Toward a Better Understanding of the Law of Unintended Consequences

We recently published a column describing a few instances of the law of unintended consequences — specifically, what happens when well-meaning legislation winds up hurting the parties it is designed to help.

I thought it was a pretty good column. But I see now where it could have been better. Alex Tabarrok, writing on Marginal Revolution, addresses the law of unintended consequences per se, something our column didn’t do. In writing a column, or even a book, there is always a struggle between how much theory to write and how much story to write. In this case, I wish we’d included a bit more theory — if, that is, we’d been able to explain it as well as Alex has.

For example:

The law of unintended consequences is what happens when a simple system tries to regulate a complex system. The political system is simple, it operates with limited information (rational ignorance), short time horizons, low feedback, and poor and misaligned incentives. Society in contrast is a complex, evolving, high-feedback, incentive-driven system. When a simple system tries to regulate a complex system you often get unintended consequences.

Unintended consequences are not restricted to government regulation of society but can also happen when government tries to regulate other complex systems such as the ecosystem (e.g. fire prevention policy that reduces forest diversity and increases mass fires, dam building that destroys wet lands and makes floods more likely etc.) Unintended consequences can even happen in the attempted regulation of complex physical systems (here is a classic example involving turbulence).

The fact that unintended consequences of government regulation are usually (but not always or necessarily) negative is not an accident. A regulation requiring apartments to have air-conditioning, for example, pushes the rental contract against the landlord and in favor of the tenant but the landlord can easily push back by raising the rent and in so doing will create a situation where both the landlord and tenant are worse off.

More generally, when regulation pushes against incentives, incentives tend to push back creating unintended consequences. Not all regulation pushes against incentives, some regulations try to change incentives but incentives are complex and constraints change so even incentive-driven regulations can have unintended consequences.

Does the law of unintended consequences mean that the government should never try to regulate complex systems? No, of course not, but it does mean that regulators should be humble (no trying to remake man and society) and the hurdle for regulation should be high.

It is only Alex’s last paragraph that significantly overlaps with what we wrote. Everything else is value-added. I’m not saying we should have loaded up our column with theory, but having read Alex’s very good take on the issue, I’m grateful that these interwebs exist in order to allow people to so efficiently and gracefully build upon existing writings.

Lyn LeJeune

Uh, what is happening here?!!! I'm going to join Jonathan Swift with a "Modest Proposal for Preventing the Children of Poor People in Ireland from Being a Burden to Their Parents or Country, and for Making Them Beneficial to the Public" (1729) and keep my 300 hundred bucks and eat a few little children. There's meat for ya! Have we also lost the gift of satire? Anti-foreign? Are you kidding? Who the heck you think is picking those stawberries down in Plant City. Channeling Ignatius? Of course, you bunch of confederacy of dunces! But, what a good conversation we got going!
Thnaks. More, more!
Lyn LeJeune -
"When Ignatius J. Reilly Worked at the New Orleans Public Library and I Went Crazy at the Port-O-Call."


Who got all this money?(except sales tax). All the items came from outside of the USA. Did any of this stimulate our economy?

Every employee at every store you went to gets a piece of that money. Since the products were acquired from foreign sources at less cost, one or two things happen. One option is those employees receive a higher percentage than if the products had been produced in America and bought at higher cost.

Another option is that since your prices were kept lower, you have more money left over to spend elsewhere, on other things. That money pays for those employees, who used to produce the things we now get from foreign sources.


Congress banned horse slaughter in the U.S. so as to protect horses from those hungry Europeans.

Sadly, horses are now abandoned to starve to death, because there is no economical way to dispose of them.


All policy-makers should be required to pass a Computer Programming exam. There would be much fewer unintended consequences.


The problem with this POV is that it oversimplifies and is over broad. Name a complex system that regulates a simple system. Can you? Can you name a simple system that regulates a simple system? Sure, things like gears that turn gears but that's only simple if you look at the specific interactions and not at what's impelling the first gears or the effect of the moved gears. Saying it's as easy as "simple to complex" is really only verbiage.

Government is not a simple system. It is highly complex. One can take the comments about the limitations of government above and flip them with the comments about the market-oriented society and lose only a little bit of truth. Government policy develops in a market and responses to government policy develop in markets.

The interface between government and the rest of the world is through specific, not often simple reductions or levers. These levers, metrics or whatever you call them may not be well aimed but that's because the problem is complex not because the generating system is simple.

A more accurate description would be that you have two complex systems, the government and the market intended to be impacted or regulated. These systems interact inefficiently because they are themselves in a complex system whose character is only partly known. The limited information and thus the misapplication of a solution is a function of all three complex systems.

You can calculate how the odds work for generating a shared set (meaning a solution which actually works). There will be odds that the first complex system (government) will generate a wrong answer or a right answer. The second (intended) system's odds of responding as the first system wants or predicts depends on how that system is affected by the complex other system in which the intended system fits. If you put that over time, you get an evaluation of the present value of the choice (and thus a measure of how long the solution might last). Game theory has ways of filling in the variables.



Am I the only one that mistakenly saw "Alex Trebek" the first time they read "Alex Tabarrok"

bev smith

I got interested in this topic when I learned that an AIDS program administered by an international NGO in South Africa was having the perverse and unintended consequences of motivating people to test positive for HIV, since HIV+ women in the NGO program had become an island of prosperity in a sea of poverty. I think some of these seemingly simple fixes are actually attempts at avoiding the real problem. The problem the ancient Jewish law addresses is the harvest every 7th year, but the ongoing poverty is ignored. The NGO that gave HIV+ women jobs does nothing to deal with the high rates of HIV+ in South Africa, since it does nothing to deal with the poverty that is responsible for the transmission of HIV.


Where can I find the research upon which you based the interpreter for the deaf's fee? I wonder if the number you cited is the norm in LA or not. It certainly seems inflated to me,but then I don't live in LA.

Most interpreter agencies will charge an extra fee if the services are not arranged with sufficient notice. Perhaps that is the case in the situation you cited. As an interpreter in private practice, I do not work through an agency and can set my own rates and policies. Two hundred forty dollars is more than twice what I would charge for a two-hour minimum.

I appreciate your explanation of the theory behind Unintended Consequences -- it certainly has applications in a wide variety of settings.

Consider the consequences for a deaf person in the Emergency Dept if a qualified interpreter is not available to assist the medical staff assess the patient's medical history, symptoms, presenting complaint, medications and allergies.

Consider also, the potential legal consequences for the hospital and staff who treat the patient without properly communicating with the patient, potentially resulting in inadequate diagnosis and treatment, or unnecessary tests and procedures.


Certified Interpreter for the Deaf
Medical Specialist



I strongly disagree with your post Mr. Dubner becaue it has an anti-government bias that is unsaid and runs through it so deeply it's almost invisible.

There are no examples given of an individual taking actions that have unintended consequences, and there are no examples of businesses doing the same. Children, parents, friends, mobs, social groups, community organizations, and corporations - apparently these are all blessedly free from ever having less than perfect and all-encompassing visions of future consequences.

Even animals never make mistakes! (those cute internet videos really are just for our amusement.)

"All policy-makers should be required to pass a Computer Programming exam. There would be much fewer unintended consequences. "
If only it were so. I've worked in I.T for ten years, and I've seen unintended consequences come out of my and other I.S. departments on at least a weekly basis.



My grandchildren have benefited from a state program for speech that they would not have been able to afford. This has been a benefit for their whole life.
Every day we benefit from government. How do we choose which will be beneficial and which won't. Should we not try?

Keith Cornelison

H.L.Mencken said it in a few words. and I paraphrase, For every problem there is a solution that is easy,simple, inexpensive and Wrong.


Jonathon almost got there. Unintended consequences intersects with and is arguably equivalent with a suitably generalised statement of the problem of future generations. To wit-we like regulations, laws and policies to be non-specific in terms of the intended beneficiary class. But this seems to entail that any such policy will always be underdetermined relative to a set of relevant and unmentioned evaluative criteria-dos the rule apply to all and only named victims(crudely, no. see the third sentence), all and only victims identifiable prior to the first applicatiion of the rule, all and only victims for all time, all and only victims identified in the first ten minutes?

Not only are resources limited, but so is the scope of our moral interest. If the latter were false, a large segment of cases of the law of unintended consequences could no longer obtain. Of course, you and I would no longer recognise the moral impulse behind our policy making, nor wold we understand the policies generated by them.



Re #1:
What's your point? That if you choose to buy things that are made in another country, you have to buy stuff from another country? A lot of the things you complain about, there simply isn't enough suitable land in the US to grow it all domestically (e.g., coffee). Interesting, too, that you decry as too expensive the (domestic) ground beef you can get for $2.50 a pound, but spend $75 on imported (rather than domestic) vodka and shrimp. And don't forget that even if you're paying $10 for that bag of shrimp, only a fraction of that is going to China--the supermarket takes some of that, as does the trucking company that brought it from the nearest port, and the (possibly American-owned) shippers--and don't forget that since the company whose name appears on the package is American-owned, any profits after the fishermen are paid comes back to America.

Larry Keiler

@ John (#14)
That was just an unintended consequence.


Nice straw man, Sharper. Who said anything about criminalization? Or intrusive monitoring? Your idea certainly is creative, but not like anything that would ever be on the table in this country. The ADA is not, and never will be a criminal statute. Nor do we have the resources or desire for such an intrusive system.

To give you a fair example of what could be done to strengthen the ADA, consider this: It appeared from the original anecdotes in the article that there might be a lot of people that don't understand the requirements of the ADA. Therefore, they either continue to discriminate in ways that are against the law, or accommodate disabled people more than they actually have to accommodate them by law (and of course, some people actually follow the law). Thus, perhaps distributing pamphlets in plain English to those governed by the ADA would help them understand what is required of them, that is, how to actually comply with the law.

If the ADA truly is "failed," the problems can be taken into account in forming the appropriate solutions that address the problem. I'm confident that whatever the answer, it has nothing to do with the criminalization and intrusive monitoring you suggest I would suggest.

Oh, and by the way, choosing not to interview anyone disabled is still discriminatory hiring. It's just at an earlier point in the process...



great article


On page 183 of Freakonomics you say the the beginning of uncommon names for black children (i.e., K'neah, T'wan) was intertwined w/ the Black Power movement. No - I remember very distinctly, because my last two children were born during the mid- and late-seventies - the mini-series "Roots" ran in 1977, and it was immediately after that that many, many black women gave their children names that they thought sounded African because the series gave them a huge jolt of pride in being black for the first time. Unfortunately, the names were not at all African, and the spelling was often left up to the obstetric floor nurses at the hospitals.


Here's the caveat to the ADA business: The climate in which the ADA was developed arose out of a body of legal work -- Brown, Loving v. Virginia, state laws on special education arising out of other court decisions, etc. etc. -- that cleared the way for the legislative branch to embrace the notion that it was inherently immoral to discriminate on immutable factors that were irrelevant to the task at hand. The whole business is itself essentially an unintended consequence of the original legal push against other forms of discrimination and unequal treatment. (If I'm remembering the historical context correctly.) I think the ADA is perhaps not the best example, because I think that like some of actions taken based on other court decisions and/or activism, the initial unintended consequences may be severe, but they diminish (not necessarily disappear altogether) over time. Even special education, which is a mighty complex system regulated repeatedly at different levels, is effective enough that school systems would be thrown for a loop if it were to vanish.

The animals are another story. If they're gone, they're gone. I find that example more persuasive.



Just to clarify my earlier post -- I assume unintended consequences may be positive or neutral as well as negative. Not surely whether the theory does! I.e., the ADA is a positive/neutral unintended consequence of earlier actions.


Alex's definition of the Law of Unintended Consequences is a masterpiece. For the first time I see that this Law is a corollary of (or vice-versa) Murphy's Law. (OK, in mathematical terms, they map to each other).

The problem underlying both is that our concious minds are too simple to grasp complex reality in full. The lessons are that we need to test, pilot, simulate, think around and cross our fingers before we commit to a new regulation or a new design; and to always plan for its review and amendment. Neither our first nor our second version is likley to meet all our objectives.