≡ Menu

“Cost Accounting is Productivity’s Public Enemy Number One”

This quote comes from Thomas Corbett’s book “Throughput Accounting.” Corbett quoted this from the godfather of Theory of Constraints (TOC), Dr. Eliyahu M. Goldratt. This is a bold statement but is true. Cost accountants and business managers are not stupid, so how can this be? This blog explores these questions and proposes a solution.

Why use cost accounting?
Ever since people have owned businesses, there have always been a need to measure how they were doing. As businesses have become more complex, it became more difficult to see if the business is doing well or in serious distress. An ideal cost accounting system would:

  • show how the business is performing
  • where to make true improvements
  • make things clearer, rather than make them murkier.

Unfortunately, traditional cost accounting falls down on all three requirements. Here’s why.

Problems with cost accounting
There are a number of problems with typical cost accounting. Here are a few of the worst problems.

Based on 100 year old assumptions
When we look at the financial statements of companies, things like labour and utilities are included in cost of goods sold and inventory. This was a good assumption in the industrial revolution, where these were truly variable costs. Now manufacturing is more dependent on machine uptime and speed than the number of people working. We don’t send people home when the factory has downtime, but cost accounting assumes that is exactly what we do.
Back in the industrial revolution, managers didn’t have access to information to the extent we have today. They might know a little bit about what was happening upstream and downstream, but nothing else. The strategy of being as productive as possible in each step of the process therefore was the best one available.

Arbitrary allocation of overhead
Instead of explaining this, let’s look at an example. A company has three divisions and a head office. The following table shows the year’s results.

Division A B C Total
Sales $100 $120 $150 $370
Costs 80 85 95 260
Overhead 25 30 38 93
Total (-$5) $5 $17 $17

If looking from a purely cost accounting standpoint, division A is losing money and is a candidate for closure. If we close the division but keep the same overhead, the following would result.

Division A B C Total
Sales $0 $120 $150 $270
Costs 0 85 95 180
Overhead 0 41 52 93
Total 0 (-$6) $3 (-$3

So by shutting down the money losing division, we have decreased the profit from $17 to -$3. But wait, it gets worse. Now division B is not profitable. What happens if we shut this division down?

Division A B C Total
Sales $0 $0 $150 $150
Costs 0 0 95 95
Overhead 0 0 93 93
Total $0 $0 (-$38) (-$38)

Although this example is very simple, it shows how cost allocation can lead to wrong decisions.

Encourages local optimization
Managers for a department or a facility inside a large organization will normally be measured and paid based on their individual department performance. This will lead to the manager to decrease costs inside their department. The one way all managers will try to decrease costs is to produce more. This makes sense because you can spread your fixed cost over more production, making your costs look lower. The obvious problem with this logic is if the extra production is not sold, it goes into inventory as work in process (WIP). In fact, the only place that this strategy works is in the constraint of the business. That is usually only one step in the process. If the constraint is sales, this is an especially disastrous strategy.

Encourages building inventory
In addition to the local optimization problem, cost accounting gives the wrong impression that more inventory is better. The reason for this is how inventory is measured and counted in financial statements. If inventory is not sold during a reporting period, the cost of the raw material along with the other costs are reported as WIP on the balance sheet. Inventory is an asset and a higher asset base is usually associated with a healthier balance sheet. If costs were unusually high during that period, this WIP value is higher. Some managers may not want to sell this inventory out of fear of the balance sheet and income statement implications. Talk about bizarro strategy.

Is this your cost accountant?

Obscures the real picture
Perhaps the biggest problem with cost accounting is it does the opposite of what it sets out to do – make decisions easier. The allocation of overhead, local optimization and the over-valuation of inventory make it difficult to see what is really happening in a company. Worse, it hides how to make improvements.

There is a much better way
Fortunately there is a simpler and much better way to measure company performance. It is called Throughput Accounting.

Throughput accounting
The first notable thing about this method is how simple it is. Instead of learning the rules of cost allocation, WIP and internal scorecards, there are only three measurements.

Throughput
Throughput is by far the most important measurement of throughput accounting. It is defined as sales minus true variable costs at the constraint of the business. True variable costs are things such as raw materials, freight, packaging etc. It does not include labour and other expenses that don’t track with throughput. The important part of the definition is throughput through the constraint of the business. This means if you increase throughput, the business will improve. There is nothing simpler than that.

Expenses
Expenses are all other costs to the company. Overhead, interest, depreciation, labour are all expenses. Throughput accounting doesn’t waste effort on finding little boxes to allocate costs. This isn’t nearly as important as throughput because there is a definite top level of cost performance – zero costs. If you meet this goal, it means you are out of business.

Investment
Investment represents how much cash the firm has tied up to operate. This is inventory, equipment and other fixed assets. It is treated as something necessary to support the business, not something with its own inherent value. The main advantage to this is it leads business managers to keep enough inventory to support throughput, but eliminate everything else. This way they avoid problems like shrinkage, obsolescence and problems when high cost WIP is finally sold.

Based on real world
The real beauty of throughput accounting is it takes advantage of the information available today to provide real guidance. If a manager makes an improvement in any of the three categories, the results will be apparent on the income statement, balance sheet and cash flow statement. The results are real and unambiguous. If you use this system properly, decisions will become easier and your success rate will increase dramatically.

I have only introduced the concept here. For more information there are two excellent books. Throughput Accounting by Thomas Corbett and Throughput Accounting – a guide to constraint management by Steven M. Bragg explain the concept and prove effectiveness with lots of examples. If you are struggling with your cost accounting system or want to improve your processes, these are must-reads.

Comments on this entry are closed.

  • PT April 9, 2010, 9:28 pm

    Good blog.

    But I think you have oversimplified and glorified TA and have made ABC look like a “stone aged” accounting method. I am personally not totally sold on TA yet and find that the accounting method (TA or ABC) should be used depending on the company's goals.

    TA biggest flaw is that it does not account for overhead and labor expenses (variable?), basically stating that the enormous capital invested on plant/building and technologies cannot be improved upon.

    In my opinion, TA is too focused on short term profits and shy's away from the aforementioned creating a distorted view on the reality of the situation. I think both have a place in the current times, but I can also see why companies who should use TA end up towards ABC, because of the current industry norm and trying to do a industry comparison when using 2 different methods of accounting is not practical or smart.

  • Sean April 10, 2010, 9:00 am

    Hi PT,
    In fairness, I said that cost accounting uses industrial revolution-aged assumptions, but I think they would work equally well for the stone age!
    I’ve read your criticisms of TA in other forums and would like to tell you my understanding of them. First you say that not accounting for overhead and labour by activity is a bad thing. It’s true that TA doesn’t parse costs into activity categories like ABC. Dr. Goldratt refers to that as being “approximately right instead of precisely wrong.” The problems I see with ABC are it encourages high productivity everywhere, leading to excess inventory. You also state that TA assumes that capital expenditures dictate business results and cost cutting can’t be improved upon. This may come from looking at snapshots of TA systems. In TOC, we determine the constraint to the goal, protect the constraint, subordinate everything else to protect the constraint, elevate the constraint, then finally if the constraint moves, go back to step 1. If the constraint changes, the TA system needs to be changed to reflect the new situation.
    You make a good point around the benchmarking aspects of ABC. ABC allows for comparison between plants, divisions and companies. Benchmarking isn’t a part of TOC because each system has different constraints and applying benchmark standards across the board usually doesn’t do much. In addition, ABC can work in a decision to shut down a plant, when all of the activity costs are truly variable.
    I’m not sure about the criticism of TA for long-term decisions. Maybe it’s due to companies that implement TOC see very large and rapid improvements early, and further improvements are more difficult and slower in coming. If they use TOC properly, the improvements still come.
    Finally, your point that the decision to use TA versus ABC should reflect the goals of the organization is right on point. Applying TA to a business that doesn’t practice TOC will only cause confusion. We only need to look at the names of the systems to gain insight to the behaviour they promote. Cost accounting and ABC have cost in their title; cost per unit of production is what they are designed to improve. Throughput accounting uses throughput; throughput at the constraint is what it is designed to improve. While costs can only go to zero, throughput has no upper limit.
    Thanks for the comment. It’s good to have healthy skepticism around different ideas.

  • Sean McAlpine April 13, 2010, 9:30 am

    Hi PT,
    In fairness, I said that cost accounting uses industrial revolution-aged assumptions, but I think they would work equally well for the stone age!
    I've read your criticisms of TA in other forums and would like to tell you my understanding of them. First you say that not accounting for overhead and labour by activity is a bad thing. It's true that TA doesn't parse costs into activity categories like ABC. Dr. Goldratt refers to that as being “approximately right instead of precisely wrong.” The problems I see with ABC are it encourages high productivity everywhere, leading to excess inventory. You also state that TA assumes that capital expenditures dictate business results and cost cutting can't be improved upon. This may come from looking at snapshots of TA systems. In TOC, we determine the constraint to the goal, protect the constraint, subordinate everything else to protect the constraint, elevate the constraint, then finally if the constraint moves, go back to step 1. If the constraint changes, the TA system needs to be changed to reflect the new situation.
    You make a good point around the benchmarking aspects of ABC. ABC allows for comparison between plants, divisions and companies. Benchmarking isn't a part of TOC because each system has different constraints and applying benchmark standards across the board usually doesn't do much. In addition, ABC can work in a decision to shut down a plant, when all of the activity costs are truly variable.
    I'm not sure about the criticism of TA for long-term decisions. Maybe it's due to companies that implement TOC see very large and rapid improvements early, and further improvements are more difficult and slower in coming. If they use TOC properly, the improvements still come.
    Finally, your point that the decision to use TA versus ABC should reflect the goals of the organization is right on point. Applying TA to a business that doesn't practice TOC will only cause confusion. We only need to look at the names of the systems to gain insight to the behaviour they promote. Cost accounting and ABC have cost in their title; cost per unit of production is what they are designed to improve. Throughput accounting uses throughput; throughput at the constraint is what it is designed to improve. While costs can only go to zero, throughput has no upper limit.
    Thanks for the comment. It's good to have healthy skepticism around different ideas.

  • Dennis T. April 23, 2010, 9:06 am

    Two more big problems with cost accounting: #1 – It judges past performance but is not very well suited at making future decisions. #2 – It is inconsistent. Sometimes it give the right answer, sometimes not. So the trust level goes down.

  • Dennis T. April 23, 2010, 4:06 pm

    Two more big problems with cost accounting: #1 – It judges past performance but is not very well suited at making future decisions. #2 – It is inconsistent. Sometimes it give the right answer, sometimes not. So the trust level goes down.