Sunk costs are costs that were paid. Since economic decisions are based on the marginal cost and the marginal benefit of a proposed action, the primary characteristic of sunk costs is that their marginal cost is zero, regardless of the initial cost. Hence, any economic decision based on sunk costs will depend on whether there is any marginal benefit remaining in the purchased item.
For example, suppose you buy a ticket for an outdoor concert that cost $1000. After the transaction is completed, the marginal cost of the ticket is zero. Whether you can get a refund or resell the tickets does not change their marginal cost, which is always zero. To see this more clearly, consider the case where the ticket seller will only refund 10% of your money. Since you will only get $100 of your money back, you will decide whether the concert is still worth the $100 or if you can sell it for a higher price. If neither case applies, then you will apply for the refund even though it is only 10% of what you paid because that is $100 more than your marginal cost and because the marginal benefits or opportunity costs are less than the $100 refund price. Suppose that you fully intend to go to the concert, but the weather turns ugly on the day of the outdoor concert. Whether you decide to go to the concert will depend on whether the marginal benefit is still greater than zero. If it is, then you will go; if not, then not.
Hence, any economic decision regarding sunk costs will depend on whether there is any marginal benefit to the purchased item. The initial purchase price is totally irrelevant.
The economics of sunk costs can also be considered by looking at how market demand is related to the market price. It is well-established in economics that demand increases as the price of the product decreases. Because the marginal cost of a sunk cost is zero, it is likely that there will be some demand for it even if conditions change that reduces the desirability of the product.
Even though economic decisions based on sunk costs should not be affected by the amount of the costs, the amount originally paid does indicate the amount of the desire and therefore the demand for the purpose of the sunk cost. For instance, suppose that in a previous year, you paid only $10 for an outdoor concert, and, coincidentally, the weather turned just as ugly on the day of the concert. Even though the marginal cost of both the $1000 ticket and $10 ticket are both zero, the difference in the sunk costs reflects your willingness to pay, which reflects your desire to see either concert. Therefore, there will probably be a greater marginal benefit, even in ugly weather, to seeing the $1000 concert than for the $10 concert. So even though the marginal cost for both tickets is zero, their original cost indicates a difference in their marginal benefit, which is why people tend to complete projects or enjoy the benefits for which large sunk costs were incurred.
In business, all costs that have already been paid are sunk costs, including all fixed costs, and variable costs that have already been paid, or for which a legal liability to pay them has been incurred. Only current or future variable costs can be adjusted according to current market demand.
Many times, sunk costs do not affect future economic decisions at all because there is no marginal benefit. For instance, if a drug company spends millions of dollars developing a drug and it turns out to be unsafe, then the drug company will probably lose all its investment in the drug.
Sunk costs can also be considered in terms of opportunity costs. For instance, suppose a business prepays a lease for 2 years because of the steep discount given for prepaying. If, after one month, the business is deciding whether to move to a bigger space, then it must consider the marginal benefit to remaining in the same space since that is the opportunity cost of the prepaid space. If the business moves to larger space, it will have to pay for the new space while forfeiting the marginal benefit of the old space. This makes the comparison different than if the business was deciding between the 2 leases before prepaying any lease. Note, also, that the opportunity cost of the old space is the remaining marginal benefit — not the pro rata value of the remaining term of the lease. For instance, if the business decides it needs machinery that requires special operating conditions that can only be provided by moving to a new space, then the marginal benefit of the old space — and therefore the opportunity cost — is zero: the amount paid for the lease or the time remaining on the lease is irrelevant.
Conventional economics is based on rationality, that people always make decisions to maximize their wealth or their utility. Behavioral economics questions this assumption, since people often make irrational decisions. The sunk-cost fallacy is just such a type of irrational decision people often make. By keeping in mind that the marginal cost of a sunk cost is always 0, you can avoid this mistake.