Collusion, retail petrol prices, and market activism

In New Zealand there has been an email going around suggesting that everyone should stop purchasing petrol from selected retailers (such as BP) in order to start a price war.

Now I accept that this doesn’t make much sense in the face of no collusion – which currently appears to be what is going on (h.t. No PC). However, if we did have collusion, how would this scheme work?

For collusion to work we need some way for the greater industry profits to be shared among the market participants. A common assumption in economics is that we have firms selling homogeneous products (which seems fair for looking at retail petrol) and as a result if all firms set the same price they will share the profit equally. In this case, given the firms rate of time preference, they may be able to sustain a collusive agreement.

Now the emailers strategy works in the following way. He hopes that as long as the retail petrol price is above a certain “fair rate” he will be able to organise people so that they consume in the following way. Firstly, they look at the price – any infinitesimal difference in price between retailers will lead to all the consumers shifting to buy off them. Secondly, if prices are all the same then they randomly shop at any petrol station – expect one (say BP).

In this case, a collusive equilibrium is not possible above the “fair price”. The reason for this is that if everyone set the same price, BP would want to defect and cut prices (as they are getting no clients). Furthermore, if BP cuts prices the others can’t continue to collude – as all the customers have moved to BP.  As a result, they will want to cut prices in order to get customers back. Given this, the consumer activism in this case will lead to prices falling to their “fair rate”.

Now this was a gross oversimplification, factors such as distance, brand loyalty (schemes) and service will also influence peoples shopping decisions. However, with petrol being such a homogeneous good, there is a definite possibility that “market activism” such as this could effectively change the outcome in the industry.

Furthermore, there is nothing wrong with such activism. In this case consumers are using their market power to extract surplus in the face of an oligopoly that is using its market power to extract surplus – sound fair to me 🙂

  • Interesting post matt.

    The idea of countervailing consumer power is an interesting one. If petrol companies are colluding then power to the people!

    I’m not sure it’ll work though, mainly for some of the reasons you mentioned. I think the main reason it won’t happen is because every individual consumer is likely to believe that whether or not they participate it won’t effect the final outcome (it’ll take allot of people to stop purchasing from a company for them to notice) . Thus with a small subjective probability that their decision will influence the outcome, it’ll likely only take a very small cost (having to drive further, not going to your usual petrol station etc..) to dissuade people from doing it.

    This kind of scheme also requires a lot of people to know about it, even though I don’t drive and probably would be too lazy to boycott BP I didn’t get the email…..

  • “I think the main reason it won’t happen is because every individual consumer is likely to believe that whether or not they participate it won’t effect the final outcome”

    Very true agnitio. That is why I used the whole “central planner” type language, specifically:

    “He hopes that as long as the retail petrol price is above a certain “fair rate” he will be able to organise people so that they consume in the following way”

    One way I could justify this scheme is if the cost of changing your behaviour and the expected benefit were very very low. In this case the individual is to follow the direction of the central planner – so why not do it? Furthermore, if there is a transaction cost to deciding on where to go and buy petrol, just going where this guy tells you too will be an optimal strategy 😉

    In this case the choice of petrol company goes from being random to pre-determined, which would be enough to make the scheme work.

  • I think my point was that there would be a transaction cost from going to a different petrol station. Not being a driver I’m probably making a raging generalization, but most people probably have usual petrol station they goto.

    It’s quite interesting when you think about it. If I have a planned trip to get petrol I probably have a usual station I go to e.g. there may be one that is really convenient to goto on the way home. On the other hand people run low on petrol in random places and thus will have to find a petrol station. The question then is, will people think “i’ll go anywhere but a BP”, or will they go to the first petrol station they come across.

    I think what I’m getting at is that for planned trips the place is actually likely to be pre determined and when the choice is random I wonder whether people will care enough to not just go to the first station they find.
    As I alluded to, if there is collusion I’m all for finding ways to break it up, I just have my doubts whether people power can achieve it in this case. There is also a credibility issue in who the “central planner” is and their reasons for choosing the company they single out.

  • “I wonder whether people will care enough to not just go to the first station they find”

    That is the fundamental thing with social activism – people have to receive a sufficient benefit from the belief they are doing the “right thing” for them such a system to function.

    Believe me, I don’t think that this scheme would actually – as you say I doubt people care enough. As a result, I think we agree here.

    However, the main reason I wanted to write the post is to show that the guy wasn’t an idiot for suggesting it (which some other blogs have said) – if he believed there was a sufficient feeling of fairness in the community for people to join him in the protest.

  • Agreed, if there were less lazy people like me in the world things like this would gain traction! Damn transation costs

  • “Damn transation costs”

    Don’t forget the benefit side – you don’t get enough satisfaction out of helping your society, if you did you would be able to solve all sorts of prisoners dilemmas 😉

  • insider

    surely the more likely impact will be to drive up prices?

    if everyone goes to BP demand increases but supply, or their capacity to manage supply, may not, assuming they have structured their business to efficiently service their expected market share. If market share drastically and suddenly increases, I suspect most businesses will find it difficult to cope.

    If so, prices should rise not fall at BP to maintain continuity of supply (you don;t make any money if your site is empty). If they do rise the incentive for other companies to lower theirs then disappears as they already have a price differential. They may even be concerned about a similar run on product and so increase prices to match, or they may see the opportunity to bolster margin.

  • Hi Insider,

    Your argument misses the fact about why the demand for BP petrol goes up in this model. Everyone only goes to BP if they charge a lower price than the other fuel companies – if they are capacity constrained then they can only serve a portion of the market, but they only receive that market share if they are charging a lower price then their competitors in the first place.

    As long as BP can serve a large proportion of the market (as fuel retailers tend to have large amounts of excess capacity, given that they are a peak load business, I think this is a reasonable assumption) this price competition would force a response from the remaining “cartel” to lower prices to the same level as BP.

    However, your capacity constraint argument is true, and extremely useful for describing the phenomenon. If the capacity constraint is strong enough you could end up with a situation where BP permanently offers a lower price and the rest of the fuel retailers agree to charge higher prices on the fringe of the market – this form of collusion could keep prices elevated and ultimately benefit BP. However, BP would also have to commit to not increasing capacity in this environment – something that it would be difficult to commit to (without getting into trouble for colluding 🙂 )

  • insider

    Thanks sorry I missed that. This email usually takes the form of “let’s all go to BP” with no initial price advantage incentivising the move – just a reef fish reaction.

    It would be the loss of custom that would drive down prices at the other retailers, who would want to recapture market share.

  • insider

    Also I;m not sure they are peak load businesses. If so, you wouldn;t have queues at pumps which can be a regular occurance at popular sites. I think they configure around their average demand – certainly on the supply side, otherwise they would have very expensive trucks and drivers waiting around a lot of the time.

  • “Also I;m not sure they are peak load businesses. If so, you wouldn;t have queues at pumps which can be a regular occurance at popular sites”

    Fair point. Actually I do believe they are a peak load business – but the fact that the peak is the same for all firm’s indicates that the capacity constraint may be strongly binding when it matters (I hadn’t thought of that!).

    As a result, I would need to assume that a significant amount of revenue generation also occurs outside of peak time in order to make my argument work – that is definitely a binding assumption!

  • insider

    I’m not an economist so you are beginning to lose me….

  • Sorry.

    The more capacity constrained business’s are then the less likely it is that market activism can lead to a full price war – as collusion could be maintained as long as the other fuel companies can keep enough customers.

    I was thinking that since fuel companies are peak load companies they should have spare capacity a lot of the time – and as a result this sort of activism could work.

    However, as you said the “peak” occurs during peak time every day – and so the spare capacity only exists at non-peak times and is infact a sufficiently different product.

    Since the peak occurs so often, and as there is a “transaction cost” associated with buying fuel outside peak time, it is likely that “peak time” accounts for a lot of the fuel retailers revenue and that the constraint that exists in “peak time” provides a binding constraint on the amount of the total retail fuel market that BP can take.

    In other words – BP doesn’t have the ability to serve a lot of customers during peak hours, and people will be willing to pay a bit more to get their fuel during peak hours, and so this market activism will not lead to a collapse in prices.

    In fact it may worsen the situation by creating price discrimination between people who value peak service and those willing to wait for non-peak cheaper fuel.

    Hopefully that cleared up what I meant – your capacity constraint point was actually really important!

  • insider

    Thanks and no need to apologiese for knowing more than me.

    The issue I have with your explanation is that you don’t see peak-non peak pricing differentials. YOu could argue that is a market imperfection this doesn’t happen as you would expect to see discounting to attract more customers and maximise the efficiency of your business. Take a look at what happens in Australia with their weekly price cycles. The ACCC recently said they could find no logical explanation of why the cycle exists, despite having a hard look. My explanation is, it’s an entrenched habit.

    Take your point on constraints. They are constrained in dispensing by size, pump capacity and locations, and on supply by the number of trucks and drivers and volume per load. Any shifts in market share are likely to be incremental. But they have occurred. There has been a major shift in the last 10 years with Mobil going from first to last and Shell moving up strongly into the lead thanks in large part to flybuys, and BP gaining with strong promotions.

  • “Thanks and no need to apologiese for knowing more than me.”

    I don’t think thats the case – I think it has more to do with me being unclear with my reasoning 😛

    “The issue I have with your explanation is that you don’t see peak-non peak pricing differentials.”

    Good point. I don’t understand why we don’t see peak and off-peak pricing differences – maybe the first firm to introduce such a scheme would surrender a whole lot of consumer “goodwill” (as consumers may feel that changing the price over the day is “unfair”), and other firms would be able to follow costlessly.

    This implies that there is a second mover advantage to flexible daily pricing. So even though it would be better for each firm to do it, each individual firm does not want to be the first one to do it.