As a business school academic, I have often marveled at the power of human creativity and entrepreneurship in the business sector to create enormous value and improve our world. I see no reason why the same ingenuity cannot be applied to social problems and I see no reason why the same individual passion, drive and entrepreneurship cannot be channeled to solving social problems.
Social problems tend to arise from systems that are complex and entrenched. Aiming for a complete overhaul of these systems is both impractical and impossible. Revolution is not the answer. We must therefore find ways to change people’s behaviors within the constraints of the existing system and to do so in ways that lead to significant improvements in the social problem.
How can we do this?
The answer we propose here is fundamentally different from what policy makers often do, so it’s important to explore first what people normally do and why that often fails.
More often than not, solutions to social problems tend to focus on three areas: (i) rectifying the most obvious sources of the problem; (ii) punishing or changing the people that are thought to cause the problem; and (iii) changing the incentives considered responsible for people’s bad behaviors.
None of these solutions work and it’s important to understand why.
Rectifying the “sources” of the problem
Consider the following story in a US newspaper:
[In 2007], amid much self-congratulatory hoopla, Oregon adopted the most stringent anti-meth laws in the nation – eliminating key ingredients for local meth cooks and kick-starting a national, even global war on what many consider the most addictive and disturbing illegal narcotic.
Today, it’s undeniable that Oregon’s laws were hugely successful in one area: The meth labs that endangered children and created hidden toxic waste dumps in basements and backyards across the state have been all but eliminated.
As [police] officers indicate, however, that success has borne unintended consequences – thanks to a massive influx of meth supplied by Mexican drug cartels.
Interviews with numerous local law enforcement officials and several meth addiction counselors – as well as a pending federal meth case investigated by the Community Newspaper’s news partner, Fox 12 News – suggest that Oregon’s legislative changes contributed to a radical transformation in the underground meth economy, one that in some ways is making the problem even more difficult to fight.
“The labs are gone, but there’s more meth,” said a longtime Portland Police Bureau drug cop, Sgt. Brian Schmautz, who stressed that he was only speaking for himself, not the agency.
“I’m not saying (Oregon’s meth laws are) not a good thing,” he said. “But we shouldn’t be fooled and say we have less meth, or less meth-related crime.”
… the transformation of the meth economy, shifting the source of supply to superlabs in Southern California and Mexico, has made cops’ jobs harder.
Previously, police say, meth was a “white boy” drug, and the bulk of the supply was generated by local “mom and pop” meth cooks, most of whom also were users, or “tweakers.”
Due to the mind-altering effects of the drug, local cops were more than a match for the tweaker-dealers who dominated the landscape of meth in Oregon. They were “the easiest people in the world to bust,” Walls said.
“They are just total morons,” Wonacott said. “They’ll do stupid stuff, and they’ll do it all day long.”
In stark contrast, the new meth distribution rings that dominate Oregon’s meth business are operated by Mexican organized-crime cartels. They are smarter and more creative and businesslike, according to Sgt. Pat Walsh of the Portland Police Bureau’s Drugs and Vice Division.
“They are the Wal-Mart of dope,” he said.
While neither he nor other cops said they prefer the days of local meth labs, many said that by removing the cartels’ mom and pop competitors, the Oregon law even may have increased the supply of meth. “I think most dope cops feel that way,” Walsh said.
… In short, the most successful victory that Oregon cops have scored in the war on meth hardly disrupted the flow of the drug.
“The unfortunate problem is that in the Portland metropolitan area you’ve got a (population) base here, and the traffickers are going to (service) it,” said Frank Romanaggi, commander of a Portland-based group of federal, state and local cops called the Regional Organized Crime and Narcotics task force. “So as you take a group out, another group shows up.”
He said that increasingly, in addition to penetrating a criminal underworld, drug cops also must infiltrate a foreign culture – meaning more time and expense, exacerbated by an unmet need for Hispanic informants. He estimated that cops’ interdiction efforts stop only a small portion of the incoming Mexican meth.
He was echoed by Cowlitz County sheriff’s Sgt. Kevin Tate, head of a southern Washington interagency drug task force.
“When it was meth labs that we were focusing on, it was pretty straightforward,” he said. “Now, we have the perception that all we get are the runners, the low-level figures, (and) that we’re overrun, and we can’t effect change.”
“We’re kind of bailing the ocean with a Dixie cup,” said DeLong, the Portland cop.
This story highlights one of the most fundamental “traps” that almost all of our current efforts to solve social problems fall into. Once a problem arises, we immediately search for the most obvious reason for the problem. We then focus our efforts at removing whatever caused the problem. Such linear thinking never works.
This is because the problem is part of a more complex system of interconnected and interrelated variables. Fixing one of the variables leads to a change in the other variables and they end up producing a series of unintended consequences down the road. These make the original problem even worse.
You see this problem raising its nasty head all the time. For example, the US Congress enacted the Passenger Bill of Rights to protect air travellers from abuses. One of its provisions was to fine airlines for long delays on the tarmac. The response by the airlines was to begin cancelling flights rather than face multi-million dollar fines.
As a result, passengers ended up being stranded, often for more than 24 hours, rather than being simply delayed by a few hours.
What started like a good-intentioned action to protect passengers, ended up making life more difficult for the people it aimed to help. Similarly, the UK government wanted to reduce the time that patients waited in the “Accident and Emergency (A&E)” department of hospitals. It therefore imposed a target of 4-hour limit.
Hospitals responded by keeping ambulances parked outside with the patient still inside so that the 4-hour clock would not start ticking! Once again, a well-intentioned action ended up producing unintended consequences that made the original problem even worse.
All this suggests that sorting out the most obvious sources to a problem is never the answer. The solution requires us to stop thinking in linear ways and to search for solutions to the problem away from its immediate and most obvious sources.
Changing the people will achieve nothing
Another typical response to a problem is to search for scapegoats in people. We have this inbuilt belief that if someone did something wrong, this individual needs to be punished for it. We rarely ask what prompted the individual to misbehave or whether we need to change something else to stop people from misbehaving.
For example, consider the following statement by Charles Ferguson, the Director of the movie Inside Job, while accepting the 2010 Oscar for Best Documentary: “Three years after a horrific financial crisis caused by massive fraud, not a single financial executive has gone to jail.” This is a classic example of blaming people for bad behavior without paying enough attention to the underlying reasons that led them to the bad behavior in the first place.
The sad truth is that if the underlying reasons are not corrected, the same bad behaviors will emerge again and again no matter how many people you punish or how severe the punishment.
The same bias manifests itself in our attempts to improve behaviors in a system by removing the “bad apples” and replacing them with “better” people.
This almost always fails. It doesn’t matter how “good” the new people are. If you put them in the same “situation” or “structure”, you will soon get the same bad behaviors out of them as you did with “bad” people.
For example, in April 2003, the U.S. Securities and Exchange Commission filed a complaint against Henry Blodget, an investment analyst with Merrill Lynch, alleging that Blodget committed securities fraud and violated the rules of the stock exchanges on which his firm traded.
Blodget’s website defended him by arguing that he was “keelhauled” by Eliot Spitzer, then the Attorney General of the State of New York, who later underwent his own keelhauling in a sex scandal and was replaced by his Lieutenant David Patterson who was later accused of inappropriate behaviors of his own! How did Blodget, Spitzer and Patterson defend themselves? None were very successful.
Blodget settled and paid millions of dollars in fines. Spitzer left government permanently. David Patterson declined to seek election to the Governorship.
The irony in all of this: In each instance, “culture” was identified as the culprit.
Spitzer in particular identified the culture of Wall Street as at the heart of Blodget’s behavior. Similarly, when Spitzer got into trouble, the culture of political leadership – the same culture in which Bill Clinton had an affair with Monica Lewinsky years earlier – was identified as the problem. Yet in each instance, the individuals charged paid huge personal penalties while the culture was left untouched.
We hold individuals responsible for systemic problems because privileged leaders – those in “charge” – represent the system itself.
We assume that the best mechanism for changing a system is to lead from the helm and that failure to change the system by a leader is sufficient grounds for keelhauling. But changing a system requires a much more sophisticated approach than only changing the individuals involved. Replace the people without changing the values, culture, and incentives and the same mistakes will occur over and over.
What will it take to prevent bankers and analysts like Blodget from succumbing to the conflicts of interest in their professions?
The challenge requires thinking differently about the problem—a task made all the more difficult because it requires creativity at precisely a moment in time when the mandate for change is urgent. Changing the culture of banking requires thinking differently about why Blodget and other analysts recommended stocks they didn’t believe in.
It requires understanding why bankers sometimes make loans that they know can’t be repaid and why consumers often borrow money without thinking about the consequences.
Changing the incentives will achieve nothing
The same idea applies to incentives.
We have an inherent bias that to get people to behave in ways we want, all we have to do is to give them the right incentives. This rarely works because incentives are only one element that make up the underlying structure of the system.
Even when incentives produce short-term change, they very rarely produce the lasting change in behaviors that we want. Only a change in the underlying structure of the system will achieve this.
Consider, for example, our attempts to get pharmaceutical companies to sell more of their life-saving drugs in poor, developing countries. In 1994, the World Trade Organization created an incentive for pharmaceutical companies to sell drugs in developing countries. Some 17 years later, there is little evidence that the incentive has worked. What went wrong?
To answer this question, it’s important to understand how the incentive worked.
It was based on enforcing patents on the drugs in poor countries (more precisely, in every country in the World Trade Organization). The way that patents work is that they give the seller a monopoly on the drug for a number of years – usually about 20 – after its invention.
Under the new policy, the sellers would be assured that imitators would not compete down profits even in poor countries, some of which had a history of hosting imitators.
The policy was adopted on the theory that the incentive (i.e. the patent) would make it worthwhile for drug companies to distribute their drugs in poor countries.
The goal was to create benefits that would help poor people as well as the companies themselves. The most important benefit in developing countries would be that drugs would become available.
If the incentive worked, then pharmaceutical companies would have an incentive to develop new drugs that were specifically designed for the kinds of diseases that turned up in developing countries – such as malaria, tuberculosis and HIV.
Nearly 20 years later, it seems that the incentive has not worked in the way that was envisioned when it was created.
We have not seen a flood of drugs into poor countries with the implementation of patent protection. The question is why? One might think that the issue was the ability of poor people to pay for drugs: If the recipients of the drugs cannot afford the drugs, then there is little profits for the drug companies in distributing them internationally. And, of course, without the prospect of a return, the drug companies also do not have much incentive to develop drugs for the diseases of the poor.
But a closer look at the situation might cause you to dismiss such an explanation. It turns out that agencies and organizations like the Gates Foundation and even the US government pay for the drugs most of the time—a fact that implies that the drug companies get paid even if the patients themselves cannot afford the drugs they need.
And some drug companies sometimes offered to charge low prices out of humanitarianism. Generic manufacturers made versions of drugs that went off patent available.
So what was the problem? Why were relatively few drugs distributed in poor countries? It turns out that the problems were even more fundamental than ability to pay.
What is missing, fundamentally, are the elements of a system for delivering drugs in the poorest countries of the world.
A vast shortage of doctors and medical personnel means that prescriptions for drugs go unwritten. Too few clinics are available.
Pharmacies, laboratories, hospitals and medical schools are in short supply. Distribution centers don’t exist. Refrigerated systems for moving vaccines and other fragile medicines are not available.
Medical devices are difficult to repair, which means that crucial diagnostic tests may be impossible to administer. Supplies such as X-ray films, test tubes and syringes are in short supply.
As a result, patent incentives on their own don’t work very well.
No matter how strong the profit incentive, drug companies can’t effectively get their products distributed and administered. What is needed is coordination and leadership to get the structure of the system in place so that interconnections between the elements of the system – such as incentives, for example – can work.
So what will it take to get more doctors, nurses, physician assistants, pharmacies, laboratories, medical supplies, equipment, and everything else required to treat patients effectively in developing countries?
You might think – We just need more incentives! We need incentives for students to go to medical school and incentives for patients to go to clinics and payments to supply companies. And you would be right. All of these incentives are needed.
But this analysis would be incomplete. The issue is that incentives are only one part of a much bigger picture. A focus exclusively on incentives leads you into making a number of big assumptions about what should be done first, and about the ways people will respond to different profit opportunities.
For example, imagine that you were to focus first on getting students to medical school, and imagine again that you were successful in making progress: that students from developing countries responded to your incentive program and enrolled in medical courses. You might feel that you had achieved a victory and of course it would be true that the incentive had worked.
And yet a problem would arise: The students that graduated from medical school under your incentive program would go to great lengths not to practice medicine in a developing country. Why? Because they have no supplies, medicines or pharmacies there to allow them to practice medicine!
Dealing exclusively with incentives is like trying to tell someone how to construct a house by discussing only the nails that will tie together the boards.
A much more direct and effective way is to present the builder with a complete plan that shows what the house should look like when it is constructed, and that includes a detailed architectural diagram describing the boards, joints, and other elements of the building along with ideas about how they may be joined by nails, screws, glue and other materials.
In global health, getting incentives right is essential to effectiveness, just as nails are essential to building a house. But describing in a sequential way all the elements that should be nailed together would be an enormously cumbersome process for planning the construction of a house. Similarly, a plan for global health is cumbersome if all it does is deal in a piecemeal way with incentive problems one after the other.
What is needed is a detailed plan.
Where do we start?
So, changing the incentives or the people will not produce the desired change in the behaviors of people.
Nor can we expect much progress if we aim our efforts at the most immediate and obvious sources of the problem. Where, then, should we focus our attention to achieve the desired results?
It turns out that the academic literature from two disciplines—social psychology and system dynamics—provide us with a common answer to this question.
First, we have more than fifty years of research in social psychology that explores the factors that influence how people behave.
Through a series of celebrated experiments—such as the famous Stanford prison experiment; Asch’s experiments on conformity; and Milgram’s experiments on obedience—this literature has shown that the “situation” or “underlying environment” has a huge effect on how people behave.
Second, a huge body of literature in System Dynamics has shown that “the underlying structure of a system creates the behaviors in that system,” and that small changes in high-leverage points in the structure of the system can have a big effect in how people behave in that system.
All this suggests that:
- To achieve change that lasts, we need to focus our efforts on changing the underlying structure of the system, not just on incentives or people;
- To change the underlying structure of a system, small changes can have a big effect. But these small changes must occur at the right points in the system at the right time.
- To prevent unintended consequences, we need to focus on changing the whole system not individual parts of the system.