VirginFlyer wrote:I'm wondering, could someone with a decent knowledge of programme accounting please give a comprehensive but succinct explanation of what accounting blocks are in relation to a programme such as the 787, so that when these questions come up we can point back to it and avoid the same ongoing back and forths over what it is or isn't?
V/F
Despite what accounting.net would try to make you believe, it is nothing more complex than the formula for an average, which is known to a 5th grader : N x average cost = SUM(individual unit cost) over all N units
The point is that over the 40- to 50-year lifetime of an aircraft program, recurrent production costs will vary a lot from one aircraft to another : there is a learning curve for early aircraft, there are differences due to customisation, there is a continuous flow of changes made to the aircraft design and/or to the infrastructure made to build it. Over such a long time and with such variance, it is difficult to recognize financial patterns and deduce investment decisions from said patterns - which is the whole point of accounting.
An extreme example given by Astuteman is for submarines, for which huge costs would be spent over a period of 10-15 years, against which revenues would be booked at a few discrete moments in time corresponding to down payments. So instead of showing the yearly figures, which would show huge losses in most years and huge profits in a few years, it makes more sense to smooth out the cash flow over the program lifetime by averaging the costs over the total number of units built.
In the end, what matters is the total cash expenditure, so if you build 10 units, it doesn't matter whether it is booked as ({cost of each unit} for all 10 units), or (10 x {average unit cost}). Both are equal, by definition of an average. Program accounting is just the second case : book N times the average production cost, with N the total number of aircraft built, so that production costs are almost constant and decision-making is a bit easier (and fooling inexperienced investors, TBH)
The only small issue is that in this case, neither the total number of units (N) nor the build costs are known. Therefore they must be estimated. The estimated 'N' is your accounting block.
Both estimates must be reasonable, and then of course the manufacturer will want N to be as high as possible and (average cost) to be as low as possible, so that it looks good. On the other hand, estimate the cost too low and the accounting will have to be re-adjusted by declaring a loss.
Following from this accounting method, some aircraft (mostly at the beginning) will cost more to produce than the declared average cost. The intent is that other aircraft will cost *less* than the average, and balance everything out by the time the accounting block is produced. But in the time between production of these expensive and cheap aircraft, there is an overcost that must be kept visible in the accounting - the deferred costs.
So to summarize :
- program accounting is a perfectly legitimate accounting method in principle
- it is just averaging the highly variable recurring costs over the program lifetime, to present a near-constant expenditure
- the sticky point is that it relies on assumptions about future costs & sales performances rather than actual measured cash flows
- it is NOT applicable to development costs which are 1-shot (non-recurring) costs and therefore do not vary over time (and therefore cannot be averaged)
- deferred costs are NOT sunk costs as they have been paid but not accounted for in the P&L, and therefore any future decision-making would have to take them into account
- 787 delivery cash flow is not directly convertible into free cash flow as a large chunk of it is earmarked to service the deferred costs - and thus is not free.
At this point, it is reasonable to expect that only a very very tiny part of the total revenue from each delivery can be considered as free, and that is supposing that each delivery is generating a gross margin (before deferred costs) already higher than other well established programs (usually around 20%). It is not totally unreasonable to suppose so, given the set-up of 787 production. And while we may debate the credibility on this site, the relevant authorities seem to accept that Boeing is capable of reducing the deferred costs to 0 by the end of its currently declared accounting blocks, which is what it is committed to do by law.
At June 30, 2017 commercial aircraft programs inventory included the following amounts related to the 787 program: $32,047 of work in process (including deferred production costs of $26,461 ), $2,608 of supplier advances, and $3,390 of unamortized tooling and other non-recurring costs. At June 30, 2017 , $24,246 of 787 deferred production costs, unamortized tooling and other non-recurring costs are expected to be recovered from units included in the program accounting quantity that have firm orders and $5,605 is expected to be recovered from units included in the program accounting quantity that represent expected future orders. 
