people simply doing their jobs, bad outcomes paradoxically emerge.

Contemporary Insights: How Rigid Evaluations Distort Good Intentions #

Good intentions combined with rigid evaluations can easily lead to unintended consequences. Let’s look at two contemporary examples.

In 2021, the Chinese government introduced the “Dual Control of Energy Consumption” policy for green development and energy saving/emission reduction. To meet evaluation targets, some provinces began to cut power to enterprises. You could certainly argue that the intention remained good, fully aligning with central government directives, and that it was forcing high-energy-consuming enterprises to upgrade or be eliminated.

But think from the perspective of a factory owner: I have a batch of orders that are due soon, and a single breach of contract would mean bankruptcy, yet the power grid cuts off electricity precisely at this time! What do I do? — I buy a diesel generator to produce my own electricity.

Suddenly, diesel generators were in short supply, and related concept stocks even hit their daily trading limit. The targets were met, and energy consumption on reports decreased, but those self-provided small diesel generators are much dirtier than centralized power plants. Would you say this constitutes green development?

Another issue we discussed when talking about “lemon markets” is this: Chinese hospitals have, in recent years, produced a large number of fake papers, creating a significant global impact. What happened here?

Perhaps to enhance research capabilities, as early as 1986, China’s health professional title system incorporated requirements for papers, monographs, and experience summaries into the qualification criteria for associate chief physicians and chief physicians. Subsequently, major hospitals escalated these requirements, demanding increasingly more papers. Many hospitals mandated that doctors must publish SCI papers to be considered for promotion.

But clinical doctors spend their days in operating theaters; where do they find the time and energy for basic scientific research? In desperation, they could only buy papers. Thus, “paper mills” emerged, specializing in ghostwriting services for academic papers. This phenomenon intensified, leading to the subsequent crackdown on fake papers.

Can you say this is the doctors’ fault? Some doctors have extremely high clinical skills, saving countless lives, yet because they lack papers, they cannot be promoted to senior titles, and their consultation fees are lower than others. Is this reasonable?

The system intended to select the most excellent doctors. Instead, the system rewarded the ability to feign excellence — those skilled with a scalpel were outperformed by those skilled at securing research projects.

When I used to hear the saying, “The top has policies, the bottom has countermeasures,” it was always a condemnation of those implementing them: that policies from above were always good, but their execution by grassroots officials went awry. But now, looking at it, aren’t unintended consequences an inevitable part of human nature? You create a new situation, and people find a new way out; needs that are forcibly blocked often divert, heading in worse directions.

How to Mitigate the Risk of Unintended Consequences? #

How can we align outcomes with intentions as much as possible and prevent unintended consequences? The most basic method is: don’t meddle unnecessarily.

In the West, this principle is called “Chesterton’s Fence,” originating from the British writer Gilbert K. Chesterton [10]: A reformer sees a mysterious fence across a road and immediately says, “This thing is useless, tear it down.” A wiser person, however, should stop him: “Precisely because you don’t see its use, you must not tear it down — first go back and figure out why it was put there in the first place, and only then return.”

Simply put, if you don’t understand a system, don’t pretend you can optimize it.

In fact, Chesterton’s words are too gentle. Historian Chen Xubin offers a concept more pertinent to China’s context, called “Sima Guang’s Dilemma” [11] —

“…a predicament that all reformers living in the Qin system era who cared about the common people would inevitably encounter: they knew that the old system was extremely detrimental to the populace and needed reform; yet they also knew that as long as reform measures originated from unchecked power, it would be difficult to bring true welfare to the people. Reforms could benefit the court, benefit officials, but rarely benefit the populace, and might even push them into worse situations. Driven by conscience, these individuals often found themselves simultaneously advocating for reform and opposing specific reform measures.”

To put it more bluntly, every time you stir things up, ultimately it’s the common people who pay the price.

Of course, doing nothing is not an option either. But having understood unintended consequences, before governments enact policies, they should first conduct a red team exercise to see to what extent the policy will be distorted as it transmits down the hierarchy. You must implement corresponding protective measures and provide compensation mechanisms. Let’s not even talk about whether it can bring benefits to the people; first, ensure that good people do not suffer.

In fact, during China’s reform and opening-up period, there was an excellent practice: before launching major policies, several locations were chosen for small-scale pilot programs. Then, feedback was tracked, and corrections were allowed. One must never make snap decisions and then rigidly enforce them, setting increasingly stringent targets down the chain…

The Responsibility of Frontline Implementers: The Warning of the Banality of Evil #

Those who set intentions at the top should consider how those below will implement them, while those responsible for implementation at the bottom should even more deeply consider the actual intentions of those at the top.

You might say, “I’m just a grassroots civil servant; my job is to meet the targets given by my superiors.” — Wrong. When you are responsible for implementing policy, you are not merely a tool; you are someone who, “as an official for a term, is responsible for the land under one’s governance.” You are Lipsky’s street-level bureaucrat.

Those at the top might only see reports, while you see people; those at the top might only want to know “the policy has been implemented,” while you see the true face of the policy after it has been distorted. This gives you a responsibility — don’t casually shift moral responsibility upward by saying, “This was decided by those above,” otherwise, you fall into what political theorist Hannah Arendt called “the banality of evil” [12].

Modern systems are adept at dissecting an entire monumental evil into numerous seemingly normal, or even completely dutiful, small actions. One person simply ensures lending targets are met, another merely fills out reports to look good, and yet another simply pushes down anything that doesn’t meet standards. In the eyes of a street-level bureaucrat, the person opposite might not be a “person,” but rather a “case,” a “metric,” or an “object” — changing the name from person to object in language reduces a layer of moral friction. Everyone can righteously say, “I’m just doing my job.”

But isn’t your true job the actual intention of benefiting the common people?

An ordinary person can actually do a lot: observe anomalies, record them, provide feedback, leave some room for explanation for those affected by policies, and at the very least, don’t help to smooth over the lie. More proactively, you can offer what James C. Scott calls mētis. At the very least, you don’t have to be the shiniest cog in the chain, but you can always turn back and ask: what kind of monster is this chain dragging along?

There’s a line from the movie V for Vendetta that serves as a fitting conclusion for this discussion.

A female doctor, who once led biological weapon research, conducted experiments on living subjects and created a virus. This virus was then used by a totalitarian government for self-orchestrated terrorist attacks, killing many civilians. The rebel V eventually found her, seeking revenge. The doctor brought up Oppenheimer to defend herself, claiming that committing evil was not her intention, and the current outcome was also something she hadn’t anticipated…

But V said: “I’m not here for what you intended to do, I’m here for what you did.”

Intentions belong to you, but consequences belong to the world.

You are not your intentions; you are your consequences.

注释

[1] Rebecca Diamond, Tim McQuade, and Franklin Qian, “The Effects of Rent Control Expansion on Tenants, Landlords, and Inequality: Evidence from San Francisco,” American Economic Review 109, no. 9 (2019): 3365–3394.

[2] Dean Lueck and Jeffrey A. Michael, “Preemptive Habitat Destruction under the Endangered Species Act,” Journal of Law and Economics 46, no. 1 (2003): 27–60.

[3] California Department of Justice, “Single-Use Carryout Bag Ban (Proposition 67/SB 270).” https://oag.ca.gov/consumers/bag-ban

[4] Atif Mian and Amir Sufi, “The Effects of Fiscal Stimulus: Evidence from the 2009 Cash for Clunkers Program,” Quarterly Journal of Economics 127, no. 3 (2012): 1107–1142.

[5] Robert K. Merton, “The Unanticipated Consequences of Purposive Social Action,” American Sociological Review 1, no. 6 (1936): 894–904.

[6] Uri Gneezy and Aldo Rustichini, “A Fine Is a Price,” The Journal of Legal Studies 29, no. 1 (2000): 1–17.

[7] Robert E. Lucas Jr., “Econometric Policy Evaluation: A Critique,” Carnegie-Rochester Conference Series on Public Policy 1 (1976): 19–46.

[8] 罗振宇，《文明之旅》，公元1069年：王安石理财有可能成功吗？；公元1071年：新法是怎么龙种变跳蚤的？

[9] Lipsky, Michael. Street-Level Bureaucracy: Dilemmas of the Individual in Public Services（Russell Sage Foundation）, 1980.

[10] Chesterton, G. K. The Thing: Why I Am a Catholic. London: Sheed & Ward, 1929.

[11] 谌旭彬，《大宋繁华：造极之世的表与里》，浙江人民出版社（2024）。得到电子书：https://d.dedao.cn/GIQHl4KVvWBShLzs

[12] Hannah Arendt, Eichmann in Jerusalem: A Report on the Banality of Evil (New York: Viking Press, 1963).