David R. Henderson  

What is Opportunity Cost?

Matt Yglesias's Best Post Ever... A Fragile Financial Order...

On his blog, "Conversable Economist," which, I agree with co-blogger Arnold, is excellent, Timothy Taylor gives an example of opportunity cost from Yale economist Shane Frederick:

"While shopping for my first stereo, I spent an hour debating between a $1,000 Pioneer and a $700 Sony. Perhaps fearing that my indecision would cost him a sale, the salesman intervened with the comment "Well, think of it this way--would you rather have the Pioneer, or the Sony and $300 worth of CDs?"
Wow. The Sony--and by a large margin. Twenty new CDs were too great a sacrifice for the slightly more attractive Pioneer. Although I could subtract $700 from $1,000 and was capable--in principle--of recognizing that $300 could be used to buy $300 worth of CDs, I hadn't considered that until the salesman pointed it out."

The problem is that the $300 in CDs given up by buying the Pioneer are not clearly the opportunity cost. Recall that the opportunity cost of something is the value of the highest-valued opportunity foregone. So the $300 in CDs are the opportunity cost only if the buyer had nothing better to do with the $300 than buy CDs. What are the odds that the salesman knew enough about the buyer's preferences to know that?

Comments and Sharing

CATEGORIES: Economic Education

COMMENTS (19 to date)
John Jenkins writes:

I'd say that the salesman knew everything he needed to know if he closed the sale, which is all he cared about.

mdb writes:

A good salesman know more about psychology than most psychologists.

celestus writes:

Ah, but if Sony + $300 in CDs > Pioneer, that is sufficient to show that Sony + optimal spend of $300 > Pioneer and you don't need to know anything about the buyer's preferences.

FC writes:

Maybe it doesn't matter that the salesman finds the buyer's highest-valued opportunity, as much as to give him a tangible anchor to compare to the less-tangible Sony-Pioneer gap?

jh writes:

Sure, it's possible he wouldn't (and didn't) buy $300 worth of CDs, but I think it does a fine job of illustrating opportunity costs. I guess the salesman could have said, "Well, think of it this way--would you rather have the Pioneer, or the Sony and $80 work of CDs, a $100 Blu Ray, $50 jeans, $30 shirt, $20 worth of take-out Chinese, a $10 movie ticket, $9 for popcorn and soda at the movie, and a $1 pack of gum?" That would really drive the point home of what all he could do with the extra $300, but it really belabors the point.

David R. Henderson writes:

@John Jenkins, mdb, celestus, and jh,
All good points, especially that of celestus. They don't address my point, however.

Jody writes:

What are the odds that the salesman knew enough about the buyer's preferences to know [the value of the highest-valued opportunity foregone]?


There are an infinite number of combinations of goods and services and not even the buyer has a closed form expression of his preferences (mmm... lexicographical preferences) which would allow him to solve for his optimal allocation before his preferences changed.

Local searches with randomization and heuristic rules are the best anyone can do (A is expectd to be better than B right now in this situation). But in such a space with such an ill-formed utility function that has such temporal and situational dependence, you can never know what decision would've been optimal.

(Or much shorter: an exact opportunity cost is impossible to calculate in the real-world)

guthrie writes:

It's an intuitive assessment of Shane's preferences - a guess, but not a blind one - on the part of the salesman, and it turned out to be correct. Shane's 'audio budget' topped off around $1000, so instead of pushing for the commission, the salesman drove for what he was assuming Shane would be willing to spend on audio entertainment as a whole... and provided him excellent service.

jc writes:

It actually might have been the highest valued foregone opportunity, given that our minds are often constrained by context and mental accounting (via discrete accounts).

Say this guy really loves fruit. And the salesman said, "Or, you could have this stereo...or *this* one and a bunch of *fruit*!"

In theory, the guy should snap up the fruity offer, because it is, indeed, his highest opportunity cost. In the real world, though, things may work differently, with this alternative simply not being a part of his mental choice set. Yes, he may say that's a great idea, I could have fruit too! But he may also think the salesman is crazy to bring up fruit. Or he may think he's already got fruit money set aside (in a different mental account).

Say a business has money budgeted for advertising. When discussing opportunity cost, is it correct to say that by buying television advertising they gave up the chance to buy print or whatever other form of advertising came in second during their cost/benefit analysis?

In other words, can the concept be context specific?

Or should we say, no, let's not just limit money in advertising accounts to advertising when considering opportunity cost, i.e., they could use it to cut costs, train employees, expand, etc...

ThomasL writes:

I realize that what you are saying is that the opportunity cost could be higher than foregoing 20 CDs, since 20 CDs were not necessarily the most valued use of the $300.

I think it is overly pedantic though. Anytime you buy anything you trade off some of the opportunities you could have had to buy something else. Likewise anytime you go anywhere you trade off opportunities to have gone somewhere else.

But how much time and effort are you are you going to spend attempting to determine the maximum value of every lost opportunity after it has already been lost? Seems like that is chock-full of opportunity cost, with little gain to balance it out.

rpl writes:


If celestus' comment doesn't answer your point, then I'm not sure what point you were trying to make. It's true that "300 CDs" is only a lower bound on the opportunity cost, but in the example given that's enough to establish that the opportunity cost is higher than the price difference between the two stereos. It's also true that the salesman didn't know for sure that this would be the case, but it's a reasonably shrewd guess based on what he knew at the time.

Perhaps your objection is that a lower bound would not have been useful, if the more expensive stereo were more valuable than the example that provided the lower bound. Indeed, as the second example in Taylor's post shows, the salesman might have been deliberately low-balling the example to make the opportunity cost seem lower than it actually was. Nevertheless, I think that a concrete example of the opportunity cost might be useful even in that situation. Consider this slightly modified version of the story:

... debating between the $710 Pioneer and the $700 Sony. "Well, look at it this way," said the salesman. "The price difference between the two is the cost of just one CD."

Now, it's true that there might be something even more valuable than buying a CD that the buyer could have done with $10, but he doesn't make this decision in a vacuum. He spends money on all sorts of things, including, probably, CDs. Thus, he can cut a CD purchase from his budget, or he can cut something even less valuable, if there is some such thing. However, he need never cut anything more valuable than a CD because the option of forgoing a CD is always available. Therefore, in this case one CD is an upper bound on the opportunity cost of getting the more expensive stereo.

The conclusion that I draw from all of this is that the notion of a single opportunity cost that is fixed at the value of the "best" alternative is a simplified approximation to how we actually make decisions. In reality the values of the alternatives are not fixed; they depend on a lot of other factors. Therefore, in real-world discussions of opportunity cost, a definitive "best" alternative is ill-defined. Most of the time the best you can do is to come up with a "plausible" alternative, and the salesman in the story certainly has enough information to do that.

Gabriel rossman writes:

as far as plain vanilla theory goes, celestus nailed it.

if you take behavioral economics seriously the salesman's pitch made even more sense for two reasons.

first, as Mullainathan shows, people don't necessarily think of concrete opportunity costs unless they are against a fairly hard constraint. rather we tend to think of "is it worth it" in a general sense. the salesman was invoking a frame of concrete opportunity costs to override the default frame of "worth it or not."

second, as Thaler is fond of noting, we tend to have mental accounts for different expenses. in practice this means that Professor Frederick had a $1000 audio entertainment account and opportunity costs within this account are much more cognitively accessible and commensurable than opportunity costs across it. in this sense if the salesman had said "but would you rather have a Sony and a set of new tires for your car," it would have invoked confusion rather than clarity.

Sam writes:

To make some decisions, you don't need to know the opportunity cost as long as you know a lower bound on it.

Jeremy, Alabama writes:

When I was young enough to care, I often recast opportunity cost decisions in units I understood, that being "number of CDs". It does NOT mean that CDs are the highest-value foregone opportunity, only that it is an easier currency to relate to than dollars.

It is hard to say whether the salesman was acting in the consumer's interest. Perhaps Sony was offering a sales bonus for selling its system. Perhaps the salesman saw the customer balking at the $1000 price point. It is best to assume that salesmen do not act dispassionately.

Eric Morey writes:

"I spent an hour debating between..."

At that point it is a sunk cost but he can learn from the opportunity cost of that hour to limit his debating time over consumer electronic purchases. And perhaps do his shopping online where, after shipping, overall costs are likely to be lower.

Hunter writes:

You all miss the point - both stereos were crap!

Norman writes:

As a knowledge problem, I see what you mean: nobody could know the true highest-valued alternative except the buyer. Hayek strikes again.

However, the example actually poses a conundrum for Hayek, since in this instance the buyer *didn't* know his highest-valued alternative use. If he had, he wouldn't have needed to hear that the $300 could be spent on CDs. So in this case, at least, general thinking in terms of opportunity cost (which the salesman guided the buyer to do) is far more important than the local knowledge of the exact opportunity cost (which only the buyer could know).

And I think it is exactly cases like this which highlight to students of economics how important the *way of thinking* is, even if you're not much for calculation.

Seth writes:
Perhaps fearing that my indecision would cost him a sale, the salesman intervened with the comment "Well, think of it this way--would you rather have the Pioneer, or the Sony and $300 worth of CDs?"

Or, maybe the salesman thought his opportunity cost of spending more time with Mr. Taylor was > what he would earn by selling the Pioneer. Maybe he saw a customer browsing the rear projection TV aisle with money hanging from his pocket and decided to expeditiously wrap up his time with Timothy and made the easy sell.

adirao writes:

personally, i believe there were many different factors leading to the salesman "knowing" what your preferences were, and what your definition of opportunity cost was in that scenario. from what i've learned from a few courses in economics, is that opportunity cost is defined as what you would MOST be giving up in exchange for doing something else. say getting up early to get some exercise in vs. staying in bed and getting some extra shut-eye. in your specific scenario, it was only logical for the salesman to believe that you would be willing to buy $300 worth of cd's if you bought the sony over the pioneer. you were buying a stereo, therefore it is blatantly obvious to anyone witnessing this case realize that you would use this stereo to play cd's, for it is one of its many uses. to be fair, he got a bit lucky that you liked cd's, rather than saying you should get the sony because you can get a better tv with it. it is his JOB to understand what you are thinking, and try his best to get a sale, he would have reached this scenario by the end of your conversation had he been a good salesman. as for your opportunity cost; everyone is trying their best to save money, so any logical reasoning to get a lower priced item would have helped your decision-making.

Comments for this entry have been closed
Return to top