Mental Accounting

I came across this statement while reading 'Fooled by Randomness':

"Statistics have proved that people prefer to earn $60,000 knowing that others earn $50,000 than earn $70,000 when others are earning $80,000"

Let me offer my un-expert analysis of this statement.

The Argument
At face value, it appear that it is a stupid mental accounting error made by most people because surely it would make more sense to have more savings in the bank with the $70K package than to have the satisfaction of earning more at $60k.

However, I would like to point out that it does not matter whether you earn $70k or $60K or $5. You can't eat the money, and wear it as protection from the elements nor can you live in it. What matters more is the value associated with the money e.g. how much food will $100 buy you in each of the two cases.

Let me illustrate further. We don't not have a single international currency (the dollar or Euro is a re-hashed arrangement). So we have to make all calculations by factoring the exchange rate in. As such it is much easier to be a rupee billionaire in India that to be a dollar billionaire (The rupee vs. dollar exchange rate has varied wildly in the 38-52 range in the past decade). So it is better to have $100 in your pocket than Rs 100 (at least while you are in India)

By analogy, the actual value of the money you have is the purchasing power that it generates. Hence, I can conclude that $60K when everybody else is earning $50K will give me an upper hand in terms of purchasing power (i.e. simple barter value) than $70K in a $80K neighborhood. So the mental accounting fallacy derived in this case is a fallacious argument in itself i.e. there is no mental accounting error. This was my first impression, which I was forced to alter when I realized there is more to this problem than appears at first dig.

The Googly
The twist in the tale is that purchasing power in terms of price of goods is not affected only by the earning power of a small part of the population. The people in the statistical study probably had a fixed set of people (e.g. family, friends and co-workers) and not the average of the entire population. For example, a construction laborer will arrive at a much lower 'desirable' remuneration that say a Wall Street high flyer simply because he has benchmarked himself with respect to those who keep him company. And the overall purchasing power is a function not only of the differing remuneration but also the number of people belonging to each category.

So, the real mental accounting error is committed by benchmarking to the observable rather than exhaustive sample set. And what is even more amusing is that this mental delusion of superiority (wealthier than thou) is even resistant to hard observations (e.g. having to cut down on luxuries) and is revised only in extreme situations (not being able to make ends meet) where survival rather than favorable comparison becomes paramount

Conclusion
So, a mental accounting fallacy does exist in this situation; albeit a different one from what I thought initially.

Corollarily, I realized that defining others as the observable sample would have avoided this hazard and hence wording your conclusion is pretty important.