In seventeen authored AEGIS records, a multiplication of quantities by unit prices, added over two or three lines, is the one thing that is settled, and the records make its total the figure that everything after it is measured against. The cases are synthetic settings, and what follows describes how a designed record set frames a decision, not how any budget, purchase or contract is or should be decided.
Discussion
The line item comes out of these records well defended. Hou, Line Item and Budget (2023), explains the survival of the line-item format by its advantages, convenience and conciseness. The records add, in their own voice, a reason the working paper's note does not name: they exclude every option that would merge the lines or replace them with one averaged price, on the ground that the average would hide a difference between prices or distort demand. In this study's reading the line protects information that the total alone would lose.
The same records show what a line does not carry. Noreen, Conditions under Which Activity-Based Cost Systems Provide Relevant Costs (1991), contrasts one allocation basis proportional to volume with several cost drivers. The allocation questions in the set ask, in effect, which basis to use: share of quantity or share of cost, the split of sites between two years, the number of runs a vehicle makes. The runs answer with a total. The consumable parts record says its multiplication yields shares that its output does not report, and the fuel transport record multiplies by vehicles while its question asks about runs. In this study's reading, the product per line is the right tool for the question "how much", and these records also put to it the question "on what basis", which it does not answer.
Read through Tversky and Kahneman, Judgment under Uncertainty (1974), the records place an anchor: a value is made available first, fixed into a budget, a ceiling or a reserve price, and later prices are to be recorded as departures from it. This study does not claim that any judgment is biased by it; the article describes a heuristic, and the records are settings. Flyvbjerg, Curbing Optimism Bias and Strategic Misrepresentation in Planning (2007), argues for projections based on the actual performance of a reference class of comparable actions. The observation lines that keep assumed and actual prices side by side, for use next time, are in this study's reading the beginnings of such a class. In the records the assumed total comes first, and the class of actual prices is still to be gathered.
What a line leaves out is named, not hidden. Ellram, Total Cost of Ownership (1995), looks beyond the price of a purchase to other purchase-related costs, and Greenleaf, Johnson, Morwitz and Shalev, The Price Does Not Include Additional Taxes, Fees, and Surcharges (2015), review research on prices divided into a base and surcharges. In the records the base is the line; the costs left out of an acquisition, the allowances of a personnel contract and the maintenance of a fleet are written beside it and handed to a clause, a later round or an alternative. The records show only that the division is made.
Finally, the words beside the total do work of their own. Wanat, Bases of Budgetary Incrementalism (1974), explains budget requests through the vocabulary of the budget participants, and Thaler, Mental Accounting Matters (1999), describes how money is labelled, grouped and evaluated, how often and how narrowly or broadly. The records speak of a total that fits or goes beyond a budget they do not state, of a price per dose that labels a sum, and of a yearly amount whose unit names no year. The arithmetic is right in every case; in a few places the vocabulary that says what it is about parts from it.
Contributions
This paper makes three contributions. First, it gives a close reading of how one designed decision record turns a reproducible total into a reference figure: the run's total is correct and replayed, and the recommendation fixes it into a budget, a ceiling, a reserve price, a baseline or a contract, while the doubts the record names lie in the factors of the product. Second, it shows that the same records speak of their inputs in two registers, fixed in the reasons for a recommendation and assumed in the runs' notes, and that most of their reopening lines send a change back to the arithmetic rather than to the question. Third, it reads the records beside five debates, on the line item and the vocabulary of budgets, on anchors and reference classes, on the cost beyond the price, on allocation and on mental accounting, and shows where each debate's question meets a sentence of the records.
Limits
The cases are seventeen authored synthetic settings from one programme of one studio. Every quantity, price, unit and schedule in them is the author's assumption, so nothing here is a finding about real budgets, procurement or contracts. The English is this study's translation of Korean records, made once and labelled, and a reader of the Korean may weigh a word such as fixed, set or assumed differently. The counts are rules on Korean words, stated in numbers.md; other rules would give somewhat different counts. The works are read through their published abstracts, and the readings that connect them to the records are this study's own. The frame measure finds two fields, the uncertain line and the counter-evidence of links, to be frame; this paper leaves them aside, and its reading of the rest may still miss repetitions that a measure on Korean words cannot see.
Further research
Three lines of work would test what this paper only describes. The other 143 runs of the same rule, in 23 other AEGIS programmes, count in minutes, hours, indices and synthetic budget units as well as in won; reading them would show whether the patterns here belong to this programme or to the rule. A comparison with the other calculation rules of AEGIS-CALC-v1 would show whether the turn from the arithmetic to a reference figure is peculiar to totals of prices. And a reading of the same questions with the runs' outputs extended, to report shares, per-period units and the budget a total is compared with, would show which of the gaps described here are gaps in the calculation and which are gaps in the setting. Such work would build on what this paper offers: a reading of seventeen records in which an exact total, replayed and fixed, becomes the figure against which later prices are measured, while what the records do not know stays in the quantities and prices the total is built from.