Late Night Discussions
Number 3

By Eliyahu M. Goldratt

How to go from one ditch into another - a discussion on transfer prices

"Jonah, more than once you've told me that where ever we turn we'll find that we've barely scratched the surface. Not that I disagree, but I think that we should test your hypothesis."

"Fine," he says, "it is always beneficial to put old opinions to new tests. Of course, only as long as the new tests are meaningful. Otherwise such exercises tend to be just a sophisticated form of procrastination."

Beautiful. I don't know how he does it but even before we start I'm already in a bind. Now all of a sudden it's my responsibility to find a 'meaningful test'.

Jonah is looking at me but I'm stuck on the word meaningful. After a short while he says, "Alex, why don't you pick a topic which is, at least in your eyes, relatively important and at the same time you feel we've already beaten it to death."

That helps. "Measurements," I say confidently.

Jonah just seems amused. "Measurements are definitely a very important subject 'Tell me how you measure me and I will tell you how I will behave.' But Alex, what on earth gave you the impression that we have succeeded to put more than just a slight dent in that subject?"

Now I'm surprised. "We have spent days debating what should be the measurements of a company. Endless talks pinpointing the distinctions between measurements and the company's goal, bottom line and necessary conditions. We've hammered in detail the use of the measurements to judge any type of decision and action. I've lost count of the numerous discussions we've had on the topic of 'local performance measurements', trying to carefully separate between the imposed plan and the execution. And now you tell me that all of that is 'just a slight dent?"'

"Of course," he says. "Alex, we should not judge by the number of hours we put into it but by the scope and quality of our conclusions. Besides, we didn't even cover all the topics we should have covered."

"You mean, measurements for non-profit organizations? But we did discuss them."

"In a very artificial way," he comments and then continues, "Alex, will you please stop patting ourselves on the back. Wake up. Even for for-profit organizations we have left important measurement issues basically untouched."

Now he's getting to me. "Give me even one single example," I throw the glove.

'Transfer prices," Jonah replies.

"What do transfer prices have to do with measurements?"

"A lot," is the only answer. After a while, and probably because of my stubborn silence, Jonah continues. "You're right. I should not use the name of the common semi-solution but rather I should have mentioned the measurement problem."

"That would certainly help," I cannot hold myself.

Luckily Jonah ignores my last nasty comment and he calmly continues, "consider a relatively big company that has several units participating in the generation of the same products. Naturally there is transfer of goods not only between the company and the outside world but also goods are transferred between the units. In such an environment how do you suggest measuring the performance of a unit?"

He got me again. It is a very important issue that we never even touched. "Jonah, you've definitely made your point. Even when we're convinced that we have thoroughly investigated a subject we should be careful not to assume that we have exhausted it. But as long as we're on this interesting subject of transfer prices, why don't we discuss it?"

"Be my guest," Jonah gestures with his hand.

Once again the ball is in my court; I should be used to it by now. "Let's start by clarifying to ourselves the nature of the problem."

"This is usually a very good place to start," Jonah smiles in approval.

"We definitely want to measure the performance of a unit. The difficulty in doing it, in our case, is that sometimes the units are transferring partially completed products. This raises the problem of establishing a sales/purchase price between the units, or as it is commonly referred to - transfer price."

"And when you can buy these partially completed products from the outside, don't you still have the problem?"

"Yes, of course," I accept Jonah's correction. "Even when we deal with finished products we have the same problem. Sales price is determined through negotiations, through the balance of perceived need and availability. In our case, of transfer between units of the same company, the balancing act of the free market is missing. That's why companies are using internally predetermined transfer prices. This is a very touchy subject. If we establish high transfer prices the units that produce the goods will come out smelling like roses and the receiving ones will have mud all over their faces. If we establish low transfer prices the opposite will occur."

"Correct," Jonah picks up the ball. "Measuring the performance of each unit starts to be not a function of their actual performance but rather a function of this arbitrary choice of transfer prices. On top of it, let's not forget the additional problem of determining when the internal sale took place."

Even though I don't fully understand his last comment I cannot help but tease him. 'Wait a minute Jonah, why do you refer to the way we determine the transfer prices as an arbitrary choice? As far as I know most companies are investing a lot of effort to calculate them as precisely as possible."

"How do they calculate it" he asks with a sneer.

"Using cost accounting of course," I answer cheerfully.

He throws me a disgusted look, but he is trapped; he must continue the discussion. "That is exactly what I mean by arbitrary way."

I cannot help but laugh. I know just too well Jonah's view on cost accounting. That got him. I'm now anxiously awaiting a long and vicious attack on cost accounting, with a particular emphasis on the futility of trying to determine sales prices through the calculation of product cost. Jonah doesn't do it, he calms down and asks, "Alex, don't you realize that determining transfer prices through cost calculations encourages inefficiencies?"

After a while he sees that I'm still thinking about his last remark, "determining transfer prices through cost accounting leads to the desire to inflate production costs by slowing down non-constraints. With this devastating trick an internal unit can drastically improve it's measurements at the expense of the company as a whole."

"Yes, I see your point," and now it dawns on me. "That's why lately, as cost accounting is no longer the sacred cow it was, companies have started to involve the free market forces in these issues."

"Please elaborate," Jonah encourages me.

"Lately many companies have given to their units an unprecedented freedom. They allow their units to buy components from the outside even when these components are produced by their sister unit. These sister units no longer have an internal captured market. Even internally they must compete with the external world. Likewise, units that were producing for internal consumption only, are encouraged to sell their goods directly into the market. The car companies are a startling example to the speed and scope that this method is spreading and they are definitely not the only ones."

"Yes," Jonah agrees, "But in the way that it is implemented I'm afraid that once again we face a situation that the medicine is worse than the disease."

Now it's my turn to say "Please elaborate."

He ignores me and mutters to himself, "I wonder why people have the tendency to swing from one extreme to the other?"

"Jonah!" I cry in a load voice trying to stop him from diving into his private shell. "From what extreme to what extreme?" I desperately ask, just to get the all too familiar answer, "try to analyze it yourself."

Well, what other choice is left? So slowly I try to solve the riddle. "What extreme have we used in the past? We established transfer prices looking only internally, to our costs, ignoring the outside world. This is definitely an extreme. And what are we doing now? We are allowing free trade between the internal units. That doesn't look like an extreme to me. Jonah? Jonah!"

"Alex, you forget that these units do belong to one single company," he says in an impatient voice.

"Yes, of course. If we allow the units free trade with the external world, we may easily ruin the entire company. Just suppose that we decide that any thing that is cheaper to buy than to produce, we are going to buy and then trim internally. We may be able to trim people but what is the point in trimming machines or know-how. Going in this direction will just inflate the internal prices of the remaining items. They will have to carry the burden that was previously shared with the items that were outsourced. It's very easy to see that this way leads to rapid shrink, first to closure of the components plants then the assembly plants and finally what will be left is a trade company. What do we call the outcome? The hollow company syndrome?"

"...At the same time, how can we justify a unit shutting down its downstream sister unit just because someone (maybe the competitor) offered two cents more per piece. Total freedom to trade with the external market prevents the company from enjoying and capitalizing on the strength inherent in its existing vertical capabilities. It looks like we have no other way, and rather than avoiding the issue we must try and solve the puzzle of the appropriate measurements for internal units. At the end, what is the purpose of measurement if not to induce the parts to do what is good for the company as a whole?"

"It looks like there is no suitable way to fix transfer prices. Should we even have internal prices? Do we have to use dollars as a means to measure, or maybe there is another mechanism? And what about your remark on determining the point in time in which the sale took place? Questions, questions. Is there a practical, good solution? I know your answer, 'of course there is but in order to find it we'll have to check our basic assumptions.' So Jonah, let's do it now. Jonah... hey, Jonah!"

By now Jonah's mind is definitely not in this room. I'll have to remember to bug him about this subject some other time.

This "Late Night Discussion" is Copyright © 1991 Dr. Eliyahu M. Goldratt