Answers are grounded in this paper and cite their sources.
A vendor who sells apples by the piece charges the same for fruit of different weights. A vendor who sells by weight charges in proportion to what each buyer takes. This paper compares the two methods for apples with a stated weight distribution, a stated price per kilogram, and a stated pattern of buyer choice. When fruit is allocated at random, the methods return almost the same revenue. When buyers choose the largest fruit, piece pricing transfers a substantial sum from vendor to buyer: the vendor receives $109.50 for fruit that would return $124.49 by weight in a day, a loss of $14.99, or 9.99 cents per apple sold. The same piece price charges different buyers between $3.07 and $5.74 per kilogram. Splitting stock into two sizes at the median weight and pricing each size separately reduces the loss to $9.34 per day, but grading labour costs $6.00 per day, so the net benefit is −$0.35 per day. From this the paper derives a grading rule: grade into two sizes only when the coefficient of variation of fruit weight exceeds a break-even value set by labour cost, the residual loss after grading, and the value at stake. For the stated terms that value is 0.2428, about a standard deviation of 43.7 grams at a mean of 180 grams. The stated apples, with a coefficient of variation of 0.2286, fall just below the line, so grading them on these terms does not pay.
No takes yet. Share an insight, caveat, or question.
William Alexander Aldred (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: