Which method of transfer pricing is used if both the division are free to deal either with each other or in the external market?

Tweet
Inhaltsverzeichnis Show
Post
Share
Save
Get PDF
Buy Copies
Print

Editor’s note: All references are listed at the end of the article.

Imagine the following conversation in a company president’s office. He is talking with the general manager of a division selling products internally (seller), the general manager of a division buying these products (buyer), and a professor who has read everything that has ever been written about transfer pricing.

President: I wish somebody could tell me once and for all how we should do our transfer pricing. All the methods we’ve tried have problems, and often both the buyer and the seller claim they’re being treated unfairly.

Professor: The answer is really very simple. When the product of the selling division is sold in a perfectly competitive market, the buying division should pay market price, with perhaps a discount to recognize that selling costs and other such expenses are saved on internal transactions. The buyer should be free to obtain this product either internally or from external suppliers.

Seller: That’s a pretty fair arrangement. Of course, selling internally is usually more trouble than selling externally, so I don’t think a discount is appropriate. After all, the buyer treats me in ways he wouldn’t dare treat an outside supplier. He delays and cancels orders and demands product with short lead times. What’s more, why should the company lose out on profits by buying outside when I have spare capacity?

Buyer: In the real world, Professor, there are very few perfectly competitive markets. I don’t think market price is necessarily the best idea, although I agree that I should be free to buy from whoever I want. After all, my performance is measured by how well my business does. It isn’t fair to hold me responsible for profits and then interfere with those kinds of decisions. I can usually get better prices, sales support, service, and technical cooperation from outside suppliers.

Professor: Well, you’re right that market price isn’t the best approach in imperfectly competitive markets. In that case the best method, when there is spare capacity, is to use marginal cost and require that you buy inside. That’ll maximize company profits.

President: That sounds good to me.

Buyer: Our accounting system doesn’t measure marginal costs, but if they’re something like variable costs, I think it’s fair. The price will be so much lower that I’ll live with the inferior product and service I’m used to getting.

Seller: You’ve got to be kidding! I have fixed costs to pay for, and I should get a fair profit on internal sales, just like I do on external sales. I’m evaluated and rewarded on the performance of my business. My reported earnings would be unrealistically low, which would lower my people’s morale. Do you mean to tell me that the products I make especially for you, that use up my capacity, should be transferred at marginal cost?

Professor: Well, now you’re talking about a different situation. If the amount is fairly small, you should get a markup on full cost to approximate what the market price would be, so that you get a fair return.

Seller: That’s more like it. I always knew you professors could reach intelligent decisions if given enough information.

Buyer: Wait a second. How are you going to determine a fair market price on something that isn’t sold in the market? You professors have no idea how much time we spend arguing about what a fair market price is, even on products that are sold on the outside. But you’ve changed the subject. Let’s get back to products sold inside.

Professor: My theory answers that problem as well.

Seller: It had better answer for situations where I don’t have spare capacity. There are times when demand for my product is very high, and he’s lucky to have an assured source of supply. Are you telling me I should get only marginal cost when I have customers knocking down my door, willing to pay premium prices so they can keep up with their own demand? Some of these customers are his competitors. If I sell to them, he’ll lose sales. I think my product is worth a whole lot more than your stupid marginal cost.

Professor: Oh dear, you’re making things more complicated, but that’s no reason to get upset. My theory addresses this problem, known in academic circles as “demand dependence.” Here it is necessary to use sophisticated mathematical programming techniques. While the procedure is too complicated to explain, shadow prices can be used to derive the transfer price that maximizes profits. It’s somewhere between marginal cost and market price.

Buyer: Look, Professor, we’re talking about real products and real costs—not this shadow world of yours. What you’re ignoring is the fact that I’m one of his largest customers, and without my volume his costs would be much higher, which would make his profits much lower. Somehow your theory had better take account of the profits I’m making for him.

Professor: Well, I’m afraid you’re now talking about what we call “technological dependence,” a much more complicated situation. To be honest with you, I’m not sure it’s always possible to find a mathematical solution when this condition holds. However, I’m sure that given enough time and a sufficiently large computer, I could develop the necessary equations and work out their solutions. Of course, we would have to specify the constraint equations for the algorithm used to solve the objective function, and a global optimum exists only—

President: Hold it! I haven’t the faintest idea what you’re talking about. It seems to me that if your solution is too complicated to explain, it’s too complicated to use. And all this talk about time and computers—we have to make decisions today. Until you professors come up with something useful in practice, we’ll just have to muddle along as best we can.

Attack on the Issue…

Since the invention of profit centers more than 60 years ago, corporate managers and academics alike have been looking for a solution to the nagging transfer pricing problem. It has two parts: the sourcing decision and the pricing decision. How should a business unit choose between internal and external suppliers, and what price should be placed on the good if the internal supplier is used? (“Business unit” refers of course to any part of an organization acting as a buyer or seller in internal transactions and includes divisions, manufacturing and sales functions, and product-business-program dimensions in a matrix structure.) The transfer price may be determined first and become an input to the sourcing decision, or the decision to buy inside may be made before the transfer price is decided on.

Most of the research academics have done on this problem has been of little use to executives.1 Studies usually recommend marginal cost and mathematical programming solutions, but the few surveys that have been conducted show that these approaches virtually do not exist in practice.2 For one thing, they are simply too cumbersome. And, ironically, in spite of their complexity, they require simplifying assumptions that ignore the many factors managers must take into account.

I decided that the best way to study transfer pricing and to develop a theory that would have clear implications for practice would be to focus on managers’ experience. In pursuit of this objective, I interviewed nearly 150 executives (CEOs, CFOs, group and business general managers, financial managers, and others) in 13 companies. Their industries are chemicals, electronics, machinery, and machinery components.

These interviews left no doubt that the key to the transfer pricing problem is strategy. Transfer pricing schemes are a means of generating information and control for implementing corporate, business unit, and product strategy. The managers involved consider these schemes to be more or less fair. It is not always possible to satisfy both the control and the fairness objectives through the transfer pricing scheme alone.

There is no simple solution to the problem. It requires continuous attention through management processes. A company’s transfer pricing policy and the processes for administering it depend on the particular situation and the direction the organization is going in.

Analysis in two dimensions.

To help managers think analytically about the problem, I have developed a simple framework that recognizes transfer pricing’s complexity in the real world. Two broad dimensions of strategy, applicable at the corporate, business unit, and product levels, are the basis for the framework I call the manager’s analytical plane (MAP).3

The first dimension of the MAP is vertical integration, that is, the extent to which the company carries on production and distribution activities that other companies could perform. Vertical integration results in interdependence between profit centers when each stage of the production and distribution processes is evaluated on the basis of profitability

The second dimension is diversification, that is, the extent to which the company is engaged in different businesses. Diversification is determined by the extent of product-market segmentation. It results in an emphasis on the independent contributions of each business when they are separated as profit centers.

Although the dimensions can operate simultaneously, they are distinct.4 A company may integrate backward without diversifying by manufacturing a component that is not sold to external customers. Conversely, a company may add a product that requires no change in manufacturing and distribution capabilities. Often a company will use several transfer pricing policies depending on the strategy of the groups, business units, and products involved.

An executive can use the plane defined by these two dimensions to analyze his or her company (or business unit), since each location determined by the two strategic dimensions is associated with particular organizational characteristics. Five especially important characteristics are the nature of corporate strategy and the strategic planning process; the primary means of control by top management; the criteria for performance measurement, evaluation, and reward; the definition of fairness in the company; and the nature of the managerial processes. The framework appears as a plane to emphasize the difficulty of categorizing organizations as discrete types and to acknowledge the complexity of the real world.

For analytical purposes, however, I present four “pure” types to help a manager think about the situation. These four types correspond to the corners of he MAP as shown in Exhibit I. Three of these, the competitive, cooperative, and collaborative types, are important for understanding the transfer pricing problem. (There is no transfer pricing in the collective organization.)

Exhibit l MAP and the four ‘pure’ organizational types

Exhibit II summarizes the characteristics of the three types. A company’s transfer pricing policy (I use this term to include internal or external sourcing policy as well) depends on the organization’s location on the plane and the direction in which it wishes to move. I’ll take up each of these types in turn.

Exhibit ll Characteristics of competitive, cooperative, and collaborative organizations

…in Competitive Organizations

Highly diversified organizations that have little vertical integration between business units represent the competitive type. The most dramatic examples are the conglomerate and the holding company. But more common are companies with several relatively autonomous divisions having all the necessary functions of a complete business and having little interdivisional dependence. Corporate strategy is the sum of the business unit strategies, and the strategic planning process is bottom-up.

The main control mechanism is management systems that measure the results of business unit actions, although organizational structure is also important. In these organizations, strategy implementation is most clearly reflected in these measurement systems. Since the business unit general manager has great authority over resources, upper management’s control is exercised by measuring performance and comparing it with objectives. The entrepreneurial responsibility of business unit managers receives emphasis through measurement of the units as profit or investment centers.5

The criteria for performance measurement, evaluation, and reward stress comparison of business unit performance with a budget or plan, similar competitors, and even sibling units. Use of similar performance measures for all units generates enormous internal competition. Since the measurement and evaluation of the use of resources and their allocation in a competitive organization resembles resource allocation in a market, the definition of fairness is similar to judgment meted out by an impartial spectator, namely the market.

Competitive organizations usually decentralize decision-making responsibility to the business unit level. This results in a bottom-up process for handling corporate-business unit relationships and a distributive bargaining process in business unit-business unit relationships.6 The latter is characterized by “win-lose” bargaining, which results from the attempt of each business unit to maximize its objectives, even at the expense of the other.

When the business units do not depend on each other very much, the costs of local decisions that are not best from a corporate perspective are small compared with the problems created by a transfer pricing policy that interferes with the stance of the impartial spectator and with the bottom-up and distributive bargaining processes. In its purest form, a competitive organization would have no transfer pricing policy, leaving the business units free to establish trading relationships as companies in the open market. A “no policy” default on top management’s part, however, frustrates vertical integration because it creates disincentives to internal trading.

The general manager of the semiconductor group in one of the electronics companies I studied explained the problem: “In my experience, sister divisions are more antagonistic to each other than to outsiders. Basically this is due to competition as to who is the best performer. There is the suspicion that other divisions are doing something to get more than their share, especially when a product is forced on a division. Sister divisions are under more pressure because word about the first missed step could be picked up by the board.”

Market-based pricing.

A response to this problem is a policy of market-based transfer prices with constrained sourcing, such as giving inside suppliers a “last look” chance to meet outside quotes. In competitive organizations, transfer prices are used to select suppliers and customers and to value internal exchanges in a way consistent with the measurement of business unit performance.

One kind of market-based transfer price is cost-plus-profit markup. A buying unit may source a unique or proprietary good from a selling unit for which there is no market price since it is not sold externally. The seller establishes a cost intended to approximate the price if the product were on the market. Methods for determining the markup include comparable products, average gross profits, and average return on assets of the selling unit. Several of the electronics companies in my study used the cost-plus-profit method for proprietary products.

In competitive organizations, headquarters normally leaves the determination of market prices up to the business units involved. The result is distributive bargaining processes as each unit tries to negotiate the most favorable price. When top management does set policies that influence the transfer price, such as “list price less 25%,” the ability of the units to establish “true” market price obviously is hampered.

A reason often given for using market-based transfer prices less some discount is the presumed absence of selling costs, receivables carrying costs, and credit risk. But as internal transactions come to resemble external ones, through the use of sales-people assigned to internal accounts, for example, the less justified this discount becomes in the eyes of the selling business unit.

One of the companies I studied that had the characteristics of the competitive type gave its divisions autonomy on sourcing decisions—they were free to buy inside or outside—and they were under no obligation to sell internally if the terms were unfavorable. Regardless of the size of their purchase, internal customers received goods at the most favorable price obtained by outsiders with the largest quantity discount. Nevertheless, in a number of instances, external vendors were chosen over internal ones. To top management and the division heads, these outcomes were in the best interests of the divisions and the company as a whole. As one manager put it, “We don’t want to do anything that would give a division general manager an excuse for not meeting his plan.”

There are real advantages to leaving the decision to the business units. A manager of the semiconductor division of a second electronics company had recommended against a division’s replacing its outside semiconductor supplier with the internal supplier even though the internal source stood to gain an additional $4 million to $6 million in sales. He explained why: “As long as Intel’s product line is more comfortable and they’ll jump through hoops to give us good quality, it’s in the corporation’s interest to stay with them.”

When top management establishes constraints favoring internal transactions, as did a third electronics company by requiring that a “substantial percent” of semiconductor purchases be made internally during recessionary periods, both business units may find it unfair. A year after this policy took effect, a division manager who bought semiconductors noted with a wry smile: “I think we’re still in a recessionary period. They raised prices a lot and were willing to lose a share of their business with us. The president deemed that the prices should hold and that we had to buy inside. They aren’t happy, and we aren’t either. They’d rather make investments in other areas that are more profitable for them and growing faster, some of them at 100% a year. But they had to continue to invest in areas that support our products.”

Unit heads think that the no-policy default is the fairest situation because top management does not interfere with their autonomy—which is consistent with exercising control through results. As the head office gets more involved with interunit relationships, complaints of unfairness may rise. Nevertheless, top management may want to increase internal transfers because of excess capacity in the selling unit or to take advantage of proprietary technologies. One way of accomplishing this while attempting to preserve fairness is through a dual pricing policy.

Dual pricing.

This policy combines the advantages of market-based pricing (the profit incentive for the selling unit) and of mandatory internal sourcing (e.g., the buildup of volume of the transferred good to reduce unit manufacturing costs). The buying unit receives the transferred good at cost, and the selling unit is credited with the market price. (The double-counting of profits is eliminated at a higher level in the organization.) As compensation for the interference with its flexibility, the buyer receives the product at less than market price, which enables it to show greater profits.

Just as a transfer pricing policy can be used to maintain a company’s location on the MAP, it can be used to move a company in the direction of a desired future position. A competitive organization can use a dual pricing policy to increase its vertical integration, moving the company “north” on the MAP. Three of the electronics companies I studied that were located in the competitive region of the MAP used this policy from time to time.

A dual pricing policy has its problems too. (Two of the three companies that had adopted it reverted to market-based transfer prices with sourcing freedom.) If the double-counting elimination is not forecast properly, the net income of the whole company will be disturbingly less than the sum of the net profits of the units. This is especially likely if the financial and control systems are inadequate, or if business is poor and the selling unit can’t meet its external quota and so generates excessive internal sales.

Moreover, since the buying unit gets the product at cost, it has little incentive to negotiate the most favorable market price on the transaction. Actually, neither division has the same incentive to monitor the performance of the other, which is a central characteristic of the competitive organization.

Sometimes the cost can be even greater than the market price, as it is at the beginning of the product life cycle for products with significant learning curve effects. This occurred at an electronics company, which had to make an exception for semiconductors.

Because of the problems, companies probably cannot use dual pricing for a long period of time for all products, although they may employ it for a few strategically important items. One electronics company, for example, set a dual pricing policy to enable a division selling batteries as a replacement part to price more competitively and regain lost market share.

Dual pricing is also used occasionally to bolster internal sales. An executive of an electronics company that used this policy for a few years described the experience this way: “The bookkeeping system didn’t have the ability to handle it. It didn’t work out well since it was difficult to administer. But it did get more transfers going, so it was the right move at the right time.”

…in Cooperative Organizations

In competitive organizations, the business units must compete with each other; in cooperative organizations they must cooperate with each other. An example of the latter is a highly vertically integrated chemical company, organized on a functional basis, where all the units are cost centers except for a single sales force that acts as a revenue center.

More generally, business units turning out intermediate products have two roles, as a profit center for external sales and as a cost center for internal transfers. Corporate strategy is established for the company as a whole, and business unit strategies are derived from it.

The primary mechanism of control in cooperative organizations is the organizational structure. Since responsibility for many decisions rests at the top of the organization, especially for those involving exceptions to corporate policy or trade-offs between functions, the authority vested in hierarchical position or generated through expertise is especially important for specifying the actions of business units.

Business unit managers in cooperative companies do not have nearly as much autonomy and as wide a range of responsibility as their counterparts in competitive organizations. Top management is more directly involved in day-to-day operations and exercises control directly through interactions with subordinates. It relies less on management information systems. In these organizations, executives pay a great deal of attention to learning the unique aspects of the company.

The criteria for performance measurement, evaluation, and reward are also very different in cooperative organizations. Although managers use objective quantitative criteria—budgeted costs and revenue—these are largely based on comparisons with historical performances or budgets. Because of the interdependence between the units, top management uses more subjective criteria to determine whether a unit manager made her “numbers look good” at the expense of other units or whether, for example, the manufacturing unit stayed on a long production run rather than interrupt it and give priority to an important customer.

In cooperative organizations, because measurement and evaluation criteria are different for each unit, units compete with each other much less than they do in competitive organizations. Also, bonuses, salary increases, and promotions for individual managers and budgets for the business units are based primarily on total corporate performance. Thus, the definition of fairness in cooperative organizations is that of shared fate.

The philosophy in cooperative organizations is to centralize those decisions involving the interdependencies between business units that affect total corporate performance. Consequently, corporate-business unit relationships are managed through top-down processes, while integrative bargaining occurs between business units. In its win-win bargaining, each business unit tries to maximize corporate objectives, even at the expense of its own performance measures. Because a unit’s contribution is not determined solely through quantitative means, the units have an incentive to perform well.

Because transfer prices strictly refer to exchanges between profit centers, in its purest form a cooperative organization would have no transfer prices. Many organizations, however, while closely approximating the pure cooperative type, have units that sell substantial quantities both inside and outside, forming quasi or partial profit centers. The transfer pricing policy for these organizations follows from the structure and systems. Internal transfers are mandated for both buying and selling units and the transfer price is full cost. Unlike in competitive organizations where the transfer price is a market price that determines whether sourcing will be internal or external, the transfer price in cooperative organizations is determined after the decision to source inside has already been made.

The mandating of transfers is a direct consequence of the vertical integration strategy whereby the large capital investments involved make any other approach impractical. Cooperative organizations measure performance of the total product flow across units, not of units as stand-alone profit centers. The purpose of transfer prices is to accumulate total costs as if the end products were manufactured completely within a single business unit.

Actual full cost.

When a company transfers actual costs and its units share each other’s fortunes, it approaches a pure cooperative type. Actual full costs are calculated by dividing all fixed and variable expenses for a period into the number of units produced. Because product costing involves valuing inventory, allocating joint product costs, and costing by-products, none of which is a precise science, costs are defined and calculated through management judgment and consensus.

The major difficulty with actual full cost transfers is that the price of the intermediate good fluctuates. If either internal or external demand falls and the volume of the selling unit decreases, the transfer price to internal buyers increases. Also, buying units do not know the price until the period is finished and the selling unit can calculate the actual costs.

A multi-billion-dollar chemical company with a sophisticated management control system used actual full-cost transfer prices. (Actual full cost is not the chosen method of only the unsophisticated.) In this company, buying divisions were told which plant to purchase from. Top management made these complex sourcing decisions to minimize transportation costs and to balance plant loadings for greatest manufacturing efficiency.

Because product costs varied according to the producing plant, some managers were unhappy with this approach. Overall, however, managers showed little desire to change to standard-cost or market-based transfer prices. The company’s controller explained: “The alternatives, such as using market-based pricing or keeping track of who gets favorable and unfavorable variances, are worse than what we have. Financial information is important but it doesn’t give managers yes or no decisions. We still pay them to run the business.”

Performance measurement, evaluation, and reward at this company were consistent with the actual-cost transfer pricing policy in that they emphasized the business unit’s contribution to the company rather than its own financial results. Subjective judgment played a large role in evaluating a business unit manager. Company performance, business unit performance, and especially personal performance determined bonuses.

Standard full cost.

If a company uses standard full-cost transfers, it can substantially reduce the variation in price the buying unit receives. This approach makes it possible to measure more precisely each unit’s contribution, especially when variances are charged to the unit responsible for them. It also makes it possible to more clearly separate the profit-center (external sales) and cost-center (internal transfer) roles of the selling unit. Although nothing can eliminate it, standard full-cost transfers reduce the effect interdependence has on performance measures.

Given the assumptions about volume, raw materials, and other expenses, standard costs are the expected unit costs of producing a product for some period. Standard costs are usually based on what it would cost to produce the good if the plant were running at nearly full capacity. A good standard-cost system attempts to identify the changes in volume or raw material costs that make the actual costs either higher or lower than the standard cost.

A standard-cost system can also isolate the effects the buying unit has on the selling unit, such as when the former buys less than anticipated. Some systems include a “take or pay” provision in internal transfers. When buying units are responsible for negative volume variances if other customers don’t turn up, a portion of the selling unit’s capacity has to be identified as dedicated to a particular buyer. Disputes can arise about whether a buying unit’s capacity was used to meet the demand of another customer.

Determining whether variances are the responsibility of the buying or the selling unit is not easy. When selling unit managers have to interrupt long production runs for special orders, they may complain that the buying units are responsible for the resulting inefficiencies. Buying divisions may argue that because cost decreases such as those for raw materials or utilities are not due to the purchasing acumen of the selling unit but to changes in market prices, the positive variances should be credited to the buying unit.

The integrity of the standard-cost system depends on how the system is designed and operated. At issue is whether managers believe that the standard costs are what the costs really would be for an efficiently run operation. Buying unit managers have to believe that selling unit managers are doing their best to forecast raw material costs accurately. A common complaint is that these estimates are purposefully conservative so that the selling unit can show positive variances.

No ultimate accounting method can solve the problems inherent in standard full-cost transfers. These problems must be managed through processes that recognize the limitations of quantitative measures of performance.

A chemical company in my study illustrates how a standard full-cost transfer pricing policy can be effective even when it does not include take-or-pay provisions. The company’s transfer pricing approach had evolved out of a standard-cost system that had been implemented in the early 1960s. In the mid-1960s, as part of a reorganization to a matrix structure, a task force recommended a market-based system to allocate variances. Top management preferred the standard-cost approach and rejected the team’s proposal.

At the time, the CEO explained his reasons this way: “I can tell how well the entire company is doing based on the performance of a few of our key chemicals. But in order to do this, I need to see all of the variances in one place, not scattered across 50 to 60 downstream products.”

Twenty years after the implementation of standard-cost transfer prices the sentiment is pretty much the same. One manager acknowledges that market-based transfer prices would have some advantages: “There is a lot of merit to a market-price system. It shows on one accounting document the relative contributions to profits of the individual products. If we were to evaluate on this basis we would have to have a better way of showing the products’ contributions.”

In this company, however, relative profitability doesn’t count for much. Managers are evaluated on the basis of their contribution to the company in ways other than the profitability of their business units.

The company is located in a small town; nearly all of its managers are chemists and chemical engineers who were recruited right out of college and promoted within the company. They know each other well and have an intimate understanding of the company’s businesses and strategy, which have been constant for years. One senior executive explained how managers are assessed: “They are measured on profit before taxes, but we evaluate them more based on the quality of their effort. It is a largely subjective judgment and includes a poll of their colleagues.” Bonuses are based on total corporate performance and on an individual’s performance appraisal—not unit results. One manager explained this shared-fate concept of fairness: “Our objective is total corporate profits. I can make a decision that costs my division money, but if it makes the company money, I’ll be okay. Its an inherent philosophical and intellectual thing. We react to overall optimization.”

Cost plus investment.

Even when managers mostly agree about what fair standard costs and variance allocations are, the standard full-cost method still causes problems. The selling unit does not receive a profit on internal transfers, resulting in lower profit margins and return on investment percentages. Even though top management does not attach much meaning to unit comparisons of net profit and ROI, some selling unit managers may feel that some of “their” profits are showing up in the performance measures of buying units.

One solution to this and other problems is to use a cost-plus-investment method for establishing the transfer price while retaining mandated internal sourcing. Several chemical companies, an electronics company, and a heavy machinery company in my study used this approach at some point. This method transfers the intermediate good on a full-cost basis, but it also “transfers” the portion of the selling unit’s assets used for internal needs to the books of the buying unit. This approach essentially splits up the selling unit into an investment center based on external sales only and a cost center for internal transfers, and makes the buying unit responsible for profits and ROI on all internal resources used to manufacture its products.

Cost plus investment plays a function in cooperative companies similar to the one that dual pricing plays in competitive organizations. In both cases the objective is to obtain some of the advantages of the other organizational type. But whereas dual pricing accomplishes this by using both of the two basic approaches, cost-plus-investment pricing is a financial technique that “eliminates” the structural arrangement that requires a transfer in the first place.

Cost-plus-investment transfer pricing also has its problems, many of which are related to allocating costs and investments. As conditions change, allocations, which are inevitably somewhat arbitrary, may not fairly reflect the balance between internal and external sales.

One company took nearly three years to make a cost-plus-investment approach work, and even then some key managers remained vehemently opposed to it. The problems included a selling unit manager who wanted his unit to remain a profit center and so purposely set standard costs that would result in a positive variance; a take-or-pay provision combined with spot market-price transfer prices on products desired in excess of budget; disagreements about which unit should benefit from purchasing and energy variances; and some buying unit managers who felt that some of the selling units’ plant investments were excessive.

…in Collaborative Organizations

Companies that are collaborative emphasize both the interdependence of vertical integration and the independent contributions of the business units as diversified businesses. They combine characteristics of both competitive and cooperative organizations. As individual profit centers, the units compete; but as a result of high interdependence, they must cooperate. The inherent tension in this situation makes it possible for companies to obtain the benefits of the cooperative and competitive types but also makes them vulnerable to their problems. For collaborative organizations, the transfer pricing problem is most complex.

It is not clear on the MAP where an organization passes from competitive to collaborative or from cooperative to collaborative. As an organization moves toward the collaborative area—either through changes in transfer pricing policies or through increased strategic emphasis on interdependence through vertical integration or on independence through diversification—the transfer pricing problem becomes more difficult to manage. As this occurs, management processes become increasingly important in designing the transfer pricing policy.

In collaborative organizations the interplay of top-down and bottom-up planning approaches establishes strategy. As the balance between the objectives of the units working together and their individual objectives changes with conditions, the corporate strategy shifts.

Collaborative organizations are usually organized as matrix structures or as multidivisional structures that are substantially interdependent. Business units might share such resources as corporate manufacturing or sales units. The structures of collaborative organizations are more complicated and messier than the clean multidivisional form of the competitive organization or the clean functional form of the cooperative organization. Supporting the diversification dimension of strategy are the results-oriented systems of the competitive type for top management control, while the hierarchical structure of the cooperative organization gives top management control in support of the interdependent vertical integration dimension of strategy.

Because of the tension between structure and systems in collaborative organizations, the primary mechanism of control is through management processes. This is a higher order form of control than in either of the other types. Top management must constantly guard against an excessive use of the functional structure or of the business unit systems; either can lead to the deterioration of the dimension of strategy associated with the other control mechanism.

The management processes shaping corporate-business unit relationships are iterative in nature. They combine both the bottom-up processes of the competitive type and the top-down processes of the cooperative type. Similarly, the processes shaping inter-business-unit relationships are those of mixed-mode bargaining. These processes combine distributive bargaining of the competitive type with integrative bargaining of the cooperative type. Iterative and mixed-mode bargaining processes are complex and require that the managers involved have sophisticated conflict resolution skills. They also increase stress, which, if excessive, can hurt performance.

Iterative processes are used in strategic planning: guidelines are issued from the top down; the units then communicate up in such a way that alters the top level’s definition of strategy. Iterative processes make possible shifts of influence and control between levels and units as circumstances dictate.

Mixed-mode bargaining processes between business units are a consequence of their simultaneously competitive and cooperative relationships. Due to the emphasis on systems for measuring quantitative results, business units have an incentive to win at the expense of others. Structural interdependence, however, acts as an incentive for each business unit to be concerned with the performance of the other as it affects total corporate results.

The dual focus that exists in measuring, evaluating, and rewarding performance reflects the mixed-mode bargaining. On the one hand, collaborative organizations use largely quantitative, objective criteria to assess individual business unit performance and to reward this performance. Executives take great efforts to establish objectives that contain measures of profitability and return on investment and to compare performance along these dimensions with similar external competitors.

On the other hand, given the very real dangers of managers attempting to improve unit performance at the expense of total corporate performance and the difficulty of finding comparable outside competitors, management tempers the quantitative approach with internal criteria such as cost-efficiency, total product performance, and other, more subjective, assessments. Bonuses for managers and capital budgets for units take into account both independent and interdependent contributions in a necessarily difficult-to-define combination of cooperative and competitive criteria.

The sense of fairness in collaborative organizations is based on a rational trust the unit managers have in top management. The rational component signifies the impartial spectator standards of fairness appropriate for competitive organizations. The trust component signifies the shared-fate standards of fairness appropriate for cooperative organizations.

Management processes & conflict.

In collaborative organizations the required internal sourcing part of the transfer pricing policy of cooperative organizations is combined with the market-based transfer prices (including cost-plus-profit-markup) part of the transfer pricing policy of competitive organizations.

Some companies move from the cooperative to the collaborative area because of increased diversification and a desire to put greater emphasis on individual unit performance. Others want to increase entrepreneurial spirit to prepare for and facilitate increased diversification. In companies that place a large emphasis on decentralized profit responsibility, selling unit managers are not likely to accept the argument that because cost transfers are built into the budget and objectives, they do not distort performance measures. As one manager in this situation put it, “I want to be able to compare my performance with competitors’ and internal divisions’ so everyone will know what this business is contributing to company profits.”

When companies move from the competitive to the collaborative area it may be because they want to increase vertical integration and emphasize internal cooperation and the total product flow.

In collaborative organizations, market-based transfer prices play a different role than do prices in external markets and competitive organizations. In economic theory, price is a single-number transmitter of a great deal of information. Companies allocate resources and make decisions on trading relationships according to this number. Market-based transfer prices play a similar role in competitive organizations. Although they have less impact on capital allocation decisions, they do determine whether a buying unit sources internally or externally.7

When top management requires internal sourcing, market-based transfer prices become a source of conflict through which management can gather information and obtain control. The conflict occurs because there is no completely satisfactory way to determine a fair market price. Very large volumes of internal transactions do not have external counterparts, and the specifications of the product sold internally may vary from those of the product sold externally. Both units can argue that the transfer price is not a true market price because changes in transfer prices can significantly affect the performance measures of business units as profit centers and thus perceptions of their relative contributions to company profits.

In making these arguments, the unit managers are likely to present information on roughly comparable external transactions or bids, differences in the cost of product sold internally compared with product sold externally, lack of cooperation in the relationship (poor R&D and service support or short lead times and canceled orders), inadequate or inflated gross margins, and so forth.

This information gives top management the control it needs to manage the business from a total product flow perspective. It is the kind of information that top managements in cooperative organizations have ready access to because their direct involvement is more accepted. When internal sourcing is required, both units actively seek the involvement of top management to solve the transfer price conflict. Control comes from top management’s using each unit to keep the other on its toes.

Conflict is an inevitable element in the iterative and mixed-mode bargaining processes top management uses to exercise control in collaborative organizations. Because it can arise when changes in external conditions put pressures on existing objectives, conflict can warn top management to reexamine strategy. Just as time spent on budgeting and allocating capital is productive to the company, so may be time spent on resolving the conflict related to transfer pricing. Suppressing or alleviating this conflict beyond a certain point actually becomes undesirable. Top management may even want to take steps to increase conflict in order to overcome any reluctance managers may have about bringing disputes—and the attendant information—to the top.

In collaborative organizations, it is impossible to find a transfer pricing policy that will satisfy top management’s needs for control and be perceived as fair by the units involved. Indeed, achieving fairness between units would reduce the utility of transfer prices as a mechanism for generating information and control. Nevertheless, to keep business unit managers motivated, management must still try to achieve fairness in its relationships with business units. When unit managers perceive that their individual rewards and units’ resource allocations are independent of the numbers generated by transfer prices, they’ll see the system as fair. It is a mistake to tie individual rewards to individual unit performance, as is done in competitive organizations. As in cooperative organizations, top management must allocate rewards—bonuses, praise, symbols of status, and so forth—subjectively to a certain extent.

A collaborative company’s approach.

One of the chemical companies in my study provides a perfect illustration of a collaborative organization. One of its major businesses is the largest producer in its industry. Its two largest customers are internal users. All three businesses are critical to the success of the company. These businesses are capital intensive and their managers accept internal sourcing without question. Transfer prices are market based, less a small discount to reflect the fact that sales effort and other expenses are not required.

The selling unit’s product comes in several grades and in several degrees of finish. The largest internal user buys the product in a degree of finish less than for that sold on the outside, which makes finding comparable market prices impossible. The oligopolistic nature of the market for the intermediate good and the extent of vertical integration in the industry also make it difficult to determine a fair market price and an appropriate volume discount, if any.

Disputes about the transfer price of this product have raged off and on for nearly 20 years. Recently, the argument heated up again. The price of the intermediate good rose on the external market and the selling unit general manager set a new transfer price on the good internally. The result was a conflict that generated a series of memos and internal studies providing a great deal of information. In presentations to the president, the buying unit managers reported their performance using both cost- and market-based transfer prices.

The CEO has managed all three of these businesses at different points in his career and understands them very well. When managing these businesses he had shifted to a cost-plus-investment approach because he said he wanted to improve the profits and motivation of the buying units. Others speculated that doing so helped him gain information he needed for capital allocation negotiations. On becoming CEO, he shifted back to a market-based approach for these businesses.

All three managers of these businesses are key members of his management team. They report to group general managers (the selling unit manager to one and the buying unit managers to another) and so are one layer removed from the president. These group general managers are due to retire soon.

The conflict the market-based approach has generated gives the CEO the necessary information and control he needs to be sure that the right balance between vertical integration and diversification is maintained. This is especially important as the company increasingly emphasizes the selling unit’s role as an internal supplier over its role as an external supplier. Another strategic concern is the poor performance of one buying unit due to industry conditions. At the time I did the study, the company was about to make a major capital investment to lower this unit’s costs and make it more competitive.

None of the managers feels he is being treated unfairly by the chief executive officer. All of them think that he fairly evaluates their performances and that they get the recognition and financial compensation they deserve. They also think that the appropriate capital investments are being made in their businesses.

The CEO has made special efforts to let managers know that he recognizes their individual contributions to the company. What’s more, his word carries weight because he receives much information and also has a deep understanding of all three businesses. Because the company is placing more emphasis on measuring unit profitability and return on investment, the CEO’s knowledge and judgment are especially reassuring to the unit managers.

Managers feared that others in the company who lack the president’s depth of understanding would underestimate the contributions their businesses make; such feelings of unfairness were directed toward each other. Perceptions of unfairness go far beyond mere financial compensation to include pride, concern with others in the unit, and personal values about what is just. In this company these feelings were exacerbated by the distrust that existed between the managers of the buying and selling units due to differences in personal style.

Using the MAP

Managers can use the MAP, which associates organizational characteristics (including transfer pricing policies) appropriate for various combinations of the two strategic dimensions, to analyze their existing transfer pricing policies. This analysis takes five steps, each of which requires substantial managerial judgment to complete. Because the strategic relationships between business units vis-à-vis vertical integration and diversification depend on the business units involved, this analysis is best performed on a product (or group of similar products) basis.

Step 1: Locate inter-business-unit relationships on the MAP.

Define measures of diversification and vertical integration appropriate to the company. Place inter-business-unit relationships on the MAP according to strategies of business units and strategy for their relationship.

Step 2: Compare location to existing transfer pricing policy.

Determine what transfer pricing policy is most appropriate according to Exhibit III. The exhibit summarizes the transfer pricing policies appropriate for varying degrees of emphasis on interdependence through vertical integration and independence through diversification. Compare to existing transfer pricing policy, and consider a change in either the strategy or the pricing policy if there isn’t a fit.

Exhibit III Transfer pricing and strategic objectives

Step 3: Analyze organizational characteristics.

Determine which organizational characteristics from competitive, cooperative, or collaborative types (Exhibit II) are most appropriate given the corporate strategy. Compare to existing organizational characteristics; consider making some changes in organizational characteristics if there isn’t a fit.

Step 4: Look for special problems.

If strategy, transfer pricing policy, and organizational characteristics are consistent but managers still complain, they may do so because of one of the five special problems noted in the next section. Resolution of the conflict will depend on its underlying source and will have little to do with transfer pricing.

Step 5: Consider future strategy.

Anticipate that changes in strategy may require changes in transfer pricing policy. Make a decision about the timing of this change. A change in transfer pricing policy prior to a change in strategy may signal and facilitate this new strategy, but expect and accept problems during the transition. Or a change in policy may occur after the new strategy is in place; the inevitable problems that will result as the old policy becomes increasingly inappropriate for the new strategy can help spur the change.

While providing a useful framework for thinking about transfer pricing, the MAP must be supplemented by management judgment. For example, the MAP enables a manager to determine if conflict is useful, as it is in collaborative organizations, or not, as is more likely in competitive or cooperative organizations. But by itself the MAP does not enable a manager to determine if the conflict is a result of other factors having little to do with transfer pricing.

Common Transfer Pricing Problems

Even if the company has a transfer pricing policy appropriate for its strategy, five common problems can lead to conflict. Here, too, management processes are far more important than any attempt to find technical solutions.

1. Performance problems.

When either unit is having trouble meeting its profit objectives, changing the transfer price is one of the easiest fixes for the bottom line. If both businesses are doing poorly, as in times of recession, conflict will be especially severe, but at these times no change in the transfer price approach will alleviate the difficulties since the underlying problems are due to external factors that are independent of the transfer price.

2. Interpersonal disputes.

Disputes over transfer prices may simply be the mechanism by which managers act on an interpersonal conflict unrelated to transfer prices. Arguing over transfer prices seems a more legitimate way of conducting this dispute than name-calling. The interpersonal and the substantive issues can get confused.

3. Power imbalance.

A unit manager who feels there is a power imbalance due to his greater dependency on the other unit (such as when a buying unit sources a product internally that is a significant part of his cost of goods sold but in amounts insignificant to the selling unit) may initiate conflict to try to correct this imbalance.

4. Demand fluctuation.

When demand for the intermediate good is high, the buying unit is glad to have assured supplies on a full-cost basis that will be much below market price. This is precisely the time, however, when external customers are especially attractive to the selling unit. The reverse situation occurs when demand for the intermediate good is low and outside suppliers are willing to sell at market prices below full cost.

5. Product pricing.

A complaint about market-based transfer prices is that the final good’s price becomes uncompetitive. The response to this is that because external suppliers of the intermediate good would receive a profit, profit should be earned at each stage of the production process. A complaint about cost-based transfer prices is that the final good will not be priced aggressively enough, resulting in lost profits or pursuit of market share at the expense of profitability. The response to this is either that vertically integrated competitors supply the intermediate good on a cost basis or that external suppliers have lower cost structures due to specialization and lower overhead and can make a profit at a lower price than would be required for internal transfers. In both of these cases, the problem requires recognition that product pricing involves issues other than cost. It is a complex problem in and of itself, and its analysis includes estimates of customer value and competitors’ prices in addition to costs set up by transfer prices.8

Although no simple solution to the transfer pricing problem exists, it can be effectively managed. The ability to do so rests on recognizing that the nature of the transfer pricing problem varies according to a company’s (group’s, unit’s, or product’s) location on the MAP and the direction in which management wishes to move. Effective management of transfer pricing requires policies and organizational characteristics that are consistent with strategy. Appropriate processes are essential for achieving the desired objectives of fairness and control.

References

1. Two exceptions are Robert N. Anthony and John Dearden, Management Control Systems, 4th ed. (Homewood, Ill.: Richard D. Irwin, 1980); and Joel Dean, “Decentralization and Intracompany Pricing,” HBR July–August 1955, p. 65.

2. They also show only limited use of variable cost. See Richard F. Vancil, Decentralization Managerial Ambiguity by Design (Homewood, Ill.: Dow Jones-Irwin, 1979); Roger Y. Tang, Transfer Pricing Practices in the United States and Japan (New York: Praeger, 1979), and Roger K. Mautz, Financial Reporting by Diversified Companies (New York: Financial Executives Research Foundation, 1968).

3. The view of corporate strategy used here is taken from Kenneth R. Andrews, The Concept of Corporate Strategy, 2d ed. (Homewood, Ill.: Richard D. Irwin, 1980).

4. An often used classification of diversification strategy that essentially combines a specialization ratio and a relatedness ratio and a single dimension is that of Richard P. Rumelt, Strategy, Structure, and Economic Performance (Boston: Division of Research, Harvard Business School, 1974).

5. Although in theory division general managers are autonomous and have the authority to control their own destinies, a study of nearly 300 companies found that profit center managers’ responsibility nearly always exceeds their authority. For fuller discussion of these concepts, see Richard E.Vancil, “What Kind of Management Control Do You Need?” HBR March–April 1973, p. 75.

6. See Richard E. Walton and David McKersie, A Behavioral Theory of Labor Negotiations (New York: McGraw-Hill, 1965).

7. For a discussion of how capital investment decisions are actually made, see Joseph L. Bower, Managing the Resource Allocation Process (Boston: Division of Research, Harvard Business School, 1970).

8. For further discussion of the product pricing problem, see E. Raymond Corey, Industrial Marketing: Cases and Concepts, 3d ed. (Englewood Cliffs, N.J.: Prentice-Hall, 1983); Benson P. Shapiro and Barbara B. Jackson, “Industrial Pricing to Meet Customer Needs,” HBR November–December 1978, p. 119, and Frederick E. Webster, Jr., Industrial Marketing Strategy (New York: John Wiley & Sons, 1979).

A version of this article appeared in the November 1983 issue of Harvard Business Review.

Which pricing method is useful when the selling division is operating below capacity?

Variable cost-based pricing approach is useful when the selling division is operating below capacity. The manager of the selling division will generally not like this transfer price because it yields no profit to that division. In this pricing system, only variable production costs are transferred.

Which method of transfer pricing considered when the supply division is a monopoly producer?

Opportunity Cost Such pricing may also be required where the supplier division is a monopoly producer or the user division is a monopoly consumer. The transfer price may be fixed at a level which equal the opportunity cost of the supplier division and the user division.

What is cost

Cost-based transfer pricing is a method of setting prices when selling products to divisions within the same company. Several factors affect the price, including: Production costs. Managers' reviews. Taxation.