Advertisements follow you everywhere. From social media to online shopping, we see more and more websites using online advertisements to generate revenue from businesses who are trying to reach the ever-growing online audience. They can track your history to great lengths to provide you with 'what you want to see'. Social networking giants tapped into the profit-churning machine long ago, an example being Facebook in 2012. It obtained a total revenue of over $12 billion in 2015, most of which was derived from advertising. More recently, Snapchat introduced adverts - now almost half its revenue is derived from it. People have long found pop-ups and adverts a pain, yet more companies (such as Reddit) are following suit in implementing them. Therefore, I have come to ask a question: is advertising beneficial for all online websites?
I realised the question can be answered using game theory. If you are not yet familiar with the concept of game theory, don’t worry – A Rational Econ is working on a guide to it. For now, just see if you can follow. The payoff matrix below is used to illustrate the options of two large players in the social media industry - Firm A (player 1) and Firm B (player 2). Each box shows the revenue players get when they choose a particular strategy, with the first number (top right) indicating the payoff to player 1 and the second number (bottom left) the payoff player 2 would receive. Let’s assume that for these two businesses advertising increases revenue; therefore, when both players advertise, the total payoff is higher than when only one or neither advertise. Although adverts go against consumer’s best interest, there is high demand for it on social media from businesses as they have the ability to reach a massive audience. Thus, as soon as either form of social media starts using adverts, businesses will pay money to promote on the platform.
Initially, Firm A had a larger user base than Firm B and so had a revenue which was greater. Firm A may have assumed demand to be inelastic due to a large loyal user base and decided to advertise in a bid to increase revenue. Their gamble worked and revenue shoot up from 5 to 20. However, in the process some of the newer users may have disliked these adverts and instead used Firm B’s website, boosting its revenue to 6. If we jump forward in time, Firm B also has an established user base and it wants to maximise revenue. Firm B now believes the strategy of advertising to be safe and lucrative; thus, Firm B also follows the dominant strategy which is to advertise and introduces adverts. Therefore, in the social media industry when both of these firms act rationally using their dominant strategies, we reach a Nash Equilibrium where they advertise. In this scenario, we assume the industry to be a duopoly, however the conclusion only strengthens as we add more firms. If one firm starts advertising, they expand their own income. In order to increase revenue too, all other firms will likely follow. Over time, all major players in the social media industry will have introduced some form of advertising and followed what game theory suggests would happen.
The concept can be applied to other industries where the game consists of players who are large firms. They will likely have inelastic demand for their good (especially if it is established and has minimal substitutes), meaning most customers won’t be put off by adverts. This is the reason behind most online websites choosing to provide adverts.
However, when looking at smaller online businesses there is a different outcome. To illustrate my point let’s take two small online grocery stores: Shop A and Shop B. Most businesses use AdSense in order to put adverts on their website but this is where the first problem arises. AdSense scans the current page and looks for key words, displaying advertisements related to what the consumer is looking for. If the adverts display products of Shop B then it may drive customers from Shop A to Shop B, leading to a reduction in sales (from 5 to 3) and vice versa. However, if both businesses use adverts, there would be advertising revenue and potential increased spending as consumers may buy products from both stores instead of just one. Therefore, the payoff table in this scenario looks different to before:
For simplicity, I am going to assume that the online grocery market is a duopoly. To get from the current Nash Equilibrium (5,5) to the optimal position (10,10), both players would have to abandon their dominant strategy of not advertising (it has lower risk of decreased revenue). If the firms cooperated then yes, both grocery stores would advertise and achieve higher revenues. However, the other company doesn’t know for sure that advertising will increase revenue for both firms with no communication and may rather stick with a revenue of 7 until the other shop removes adverts again. Therefore, without cooperation or contact many small firms may not advertise even though it seems to be in their best interest.
It’s true wherever you go – a typical profit-seeking firm will advertise. It's almost always in their best interests. The point about AdSense clocking similar products I made earlier – there are ways of preventing this from happening through blocking ad content. Sometimes this doesn’t work, however. It is always important to consider that many assumptions are made in these models, and specific scenarios may “prove to disprove” the game theory behind advertising.
Written by Harshith Chakka
Edited by Omkar Dixit