Forcing somebody else’s model onto a company hoping it will work is a dangerous game. John Darlington warns against the risk of adopting prescriptive approaches to lean accounting, which could mislead many practitioners who do not have the advantage of demand stability and volume enjoyed by automotive companies.

Although book keeping and accounting record keeping goes back many hundreds of years, management accounting is a relatively new subject and really began to “take off” during the era of the Industrial Revolution.

Until then “piece rates” were a popular method of paying people and as labour and material dominated the costs of running an organisation, overhead departments were small, you could argue that a product cost did truly represent the real cost of producing an item. For example, there was real market traceability between the sell price and the money the business had parted with to make the product.

As a consequence of the Industrial Revolution, owners, attracted by the gains to be had from economies of scale, were inclined to invest significant sums of money in capital and hired workers longer term to develop skills rather than hire them “spot” and pay by the piece produced.

Add to these changes the propensity of organisations to integrate vertically and market traceability began to diminish between the costs of running production and the revenue streams from product sales. This was the “fuel” needed to explore management accounts for internal decision making as opposed to financial accounts for external reporting.

The result of these forces, to provide useful information for managers to make better decisions against a backdrop of distance from markets, was ultimately the design of the standard costing system.

The mechanics of the system had one last twist worthy of note to it. The growth of companies around the 1920s and the pressure on capital markets was the motivation for public accountants to agree standard well-defined procedures that any audit firm could, well, audit fairly.

The public accountants could not cope with the diversity of methods they encountered when contemplating the costing practices of the engineers in their attempts to trace costs to products so this was given up to a broader method of allocating cost to inventory for periodic profit reporting – a financial reporting response.

The result was “an invention” hinging around double entry book keeping that inflated inventory value in terms of work in progress and finished goods for periodic profit reporting that is with us to this day. And inadvertently it favours profit during periods where production outstrips sales. And worse it penalises profit if you happen to stop replacing unwanted inventories.

Ever heard the phrase, “We have not recovered enough overhead.” Well, now you know what it means: “We haven’t produced enough things”.

Never mind that we have satisfied demand and reduced our lead times. The double entry book keeping is unhappy!

So, most of the innovation in design of standard costing had been developed by 1925 and increasingly the administration was computerised and integrated from the late 1960s with the rise of MRP culminating in today’s ERP systems.

What we have is a case of uncorrected obsolescence, not original error.


How come 50 years plus went by in cost accounting stagnation until Johnson and Kaplan and Goldratt came up with alternative if rival systems? It isn’t as though there were no other warnings.

Taaichi Ohno: “It was not enough to chase out the cost accountants from the plants. The problem was to chase cost accounting from my people’s minds.”

And Deming: “…using cost accounting metrics to make decisions is like driving by looking in the rear view mirror… We need good results, but management by results is not the way to get them.”

How come the academics did not spot that these systems were increasingly redundant? The only discussion I recall when trying to qualify through the Chartered Institute of Cost and Works Accountants was a facile debate on fully absorbed or marginal costing; really a discussion about which costs were variable and which costs were fixed.

Answer: neither or both, it depends on the time frame being considered.

But this was not the way exam questions were structured. What we got were exam questions that posed seductive if entirely simplistic and useless models compared with what real companies were actually grappling with in real life.

The questions posed always had just a handful of products, a single set of resources; an assumption that there was no starting inventory or ending inventory for fear it distorted the variance analysis so favoured by the Institute yet invented circa 1925!

So flow and lean accounting are united in the need to change:

  • Conventional accounting systems encourage overproduction by their clumsy interpretation of the “value added” principle;
  • They do not recognise the importance of bottlenecks, constraints or pacemaker processes;
  • They encourage local efficiency creating “islands of excellence”;
  • They have little or nothing to say about lead times;
  • They promote the idea that the bigger the batch the lower the unit cost;
  • They encourage “cost reductions” which often prove to be “mirages”.

The list is not meant to be exhaustive. it is frightening to consider the amount of real money being spent by people continuing to design and administer these old out-of-touch systems.

Installing or upgrading a modern ERP system is eye-wateringly expensive and yet in the course of doing it you embed standard costing invented circa 1925!


If marginal costing and fully absorbed costing vied to answer the question of which system best served management decisions up to the 1980s, two new rivals appeared with Johnson & Kaplan and Goldratts writing. We got activity based costing (ABC) and throughput accounting (TA).

And in many respects these were a polarised version of the earlier rivalry.

ABC put product cost calculations as a central platform but recommended cost pools and cost drivers to deal with overhead allocation in a discernibly different form than standard costing. A more accurate product cost was the promise.

TA, which evolved from Theory of Constraints, suggested that very few costs were truly variable with one more or one less unit sold but a thorough understanding of constraints i.e. limiting factors was required. Product costs were ignored in favour of contribution often linked to the consumption of constraint resources. Goldratt’s throughput inventory and operating expense (TI&OE) was direct, simple and very demanding and we use it to this day.

Both rival schools of thought had their “champions” and claimed their successes.

And despite the fact that I personally favoured TA I think it could be said that “ABC won”, at least in the UK. It certainly entered the mainstream vocabulary and more companies claimed to be applying ABC than TA.

I should know, I was quite often invited to ABC conferences as the “token heckler”!

To summarise, I think you can see these antecedents in the new rivalry emerging as the question of what sort of accounting you need for a lean organisations is asked.

Flow accountings roots are definitely traceable to the TA approach. Knowing some of the main players in lean accounting I would argue that ABC is still influencing their approach.


We have a pretty unique opportunity to establish afresh the accounting data we need to make better decisions in a lean organisation.

I have a number of concerns about the current direction of lean accounting: the movement places a great deal of emphasis on the value stream. And you can appreciate why this is given so much preeminence: it is mentioned in the 5 lean principles established by Dan Jones and Jim Womack in their seminal book Lean Thinking.

It is certainly possible to reduce the need for scheduling synchronisation when resources are co-located and I am not naïve enough to say the benefits are limited to this alone. However shared resources do exist. In fact shared resources appear to be more common than dedicated value stream based ones.

And as if to emphasise their dislike of shared resources lean accounting often labels them “monuments” and not in a flattering sense.

Well, this is ignoring some fundamental production organisation structures. I learnt about VAT analysis not in the accounts class but via constraint based management. There are three fundamental structures which tend to reflect the dominant characteristics of the production bill of material and route.

“V” plants will typically be process orientated with little or no assembly. Steel or paper mills, pharmaceuticals at least at the primary end etc., would come under this category. They start off with common processes but points of differentiation allows the product offering to explode into many end items.

“A” plants will have deep bills of material with many layers. Component and sub-assemblies eventually end with few unique production items. Aerospace bodies, wings, engines and the final aircraft itself provide good examples.

“T” plants are the modern day assembly operation. The “bar” represents the common components either bought out or made in house which can be put together in an almost infinitely large range of end items. I worked in AlliedSignal Turbochargers for many years and we were predominantly “T” orientated.

And the point is that as you move from left to right on the diagram above you can plot the likelihood of being able to dedicate into value streams. It can often be achieved in a light assembly plant but a steel mill, a large forging or casting business, no, not desirable let alone feasible.

Even in “A” plants where the final assembly is almost certainly arranged by stream, for example by type of aircraft, the further you go upstream in the bill of material the more you find manufacturing centres that feed all the different types of jet assembled with sub-assemblies or components, like panels. You are not going to lay down extra autoclaves or chemical milling equipment just because you make five different types of aircraft; one for each?

And if we start to discuss service-based organisations like hospitals then the shared resources become even more critical and often dominant. There is a possibility of excluding a whole plethora of organisations that also need accounting assistance with this dogmatic approach.

Maybe we should apply value streaming to the M6 motorway using type of vehicle? One M6 for cars, another M6 for trucks and a third for buses!

The point is that the lean accounting movement is in danger of taking a concept like value streaming and forcing it into situations where it does not fit as it might in a lightweight assembly environment where dedication makes sense and is relatively cheap to accomplish.

We certainly need to understand how to treat shared resources better but the solution is not always to double up to the point of dedication.

Flow accounting believes that we need to reflect the current physical flow to the boundary walls of the organisation which is the only place where real money changes hands. This will reveal V, A or T. Record the potential delays and mismatches and this can often be traced back to shared resources. Dedication is but one option on how to improve the flow.

The lean accounting concept of a maturity path arises because you undoubtedly find that not all resources can be packaged into dedicated value streams. So allocation becomes the countermeasure “in the interim”. You are likely to wait a long time if you accept the flow accounting VAT analysis. Of course allocation does not lessen the total expense, but it is comforting home territory to some accountants.

Lean accounting wants to get rid of “wasteful transactions”. Who wouldn’t? Inventory recording, valuation and other associated transactions like work orders are often in their sights. The fundamental idea is that once the inventory is beneath a certain level, for example one month’s worth, there is no need to record it separately as raw material, work in progress or finished goods. You can simply expense it.

Now I like the Gregorian calendar as much as the next person, but is this really sensible?

Firstly, raw materials, work in progress and finished goods do three totally different jobs for manufacturing organisations and we need to understand how much we have of each in terms of time. Suggesting we can do away with the transactions that allow us to understand the different elements does not make common sense.

Flow accounting wants to know all about inventory in its different forms:

  • Because the work in progress and finished goods reveal how well the company is really using its capacity. Only capacity puts WIP and finished goods in place.
  • Because, as Ohno said, overproduction is the worst form of waste and “earlies and lates” in stock reveal it.

So in an effort to remove “wasteful transactions” we might lose these fantastic insights. Reducing the waste in an accounts department, whilst commendable, is not valid if the business as a whole suffers.

There are some painfully naïve statement being made regarding capacity and the ability of the organisation to take on more business and to be fair this is not exclusive to lean accounting. Flow accounting is clear, when utilisation of constraint resources start to reach 80% to 85% you are approaching bottleneck territory. And if you continue to take on additional orders your delivery promise dates will start to deteriorate and lead times will begin to grow in an unpredictable manner.

Flow accounting is completely linked to time-based capacity planning with the above levels of utilisation at the forefront. Unfortunately, lean accounting is reticent on the same subject and as part of the value stream organisation advocates dividing the cost of the value stream by the number of items it has produced or sold. The result of this is a different unit cost but still one that encourages you to take on more business with scant regards for time based capacity planning.

Unit cost thinking was a major contributory factor in getting us into this mess in the first place, and I can see little or nothing coming out of lean accounting which is flagging the issue of linking time and money closer together.

And this is really unforgiveable because clearly the sell price is at least a tactical if not strategic decision and should not be left to unit cost thinking.

So why have a unit cost? We do not need one if we are limiting ourselves to management accounting. Only the desire to retain consistency with Generally Accepted Accounting Practices (GAAP) and financial accounting is holding us back.

Flow accounting wants good financial accounting to stay in the accounts department. We find GAAP generally unacceptable (GAAPGU). Management accounting practices offer wider opportunities to explore how to assist lean practitioners make more effective interventions.


There is a historical precedent for dividing financial accounting from management accounting.

In the same way that standard costing was applied long after the “model” had outlived its usefulness there is a danger of lean accounting “straight-jacketing” us into the wrong direction.

The reservations expressed here are symptomatic of the lean movement as a whole. Trying to force tools and “models” which worked well in one environment onto others indiscriminately will ultimately fail and damage the reputation of lean as well as the organisation on the receiving end of it.

We need a less prescriptive approach whilst retaining our desire to create and maintain and organise so that the flow of goods and services gets better to end customers.