Throughput accounting – Taking decisions using financial metrics
We had discussed, the problems that the current system of financial decision-making poses, in an earlier episode. We had established the need for local decisions with clear assessment of organisational impact. We now discuss how the new method overcomes the pitfalls of using the absorption-based costing method. We also discuss the areas where the marginal-costing method fails. The episode talks about the need to consider uncertainty of demand and impact on constraint to get the holistic financial impact on organisation. For more details on this topic, you can check out the article at: https://www.vectorconsulting.in/blog/nuances-of-toc-concepts/throughput-accounting-an-introduction/
|Shubham Agarwal :||Hello and welcome to the CounterPoint podcast. In an earlier podcast we had discussed how decision making in organisations using absorption based costing can lead to bad decisions. Now one of the major reasons of failed organisations is bad financial decision making. The way out is to have a framework where one must understand the impact of local decisions with the bottom line and cash flows of overall organisation. Now, this must be created for operational managers without any dependence on finance or the accounting department. For the discussion today, we have a very esteemed guest with us today. Eli Schragenheim, he’s one of the thought leaders in TOC, or Theory of Constraints, and he has authored six books, including necessary but not sufficient, manufacturing at warp speed, among others. His latest book, throughput economics addresses the problem of decision making using financial metrics, and it was written along with Henry camp and Rocco Surace. So let us welcome him. It is a true honour and pleasure to have him here. Hi, Eli, welcome to the CounterPoint podcast. How are you?|
|Eli Schragenheim :||Good, thank you Shubham. This is real pleasure to talk with you.|
|Shubham Agarwal :||Wonderful to have you Eli here. So Eli, we’ll start with some of the questions that we had in terms of the financial and accounting methods, the wrong decisions or the wrong steps that people across organisations have been taking. So I want to understand first, what do you think is the major flaw in you know decision making using the absorption based costing methods. What are the assumptions that are wrong. We’ve been using it for years now.|
|Eli Schragenheim :||Oh, yes. Quite a lot time a it’s actually two key flows I will start with the first one, which is that we assume that the cost of capacity is linear. Actually, what this means is if you assign a cost to one unit of a product, somehow you assume that the cost of 100 units will be a simple 100 times the cost of one unit. Sorry, this is simply bullshit. I’m sorry, I don’t know how to call it differently.|
|Shubham Agarwal :||to elaborate for our audience, if one produces more with existing fixed costs, it seems as if the costs are coming down and vice-versa. Like I saw in a pharma company, calculating the cost of a doing a test by allocating the salaries of lab personnel, which is mostly fixed, on to the test. The cost per test obviously varies widely with level of production and gives a bad information about what is exactly happening in the department. If production comes down in a month, the volume tests in QC lab comes down, and it seems as if the costs of QC department is going up.
But Eli finance guys understand the behaviour of costs with respect to time and in the long run fixed costs are also variable
So okay, fixed costs are not fixed. And so, sometimes accounting are a dividings in a relative to how fixed they are. But a fixed costs are always also dependent on some kind of quantities, but not in a linear way. And this is a very important point to remember. Because when the organisation belches let us say resource could be hiring a new employee or a machine, it comes with a certain amount of weekly or monthly capacity. How much you feed you succeed to a drove value for it is up to you. In so as long as one resource is loaded, 75% of its time and now you aid more work to this resource. The cost of extra capacity is zero. This is simply something that if you can you use it and if you don’t have something to do with it, you don’t use it. It’s still the same cost.
|Shubham Agarwal :||Yeah so in the example of pharma company, the cost of doing an additional test, with respect to salary of lab personnel is zero, as long as the lab personnel are not the constraint.
Of course, once you touch the limit, well first of all, you have a problem if you touch the limit with all your commitments, but let us say that your option is to increase the capacity,
and this will cost extra but this extra is usually usually it’s much more expensive than just takes a current capacity and divided it into how many units it can do.
|Shubham Agarwal :||One counter view that some people take is – it is important to consider every cost as variable so that the wastages of capacity is penalized
there is a belief that well managed organisation is able to utilise every resource For 100% of its available capacity
|Eli Schragenheim :||I did a huge amount of a simulation just to show this point, what does it mean. And first of all, it has to mean that there is a perfect match between the demand and the capacity. You cannot control release if a demand and when it comes in when some much less is coming. And second, it also ignores the internal relationship between the different resources that are required to deliver the value. Because if you push even two of the resources to the limits, but they interact with each other, what you get is somewhat of a chaos because again, there is a lot of statistical fluctuation in the performance of each of them and each one immediately impacts the other and you don’t get the 100% of both, you simply don’t. So we have to be much more realistic than this.|
|Shubham Agarwal :||We are coming to an interesting insight
capacity, in many cases, can be purchased in chunks and not in terms of per unit of product , which implies the cost of Capacity is not linear, It is more a step function. At the same time, it is impossible and undesirable to get 100% utilization of all capacities and hence the attempt of treating fixed costs as variable per unit is not only bad for short term decision but also long term.
But my question is , why are companies using it.
|Eli Schragenheim :||This is really a great question. Look regarding a cost per unit in general I remember my professor in my MBA studies and this was already in 1976 for me, it was long before I met Dr. Goldratt. And this professor told the class something like this every manager who takes cost per unit is an idiot who eventually will get what he deserves.|
|Shubham Agarwal :||Nice|
|Eli Schragenheim :||It impacted me and in ways a student asks the professor. Well, if in such kind of case, how do you make decisions? And his answer was every decision needs to be analysed on its own terms. Shubham it took me years to understand that the second answer is actually wrong. Because manager cannot afford to make decision each one in a different way according just to their intuition and intelligence. Because managers are exposed to after the fact criticism. So actually what managers all over the world look is to a book, some things that they can say we follow the current knowledge, we follow the cost per unit we have we have it and so we follow it. So why are you criticising me?|
|Shubham Agarwal :||So, that implies that a new book has to be officially declared. So tell us about the new approach|
|Eli Schragenheim :||Okay, I think Goldratt showed us the direction but I think I and my co writers have looked into it that we need to come up with a new book. And the first thing is to recognise the power of the term throughput. You can call it contributions that is somewhat known in for accountants, I mean, it exists but it is not really used as much as it should and not always in the right way. But it is a similar concept. But the key insight was to separate the linear part which are sales minus the truly variable cost that are linear from the part of the operating expenses that are basically nonlinear|
|Shubham Agarwal :||So trully variable expenses like material costs, transportaion etc where one is paying per piece is perfectly linear. The same has to be subtracted from sales to give us contribution or throughput. The non-linear costs or the costs with a step function , also called the operating expenses, should be put into a different bucket and not forcefully made linear.
So throughput minus operating expenses, we get the operating profit. Shubham
This seems like the marginal costing. Some smart organizations tend to use marginal costing from time to time while considering some deals.
|Eli Schragenheim :||So sometimes we see cases in which basically, if you take the cost per unit, we should not take this deal. But actually, it’s money and we see it so suddenly we use marginal costing. But the flaw in it is you are looking only on an individual deals, you don’t take in and ask a general question how this will impact throughput, operating expenses and ‘i’. And if you lose sight of it, and you lose sight of how it will impact in the future, you get a problem.
Can you elbaroate ?
And so, the new way is to come up with a new book of instruction of how to make decision and really from the start, I like to ask, taking this decision on top of everything is of all the other the existing already commitments and actions that we already take or going to take. Right, you will see critical elements to evaluate the impact of the new potential decision how it will impact your bottom line. One is how it impacts the demand. You need to send the impact on the demand. And this is the most difficult element because of the it’s really uncertain. Okay, we guess a lot when it comes to the demand and guess us a good I am the last one to say that you should not guess you need to guess but you need to understand also the boundaries of your guesses.
|Shubham Agarwal :||So what you are saying is – Just using marginal costing principles without doing impact analysis on capacity and future demand can lead to erronoeus decisions.|
|Eli Schragenheim :||Now you might get all kinds of answers including no you cannot do everything. But do you know what if you give up something else you can. Should you or should not go this way. Or you have another way. Do you know what all the normal available capacity we don’t have enough at least from one resource and this is enough. It’s enough that we lack resource lacks the capacity of one resource and we already cannot deliver everything. But if we add some Delta operating expenses and use some overtime or something like this, probably we can but we have also some extra cost. Then we come to the deciding element eventually this is a finance now let us translate. Let us translate what is the Delta throughput and what is a Delta operating expenses. Maybe we need also Delta ‘i’, if part of the solution sort of is to hire another employee or buying something or something new that we didn’t do it so far. And then you come whether you like the result or not. And see if you don’t like the results so much or you think you can improve it you can go back, back to the demand maybe we should change the offering a little bit. And then again, check the operation and then check against the finance you have a kind of a loop here between this three elements demand, operations, finance.|
|Shubham Agarwal :||But such simulations and impact analysis implies, one has to guess the impact of demand, which is not a firm number. There is uncertainity there. One has to account for it.|
|Eli Schragenheim :||But such simulations and impact analysis implies, one has to guess the impact of demand, which is not a firm number. There is uncertainity there. One has to account for it.|
|Eli Schragenheim :||this is to deal with what I call common and expected uncertainty you see Shubham I try|
|Shubham Agarwal :||Common and expected uncertainty. I like the paradox.|
|Eli Schragenheim :||Yes because this is you know sometimes you call it the noise which means some things that nothing within it will surprise you it’s not one you know it’s not suddenly the emergence of COVID-19 not like this we already know COVID-19 and yes what he did is it made common unexpected uncertainty much larger because we are now used to all kinds of fluctuations maybe in the market certainly with cash and other things that we need and so, you know about what can be expected and what is well could still happen but very rare. And I think we should put some kind of line between the two and between you and me Shubham common and expected uncertainty is something we all know but the vast majority of the organisation have no clue how to deal with it.|
|Shubham Agarwal :||Can you elaborate on how this framework of looking at global picture happen in a typical organization. Let us look at a decisions of product mix involving sales and operations.|
|Eli Schragenheim :||So if I’m a sales guy, and I see a new opportunity, I need to look on the plain load to see well, the responsibility to deliver will be operational, do they have enough time to digest the additional load and react on time for it, and yes, this one information, a item shows you whether it can easily be done, or it could be problematic, and maybe you will really call and speak directly with your operations people. S From the top management point of view, the new approach, the new book is really to bring together the people in charge, at least of the three elements, sales, operations and finance. Actually, I like the finance manager to somehow facilitate between those because each one of them brings intuitive information and knowledge and sort of readiness to try to see can we solve the problems together? Not just the problems, also the new opportunities that we like to discuss whether to take or not? So generally speaking, this is a frame.|
|Shubham Agarwal :||So one should see the load profile across critical work centers and take a call on the additional order with impact analysis on chanage in overall throughput and overall OE. Because this might involve letting go of some other orders or even decion on outsourcing.
In this situation pre-identifying the critical work centers is important.
|Eli Schragenheim :||critical resource means a resource that if there will be even a medium change in the sales or in the product mix, I already need to check because could be this resource we lack capacity, if this is a real option, I need to check this is one of the critical resources we just means constant checking.|
|Shubham Agarwal :||Alright|
|Eli Schragenheim :||So marketing and sales are to suppose to look all the time for new opportunities to grow sales. And by this increased throughput by there is a need always to check the situation with critical resource.
if you have elevated the current constraint, whoa, quite a lot has been changed. And it could be also within the group of the critical resources so you need to check several scenarios eventually to really see what really makes a very what should be the constraint that you would like, where and how much throughput you can pull out of it before you really consider investment in much more capacity. And this brings me a little bit into the role of software. Because the great thing about software is that it can make huge calculation in very short time in which we humans cannot do. And so if we have software with very friendly, what if capabilities, we can change a lot of scenarios that coming from reality from sales people telling us what options they see and what is the relative size of the option. And then operation can tell us what they can do is it and then the purchasing will tell us, okay, where we can get all the materials that we need? This is the kind of a brainstorming together with this tool that is actually critical to make our life so much more successful
|Eli Schragenheim :||Vector developed a first software that really laid out a, it laid out the possibility for what if scenarios that can be done and if something is not sufficient enough or not satisfactory enough. It’s so easy to make a change. So if suddenly product two gives us a little bit too much promise with a certain capacity, so maybe we will use product four instead, let us see what this means with our offering to the market. This easy, this is friendly, it can be done within a normal a meeting, all I need is a computer young guy to tell him what to do, and he will do it to us and then the computer makes the calculation and then we see whether it was works or not.|
|Shubham Agarwal :||Great. What you have added to the realm of financial decision making is an appreciation of uncertainity, more so when demand nos are involved.|
|Eli Schragenheim :||Yes, it is extremely important, because this is really important through out the book you will see you see one number forecast are actually useless. Because you miss a huge amount of information about what is an acceptable deviation and what is not. Because between you and me, I mean if the focus tell me I will sell 100 units, I will not sell 100 units, maybe I will say only 98, 98 is accepted or not, if it is accepted and if I sell only 88 was it already expected when the forecast was made and so, we have to understand that forecast are actually ranges its mathematically true, because in mathematics, you are giving expected value and a forecasting error and then you can deduce it by yourself and so what I came by for this kind of decision making is to create at any given this for at any given decision, two scenarios, one I called conservative scenarios in which everything we put is conservative, still reasonable, but very conservative and optimistic one. And Shubham, we need both. Because say a conservative tells you is there risk, should we lose money if we try this way. We definitely don’t want to be in a scenario that we lose too much. This probably will say we are not going to we don’t want this decision. But we need to check the optimistic as well. Because eventually you cannot afford to disappoint your market. So you need to make sure usually on the more optimistic one, you may need more capacity. So fellow you need to check it. Eventually I like to get to the level in need. This feelings that we are gambling will not be there at all. We are taking decisions that will lead us to very big success, almost without any hint of gambling. Because which really checked both sides very well. This is a robot economics way.|
|Shubham Agarwal :||Great. Thanks a lot Eli for your time and it was a pleasure to have you. As usual you can send in your comments on this episode and you can also write to Eli with your queries.|
Financial decision making – mistakes and fallacies
Why financial know-how and know-why is critical for every manager and what happens when wrong decisions are taken
Why individual targets do not drive accountability
Targets don’t work! The most popular method by which organizations try to achieve desired behaviours is by setting individual targets and by using a system of rewards and punishments to reinforce them.