It is not a fallacy to consider sunk costs when they can help you estimate the marginal return on a new investment. This is likely when you are investing in som

It is not a fallacy to consider sunk costs when they can help you estimate the marginal return on a new investment. This is likely when you are investing in something which is hard to value, with diminishing returns, like growth marketing. Making another $1 billion loss next year might be rational, but if profitability keeps getting deferred year after year, eventually it will become irrational to continue throwing good money after bad.