Artificial Intelligence (AI) is transforming the way we live and work, with many of us knowingly or unknowingly using some form of AI daily. Businesses are also adopting AI in increasingly innovative ways. One example of this is the use of pricing algorithms, which use large datasets on market conditions to set prices.
While these tools can drive innovation and efficiency, they can also raise significant competition concerns. Subsequently, competition authorities around the world are dedicating efforts to understanding how businesses are using AI and, importantly, the potential risks its use may pose to competition.
How AI pricing tools can enhance competition
The use of AI pricing tools offers some clear potential efficiencies for firms, with the potential to reduce costs that can potentially translate into lower prices for consumers.
Take, for instance, industries with highly fluctuating demand, such as airlines or hotels. Algorithms can enable businesses to monitor demand and supply in real time and respond more quickly, which could help firms to respond more effectively to changing consumer preferences. Similarly, in industries which have extensive product ranges, like supermarkets, algorithms can significantly reduce costs and save resources that are usually required to manage pricing strategies across a large range of products.
Furthermore, as pricing algorithms can monitor competitors’ prices, firms can more quickly respond to their rivals. This could promote competition by helping prices to reach the competitive level more quickly, to the benefit of consumers.
How AI pricing tools can undermine competition
However, some of the very features that make algorithms effective can also facilitate anti-competitive behaviour that can harm consumers. In economic terms, collusion occurs when firms co-ordinate their actions to reduce competition, often leading to higher prices. This can happen both explicitly or implicitly. Explicit collusion, commonly referred to as illegal cartels, involves firms agreeing to co-ordinate their prices instead of competing. On the other hand, tacit collusion occurs when firms’ pricing strategies are aligned without a formal agreement.
The ability for these algorithms to monitor competitors’ prices and react to changes quickly could work to facilitate collusion, by learning to avoid price wars to maximise long-term profits. This could result in harm to consumers through sustained higher prices.
Furthermore, there may be additional risks if competitors use the same algorithmic software to set prices. This can facilitate the sharing of confidential information (such as pricing strategies) and, as the algorithms may be able to predict the response of their competitors, can facilitate co-ordination to achieve higher prices to the detriment of consumers.
This situation may resemble what is known as a ‘hub and spoke’ cartel, in which competing firms (the ‘spokes’) use the assistance of another firm at a different level of the supply chain (e.g. a buyer or supplier that acts as a ‘hub’) to help them co-ordinate their actions. In this case, a shared artificial pricing tool can act as the ‘hub’ to enable co-ordination amongst the firms, even without any direct communication between the firms.
In 2015 the CMA investigated a cartel involving two companies, Trod Limited and GB Eye Limited, which were selling posters and frames through Amazon (see linked CMA Press release below). These firms used pricing algorithms, similar to those described above, to monitor and adjust their prices, ensuring that neither undercut the other. In this case, there was also an explicit agreement between the two firms to carry out this strategy.
What does this mean for competition policy?
Detecting collusion has always been a significant challenge for the competition authorities, especially when no formal agreement exists between firms. The adoption of algorithmic pricing adds another layer of complexity to detection of cartels and could raise questions about accountability when algorithms inadvertently facilitate collusion.
In the posters and frames case, the CMA was able to act because one of the firms involved reported the cartel itself. Authorities like the CMA depend heavily on the firms involved to ‘whistle blow’ and report cartel involvement. They incentivise firms to do this through leniency policies that can offer firms reduced penalties or even complete immunity if they provide evidence and co-operate with the investigation. For example, GB eye reported the cartel to the CMA and therefore, under the CMA’s leniency policy, was not fined.
But it’s not all doom and gloom for competition authorities. Developments in Artificial Intelligence could also open doors to improved detection tools, which may have come a long way since the discussion in a blog on this topic several years ago. Competition Authorities around the world are working diligently to expand their understanding of AI and develop effective regulations for these rapidly evolving markets.
Articles
Questions
- In what types of markets might it be more likely that artificial intelligence can facilitate collusion?
- How could AI pricing tools impact the factors that make collusion more or less sustainable in a market?
- What can competition authorities do to prevent AI-assisted collusion taking place?
Policy makers have become increasingly concerned about what the US Federal Trade Commission (FTC) describe as ‘negative option marketing’. These are marketing deals that contain the following feature:
a term or condition that allows a seller to interpret a customer’s silence, or failure to take affirmative action, as an acceptance of an offer.
For example, companies such as Amazon, Apple, Spotify and Netflix may offer students a 3-month free trial or 3-month introductory offer (at a special lower price) for movie and music streaming services. However, many of these subscription contracts contain an example of negative option marketing – auto renewal clauses.
Problems with auto-renewal contracts
The inclusion of an auto-renewal clause means that if a customer fails to cancel the subscription at the end of the three-month period, the subscription automatically reverts to its full price. The full-price contract then continues to roll-over indefinitely unless the customer takes a pre-specified action to terminate the deal. Inattentive consumers could end up paying subscription prices that far exceed their willingness to pay.
Auto-renewal contracts are not just an issue with free trials/introductory offers. Some people may purchase subscription contracts at the full price and then forget about them. These consumers could end up paying fees for months after they have effectively stopped using the service.
Another potential problem with the use of auto-renewal contracts, is businesses deliberately making the cancellation process more complex than it needs to be. In many cases it takes just one click to sign up for the subscription, but multiple clicks through a series of menus to cancel. Some businesses do not provide consumers with the option to cancel online and, instead, they are forced to phone a number that is often very busy.
Effects on consumer welfare
To what extent do these problems caused by auto-renewal reduce consumer welfare? What evidence do we have?
Research by Citizens Advice found that just over one in four people (26 per cent) had signed up to a subscription by accident. 58 per cent of this group forgot to cancel a free trial, while 21 per cent did not realise that the free trial would automatically roll-over to a full-price subscription. This seems to be a particular issue for those on low incomes with 46 per cent of people on Universal Credit signing up to a subscription by accident.
Analysis by the Department for Business and Trade (DBT) has tried to estimate the value of these unwanted subscriptions. The study found that consumers spent £602 million on unwanted subscriptions where a free or reduced-price trial had been rolled over to the full price. The same study also found that £573 million was spent on subscriptions that people had forgotten about.
One in five people in the Citizens Advice study who tried to cancel a subscription found the process difficult. The DBT estimates that cancellation difficulties led to £382 million being spent on unwanted subscriptions.
UK Government response
In response to these findings, the government introduced the Digital Markets, Competition and Consumers Bill into Parliament in April 2023.
Provisions in the Bill seek to standardise the information that businesses must provide consumers before they sign up for subscription contracts. For example, in the future, firms will have to display prominently (a) any auto-renewal provisions, (b) whether the price increases after a specified period, (c) details about how consumers can terminate the contract and (d) cooling-off periods.
The Bill also stipulates that businesses will have to provide consumers with reminders when a free/reduced-price trial period is about to end and/or a subscription is about to renew automatically. They must also make it easy to exit contracts and remove any unnecessary steps.
The government initially considered an additional measure that would force businesses to provide consumers with the option to take out any subscription without auto-renewal.
Citizens Advice strongly supported this policy. They argued that not only should consumers be given the choice, but that auto-renewal should not be the default i.e. people would have to opt-in to auto-renewal subscriptions.
However, after the consultation process for the Bill, the government decided against introducing this additional measure. Businesses have also argued that the other elements of the policy are too prescriptive.
Articles
Government documents
Questions
- Outline some theories from behavioural economics that might help to explain why people sometimes end up with unwanted subscriptions.
- Discuss some of the potential benefits of auto-renewal subscriptions for both consumers and firms.
- Using behavioural economic theory, explain some of the potential disadvantages for businesses of using auto-renewal subscriptions.
- When businesses deliberately make the cancellation process more complex than it needs to be, it is referred to as an example of ‘sludge’. Explain the meaning of ‘sludge’ in more detail, referring to some different examples in your answer.
- What difference do you think it would make to the number of people signing up for auto-renewal subscriptions if you had to opt-in as opposed to opting out? Explain your answer.
- Another policy would be to force firms to cancel subscription contracts if there is evidence that consumers have not used the service for a long period of time. Discuss some of the advantages and disadvantages of this measure.
- Explain what are meant by ‘dark patterns’. How may the choice architecture on some sites actually hinder consumer choice?
When building supply and demand models, the assumption is usually made that both producers and consumers act in a ‘rational’ way to achieve the best possible outcomes. As far as producers are concerned, this would mean attempting to maximise profit. As far as consumers are concerned, it would mean attempting to achieve the highest satisfaction (utility) from their limited budget. This involves a cost–benefit calculation, where people weigh up the costs and benefits of allocating their money between different goods and services.
For consumers to act rationally, the following assumptions are made:
- Consumer choices are made independently. Their individual choices and preferences are not influenced by other people’s, nor do their choices and preferences impact on other people’s choices.
- The consumer’s preferences are consistent and fixed.
- Consumers have full information about the products available and alternatives to them.
- Given the information they have and the preferences they hold, consumers will then make an optimal choice.
Black Friday can be seen as a perfect occasion for consumers to get their hands on a bargain. It is an opportunity to fulfil a rational need, for example if you were needing to replace a household appliance but were waiting until there was a good deal before committing to a purchase.
The assumption that people act rationally has been at the forefront of economic theory for decades. However, this has been questioned by the rise in behavioural economics. Rather than assuming that all individuals are ‘rational maximisers’ and conduct a cost–benefit analysis for every decision, behavioural economists mix psychology with economics by focusing on the human. As humans, we do not always behave rationally but, instead, we act under bounded rationality.
As economic agents, we make different decisions depending on our emotional state that differ from the ‘rational choice’ assumption. We are also influenced by our social networks and often make choices that provide us with immediate gratification. Given this, Black Friday can also be viewed as a great opportunity to fall prey to irrational and emotional shopping behaviours.
Black Friday originated in the USA and is the day after Thanksgiving. During this annual shopping holiday, retailers typically offer steep discounts to kick off the holiday season. The Black Friday shopping phenomenon is less than a decade old in the UK but it’s now an established part of the pre-Christmas retail calendar. Between 2010 and 2013, Black Friday gradually built up momentum in the UK. In 2014, Black Friday became the peak pre-Christmas online sales day and many online retailers haven’t looked back.
Arguably, from a behavioural economist’s perspective, the big problem with Black Friday is that all the reasons consumers possibly have to partake can be largely illusory. Consumers are bombarded with the promise of one-off deals, large discounts, scarce products, and an opportunity to get their holiday shopping done all at once. However, on Black Friday, our rational decision-making faculties are tested, just as stores are trying their hardest to maximise consumers’ mistakes.
There are many ‘behavioural traps’ that consumers often fall into. The following two are most likely to occur on Black Friday:
- Scarcity and loss aversion. Shoppers may fear that they will miss out on the best sales deals available if they don’t buy it now. Retailers commonly spark consumers’ interest by highlighting limited stocks available for a limited time only, which raises the perceived value of these goods. This sense of scarcity can further trigger the need to buy now, increasing the ‘Fear of Missing Out’. Consumers therefore need to ask themselves if they are really missing out if they don’t buy it now? And is the discount worth spending the money today, or is there something else I should be spending it on or saving for?
- Sunk cost fallacy. Once consumers have started to invest, they often struggle to close out investments that prove unprofitable. On Black Friday, customers have already made the initial investment of getting up early, driving to the shops, finding parking and waiting in a queue, before they have purchased anything. Therefore, they will be inclined to buy more than they initially went for. It is important therefore to think about each purchase in isolation.
This year, however, there is also the added complication of the rising cost of living. Whilst this may deter some consumers from unnecessary, impulse purchases, some consumers are using Black Friday as an opportunity to stock up on expected future purchases, hedging against likely price rises over the coming months.
It is thought that more consumers will be looking for a combination of high quality but low price to make sure their purchases are affordable and can last for a long time. According to PwC, many consumers have closely monitored their favourite brands in anticipation that big-ticket electronics, more pricey winter wear or Christmas stocking fillers will be discounted. Consumers are also in search of bargains more than ever given rising inflation. This would suggest a shift in attitude, meaning consumers will be more aware of what they cannot afford rather than giving in to emotional temptation brought on by Black Friday.
Retailers are fully aware of the cognitive biases that surround Black Friday and take full advantage of them. ‘Cyber Monday’ follows right after Black Friday, giving retailers an extra opportunity for them to keep those ‘urgent’ or ‘unmissable’ sales going and increase their revenues.
Black Friday is one of the biggest shopping days of the year. However, the way retailers approach it is growing increasingly mixed. Stores such as Amazon, Argos, Currys and John Lewis have started offering Black Friday deals much earlier in the month, leading some to refer to the event as ‘Black November’. Other stores, such as M&S and Next, didn’t take part at all this year.
Ultimately, Consumers can use insights from behavioural economics to empower them to make more rational decisions in such circumstances: ones that better align with their individual budgets. Nevertheless, the Black Friday sales mania can trigger our deepest emotional and cognitive responses that lead to unnecessary spending.
Articles
Video
Questions
- Discuss what is meant by the term ‘rational consumer’. Is it a useful generalisation about the way consumers behave?
- Discuss what is meant by the term ‘rational producer’. Is it a useful generalisation about the way firms behave?
- What is cost–benefit analysis? What is the procedure used in conducting a cost–benefit analysis?
- In addition to scarcity and loss aversion and the sunk cost fallacy, are there any other reasons why consumers may not always act rationally?
- Are people likely to be more ‘rational’ about online Black Friday purchases than in-store ones? Explain.
On 15 March 2019, the ‘Organ Donation (Deemed Consent) Bill’ was passed into law. (See the blog, Organ donations – Changing the default option vs active choice.) The government has just announced that this will come into force on 20 May this year. Under the scheme, ‘adults in England will be considered potential donors unless they chose to opt out or are excluded. The act is known as Max and Keira’s law in honour of a boy who received a heart transplant and the girl who donated it.’
This change from an ‘opt-in’ to an ‘opt-out’ system follows a similar a move in Wales in 2015. Since then, Wales has seen a significant increase in potential donors, with the consent rate rising from 58% to 77%. A similar move in Scotland will come into force in the autumn of this year. The government expects there to be an additional 700 transplant operations per year available for transplant by 2023.
These moves from an opt-in to an opt-out system are consistent with ‘nudge theory’. This maintains that positive reinforcement or making a decision easy for people can persuade them to make a particular choice. They are ‘nudged’ into so doing.
Opting out and nudge theory
In the case of having to opt in to a scheme such as organ donation, people have to make the decision to take part. Many, as a result, do not, partly because they never seem to find the time to do so, even though they might quite like to. With the busy lives people lead, it’s too easy to think, ‘Yes, I’ll do that some time’, but never actually get round to doing it: i.e. they have present bias and hence behave in a time-inconsistent manner.
With an opt-out system, people are automatically signed up to the scheme, but can freely choose to opt out. In the case of the new organ donor schemes in the UK, it is/will be assumed that organs from people killed in an accident who had not opted out could be used for transplants. If you do not want your organs to be used, you have to notify that you are opting out.
It could be the same with charitable giving. Some firms add a small charitable contribution to the price of their products (e.g. airline tickets or utility bills), unless people opt out.
Similarly, under UK pension arrangements introduced from 2012, firms automatically deduct pension contributions from employees’ wages unless they opt out of the scheme. Opt-out pension schemes like this retain between 90 and 95 per cent of employees. Opt-in pension schemes, by contrast, have much lower participation rates of around 60 per cent, even though they are otherwise identical.
This type of ‘nudging’ can improve the welfare of those who make systematic mistakes (i.e. operate in a time-inconsistent manner), while imposing very limited harm on those who act in a time-consistent manner. If it is in the interests of someone to opt out of the scheme, they can easily do so. Policies such as these are an example of what behavioural economists call ‘soft paternalism’.
Articles
- New opt-out model for organ donation to come into force in May
Pulse, Helen Quinn (25/2/20)
- Max and Keira’s law: New ‘opt-out’ organ donor system to be introduced on 20 May, government plans
Sky News, Greg Heffer (25/2/20)
- Adults to be automatically enrolled as organ donors under new law
Independent, Peter Stubley (26/2/20)
- Max and Keira’s law: New ‘opt-out’ organ donor system that presumes all adults agree to donate when they die will be introduced in May
Mail Online, Ben Spencer (25/2/20)
- Opt-out organ donation law: Your questions answered
New Scientist, Clare Wilson (27/2/20)
- Should organ donors be paid? The heavy toll of US kidney shortage
BBC News, US and Canada, Henri Astier (18/2/20)
- What Spain can teach Scotland about organ donation
BBC News, Scotland, Susie Forrest (18/10/19)
- A little nudge goes a long way in increasing organ donor registrations
The Conversation, Nicole Robitaille (2/5/19)
- How nudge theory is ageing well
Financial Times, Julian Baggini (19/4/19)
- Richard Thaler: ‘If you want people to do something, make it easy’
Financial Times, Tim Harford (2/8/19)
- Why a nudge from the state beats a slap
The Observer, Richard Reeves (30/7/08)
Official Information
Questions
- Why do opt-out schemes have a higher take up than opt-in ones? Would this apply if people behaved in a time-consistent manner?
- What is present bias? How does it differ from simple impatience? Explain how present bias might help to explain the impact of changing the default option.
- What are the arguments for and against nudging people to make decisions that benefit them or are in the social interest?
- Give some example of nudges that are used in public policy or would be a good idea to use. Consider how effective they are likely to be. (You might refer to the work of the Behavioural Insights Team.)
- What are the possible drawbacks of presumed consent in organ donation?
- What are the arguments for and against paying live people to donate organs, such as a kidney?
- How might people be encouraged to behave in the right way during an epidemic, such as corona virus?
- To what extent was nudge theory used during the Brexit referendum campaign and in the two subsequent general elections?
Economists are often criticised for making inaccurate forecasts and for making false assumptions. Their analysis is frequently dismissed by politicians when it contradicts their own views.
But is this fair? Have economists responded to the realities of the global economy and to the behaviour of people, firms, institutions and government as they respond to economic circumstances? The answer is a qualified yes.
Behavioural economics is increasingly challenging the simple assumption that people are ‘rational’, in the sense that they maximise their self interest by weighing up the marginal costs and benefits of alternatives open to them. And macroeconomic models are evolving to take account of a range of drivers of global growth and the business cycle.
The linked article and podcast below look at the views of 2019 Nobel Prize-winning economist Esther Duflo. She has challenged some of the traditional assumptions of economics about the nature of rationality and what motivates people. But her work is still very much in the tradition of economists. She examines evidence and sees how people respond to incentives and then derives policy implications from the analysis.
Take the case of the mobility of labour. She examines why people who lose their jobs may not always move to a new one if it’s in a different town. Partly this is for financial reasons – moving is costly and housing may be more expensive where the new job is located. Partly, however, it is for reasons of identity. Many people are attached to where they currently live. They may be reluctant to leave family and friends and familiar surroundings and hope that a new job will turn up – even if it means a cut in wages. This is not irrational; it just means that people are driven by more than simply wages.
Duflo is doing what economists typically do – examining behaviour in the light of evidence. In her case, she is revisiting the concept of rationality to take account of evidence on what motivates people and the way they behave.
In the light of workers’ motivation, she considers the implications for the gains from trade. Is free trade policy necessarily desirable if people lose their jobs because of cheap imports from China and other developing countries where labour costs are low?
The answer is not a clear yes or no, as import-competing industries are only part of the story. If protectionist policies are pursued, other countries may retaliate with protectionist policies themselves. In such cases, people working in the export sector may lose their jobs.
She also looks at how people may respond to a rise or cut in tax rates. Again the answer is not clear cut and an examination of empirical evidence is necessary to devise appropriate policy. Not only is there an income and substitution effect from tax changes, but people are motivated to work by factors other than take-home pay. Likewise, firms are encouraged to invest by factors other than the simple post-tax profitability of investment.
Podcast
Article
Questions
- In traditional ‘neoclassical’ economics, what is meant by ‘rationality’ in terms of (a) consumer behaviour; (b) producer behaviour?
- How might the concept of rationality be expanded to take into account a whole range of factors other than the direct costs and benefits of a decision?
- What is meant by bounded rationality?
- What would be the effect on workers’ willingness to work more or fewer hours as a result of a cut in the marginal income tax rate if (a) the income effect was greater than the substitution effect; (b) the substitution effect was greater than the income effect? Would your answers to (a) and (b) be the opposite in the case of a rise in the marginal income tax rate?
- Give some arguments that you consider to be legitimate for imposing controls on imports in (a) the short run; (b) the long run. How might you counter these arguments from a free-trade perspective?
