Talk:Perverse incentive/Archive 1

Windows
I deleted this, as it is not an example of a perverse incentive (not to mention the fact it's rather silly):

"A glazier can generate business by encouraging vandals to smash windows."

If i may respond: If a glazier creates an incentive for vandals to smash as much glass tonight as they can, then the likihood is that the vandals will visit establishments whom have the most glass. For example, the glazier's place of business. If the glazier indended his glass to be smashed (for example, to make a bogus insurance claim) then it is not a perverse incentive. If he did not expect his own glass to be destroyed, then it is. So long as the incentive leads to an unexpected consequence for the glazier, then it is a perverse incentive. Acssm 11:07, 7 May 2006 (UTC)

I'm seeing some examples on this page that would be better suited for "Side-Effect" or "Unintended consequences".

With respect to the question re: the worthiness of the Glazier statement as an example of perverse incentive:

If the statement were to read, "A glazier can generate business by paying vandals to smash windows" then I believe you would have a perverse incentive. --Dewbrain 19:28, 18 October 2006 (UTC)

Agree 20% with the tags Examplefarm and Prose
Alerted by the tags I agree, some of the examples seem not much illuminating. While I was about to remove a few, I'd add others. Checked on the tag "Examplefarm" and found 7 occurrences in all of Wikipedia. Although that tag links to Writing better articles, I find no support there for having fewer examples. As for the remaining 80%, I do not see how fewer examples would improve readability and support understanding for our readers. Or changing this article to prose, for that matter.

I'd add (not sure I got the citation syntax right)

Paying architects and engineers for what they spend, not for what they save.

Paying medical professionals and reimbursing insured patients for treatment but not for prevention. —Preceding unsigned comment added by Bernd in Japan (talk • contribs) 09:01, 12 January 2008 (UTC)

Signature added Bernd in Japan (talk) 09:29, 12 January 2008 (UTC)

Capital gains vs. dividends tax rates
I don't recall where, but I've heard the argument that the practice of taxing income from capital gains lower than income from dividends creates perverse incentives. Low capital-gains tax rates are supposed to encourage people to benefit from investing in capital (e.g. stocks), but ...

First off, favoring capital gains over dividends encourages investors to prefer "growth" stocks over "value" stocks. Each investor only expects to make money when they sell the stock, rather than by holding it and taking dividends. Since you can only sell the stock when someone else is buying, this makes investing in stocks more of a zero-sum game, where one person's gain is another's loss -- better for lucky speculators and the highly diversified; worse for (e.g.) employees holding on to their own company's stock.

Second, because investors come to prefer growth stocks, it encourages companies to focus on increasing their share price rather than on consistently delivering a profit. This encourages attempts to drive up the share price through hype or creative bookkeeping to meet projections. (You can't fake dividends; you can con the market into believing you're doing better than you really are.)

The first effect produces a perverse incentive because shareholders are encouraged to do something that is bad for themselves as a group: to move from a positive-sum strategy toward a zero-sum one. The second effect produces a perverse incentive because companies are encouraged to be less honest with their investors in order to attract capital. --FOo (talk) 06:18, 9 October 2008 (UTC)

Requests for citation
Someone has placed "citation needed" notes all over this page, including in spots where I don't think they are required. While I appreciate the importance of providing citations, I think citations ought to be restricted to those passages that are either obviously a matter of opinion, or highly debated. Statements based on straight forward logic, or common knowledge, should not require citations to back them up. The statements like "Mexico lies south of the USA", "elephants are usually bigger than mice", or "World War II ended in 1945", do not need to be cited. They are common knowledge. Similarly, if a statement is based on a combination of common knowledge and simple logic, then it shouldn't need to be cited. For example, there are a lot more people living on the earth today than there were 1000 years ago (common knowledge). Logically then, we can conclude that the reason more people died in the year 2004, than did in the year 1004, is that there are simply a lot more people around today than there were a thousand years ago. We ought not to have to provide a citation for this statement. It is fairly obvious. For the same reasons, many of the examples, for which someone has requested citations, do not, in my oppinion, require citations. They are common knowledge & straight forward logic. Northern Bear 21:32, 29 August 2007 (UTC)


 * For example, someone requested a citation for the statement "When it is illegal to sell to minors but legal to sell to adults, drug dealers have an incentive to refuse to sell to children."
 * I don't think that needs any citation. It is common sense.  All it says, is that if X is illegal, there will be an incentive not to do X.  It is common knowledge that breaking the law can result in penalties, and it is common sense that people do not like receiving penalties and so have an incentive not to break the law.  Why should this require a citation?Northern Bear 21:40, 29 August 2007 (UTC)


 * In keeping with the theme of the article: this restriction on legal sale to minors could create a market of underage users who will pay more for this item.  This could very well lead to dealers targeting minors.  This is the exact opposite effect as intended.  —Preceding unsigned comment added by 71.33.68.113 (talk) 06:48, 18 November 2008 (UTC)

Incorrect reference link
Footnote 5 had the same hyperlink as footnote 4, which is incorrect. I deleted the hyperlink, but the rest of the citation remains. If you know where the JAMA article can be found please edit the citation.

The cited work in Footnote 4 "Paying architects and engineers according to what is spent on a project leads to excessively costly projects" does not provide a good example of the sentence it supports --38.117.159.162 (talk) 18:23, 16 March 2010 (UTC)

Suggested removals
"* Mandatory drug testing in schools and workplaces may encourage students and workers to use drugs which are less detectable by drug tests, such as alcohol."
 * This is not a perverse incentive, as there's no effect contrary to the purpose. Unless the undetectable drugs are much heavier. CP/Mcomm |Wikipedia Neutrality Project| 18:47, 30 September 2008 (UTC)
 * Drug tests test for well-known, readily-detectable drugs. If alcohol's the only good example of more dangerous, but less detectable drugs, then there's not enough correlation between a drug's detectability and its danger for that statement. However, there is a strong correlation between a drug being relatively safe and becoming widely used and thus well-known enough for tests to check for it. Basically, drug tests influence users to switch from relatively safe stuff like marijuana to much riskier experimental drugs. That statement could be put back in the article if everything after "workers to use" is replaced by "experimental designer drugs which are less tested". --70.41.70.6 (talk) 08:21, 5 April 2011 (UTC)

"* Where libraries, universities, and similar institutions charge a higher fee for copying than for printing, users may print multiple copies of a document, which could cost the institution more than free copying would.  "
 * Could seem right, except that copying normally costs more (often noticeably) than printing, due to more expensive equipment and paint waste. Also, printing produces higher quality copies, plus, it's just ergonomic to print copies. A reverse, though, would be almost correct. Almost - the key part here is that different pricing is not intended to make the students use copiers more. If it counts as policy at all, and not just covering costs. CP/Mcomm |Wikipedia Neutrality Project| 18:47, 30 September 2008 (UTC)

Example or opinion?
The ESA example is poor, IMO. One alternative to the moral hazard of land owners purposely making land uninhabitable to an endangered species is to simply pay the land owner for the loss of use. In fact, this is the standing opinion of SCOTUS on "takings" -- i.e., if the government takes land for an ostensibly beneficial use to greater society, it may do so, but only if the owner is compensated for it.

Otherwise, the ESA example sounds like so much right-wing rubbish, which I believe violates the rules of wikipedia, right? Might be best to look for a more neutral item to replace this "example" (opinion) with. —Preceding unsigned comment added by 68.3.56.57 (talk) 18:09, 21 September 2008 (UTC)


 * Well, if it was true, it'd be a perverse initiative. However, I doubt it: the procedure making the land uninhabitable might cost a lot more than the lost profit from restrictions due to the species. CP/Mcomm |Wikipedia Neutrality Project| 18:47, 30 September 2008 (UTC)


 * The ESA example is an excellent example, practically a textbook example of what is meant by the phrase "perverse incentive". I'm going to remove the "opponents of X claim" nonsense but try to write the claim more clearly - the problem isn't "The ESA" but *specific parts* of the ESA. Specifically things like this: if - as part of an effort to help migratory birds - we choose to impose development restrictions on, say, "wetlands", it's nearly impossible not to create a perverse incentive for property owners whose marsh/swamp *hasn't been noticed yet* to drain or fill in the swamp. Whereas if the law didn't exist, property owners might even choose to *deliberately create* wetlands to attract birds. The law has a desired direct effect - preserving existing known wetlands - and an undesired perverse incentive effect - to prevent some new wetlands from being developed or coming into existence. This is a factor that needs to be considered when implementing command-and-control policies.


 * The "procedure for making the land uninhabitable" we're talking about is to dig a drainage ditch, push dirt into a shallow to prevent water collecting there, or chop down some trees - it's not inherently expensive. The referenced Freakonomics article makes a good case for it actually happening. --Blogjack (talk) 13:36, 7 June 2012 (UTC)

Hanoi example
The Hanoi example is quite different than the same story on the page about the cobra effect. With just this info, I'm inclined to believe the story on the cobra effect page, and this here should be "fixed". Cobra_effect — Preceding unsigned comment added by 95.36.47.159 (talk) 09:49, 30 August 2014 (UTC)

Is this just a racist rant?
The perverse incentives of the welfare state have all too frequently enticed the poor, blacks included, away from finding remunerative work and toward a mentality of dependency and entitlement.[11]

Is there anything to [11] other than a rant with seeming racist undertones about welfare? As it doesn't appear to be a good source. 72.31.17.20 (talk) 01:28, 9 September 2014 (UTC)
 * I agree the source was very poor. You should have seen the text before I tried to temper it. EllenCT (talk) 06:54, 29 September 2014 (UTC)

Welfare Wish-Wash
The welfare bullet, "Welfare spending may entice the poor away from finding remunerative work and toward dependency on the state," seems too weak for Wikipedia because welfare either entices or does not. Whatever enticement welfare might offer should be stated; else, we should remove the statement. --Duxwing (talk) 02:50, 22 September 2014 (UTC)
 * Looks like the welfare trap article puts it a little less bluntly, with some sources. I've rewritten the bullet point. --McGeddon (talk) 08:28, 22 September 2014 (UTC)


 * Thanks, McGeddon. :) Could we rewrite the sentence more concisely?
 * Original: "...incentive, where the withdrawal of means tested benefits that comes with entering low-paid work causes there to be no significant increase in total income..."
 * Suggestion: "...incentive should losing means-tested benefits upon gaining lowly-paid work offset its wage."
 * Duxwing (talk) 21:58, 25 September 2014 (UTC)
 * Oh, didn't notice your username, hello again. Didn't we have this conversation months ago, that "make sentences as short as possible at any cost" was something you'd imagined you'd read in the Wikipedia MOS once, but which wasn't actually there? The phrasing "should losing means-tested benefits upon gaining lowly-paid work offset its wage" is very hard to read, and shaving nine words off of one of the article's shorter bulletpoints doesn't seem necessary. --McGeddon (talk) 22:25, 25 September 2014 (UTC)


 * Yikes! I see your point about readability--my revision doesn't introduce its ideas before using them--and it's very frustrating to see that I've made the same mistake.  Thanks for taking the time to tell me, and I would very much appreciate to e-mail with you about this. :)  Duxwing (talk) 04:09, 26 September 2014 (UTC)
 * I think the following, "should losing means-tested benefits upon gaining lowly-paid work offset its wage" can be fixed by rewording it to, "Does losing means-tested benefits in retaining low-paid work offset wages?" But yeah, that was really difficult to read at first, stemming from the continual use of progressive tense while then using an unclear, impersonal third person possessive. Rather than trying to defend something that was blatantly wrong, Duxwing stepped up to the plate and admitted he was wrong, for which I thank him. — Preceding unsigned comment added by Chewbakadog (talk • contribs) 05:15, 26 September 2014 (UTC)


 * Thanks for joining us, Chewbakadog! :) Would you please make your well-worded question into a statement?  I ask because I think my quotation misled you into thinking we needed a question instead of a conditional clause.  What do you two think of the following revision?  "People receiving means-tested benefits are discouraged from working when these benefits, which are withdrawn upon employment, exceed wages."  Also, do we have a source yet? Duxwing (talk) 23:39, 26 September 2014 (UTC)
 * There's a source at the end of the sentence. And the sentence is fine, I can't see any reason to single it out and try to make it shorter. --McGeddon (talk) 08:19, 29 September 2014 (UTC)


 * The issue with the welfare trap is that there is a specific range of earned income in which each after-tax dollar earned results in more than one dollar's worth of means-tested benefits being clawed back. This is a substantial barrier to entry to the low end of the workforce. I tried adding the info, but there is currently a vandal following me from article to article, reverting everything I contribute. I might look at this later if I find the time, as the issue is real and valid. K7L (talk) 17:36, 1 December 2014 (UTC)

Carbon Credits
I think this is another example of perverse incentives:

http://www.nytimes.com/2012/08/09/world/asia/incentive-to-slow-climate-change-drives-output-of-harmful-gases.html?pagewanted=all

Awarding carbon credits for destroying HFC-23, a greenhouse gas that is a by-product of producing the coolant HCFC-22, led to the increased manufacture of the coolant, just so that the factories manufacturing it could obtain the carbon credits. The consequence is that the coolant is itself bad for the environment (I think it is a CFC), and the increased production has decreased price drastically.

Shankar Sivarajan 13:51, 2 July 2015 (UTC)

Eliminating social safety nets can discourage free market entrepreneurs
This is not an example of a Perverse incentive. It isn't even an incentive. Removing social safety nets isn't intended to be an incentive at all.

The purpose of eliminating social safety nets would not be to increase business activity. Therefore even if this was happening, it would be a side affect but not one that is the opposite of the intended affect.

This seems to be a political statement intended to influence opinion — Preceding unsigned comment added by 70.60.146.186 (talk) 17:52, 19 January 2018 (UTC)

Should be a list
This article should be renamed and reformatted as a list. (See WP:FL for examples]] of good lists.) As it stands now, it's mostly a list anyway. Of course, all the list elements should be properly sourced and cited. I'll reformat it after a few days if no one objects. LK (talk) 03:04, 16 February 2009 (UTC)

The list should at least be split into to parts: One of simple, commonly agreed upon examples and debateable examples.

The programmer example is very obtuse. Understanding it would require knowledge of computer programming, and someone with such knowledge is likely to understand the other examples. —Preceding unsigned comment added by 69.254.154.53 (talk) 01:45, 18 May 2009 (UTC)


 * Personally, I'd plead for this article to explain the theory behind the subject, while the practical emanations and examples would find a place in a Timeline of perverse incentives. Wakari07 (talk) 11:18, 15 March 2018 (UTC)

Chinese peasants vs. palaeontologists example
I searched for this in the Bill Bryson's book (first ed.) and could not find anything similar there. Aarrcchhiimmeeddeess (talk) 16:31, 2 July 2017 (UTC)
 * Well spottted!
 * (Conventionally, new discussion page sections go at the bottom rather than the top...)BushelCandle (talk) 00:22, 3 July 2017 (UTC)

This is the first time I partake in editing wikipedia so my apologies for any mistakes I make. I distinctly remember reading, in Bill Bryson's book, something about locals destroying fossils to increase the fee they received. I don't have the book at hand so I searched the book using Google books. And I found it. It is mentioned on p. 528. I also looked up the source given by Bill Bryson, which is a book by Carl C. Swisher III called Java Man: How Two Geologists Changed Our Understanding of Human Evolution. Again I looked up the book via Google books. On page 77 it mentions the story of a palaeontologist called von Koenigswald, who paid Javanese locals 10 cents per piece of a skull. He did find that the locals were breaking larger pieces of the skull into smaller pieces, but as quoted from the book: Nevertheless, it was a great find, and celebrations was in order. In the end the skull was of crucial importance proving that Pithecanthropus was human.

The main differences:
 * It were Javanese locals, not Chinese locals
 * It were homonid bones, not dinosaur bones
 * Arguably, there was no significant decrease in scientific value of the bones

In the end, I don't think the locals breaking the bones had much of a negative impact on the scientific goals of von Koenigswald, so its inclusion in the list of examples perverse incentive is perhaps overstated.

Bill Bryson's book, p. 528 Carl C. Swisher III' book, p. 77 Randow word generator (talk) 13:06, 11 January 2019 (UTC)

Aral Sea
I've just removed an item about the Aral Sea, which is a classic and very big example of unintended environmental catastrophe but not a cobra effect. What separates the "cobra effect" from any old unintended consequence is that the incentives lead to action which directly oppose the exact goal of the incentives. In this case the result was bad but not directly opposite of what was intended.

This is sometimes a subtle distinction and the reason that the examples list is a constant struggle to stay relevant. - DavidWBrooks (talk) 21:41, 22 August 2021 (UTC)


 * I have remnoved two more examples - Texas college admissions and lawsuit payments - that produced some unintended effects but not direct cobra effects. - DavidWBrooks (talk) 21:59, 22 August 2021 (UTC)

Solutions
Does anybody know any strategies to anticipate and prevent these perverse incentives in the literature?

Right now the article is all problem and no solution. Semitones (talk) 17:07, 13 September 2021 (UTC)

Too many examples
Boy, there are way too much examples on this article that don't really inform the reader beyond "here's another one!". Also, the merger with "cobra effect" also makes it confusing that certain examples (from that article) are discussed in detail while others are just listed. - DavidWBrooks (talk) 17:27, 30 March 2021 (UTC)
 * Yes, that was the laziest merge I've ever seen; I've done it properly now. And yes this is basically a list article; but is that a problem? ··gracefool &#128172; 23:50, 30 March 2021 (UTC)
 * They're generally frowned upon unless they are a specific "List of ..." article. This is a good example of why, I'd say: it's a tedious bunch of similar samples, so long that no casual reader will go through it and therefore useless unless somebody is searching for a specific example they already know about - and google does that better. The third or fifth or tenth item don't tell you anything that you didn't already know. - DavidWBrooks (talk) 01:46, 31 March 2021 (UTC)
 * Well feel free to move them to a list article; but then this article would be rather short, so I don't see it as being enough of an improvement to be worth the bother myself. ··gracefool &#128172; 20:00, 31 March 2021 (UTC)

Frankly, I rather like that there are so many examples from various fields of human activity. It gives some food for thought in search of remedies to predict, identify and prevent the described effect. Filozofo (talk) 10:54, 18 February 2022 (UTC)

Can someone help me add a reference?
I would like to add  as a reference to the 10th option of the "other examples" section under "Examples of perverse incentives". Sorry that I am posting this here, I am new to this and couldn't figure out the help text. --Tglkwd (talk) 04:32, 29 April 2022 (UTC)

Citation #7 is a 404
that is all. — Preceding unsigned comment added by 09:29, September 6, 2021‎ 73.150.73.172 (talk • contribs)


 * Fixed. – S. Rich (talk) 03:01, 25 July 2022 (UTC)

Many bad examples
The far too long list of examples contains several that are fraud, rather than perverse incentives: for example, pretending that patients were mentally ill to get a bigger payment from the government. I am going to remove them. - DavidWBrooks (talk) 16:59, 11 April 2023 (UTC)


 * Oppose - Calling things that are clearly good examples of perverse incentive bad examples or not perverse incentive is not helpful. See WP:UNRESPONSIVE and WP:CAUTIOUS and WP:RVREASONS. Kire1975 (talk) 19:28, 11 April 2023 (UTC)

I don't see what the rationale is here. Obviously, many perverse incentives encourage people to do things that are useless or evil; that is what makes them perverse! jp×g 22:07, 8 June 2023 (UTC)