 |
|
Actions - User interviews & research to find the problem
- Next page button disrupting train of thought
- Brainstorm ideas on how to solve it
- Work with developers to create infinite scroll
- Put it through more user testing
Intended Consequences/Outcomes: - Aza Raskin’s main goal: stop disruptions to a user’s train of thought
- The infinite scroll
- Something that would focus our attention and allow us to control the tempo when browsing websites and apps
- Streamlining the clunky experience of the next page button
Unintended Consequence: - Tech companies exploiting the infinite scroll, providing endless content without the user’s consent or ability to opt out
- Social media becoming the biggest waste of time
- "…hours, even lifetimes — of mindless surfing and scrolling"
Analysis by: Cynthia Cui
|
|
|
|
Outcomes and actions
Background: The Seamen’s Act, which was passed after the RMS Titanic disaster, required retrofitting of lifeboats to passenger ships. Management also replaced hardwood floors with concrete in upper decks, adding more weight to an already top-heavy ship. Unintended consequence
• SS Eastland capsizes, killing 844 passengers. “Ship Safety” might be a starting outcome wording, but to make it measurable, you’d probably need to develop or use some safety audit methodology.
Intended consequence
• Improve the safety of passengers. “Ship safety measured by safety audit score”
• Ease of cleaning and maintenance. “Maintenance burden measured by monthly maintenance cost”, or perhaps another “Ease of cleaning measured by average turnaround time”
Actions taken
• Add lifeboats (Lever: “Number of lifeboats”)
• Replace wooden floors with concrete
Alternative action
• Detailed safety audit, including lifeboat complement
• Proper investigation of previous issues with top-heaviness
• Both of these demand another action “Implement actions suggested by reports”
Analysis by: Carl Sandrock
|
|
|
|
Actions:
• Workplace mandates.
• Limit access to social activities and travel.
• School-based mandates.
• Differential lockdowns for the unvaccinated.
• Differential access to medical insurance and healthcare.
• Mandatory population-wide vaccination with taxes, fines, and imprisonment for the unvaccinated.
• Governments announcements that promise vaccination would ensure a ‘return to normal’ if many people (especially younger people) had vaccinated.
Intended consequences:
• Increase vaccination rates.
• Reach herd immunity.
• Reduce hospital/ICU burden.
Alternative actions:
• Transparency in government metrics for policy making.
• Scientific and objective government announces.
• Social and psychological focused analysis for policy making.
• Invest more in Hospital/ICU infrastructure and personnel.
• Social cohesive instead of punitive and discriminatory politic discourse.
(Aggregated) Alternative consequences:
• Social trustiness for preventing confirmation bias.
Analysis by: Raul Navarro
|
|
According to Ben Ramalingam’s Aid on the Edge of Chaos, international development is just such an invasive species. Why Dertu doesn’t have a vaccination clinic, why Kenyan schoolkids can’t read, it’s a combination of culture, politics, history, laws, infrastructure, individuals—all of a society’s component parts, their harmony and their discord, working as one organism. Introducing something foreign into that system—millions in donor cash, dozens of trained personnel and equipment, U.N. Land Rovers—causes it to adapt in ways you can’t predict.
|
|
A must-read. One unintended consequence after another, from well-meaning, but ultimately misguided initiatives. But my conclusion is a bit less bleak than the author’s. I think that with better understanding of complex adaptive systems (including adjacent areas like knowledge transfer between domains, real-time systems modeling with adaptive learning, integrated human, evidence-based analysis, and automated decision making), we can still do good. We must.
|
|
|
|
Action: Indiana’s new religious freedom law Intended consequence: to protect religious expression Unintended consequence: Massive “blowback”: boycotts, legalized pot, and more. Analysis: This debate has raged throughout the history of the United States. It stems from what this author calls “an inability to distinguish between: 1) the universal spiritual values that underlie the American experiment in democracy, and 2) the role assigned to government to advance those same values by protecting freedom of conscience and belief.” Experiences like Indiana bring this distinction into sharp relief.
It stems from
|
|
|
|
Action: Require two A* exam scores instead of one for Oxford / Cambridge entrance.
Intended consequence: solve the problem of over-subscription at universities.
Unintended consequence: Students who receive special coaching in test-taking are the only ones who can receive two A*s, which decreases the prevalence of “out-of-the-box” creative thinkers, creating a situation where the university is populated by good test takers.
Analysis: Test scores by definition are proxies for a student’s true ability to benefit, and to benefit from, a university education. So if test score criteria are changed, it’s important to reexamine the link between those who achieve those scores, and the true outcomes for which the score is a proxy. Is Oxbridge looking for creative thinking? Are they looking for the diversity of thought that leads to great decision making? If they are trying to balance these kinds of goals with the simple goal of ensuring the right admissions numbers, then they need to take a more nuanced approach to admissions criteria than simply changing one A* to two. If you believe this article, there is a qualitative difference between the student populations that are generated by this policy, which goes well beyond what the school might be intending.
|
|
|
|
Lorien Pratt’s insight: Action: US begins bombing ISIS in Iraq and Syria Intended consequence: Reducing ISIS’ influence Unintended consequences: – this author says that blowback from civilian collateral damage will incentivize ISIS to attack the U.S. – Aleppo-based al-Monitor correspondent, Edward Dark says that US attacks would stoke anti-Western resentment, and that ISIS is “secretly overjoyed” at the attacks for this reason. – “Syria and Iraq as we know them…will disappear”
Analysis: This is an incredibly complex situation, with wheels within wheels. Every consequence, intended and otherwise, should be considered, including these.
|
|
|
|
Decision: Legislation makes government records available to the public
Intended consequence: Transparency / informed citizenry / better government participation.
Unintended consequence: Opposition research firms hired by political candidates submit lots of requests, end up forcing office workers to spend hours photocopying, at considerable taxpayer cost.
Analysis: This is the Exploit pattern: a new system is created, and certain persons discover that it can be used to achieve an outcome that was not intended by its designers. Now that the exploit has been discovered, it’s time to do a system redesign / tweak to patch the hole. Perhaps improvements to automation would help, funded by an access charge?
|
|
|
|
Action: Create a powerful new system-called MonsterMind-which can respond rapidly to cyberattacks to trigger a real-world strike.
Intended consequence: powerful deterrant to cyberattacks
Unintended consequence: If true, this massive concentration of power could create an opportunity for hackers to spoof identities in order to trigger an attack.
Analysis: This seems unlikely, but if true it illustrates an unintended consequence pattern, which is overlooking a specific kind of external factor: specifically the susceptibility of a powerful new system to intrusion/spoofing/illegal activity, and the resulting opportunity for that intrusion to have an equally strong negative consequence. Any time that information is concentrated in a single place – whether a credit card database, a facebook server, or a google server, or any it lowers the effort and raises the benefit of a criminal attack.
Attention to antifragility principles and good system modeling would probably go a long way here.
|
|
|
|
Action: A bill in the senate would ban the import of goods from countries the employ children.
Intended consequence: Reduce child labor
Unintended consequence: Children don’t go back to school or remain out of the labor force. Instead, many took jobs with unregistered workshops that have even worse conditions.
Analysis: Benjamin Powell, director of the Free Market Institute, says that economic growth, not labor laws, is the answer to this problem. As countries become wealthy, their child labor rates fall.
|
|
|
|
Action: New GMO labeling law is set to come into effect: one of 35 bills in 20 states that have been introduced this year.
Intended consequence: Consumers know what GMOs are in their food
Unintended consequence: smaller grocers must incur unreasonable expense to comply, or go to jail!
Analysis: Legislators appear to have not thought through the negative impact on small grocers that the law will create.
|
|
|
|
Better executive decisions => reduced unintended consequences.
|
|
|
|
Decision: To add safety warnings on antidepressants based on recent studies that 1% of teens and young adults have adverse reactions, actually leading to a greater level of depression.
Intended consequence: To decrease the incidence of this negative effect among teens and young adults, and thereby decrease the depression and suicide rate in this population.
Unintended consequence: In the year following the warnings, antidepressant prescriptions fell by more than a fifth. There has also been a 21.7 percent increase in suicide attempts by overdose with psychotropic drugs, and this number goes up to 33.7 percent in the young adult population.
Analysis: This study is “one of the first to directly measure a health outcome driven by the interaction of public policy and mass media.” The news about the 1% of patients with adverse consequences ended up drowning out the fact that under-treatment of depression can have a big effect, too.
This is an illustration of a cognitive bias called the “availability bias”. Our brains tend to take a short-cut: instead of analyzing the facts, we use the frequency of an incident in the media as a proxy for its true frequency. In this case, even doctors were fooled.
|
|
|
|
Thanks to Steve Omohundro for these insights: “Just like us, fruit flies dwell on difficult decisions, according to a study published in the journal Science. They spend more time choosing between a strong and a weak smell if the difference is small. The research links this deliberation to a particular gene, FoxP, and the activity of fewer than 200 neurons. Mutations in FoxP, also associated with cognition and language in humans, made flies’ decisions even slower without affecting which choice they made. Gathering information before committing to a decision is a hallmark of intelligence. If the information is unclear, the choice is trickier and the decision takes more time. We do it, other primates do it, even rats and mice do it – but now it seems that flies do too. “
|
|
|
|
Decision: to regulate the sale of prepaid SIM card mobile phones in Papua New Guinea
Intended consequence: to reduce use of “burner” phones for criminal and terrorist activities
Unintended consequences: - Loss of communication to Wantoks (i.e. friends and family).
- Negatively impacting on PNG national emergency warnings and evacuation plans in times of natural disaster such as tsunami or major volcano eruptions.
- Increased cost to telecommunication providers and consumers.
- Reduced revenue for telecommunication providers.
- Growth in black markets for fraudulently registered or stolen prepaid mobile phones
Analysis: This article names a number of effects that come from tracing the cause-and-effect links out a little further than just the crime-reduction impacts. A full systems model might discover a way to make this decision that does not have these consequences: a “sweet spot” that achieves all intended outcomes, and perhaps even more.
|
|
|
|
- Decision: Implement geo-engineering schemes, such as dumping iron filings into the sea, or building giant reflectors
- Intended consequence: Reduce climate change
- Unintended consequences: “potentially severe side-effects”, according to a research lab in Germany
- Analysis: Suzuki says that geo-engineering solutions are only part of the problem, and need to be considered as part of the whole system to avoid unintended consequences
|
|
|
|
Decision: Build roads to support fracking operations in Texas Intended consequence: More efficient fracking Unintended consequence: Better roads for drug smugglers Analysis: Missed the link between roads and drugs
|
|
|
|
Decision: Ban fisherman from throwing away unwanted fish Intended consequence: Improve fish stocks Unintended consequence: Decrease fish stocks Analysis: A key link in the chain was not understood: discarded waste from humans is valuable in the ecosystem
|
|
|
|
Economic modeling is a field that stands to benefit tremendously from decision intelligence. This article is a data point confirming that. Specifically: – We need models that include links to other social science disciplines: how does climate impact the economy? Medicine? More. – We need to embrace the fact that the future is not like the past, and take accept the black swan nature of the economic events. A forward systems model is the only way to do that: induction from past data is not enough – Explicitly tracking the impact of varying assumptions (e.g. Greenspan’s versus others’) is essential to effectively working together.
|
|
|
|
An exploration of the unintended positive social consequence of individual selfish actions: the underpinnings of Adam Smith’s Invisible Hand. Worth a read to understand the historical background of one of our most important emergent patterns.
|
|
|
|
|
Main Insight: The 2020 lockdown helped prevent COVID-19 cases and deaths. However, unintended consequences included the delayed diagnosis of new-onset type 1 diabetes (T1D), undiagnosed severe pediatric diabetic ketoacidosis (DKA), and limited healthcare access for patients with chronic diseases. ACTIONS
• Issue strict lockdown orders
• Reduce access to emergency hospitals
INTENDED OUTCOMES
• Cases of COVID-19 measured as cases per million by the end of 2020
• COVID-19 related deaths measured as deaths per million by the end of 2020
UNINTENDED CONSEQUENCES
• Cases of delayed diagnosis of new-onset type 1 diabetes (T1D) measured as cases per million by the end of 2020
• Cases of undiagnosed pediatric diabetic ketoacidosis (DKA) measured as cases per million by the end of 2020
Analysis by: Frida Rodríguez Escobar
|
|
|
|
This article has two examples of unintended consequences. The article hypothesises an unintended consequence of a decision made as a response to an initial unintended consequence. The Treasury attempted to curb the irregular spending of state funds by instituting a ban on tenders and procurement. In practice, this ban required that all procurement made by government departments, municipalities and state-owned entities. Action: Any procurement above R30k had to be signed off by the Treasury. Intended consequence: Limit the opportunity for irregular spending. Unintended consequence: Single point of failure bottleneck created a severe backlog in requisitions meaning that govt. departments, municipalities and state-owned entities could not issue tenders. The effect of this is poor service delivery. Analysis: There is a political incentive to push for broad and "effective-looking" changes to expenditure policy. A showing of "zero-tolerance" gives the impression of well-regulated expenditure, but this poorly chosen intermediary has the effect of actually locking down local level service delivery. As a response to the unintended consequence above, the issued new procurement regulations: Accounting officers now have the ability to issue deviances from the tender and must report them to the treasury in 14 days. Action: Issue regulation that accounting officers can now vary tenders. Intended consequence: Give the departments the power to vary tenders and alleviate the bottleneck of Treasury sign-off. Unintended consequence: (Potential) The safeguards for irregular expenditure are now in the hands of the individuals who might spend irregularly. Analysis: This is essentially a rollback of the action previously taken instead of re-analysing it as its own decision. This is still not a systemic solution to the problem of irregular expenditure and tender awarding.
Analysis by: Jaco Stroebel
|
|
|
|
Action: Increase the Basic Housing Allowance (BAH) to military personnel Intended consequence: an appropriate compensation Unintended consequences: 1) No incentive to save costs, so personnel will move from many-roommate homes to more expensive single-dwelling housing 2) Landlords in military areas raise rents to satisfy increasingly wealthy personnel
|
|
|
|
Unintended consequences are mentioned explicitly in this report; nice to see some semi-formal attention to this topic.
|
|
|
|
Is it just me, or is the Rube Goldberg meme taking off a bit lately? Now if we could just model complex interactions between the economy, climate, and more with this much sophistication, we’d be getting somewhere!
|
|
|
|
Action: US airstrikes in Syria Intended consequence: impair ISIS Unintended consequence: ISIS and Al-Ansari, who had previously experienced a “bitter split”, could call a truce and/or unite under one banner, according to CNN, producing a level of power that could overwhelm more moderate rebel groups.
|
|
|
|
Transparency seems like a good thing, but it can lead to unintended behaviors based on the fact that observed persons know they’re being watched. It can reduce (good) risk-taking, and the use of judgment and intuition.
Another cause here: rules in any situation cannot capture every nuance of a situation, and so the use of context-based judgment is goes beyond optional and occasional to absolutely essential. Create too tight control – as can happen with the use of cameras – and you throw the appropriate use of instinct out with the inappropriate bad judgment bathwater.
Perhaps an adjunct or even substitute for cameras would be to train good judgment more effectively through role-playing and a deeper systems understanding and then, still, to trust your officers.
|
|
|
|
Action: Creation of the NAPLAN: The National Assessment Program – Literacy and Numeracy
Intended consequence: Improvements to education through objective measurement
Unintended consequence: stress symptoms amongst students
Analysis: This is a classic example of the “side effects” unintended consequence pattern: where an intervention with largely positive results creates some negative effects in certain students. As with side effects from medicine, a side effect can be considered acceptable if it is rare enough and if the impact is small enough compared to the benefit.
Also as with medicine, a better understanding of who is susceptible to the side effect, along with design of mitigation strategies for affected populations, can increase the net positive impact of the intervention.
As always, a good visual model showing the cause-and-effect pathway to desired as well as undesired outcomes can be helpful.
|
|
|
|
Action: The National Defense Authorization Act (NDAA) was established out of concern that local law enforcement agencies were literally being outgunned by drug criminals.
Intended consequence: Drug enforcement
Unintended consequence: Unnecessary escalation in Ferguson.
Analysis: This situation illustrates four unintended consequence patterns: 1) focus on just one outcome (law enforcement) instead of multiple outcomes from the decision (potential for unnecessary escalation of a conflict). 2) Underemphasis on the impact of one lever (training in the proper use of this equipment) 3) Underemphasis on two intangibles: a) value of training b) causal flows involving psychological / social factors 4) ignoring a feedback loop / vicious cycle: (armed cops->angry citizens->more use of force->etc.)
|
|
|
|
I hadn’t realized this! Bad decisions often come from unintended consequences, and most of those can be seen as arising from a lack of ability to see the entire system. Ergo: solve them through better systems thinking.
We seem to be suffering from an epidemic of hindsight, judging the situation by the data, not the system. It’s like trying to predict where the elephant will walk based on her footprints (or maybe something else she leaves behind) rather than actually understanding elephants themselves.
Many of the bad decisions in this articlecan be seen a consequence of this lack of deep systems knowledge / systems navigational ability: the firing of Steve Jobs, Bank of America’s purchase of Countrywide Financial, Fed moves that led to the Great Depression, and many more.
Wow. Hadn’t realized. Let’s fix this together, OK, friends???!!
|
|
|
|
An analysis of the situation with the U.S. border children through the lens of unintended consequences.
|
|
|
|
Action: Raise state-mandated math and science course graduation requirements.
Intended consequence: Improve college enrollment, college degree attainment.
Unintended consequence: Increasing dropout rates
Analysis: misunderstanding of the fact that raising standards, without a commensurate improvement in education to allow students to meet those standards, can increase dropout rates.
|
|
|
|
Decision: a doctor prescribes a medication for certain symptoms
Intended consequence: relief of symptoms
Unintended consequence: support for an addiction
Analysis: For certain patients,, a short-term, salient effect swamps understanding of a longer-term, larger negative impact. Shall we name this unintended consequence pattern “short term salience swamps long-term net negative?”
Solution: from the article: “…a significant number [of doctors] have had less than half a day of training in prescription drug diversion in medical school, residency, or continuing medical education. ”
Doctors need to be trained on the drug addiction impacts of prescriptions to avoid this consequence.
|
|
|
|
Decision: Pension reform in Illinois Intended consequence: Reduce pension costs Unintended consequence: Wave of early retirements, with impacts like this festival being canceled this year Analysis: As a transition like this one is implemented, often there are transient oscillations and other temporary effects before it re-equilibrates. These transients can have unintended negative consequences that are not characteristic of the new system, but can be painful for those impacted by the transition.
|
|
|
|
Another article by Josh Kerbel. Like the last one I shared, I’ll say if you read anything this month, this should be it. Kerbel is one of our generation’s great thinkers, and this article reflects his insightful prescription for how one complex, data-rich organization – the Intelligence Community (IC) in the US – must evolve to understand context, interdependencies, systems, nonlinear effects, sensemaking, and more. This is the essence of Decision Intelligence (no wonder, since Kerbel was very influential on our early work).
Please share.
|
|
|
|
Decision: A new EU law supporting the “right to be forgotten” to protect privacy Intended consequence: Privacy Unintended consequence: Criminal activity can be forgotten, too Analysis: Probably a known negative side effect
|
|
|
|
Decision: In a 2003 ruling striking down an anti-sodomy law, Scalia says this opens the door to all laws based on moral choices, like same-sex marriage Intended consequence: To strike down the anti-sodomy law Unintended consequence: Judges in Utah, Ohio, Virginia, Kentucky, Texas, Oklahoma and Michigan have all cited the 2003 ruling in their support of declaring same-sex marriage bans unconstitutional Analysis: Scalia overlooked the fact that invoking morality would translate to an argument against stigma, which would produce the opposite effect of his intent
|
|
|
|
Good interdependency analysis allows us to avoid unintended consequences and to capitalize on synergies, in many domains. And a decision sets the chain in motion.
|
|
|
|
oops. Looks like someone needed to run a full cause-and-effect model including GP financial well-being.
|
|
|
|
In my experience, often when people say that they need a “data scientist”, they really need a decision scientist . This article contains the six key steps that decision scientists follow
|
|
|
|
A good place review of some of the key principles and myths of failure in complex systems.
|
|
|
