Need Not Greed: Bonuses, Risk–Taking And Evolution

What to do with those ‘greedy’ bankers

In the wake of the most recent financial crisis, financial regulatory bodies across the world have sought ways to curb the excessive risk–taking of bankers. Most often, this is done by capping bonuses and developing clawback policies (Gray, 2016). Such decisions are based on the widespread view that financial incentives are a factor for risk–taking. That the potential to earn large personal payoffs is so attractive that many individuals are willing to take excessive risks in order to earn them. This is the classical ‘homo–economicus’ view of human behavior, where people are considered rational actors, motivated to maximize personal utility for reasons of self–interest (Aktipis & Kurzban, 2004; Thaler, 2000). In other words, be greedy.

But…. what if bankers didn’t take risks to earn their bonuses for reasons of greed, but for reasons of need?

We are ‘needy’ humans 

Bankers are of course ‘homo–sapiens’ not ‘homo–economicus’. So they are driven and influenced by the same innate mechanisms that guide and influence us all. So why would we seemingly be willing to take risks to earn more? What actually motivates and drives risk–taking? The answer may lie in how we have evolved.

Throughout history, our ancestors constantly faced problems. Often the same problems over and over again. The particular conditions of the environment and the problems encountered sculpted specific adaptations to overcome them. Sort of like a lone tree on an unsheltered hill. It is constantly battered by winds, and over time it adapts to deal with this by curving over, allowing the wind to pass easily. Similarly, humans have evolved behavioral adaptations which exist to solve a particular problem our ancestors faced in a particular domain.

One of the problems which risk–taking solved, was ensuring we had enough energy (i.e. we ate) in unpredictable foraging environments (Caraco, Martindale, & Whittam, 1980). When foraging, our ancestors may have faced a number of choices; they could forage in the area they knew was not overly abundant but fairly consistent in terms of food levels (i.e. the low variance, low risk option) or they could forage in a more variable area that sometimes had a lot of food and sometimes not much at all (i.e. the high variance, high risk option). The choice depends on the person’s need levels. If their need, in terms of energy requirements, are fairly well satiated, then why bother going to the highly variant area? The low risk option will suffice. However, when the person’s need is high, they are close to starvation and energy levels are depleting, then taking the chance with the risky option might just payoff. It is certainly worth the risk, as the low variant option won’t offer enough energy to ensure survival.

Thus, humans have evolved to be very sensitive to need levels. Ensuring needs are met ensures survival. Consequently, we tend to assess our current situations in terms of whether it is meeting our needs. When needs are unmet, (i.e. our current state is disparate to our desired goal state) and a low risk option offers little chance of meeting our need, we are innately motivated to take risks (Gonzales, Mishra, & Camp, 2016; Mishra, 2014).

Need – satiation: the ultimate driver of risk–taking

Evolutionary perspectives operate at an ultimate level (Scott–Phillips, Dickins, & West, 2011). They seek to explain why something exists or what its function is, rather than offer a proximate explanation of how it works (Saad, 2011). Theorizing that need satiation is the ‘ultimate level’ driver of risk–taking gives such behavior an origin and a rationale. More importantly, such innate drives are human universals and so have broad application for understanding risk–taking in many different contexts (Witt, 2016).

An interesting aspect of many bonus systems, particularly those in the banking sector, is that they are contingent on meeting a particular performance level. Banks set targets and individuals have to meet those targets in order to be eligible for reward. Furthermore, not meeting those targets may be detrimental to the individual — it might limit their progression or result in them losing their job. In essence, their survival is somewhat dependent on performing and meeting those targets. Unbeknown to managers, this may be creating a ‘need’ situation comparable to those our ancestors faced.

Imagine a salesperson who has a target to sell $10,000 worth of products each month. This essentially becomes a required ‘need’ level. What if he experiences poor sales a few months in a row? This puts him in a high need situation, as the likelihood he will achieve his target is reduced and his ability to do so is compromised. Such an individual will be more likely to prefer a gamble where he has a 10% chance of making $10,000 of sales, rather than accept a certain sale of $1,000. In the finance sector, yearly rather than monthly targets are set, and many such professionals recount how people down on target towards the end of the year tend to up their risk–taking (Shapira, 2002). Furthermore, most accounts of rogue traders describe how their risk–taking increased with every loss, as they sought to recoup the losses and break–even (Abdel–khalik, 2014).

The evolutionary perspective described here provides an explanation for why this happens. Bonuses arguably don’t motivate people to take risks because they are greedy. Rather, they impose a threshold which individual have to or need to meet. By doing so, they evoke the evolutionary function of risk–taking — need satiation. Humans have ensured their survival by evolving to take risks when their current situation does not meet their desired goal state (Mishra, 2014).

What is the implication of this?

Recognizing that risk–taking is influenced by something so innate, has specific implications for interventions (McDermott, Fowler, & Smirnov, 2008). Rather than trying to cap bonuses, the focus should be on carefully calibrating performance targets. Much of the theory behind goal–setting rests on the premise that the harder a target is to achieve, the greater the effort it will motivate (Locke & Latham, 2002). Consequently, many organisations, not just those in finance, set employees hard to reach goals. However, if the target is so hard to reach, it may evoke risk–taking as individual’s perceive themselves to be in a situation significantly disparate to where they need to be. Of course, taking risks is not necessarily a bad thing. In fact, it is often required. Problems arise when risk–taking becomes excessive. From this evolutionary perspective, the greater the disparity between current state and desired goal state, the more willing someone is to take risks.


The AI Governance Challenge

Understanding this ultimate driver of risk–taking may also help people make better decisions. For example, in financial markets traders, are expected to make decisions based on ‘running their winners and cutting their losers’. However, many traders describe how difficult that it is to do. They tend to take profits too early and chase losses (Willman, Fenton–O’Creevy, Nicholson, & Soane, 2002). Such behavior is understandable when you recognise the need satiation function of risk–taking — we evolved to take risks to meet needs, but as soon as needs are met we lock in that level. We have not evolved to maximize, but to satisfice and accept good enough decisions (Naumof, 2016).

When facing difficult economic circumstances, such as the financial crisis, it is often easy to forget that human nature is involved. Designing organizational systems with the evolutionary adaptations which underlie our decision-making, provides us with novel ways to overcome further crises and enhance the functioning of many organizations.

Disclaimer: The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of The Decision Lab.

To Nudge, or Adjudge? That’s the Enviro-Policy Question

Ensuing America’s election of Donald Trump, many Canadian energy producers are feeling ill at ease about our nation’s future environmental policy. This sudden apprehension surfaced after realizing the U.S’ new position on energy. President-elect Donald Trump has pledged to withdraw from the Paris climate agreement, has repeatedly denied the existence of climate change, and has expressed his will to repeal some of America’s current environmental regulations. In sharp contrast, Ottawa announced plans to impose a controversial national carbon tax, which aims to charge $50 per tonne of carbon emitted by 2022, according to BNN. The Canadian energy industry is imploring our government to rethink carbon taxes as it will curb their leverage with the U.S, our biggest trading partner. This article uses insights from behavioral economics to propose the implementation of nudges as a more innovative solution to traditional carbon taxes.

Drawbacks of Carbon Taxes

Within the domain of environmental economics, regulating through constraints on prices and quantities are common. In both cases, regulators intend to generate behavioral change so that consumers will pollute less. Attempts of behavioral change stem from the classic carrot and stick approach, which produces desired behaviors through a combination of rewards and punishments. However, these conventional methods are riddled with several drawbacks.

From a logistical perspective, taxes are sometimes difficult to implement (Gaunt, Rye, & Allen, 2007) as they frequently meet resistance from political parties and lobby groups. Firstly, carbon taxes raise issues because of economically misinformed consumers. At the basic level, agents struggle to differentiate Pigouvian taxes (intended to correct an externality) from Ramsey ones (intended to raise revenues). As consumers are generally opposed to the implementation of taxation, the ‘social cost’ of carbon tax adoption increases, which can potentially lower altruistic motivation to reduce energy consumption (Ouvrard & Spaeter, 2016). Even if carbon taxes successfully reduce agents’ energy consumption, it is unlikely that the fiscal deterrence itself will persuade them that polluting is harmful.

Further, governments need to consider the level of specificity in the information required to properly implement carbon taxes. Regulators must perfectly know each agents’ environmental sensitivities (concern for the environment) and risk perception (evaluating global warming as a threat) to calculate the optimal carbon tax (Ouvrard & Spaeter, 2016). Clearly then, acquiring the necessary data to formulate the optimal carbon tax on a mass scale is unfeasible. This suggests that other carbon tax models will be economically suboptimal. Nevertheless, some carbon tax models are good in theory. 

Some potentially good carbon tax models offer cash rebates to low-income households to protect them from the burden of taxation hikes caused by richer agents who consume more energy. However, even this model is flawed because it precludes price sensitivity. This occurs because rebates depend on an agent’s consumption in comparison to the average energy consumption of every other agent. So, if the government cannot connect this marginal incentive to effective wealth, then a marginal dollar in incentive will over-motivate the poor, and under-motivate the very rich (Galle, 2013). Clearly, even otherwise well-intentioned models fail to adjust to varying levels of wealth — a fact that is especially troublesome considering our current wealth inequality.   

Benefits of Implementing Nudges  

We desperately need an alternative method. Nudging is one effective tool with the potential to serve as a viable solution to current challenges faced in environmental policy. By definition, a nudge improves consumers’ behavior by making them conscious of their behavior either by disclosing useful information or by using simple techniques like default options (Thaler & Sustein, 2009). In this way, a nudge is simple, affordable, and non-intrusive.


The AI Governance Challenge

One simple example is installing ‘smart thermostats’ like the Nest Learning Thermostat. The initial choice results in savings on the energy bill, and reduction in overall energy usage. In fact, independent studies from Nest have shown that users save up to an average of 10-12% on heating bills and 15% on cooling bills. Like nudges do, this thermostat shares the history of energy usage on users’ mobile devices. This positively reinforces consumers to retain their eco-habits, or motivates consumers to conserve even more. This thermostats unknowingly uses behavioral economics as it automatizes behavioral, which reduces the cognitive effort of actual behavioral change. As a result, it is more likely to equally impact both rich and poor households as the thermostat is programmed to reduce energy consumption when it is not needed.  



Fig 1. An example of the energy history data stored on Nest’s mobile app.

Energy suppliers have also implemented nudges. Opower, the American electricity company, employed social normative messaging. This nudge involved comparing a person’s electricity usage to that of their neighbours in the attempt to reduce overall electricity consumption. Empirical findings showed promise. Schultz and colleagues (2007) reported that the use of normative messages with injunctive emoticons were effective at reducing energy consumption. According to one estimate, this program reduced consumption by 2% (Allcott, 2011).


Fig 2. A draft of Opower’s Home Energy Report (HER), with comparative data and adjunctive emoticons.

Marketers can also nudge consumers to buy eco-friendly appliances by strategically designing energy efficient labelling. In a choice experiment with various labelling conditions, preferences for household appliances were analyzed. Newell & Siikamäki (2013) found that basic information on the economic value of saving energy was the most important element in promoting cost-efficient investments in energy efficiency appliances. Physical energy use and carbon dioxide emissions helped additionally but with lesser importance.


Fig 3. An example of a strategically designed energy efficiency label.

In conclusion, nudges are a step in the right direction. We should add them to our toolkit to prepare us for when the good old fashioned carrots and sticks fail to work. We should not rely on such outdated techniques. There is no room for them in our thriving age of information.

To Marry Or Not To Marry? A Behavioral Perspective

The current issue of early and child marriages 

Approximately 39,000 children are married every day (Unicef, 2013) and pushed further into poverty and despair. In developing countries such as South Asia, tribal leaders can decide marriages, with parents or children having no say in the matter, robbing girls of their future as early as the age of 13. Commonly thought of as a developing world issue, early or child marriages is a practice seen across the world, affecting girls disproportionately. The Tahirih Justice Center, a non-profit advocacy organization, finds that child marriage persists across the US, even today, legally through parental or judicial consent for children under the age of 18.  Le Strat et al, (2011) show that as many as 8.9% of women were married as children in the United States in 2011.

By definition, an early marriage or a child marriage is a marriage before the age of 18, the common legal age for marriages across the world. At the age of 18, individuals are usually considered to be legally an adult, having completed basic 12 years of schooling and able to earn a living wage. Child marriages can have harmful consequences for the individuals involved. Girls that marry underage face many hardships including family instability, incomplete education, lack of work opportunities, higher risk of domestic violence and deteriorating mental and physical health.  As devastating as the outcomes of early marriages are, it is imperative to understand the behavioral practice and its effects on economic decisions of the household.

The shortcomings of current policy interventions

The concern about early marriages is not new and is recognized by governments alike. In attempts to better opportunities for girls, various policy interventions have been tried and tested across the world. Most of these include programs that aim to increase the education attained of girls, which research proves to significantly delay the age at marriage (Duflo et al, 2015). These interventions have been in the form of education subsidies and/or cash transfers such as the Berhane Hewan Program in Ethiopia and Punjab Female School Stipend Program in Pakistan. While helpful in the short term, they may or may not entirely alter the behavioral decision and validation for early marriages that exists in certain cultural norms.

Behavioral decision theory demands a rational decision of the age of marriage for an individual. Many parents will argue that an early marriage is based on rational persuasions rooted in cultural tradition. However, these stem from incomplete information of the consequences of early marriages and a lack of understanding of the greater potential. Therefore, where many policy interventions have taken place, none have been as successful as hoped in generating a permanent behavioral response.

An opportunity for behaviorally guided policy responses

What is required is a change of belief structure, which can only come about through the understanding of the consequences of child marriages. Here, non-profit organizations can play a significant part in the development of societies, raising awareness and understanding, which is a much less costly initiative than education subsidies and cash transfers.


The AI Governance Challenge

Raising awareness needs to be structured based on beliefs and respective cultures. Some forms of this strategy have been used in developing countries, but has been entirely absent in the developed world. This absence could be rooted in the illusion that child marriages do not exist in the west. One form of acheiving this awareness in the developing countries has been door-to-door visits of lady health workers with pamphlets. Another more recent idea has been to perform moral based theatrical events or plays in the village community. This helps shift beliefs individually and collectively of the village community. For altering beliefs in countries such as the US, a possible avenue is using media in a similar manner as the theatrical plays of the villages in the east.

As completely clear as this may seem, policy and research have not been able to test whether providing more and complete information alters behavior of early marriage across the world, specifically in developing countries. Thus while the simple solution makes sense, evidence does not exist that proves this would be case.  One thing that is certain, a truly committed response is key to reducing this practice.

How Wells Fargo Nudged Their Employees To Commit Fraud

This article originally appeared in [] and belongs to the creators.

Over the course of four years, at least 5,000 Wells Fargo employees opened more than a million fake bank and credit card accounts on behalf of unwitting customers.

Although many bank accounts were deemed “empty” and closed automatically, employees sometimes transferred customer funds to the new accounts, triggering overdraft fees and hurting credit ratings.

This scandal feels different from the mortgage crisis because it was not carried out by the 1 percent – such as wealthy investment bankers indifferent to the effects of their actions on regular homeowners – but by “$12 an hour employees,” as one lawsuit alleged. Even if supervisors encouraged or directed the fraud, it was likely these low-wage workers who actually clicked the button to open those accounts.

These workers likely knew better than most what it’s like to be slapped with unfair overdraft fees or undeserved hits to their credit rating.

So why did they do it?

Situational cheating

Social science research suggests that ethical behavior is not about who you are or the values you hold. Behavior is often a function of the situation in which you make the decision, even factors you barely notice.

This makes cheating more likely to happen in some situations than others. Although many honest Wells Fargo employees realized that opening fake accounts was wrong and refused to do so, it is also the case that other employees who considered themselves honest participated in the fraud.

What would it mean to apply these behavioral insights to the Wells Fargo situation? Here, I draw from White House guidance on how to implement lessons from behavioral science into government policy to identify situational factors that contributed to the fraud.

Repeated reminders of terrifying incentives

“In cases where the goal of an incentive is to encourage a particular behavior, agencies should ensure the incentive is salient to individuals.”

As early as 2010, Wells Fargo imposed extremely aggressive sales goals on its employees. Specifically, they were told to sell at least eight accountsto every customer, compared with an average of three accounts 10 years earlier.

Unmoored from what his salespeople could realistically achieve, the CEO justified this goal on the basis of a simple rhyme, telling shareholders in the bank’s 2010 annual report:

“I’m often asked why we set a cross-sell goal of eight. The answer is, it rhymed with ‘great.’ Perhaps our new cheer should be: ‘Let’s go again, for 10!’”

These goals loomed large when supervisors threatened salespeople who failed to meet them. One former employee interviewed by CNN reported, “I had managers in my face yelling at me” and that “the sales pressure from management was unbearable.”

Another former employee told the LA Times: “We were constantly told we would end up working for McDonald’s… If we did not make the sales quotas … we had to stay for what felt like after-school detention, or report to a call session on Saturdays.”

A lawsuit against Wells Fargo alleges that “employees who failed to resort to illegal tactics were either demoted or fired as a result.”

As the guidance suggests, incentives matter a lot when they are highly salient, or foremost in the employee’s mind. It’s hard for an employee to ignore the threat of losing one’s job or even the threat of embarrassment in front of other employees.

At a bare minimum, Wells Fargo should have done a better job of investigating and stopping the coercive enforcement of its sales goals.

Cheating is contagious

“[I]n many contexts, individuals are motivated by social comparisons, such as learning about the behavior of their peers. Research finds that individuals reduce residential energy consumption when provided with information on how their consumption compares with that of their neighbors.”

While the guidance emphasizes the positive side of social comparisons, they also work the other way: watching others misbehave influences our own misbehavior. We’re more likely to litter in a park full of litter – especially if we observe someone else littering. Watching someone else on our team cheat on a test makes us more likely to do the same.

In his congressional testimony, Wells Fargo CEO John Stumpf made it sound as though the employees responsible were bad apples or lone wolves who disregarded the company’s code of ethics. Although we don’t know the identities of the terminated employees, this is an unlikely explanation for a fraud so widespread.

What is more likely is that the fraud occurred in clusters, as groups of employees rationalized their decisions. This hypothesis is consistent with the CEO’s testimony that branch managers were terminated, suggesting that entire branches may have been infected by cheating.

A lawsuit filed against Wells Fargo also claims that employees shared with one another the know-how used in the fraud. They used shorthand reminiscent of a video game hack: “gaming” referred to opening accounts without authorization, “sandbagging” meant delaying customer requests, “pinning” stood for generating PINs without authorization and “bundling” involved forcing customers to open multiple accounts over customer objections.

This euphemistic terminology allowed employees to lie to themselves about what they were doing, making it seem as though they were gaming the system rather than ripping off customers.

A victimless crime

“Consider the framing of the information presented.”

In retrospect, it seems impossible to believe that any honest person at Wells Fargo would have felt okay about opening fake accounts. But as social scientists Nina Mazar and Daniel Ariely have argued, “people like to think of themselves as honest.” But their research shows that “people behave dishonestly enough to profit but honestly enough to delude themselves of their own integrity.”

In this case, the Wells Fargo employees probably focused on the respects in which their actions were harmless and ignored the downstream implications of what they were doing. Even Stumpf was guilty of this form of self-delusion, explaining to Congress that he initially believed the practices were harmless because empty accounts were “auto-closed” after a certain period of time.

Research suggests that people are more likely to engage in dishonest conduct in which they can tell themselves they’re not stealing money. As implausible as it may seem, Wells Fargo employees may have told themselves they weren’t “stealing” because they weren’t directly removing money from someone’s account. They were just moving it from one account to another.


The AI Governance Challenge

Technology also tends to have a distancing effect. Pressing buttons on a screen feels morally different from robbing a bank, even if it achieves the same result. That’s sort of the premise of a main plot point in the comedy “Office Space,” when the main characters unleash an algorithm designed to steal fractions of a cent from bank transactions.

Wells Fargo employees may not have considered how their conduct affected customers in terms of overdraft fees or credit ratings. Even if they did, they could rationalize those consequences as outside of their control. In their minds, it was the Wells Fargo algorithm that assessed the overdraft fee. It was the credit rating agencies that make decisions about credit scores. The logic goes something like this clip from “The Simpsons” in which Bart punches the air and marches forward, warning Lisa that if she gets punched, it’s her own fault.

In this case, the customer didn’t even know the punch was coming.

‘I don’t buy it’

As early as 2011, the Wells Fargo board was informed about reports of ethics violations. The cheating continued, leading Wells Fargo to fire at least 1,000 people per year in 2011, 2012 and 2013. Any company that fires 100 people for the same type of cheating, let alone thousands, knows or should know that situational factors are contributing to the cheating.

Instead of addressing that environment, however, the bank allowed the situation to persist. In the words of Representative Sean Duffy, who dismissed the CEO’s claim that they are now “trying” to fix the problem, “We’re five years on! … I don’t buy it.”

So how to fix a culture that’s gone bad?

Although we don’t know what sorts of internal controls Wells Fargo had in place, it should have examined the patterns of cheating and made it both practically – and morally – harder to do.

Five years later, the bank is finally sending customers an email every time a new account is opened and revising its sales goals. It also needs to revisit how supervisors are evaluated and crack down on those who threaten employees over sales targets.

Software could also be used to apply “moral speed bumps” that remind employees engaged in suspicious activity like opening unauthorized accounts that the behavior is wrong and illegal.

Most of all, Wells Fargo needs to send a strong message to its employees about the moral implications of their actions. In my view, that starts with the CEO’s resignation.

How Your Workplace Might Be Making Bad Decisions For You

An environment speaks a thousand words

The objects around our homes and our offices say a lot more than we think they do. And sometimes, they don’t exactly say what we want.

Research has shown that environments themselves can convey which behaviors are considered ‘normal’ when people are in them. They even encourage certain types of behavior when people enter them.

In an age where consulting and co-working spaces are the new norm, and offices are no longer where an employee will stay from 9-5 for the next thirty years, do we want our workplaces to be conveying the wrong messages? Can we really afford for employees to be surrounded by an environment that stifles their productivity and their work?

But how do you change what message you are putting out there?

Make people feel like they belong

Belonging is one of the most important factors for a workplace. If people feel like they don’t belong somewhere, it can seriously impact how they work.

In 2009, research found that people could decide whether they wanted to join a group  just by looking at the group’s physical environment. Environments with lots of objects that are obviously linked to male-dominated areas (eg a Star Trek poster for computer science) were enough to make women feel like they didn’t belong there. And this made them not want to join teams that worked in these environments.

However, there is hope! By having objects that were similar but not so male-focused (eg a landscape poster), women’s belonging and interest was restored.

The main take home message from this is that people’s own identities are important for the environments they feel good in. Workplaces should focus on avoiding having lots of objects that prime the idea that only certain people belong there. But in practice what does this actually mean?

Rather than having a big company sign that highlights how amazing the business world is, workplaces should show off how great it is to be a part of their team. Kay et al (2004) even found that the presence of common ‘business objects’, such as briefcases and boardroom tables, made people more competitive.

So having photos of your actual team all out for lunch together or at the Christmas party could go a long way to showing people that you have a great team identity, and that they could belong in your environment.

Then, what effect does the workplace have on people once they feel like they do belong?

Give the walls a motivational makeover

Once people are within a workplace, the kind of motivation they are given can also impact the way they work. And the environment could be nudging them towards the wrong type of motivation.

There are two main ways at looking at achievement:

  1. Performance – Failure is bad. Success is good. To be good you need to do well and do better than others.
  2. Mastery – Slip-ups happen. You need to break things to make things. You might make mistakes, but you will learn from them and learn new things along the way too.

A lot of the time workplaces can be nudging people to be motivated towards performance. Yet, time and time again, having a mastery attitude has been shown to be much better for performance than having a performance attitude. It is definitely easier for people to persist at challenges with a mastery mindset.

A common example of a workplace promoting performance might be having a visual record of how many sales everyone has made and rewarding the best salesperson. Or having a list of what the team needs to achieve for the month. The perfect image of performance motivation is the ‘top scarer’ leaderboard in the film ‘Monster’s Inc.’, and how all of the employees are obsessed with who the top scarer is every day.


The AI Governance Challenge

Yet companies can use some priming to nudge workers into a mastery attitude instead. Instead of buying posters that shout ‘Failure is not an option!!’, workplaces could be kitted out with slogans that chant ‘success is how high you bounce after you fall’ or ‘you never know unless you try’. Research has shown that priming people with mastery can lead to them performing better, even if they usually have a performance attitude!

Reclaim your physical environment

So right now your workplace could be deciding who belongs there, and it could be telling some talented people that they don’t. It could also be encouraging the people that do belong there to have the wrong attitude to work.

By thinking about what the objects in your office really say about your team, you can start to help everyone around you make the right decisions about where to work, how to work, and how to overcome those challenges that face us every day at work.

Why Don’t We Save Enough for Our Retirement?

This article originally appeared in [] and belongs to the creators.

When they retire, Dutch citizens really have it good. Not only does everyone receive a basic income of approx. € 1,000 from the state when they turn 65, but most people (as many as 90% of the citizenry) also get employer-administered pensions that generously provide 85% to 95% of what their earnings. It is no surprise that the Dutch pension system is among the best in the world, and the source of much envy.

In the US, and elsewhere, however, things are a lot bleaker. While most Americans receive social security (similar to the Dutch basic income), it is nowhere near enough to replace pre-retirement income. Any other money must come from what people have saved themselves for retirement over their working years. Financial experts are concerned that a majority of Americans are not saving enough money for retirement, and that in fact, we are in the middle of a retirement savings crisis that will only grow in severity.

I have written before about how optimism can lead people to postpone saving money. In this post, I want to explore three other psychological reasons for why people do not save enough for retirement. Surprisingly, none of these reasons have to do with earning sufficient money. Instead, they are all psychological and have to do with the person’s knowledge and motivation. The idea behind this post is that if we can understand these reasons properly, it will give us some ideas about how to go about solving the retirement savings problem, at least for ourselves individually.

1. Setting a clear target retirement savings goal is really, really hard.

Although the conventional wisdom is that people retire when they turn 65, the reality is much more complicated, especially since the great recession. People are living longer, and so Americans routinely spend decades in retirement. During this time, many people continue to work full-time or part-time, or even begin a whole new career. Others retire prematurely because of health issues or because they get laid off and can’t find another suitable job. After they retire, some people increase their spending by traveling a lot or indulging in expensive hobbies they have dreamed about throughout their working lives. Others downsize significantly, spending a fraction of what they spent before. Because of these enormous variations, and the inherent uncertainties about lifespan, health, and even the financial environment in which their investments will grow, it is very difficult to figure out exactly how much money one will need to save for retirement. In other words, there is no “magic number” or specific target to aim at.

Saving goals are problematic anyway. Decades of psychological research on setting and pursuing goals shows that abstract goals are the most difficult goals to pursue. People have difficulty developing a plan of action for abstract goals, and they are more likely to procrastinate and avoid pursuing them. Where retirement savings are concerned, without “the Number” to aim at, goal pursuit is hampered for many people.

2. Getting started with retirement saving & then adding to the savings regularly are two huge problems.

When setting a monthly budget, or making purchase or other financial decisions, retirement seems like a distant milestone, somewhere over the horizon. Other nearer-term, and more specific, financial goals take priority. In their 20s and 30s, individuals pay off their student loans, and save money for a down-payment on their first house. In their 40s, saving for children’s college education, and paying down the mortgage becomes important. And in their 50s, people are saving for (or paying off) discretionary purchaseslike vacations, second homes, boats, or RVs. At every life-stage, other financial goals seem far more salient, and the resulting accomplishments far more significant, than saving money for retirement.

It’s not that people don’t have a sense of urgency about saving for their retirement. When asked, people consistently say they do have retirement saving on their radar. For example, in one survey of 2,700 millennials conducted by the financial services company Aegon, 41% of respondents indicated that they aspired to save for the future, and 37% felt “very responsible” for their retirement savings.  In another survey by Wells Fargo, 61% of respondents expressed urgency. Yet, this sense of urgency does not translate into action. In the Aegon survey, for instance, only 9% of the millennial respondents indicated that they had an actual saving plan for their retirement

3. Many of us don’t understand how exponential growth works and why it is important to saving now, without waiting for a single day longer.

One of the most significant characteristics of retirement savings (and actually, any saving) is that because of the magic of compounding, the money grows exponentially—increasing ever more rapidly as time passes and it grows in value (see adjoining figure). Consequently, starting saving money sooner leads to a disproportionately larger nest egg, and there is a huge cost to waiting to save money. However, as I have written before, people have difficulty with understanding how growth works, and can make cognitive errors.


The AI Governance Challenge

Research by consumer psychologists Craig McKenzie and Michael Liersch gives some useful insights about this issue. They showed that people are not able to accurately estimate the outcome of such non-linear processes. Instead, they believe that savings will grow linearly, and underestimate how much their current savings will be worth in the future. In one study, for example, participants estimated that saving $400 per month and getting 5% return annually would grow to $200,000 after 40 years, when in fact, it would grow to over $600,000. The authors’ conclusion is grim:

“People’s failure to recognize the power of compound interest—especially over long periods of time—leads to gross underestimation of future account balances, and by consequence, the cost of waiting to save. The unfortunate byproduct of this underestimation is that it has a negative impact on people’s motivation to save now. Since the benefit of compound interest is reaped by saving over long periods of time, this lack of motivation is particularly threatening to young people’s attainment of their retirement savings goals.”

These three factors, a lack of clarity about a specific goal, the constant competition with other shorter-term and more specific financial goals, and failing to value the power of compounding, all contribute to delay and lack of focus about retirement savings.