Tuesday, December 1, 2009

Amazon Vs Wal-Mart - Negative Sum games!

Are you following the ongoing price war (started with books, spread into DVDs & now into game consoles etc.,) between Amazon & Wal-mart? Who do you think would the winner be? Ok, before even getting to that question, let’s analyze this situation from the game theory perspective.

It all started, when Wal-Mart decided to mark down the prices of ten best-seller books to ten bucks. When Amazon, predictably, matched that price, Wal-Mart went to $9, and, when Amazon matched again, Wal-Mart went to $8.99, at which point Amazon stopped responding, but another major retailer Target matched the Wal-mart’s price leading Wal-Mart to drop the prices further down to $8.98. For now, price war in books apparently stopped at that point, but spread into other products like DVDs & Game consoles.

Now, let’s take a look at the cost structure of the book industry and analyze the impact of this price war. Book prices are traditionally around fifty per cent off the cover price, and these books are now marked down sixty per cent or more, Amazon and Wal-Mart are surely losing money every time they sell one of the discounted titles. So, the more they sell of these titles, the less they are making. (Does n’t that remind you of the Viesta case results, where the busiest person with lot of jobs in hand actually loses money?).

Given this fact, why do companies engage themselves into price wars, in the first place? The hope is that if you cut prices enough you can increase your market share, and even your profits. But this works only if your competitors won’t, or can’t, follow suit. More likely, they’ll cut prices, too, and you’ll end up selling the same share, only at a lower price.

From a game-theory perspective, price wars are usually negative-sum games: everyone loses (Of course, when companies lose, customers win!). A recent study found that, if competitors do match price cuts, industry profits can get cut almost in half. The best way to win a price war, then, is not to play in the first place. Instead, companies should actually focus on their competitive advantage (customer service, quality, coolness etc.,). Companies can also “signal” competitors that they are actually for “stable pricing”, and not interested in starting a price war. The idea is to let your competitors know that you’re not eager to slash prices—but that, if a price war does start, you’ll fight to the bitter end.

If with a little common sense and basic economics, we can demonstrate that Amazon Vs Wal-mart price war is such a bad step, why did not Wal-Mart’s strategists stop it from initiating the price war? And why did not Amazon resist from getting itself into this war? Were they just reckless? Or did this turn into an “ego” issue?

NO, apparently not!! By starting this price war, Wal-Mart is actually making a statement that it’s a player in the online world. From Wal-mart’s point of view, the real goal of this conflict isn’t to lure readers away from Amazon, and it isn’t to get people to buy one of “those ten books”. It’s to lure them online ( to their walmart.com), away from big booksellers and other retailers, and then sell them other stuff. Usually, price wars wreak havoc because they erode the pricing power of an entire business. But, because this price war involves just ten items (that too low priced books), its impact on revenue will be small, and outweighed by the positive effects of all the publicity. Hence this price war may not hurt both Amazon & Wal-mart as bad as it looks on the surface, as it potentially gets them more revenue in other areas and also gives ample publicity!! However, the remaining small players in the book industry are going to be the real losers!! (Squeezed between the behemoths).

Prohibition of Insider Trading – Fair? Really?

You may have been following the recent Insider Trading scandal(s) & the ongoing debates about whether it’s time to legalize the Insider trading. I don’t have any plans to get into that debate, but there is one thing about insider trading that really caught my attention. Is Prohibition of Insider trading really fair?

In layman terms, prohibition of insider trading means – If you are an individual with potential access to non-public information about the company, you are prohibited from buying/selling stocks of that company based on that material non-public information. (NOTE: For simplicity, I am leaving out the type of insider trading that is legal).

Now, I perfectly understand why there are such restrictions on insider trading. But I still think it is UNFAIR. Here is why – Say, I am an insider of a company with significant number of stocks. Based on the material non-public information I have got access to, I can theoretically do either of 3 actions - BUY/SELL/HOLD-from-Sell (i.e., I've otherwise decided to sell, but changed the plans in the light of insider information)


First two (Buy/Sell) options need no further explanation but the third one (HOLD-from-Sell) is a tricky one!

To elaborate - Say, if a person has almost decided to sell his stocks and then suddenly came to know certain very confidential & critical information about the company that, in near future, would contribute to an increase the stock price significantly. He/She can then put plans of selling stock on HOLD based on this insider information & benefit significantly! None of the rules/laws that are present can catch this person in this instance!! It is not even possible to implement any rules/laws to catch this scenario as it is impossible to guess one’s “intentions to sell”.

As a result, we are ending up with a system – where two out of three scenarios are traced and punished, whereas, the third one, with the same implications, is just left free! Does n’t it make the system UNFAIR? I think, it does.

Why corruption is such a tough issue to deal with? – Taking Game Theory Perspective

Corruption is one of the most difficult problems to deal with, for majority of the developing countries. Many non-profit organizations are trying sincerely to bring awareness among citizens of these countries about the magnitude of the problem and possible solutions, but nothing seems to be actually working.

Having seen this problem of corruption first hand, I always wonder why it turned out to be such a hard thing to crack! In fact, this issue has so many faces to it - social, economic, cultural, political etc., But let me highlight one aspect that stands out from the perspective of our subject of interest - Game Theory!

To better demonstrate my point, let me take an example of a person "A" who wants to start a new business in one of these countries, where corruption is prevalent. Now, "A" has all the credentials & financials to make his/her case successfully. However, in this corrupted country, things won't work unless you bribe the government authorities, no matter how good your credentials are!

That leaves "A" with 2 options:

- Don't Bribe & Try to win the license only with his/her own credentials?

- Bribe the authorities (like rest of the competitors) and start your business!!

Payoff table looks something like this for this case – Assume the value of the project at stake is worth $100. Both players A & B are having equal chances of winning the project with NO CORRUPTION in picture. So, the value for both no-bribes is $50 each. If both payers pay a bribe of $10 each, they are left with a 40-40 (in bribe-bribe scenario). However, in the case of only one player paying the bribe, the probability of winning increases by 4 folds & hence the winner has a payoff of $70 ($80 – bribe of $10) and the other player has a payout of just $20.

Expected Profit

Player B

Bribe

No Bribe

Player A

Bribe

40,40

70,20

No Bribe

20,70

50,50

From the above table, it is a classic prisoner’s dilemma! Both the payers (A & Competitors) have dominant strategy – “Bribe”. Player “A” & the player “B” (Competitors) can actually coordinate and decide not to pay the bribe, there by improve their expected profits. But both the players have mutual doubts on other’s intention to defect.

This dilemma, I think, is one of the significant influencers that encourage people with otherwise honest (Anti-bribers) intentions to pay the bribes to get things done.

Does "Game Theory" offer a solution to eradicate corruption? Answer is NO & Yes. No, it does not offer any solution to this problem in specific. Yes, it does suggest way to get players out of this dilemma.

Alter the payoff structure so that “defection” is strictly punished with higher penalties and thereby, encourage players to cooperate! Say, there is a penalty of -40 enforced in case of defection & that changes the table as below –

Expected Profit

Player B

Bribe

No Bribe

Player A

Bribe

40,40

30,20

No Bribe

20,30

50,50

This takes out “defection” as an option for both the players & hence encourages the rational players to go for a Bribe-Bribe or NoBribe-NoBribe! Since No-bribe offers higher payoff, if the players are rational, they end up with a no-bribe choice.

(NOTE: There are multiple ways to make sure this kind of ENFORCEMENT mechanism can be introduced into the system, but leaving that part out here).