When Blockbuster Vídeo entered the Brazilian market, it seemed every store was located less than one block away from the then current market leader, Hobby Video. One hypothesis was that they bought the same market/demographic data. The other is that Blockbuster never even needed that data - all they had to do was to locate their stores based on Hobby Video's use of the demographic data.
In the two shops on the street example, Hotelling's explains that while it would better serve the customers to be each located one quarter distance from each end of the street, but that neither shop would risk letting the competitor relocate to capture more of the market.
Can this be expanded to mean that Hotelling's law allows for a third competitor to approach an adjacent pair and simply steal all of the business from one side of the street?
Hotelling's law assumes businesses are mobile, which is why its often characterized as a pushcart or ice cream truck. The business in the middle would want to move to the other side of its neighbor, and the process would repeat.
As such, there's no stable equilibrium in a three-firm Hotelling's law problem.
Simplest explanation: the first Starbucks had enough business to justify a store twice as large but didn't have room to expand. So they opened a second store across the street.
It also has some other benefits, where you can service traffic coming from different directions. When you drive on the highway and get off on an exit to get gas, most of the time you'll stop at the one on your side of the road so it's easier to get back on the highway. You also avoid having to make traffic modifications to deal with the queue of cars trying to make a left into the Starbucks.
I imagine so. I believe most (if not all) Starbucks are franchise stores. Franchise stores -- where the products are indeed identical and not just similar -- are the perfect-case scenario for this effect.
Why can't one of the pharmacies open up a second store on the other side of their competitor? Then instead of the service areas <-AB->, shop A could have <-ABA->.
Presidential candidates and choices in the "I am thinking of a number between one and ten" game can't expand to second locations, but businesses can. In addition, if the pharmacy can't afford an expansion, a competitor could start up and secure n-1 of B's traffic by being slightly rightmore.
I recently started reading a book called "The Joy of Game Theory: An Introduction to Strategic Thinking" which touches on explaining the grouping of similar shops (and politicians tending towards the centre!) as an effect of Game Theory. Very interesting read so far.
Isn't it the same with software? Office suites, map-apps, email-apps, Android (from different vendors), etc. try to more or less mimic the most successful one.
Entropy is the idea that all information will become uniformly distributed.
In a market space given enough competition all products become similar.
These are statements of the same fact. Just instead of the universe we look at a market space, time becomes competition, and feature set becomes entropy count.
I'm familiar with a busy intersection that has two Starbucks across from each other and after some investigation I found that they both served different markets - one for morning commuters (opened early, closed earlier) and the other for afternoon commuters with evening meetings (opened later, closed later). Traffic was so high if you tried to cross to the other side's Starbucks that everyone suffered. So even though it seems wasteful to have redundant use of resources, this is a form of load balancing (perhaps closer to sharing) in the real world that addresses one of the major difficulties of services being adequate distribution of it rather than the supply. The reduced hours for employees at each location are another matter but given the actual demand was about right for a fully staffed single Starbucks it doesn't mean we "lost" jobs with reduced work either.
To the contrary, this is a great example of how regulated markets with high barriers to entry can have adverse consequences, and why we need free markets.
In a free market, Hotelling's law does not hold because a third competitor can enter the market, position themselves slightly to one side of the duopoly, and claim almost half of the market. In order to prevent this, the incumbents will avoid placing themselves right next to each other -- and in fact, acting to minimize the threat posed by new market entrants produces the solution which is optimal for customers.
The situation described by Hotelling's law only arises in a market where the duopoly can be confident of remaining a duopoly.
EDIT: The deleted comment was arguing that Hotelling's law illustrates a failure mode of the free market and a consequential need for regulations to further the public good.
I've seen both good and bad societal outcomes from low barriers to entry. Similarly, I've seen both good and bad societal outcomes from unregulated free markets.
I would argue that there is a distribution of outcomes as well as a distribution of "degrees" of free market possible in real-world markets.
To me, it makes far more sense to look at the outcomes desired by our society and create policy to "force" those outcomes - regardless of the degree of philosophical advocacy for or against "low barriers to entry" or "unregulated free markets". It just makes more sense to focus on the desired outcomes for our society.
I'm not sure I buy this because even if a third competitor entered and claimed nearly half of the market, the people far out in the tail of that claimed half would not experience any improvement in service. To them, they would be nearly just as bad off. Their relative position wouldn't improve unless many, many entrants just kept repeating the exact same formula, which is highly speculative because the market could become saturated long before a chain of entrants is built up to such a degree as to effect the people in the tails in any meaningful way
It's also worth mentioning that in most real life scenarios, the risk of a cabal is incredibly high. Without regulation, what stops the first two market participants from using wealth created during their duopoly phase to "buy out" the third entrant, and then all three buy out the fourth, and so on? Each new entrant would have to value the discounted future stream of returns from their market share at something higher than the buyout, which again is really speculative.
I understand that regulations present their own problems: what stops the duopoly from purchasing regulatory capture and raising barriers to entry?
I'm just saying I don't think for a second that any quick and simple "but low barriers to market entry would solve this" thinking is even slightly convincing here. The Hotelling problem does really present a significant problem to natural equilibria that free markets arrive at, at least from a social welfare optimization point of view. It doesn't mean any given proposed regulation is the "right" answer, but it also doesn't do anything to build assurance that less regulation is acceptable either.
I'm not sure I buy this because even if a third competitor entered and claimed nearly half of the market, the people far out in the tail of that claimed half would not experience any improvement in service. To them, they would be nearly just as bad off.
You're missing my point. A rational duopoly which is concerned about the possibility of a third competitor entering the market will avoid satisfying Hotelling's law. It isn't the third competitor entering the market which improves things; it's the threat of a third competitor entering the market.
I understand that regulations present their own problems: what stops the duopoly from purchasing regulatory capture and raising barriers to entry?
That's a political problem, not an economic one. ;-)
I don't think I missed your point. I think I said that a rational duopoly won't bother worrying about what economical solution best prevents a third entrant, when there is always the possibility of simply paying from established wealth to compensate a would-be entrant for their willingness to not enter.
I don't see it as a political problem. It's such a ubiquitous part of any real market that I just see it was a way in which typical econ toy problems fail to correspond to reality and fail to produce useful predictions. It's like trying to use Newtonian physics when quantum effects really matter, except you've got so many economists employed because of the Newtonian-like toy models and they have a real need to keep hand waving and trying to convince everyone that the Newtonian approximations matter, that deriving theoretical results like this matter to "build intuition" or "create the mathematical framework" or whatever, but it's just clearly not true or useful to anyone but those researchers themselves.
My theory was that in order to get a bank loan one must demonstrate an existing proven business model and the math behind it. This means things get copied past the point they don't work anymore.
We had someone from a local bank come in to our High School "Career and Life Management" class. Half the class was assigned a script and had to go looking for loans. The other half was given fake money to open a bank and give out loans.
My group struggled to attract people while paying out reasonable interest rates. All the other banks were giving better offers. To actually make a deal, we ended up having to accept uncomfortably risky sounding customers at razor-thin margins.
When the time came, we looked on the back of the customer cards and discovered that thankfully they were not all as risky as we feared. We rolled the dice, but unfortunately rolled poorly and had a partial default by one of our borrowers.
We were a little unlucky. Without that default we'd have turned a profit, and most groups didn't have a single default. In the end, we suffered a moderate net loss.
However, not a single group made a profit. All our competitors were giving out loans at a loss after accounting for inflation.
The lesson was supposed to be about banking and interest. Instead, it was a lesson on business. I'd never realised quite how destructive idiots with money could be.
This is better explained as a Nash equilibrium. The example of shops on the street encourages similarly located shops only when there are two shops. If there are three or more shops each shop owner no longer wants to locate their shop in the center of the street. Instead, the final result is unstable (no Nash equilibrium). so the keep jockeying for the best location, thus speading out.
If you think of the distance any individual consumer needs to travel as a penalty/loss to that consumer, locating both shops right next to each other at the center clearly does not minimize the sum penalty to the set of consumers.
Moving each shop 1% of the distance away from the center will reduce by that amount the distance travelled by nearly every consumer. (Only those consumers located at or between the shops will be harmed, and everyone else will gain. As you continue to move the shops apart, this set becomes larger until you reach an equilibrium.)
Free markets are stochastic Nash-equilibrium-seeking optimization processes. By which metric do they "not work"? It's not as if governments are any better at reaching Pareto optima.
Geographic regulation is unnecessary, but the government may decide it is responsible for lowering barriers to entry -- supporting a healthy financial market and streamlined licensing/inspection. Competitors entering the market will destabilize the two-firm Hotelling equilibrium.
