Using AI and data-driven tools, companies can change the price of a good or service based on who is buying, when they’re shopping, and myriad other factors. This power raises a question: Should ethics and societal values factor into these prices — and how? Companies should consider what they’re selling, who they’re selling to, and how they’re selling their products, and consider the range of possible impacts. Then, they should 1) get creative about reducing negative externalities, 2) reach out to stakeholders, and 3) build proactive filters into their pricing deliberations.
More than ever, companies are able to tailor prices across people, places, and time. They do this to maximize profit, and sometimes simply to survive. We’re in a new era of supercharged price discrimination, made possible by two major scientific and technological trends. First, AI algorithms — often trained on highly detailed behavioral data — enable organizations to infer what people are willing to pay with unprecedented precision. Second, recent developments in behavioral science — often invoked with the tagline “nudge” — provide organizations greater ability to influence their customers’ behaviors.
Not long ago, such tactics might have been taken in stride. But no longer. Today’s overlapping crises and changing attitudes highlight a truth that’s often been hidden in plain sight: Such supercharged pricing strategies can cause real harm to individuals, organizations, and societies. When companies deploy such strategies without considering their potential societal impacts, they risk harming people, inviting customer outcry and public outrage, and even reduced prospects for long-term survival.
Given all of this, organizations must consider how to use their supercharged pricing capabilities in ways that reflect important societal values and avoid causing harm and amplifying inequities. Going further, farsighted organizations would do well to seek profitable data-driven pricing strategies that not only avoid harm but also help advance societal wellbeing.
To do this, companies need to embed societal considerations and ethical deliberations in their pricing calculus. The fields of medicine, data science, and AI have undergone similar evolutions, embracing the core ethical concepts of avoiding harm, promoting wellbeing and fairness, and not manipulating people’s behaviors and decisions. In a similar spirit, we offer a framework comprising three easy-to-ask, easy-to-answer, and easy-to-apply questions that together reduce the chances that your pricing strategy will have negative social impacts. Think of each of these three questions as layers of a “filter” that can reduce potential harms to your customers, your company, and society.
Three Questions Companies Should Ask
What am I selling — and can these prices impede access to essential products?
Think about your products and services. Are they essentials, such as food, shelter, medicine, transportation, and internet access? During the pandemic, high prices blocked access to personal safety equipment like N95 masks and hand sanitizer, as well as to must-have products such as toilet paper. In each case, not being able to get these items exposed people to risk or harm.
Of course this question remains relevant beyond times of crisis. High pricing by pharma companies has historically blocked access to drugs for chronically ill patients — harming people they could help and making these companies targets of social outrage. Similarly, AI algorithms enable surge pricing for taxi rides. Surge pricing can be a useful tool – for example it can both increase the availability of transportation and enable drivers to earn more. But context is important. For example, significant price increases during emergencies such as severe snowstorms or hurricanes can prevent vulnerable consumers from moving to safety.
Consider the recent cold wave in Texas, where people suffered ruinous spikes in their utility bills due to a pricing scheme and algorithm that didn’t anticipate the harms that could result from a single-minded focus on economic efficiency. As one customer lamented, “It’s a utility — it’s something that you need to live. I don’t feel like I’ve used $6,700 of electricity in the last decade. That’s not a cost that any reasonable person would have to pay for five days of intermittent electric service being used at the bare minimum.” This is a sobering illustration of the social consequences of pricing that is narrowly framed in a way that crowds out such ethical considerations.
When your offerings become necessities, beware: Higher prices can block access to essential needs, magnifying the possibility of societal harms. The more essential the need and the larger the price increase, the greater the potential harm.
Who am I selling to — and can these prices harm vulnerable populations?
Are some of your customers members of vulnerable populations? Are you selling to the elderly, to customers with chronic medical conditions, or to customers living on limited incomes? Are you selling to customers in social groups that have traditionally been discriminated against? Analytics, big data, and AI technologies supercharge companies’ ability to address these questions in granular detail.
Price discrimination that reinforces societal discrimination is unacceptable. Insurance companies have long confronted this issue: It’s a business necessity to set rates using ever more sophisticated applications of actuarial science and AI, but balancing actuarial fairness with societal fairness requires ethical deliberation that can’t be reduced to a formula. For example, some insurers have incorporated such information as policyholder profession into their pricing algorithms, leading to lower insurance rates for policyholders in elite professions and higher rates for policyholders in less well-paid professions. This in turn has led to negative media coverage, public outrage, lawsuits, legislative investigations, and government regulatory changes — putting the brands, reputations and trust in these companies at risk.
Price discrimination is a common business practice. But when used in ways that reinforce social discrimination, the term takes on a new meaning to those discriminated against and the communities they dwell in. The more vulnerable the population and the higher the degree of discrimination experienced, the more damaging the consequences of ill-conceived pricing practices.
How am I selling — and can these prices manipulate or take advantage of customers?
Are you using information or behavioral data in ways that advantage your company but disadvantage your customers? Are you withholding or complexifying price information that could help customers make better decisions? Are you taking advantage of their bounded options, information, or rationality? Or are you designing pricing strategies in ways that empower customers to make better decisions?
A few examples:
- Disney, early in the pandemic when its parks were closed, continued to collect automatic monthly payments from pass-holders. Customers, as you can imagine, were less than thrilled.
- Long before the pandemic, many small third-party sellers alleged that Amazon had used its data and analytics to directly compete with these small sellers through lower prices by buying in bulk from suppliers at lower costs.
- Uber, rather than applying its analytics, AI, and behavioral insights to help drivers to make more informed decisions, instead used gamification and behavioral nudges to prompt them to travel longer distances to locations that were less profitable for them.
- McKinsey recently apologized and settled a lawsuit for its role in helping Purdue Pharma “turbocharge” opioid sales.
When your pricing strategies take advantage of information asymmetries, customers’ personal data, and challenging situations, the risk of societal harms becomes magnified. Your pricing risks becoming a barrier to individual choice and personal autonomy.
Each of these situations illustrates how pricing strategies that do not take these questions into account can negatively impact society and generate significant negative consequences for companies, customers and communities.
If the answer to each of the above three questions is “no” then you can feel more confident in proceeding with your pricing plans, given the steps you’ve taken to ensure that your pricing will be better for customers, for your company and its reputation and brand, and also for society.
But if you answered “yes” to any of the above, how should you proceed?
How to Price More Ethically
1) Pause, step back, and get creative.
Consider the tradeoffs at stake and the potential courses of action. Many companies have found ways to maintain profitability while limiting negative social consequences.
For example, the Norwegian retail chain Meny was in a bind with its hand sanitizer pricing. Caught between offering accessible prices and buyer hoarding at those prices, many retailers have turned to imposing restrictions on the number of bottles customers can buy — creating hassles for employees and customers. Meny priced its hand sanitizer at normal price for one bottle, $100 for each additional bottle. This clever strategy simultaneously maintains accessibility and reduces hoarding.
Hyundai provides another example of creativity. During the 2008 Great Recession, Hyundai focused on a key vulnerability for working-class customers: the risk of losing one’s job. Rather than lowering prices across the board, Hyundai creatively offered customers a car-return guarantee in the event that they lost their job. This worked — customers bought Hyundai cars in droves, making Hyundai the only car company to be successful during the Great Recession. In the midst of today’s pandemic, automakers have taken this lesson to heart, and are all pricing their cars more creatively.
AI can be harnessed as a force for good along similar lines. For example, insurers can use telemetry data and usage-based pricing to both economically incent and “nudge” safer driving behaviors. Most drivers believe themselves to be better than average — a dangerous example of overconfidence bias. Data-rich feedback reports comparing drivers to their peers would likely prompt safer driving. Such interventions have the potential to simultaneously make insurance more affordable by reducing claim frequency, while making the roads safer.
2) Be willing to compromise and communicate this with stakeholders.
If no such creative pricing options are available, then explore ways to compromise. You may have to give up some profitability to avoid or ameliorate negative social consequences. At the same time, customers may need to endure some negative consequences to enable the company’s financial survival. Recall the consumer outrage resulting from Uber’s use of surge pricing. Uber could not altogether eliminate surge pricing, given that incentivizing drivers is a key part of its business model. So, it worked with regulators to cap its prices during emergencies and disasters at a level below the three highest-priced days in the previous two months.
In addition to engaging in creativity and compromise, it is also necessary to clearly communicate to key stakeholders. Uber also added functionality to its app to let consumers know ahead of time how much extra they would pay and get their consent, and they further announced that the commissions earned during the crises would be donated to the American Red Cross – thus tapping into customers’ intrinsic pro-social motivations. This strategy reflects an insight from behavioral economics: Customers can be willing to give up economic benefits in order to promote fairer social outcomes. Uber’s revised surge pricing strategy and communications strikes a balance between achieving corporate goals and warding off bad outcomes.
3) Incorporate this filter into your pricing deliberations.
The actions of many prominent organizations suggest that integrating ethical deliberation into pricing strategy is an idea whose time has come. As mentioned above Uber now proactively incorporates the capacity for societal deliberation into its surge pricing activities. This enables local flexibility to allow a quicker trigger finger for eliminating surge pricing entirely in emergency situations. Microsoft provides an analog outside the pricing domain: A recently formed ethical development team reports to President Brad Smith. The team is responsible for guiding technical and experience innovation towards ethical, responsible, and sustainable outcomes.
Furthermore, incorporating this filter into your pricing deliberations can spur innovations in harnessing big data, AI, and analytics that are both pro-social and good for the company’s long-term competitive position. Customers are understandably wary that companies might use their data in ways that benefit shareholders rather than the individuals who generated the data in the first place. So it behooves companies to explore how to use big data and pricing in ways that address such concerns.
For example, grocers can apply analytics to granular club card data in ways that inform, reward, or incentivize healthier food purchasing options. Doing so can increase access to essential needs, of particular value to vulnerable populations living in food deserts. Analogously, insurers can use policyholder data not only to build actuarial models, but also to use pricing to incent safer risk behaviors and risk management on the part of policyholders. Online retail platforms can use their data to inform customers when prices for essential goods are excessive. These are all examples of using data and AI in ways that help rather than exploit customers. Such pricing strategies are at once pro-social – in each case the customers’ data is harnessed in a way that helps the customer better achieve his or her own goals – and pro-business in that they enable brand differentiation and increased customer loyalty.
Refining pricing capabilities in ways that keep ethical considerations front and center, exercising creativity and judicious compromise, and communicating with forthrightness and transparency enable organizations to pursue the profit motive in ways that help — rather than harm — the populations they depend on and serve.