Use Weighted Averages to Determine Transfer Pricing

A prize-winning paper presents a real-world method for transfer pricing that incorporates both management and tax considerations.

October 01, 2004

| by Frederick Rose

How should companies price goods that they ship between their own divisions correlated companies?

Internationally, that quandary confronts the producers of nearly half of all U.S. imported goods, a third of all U.S. exports, and a huge proportion of global trade elsewhere. By definition these aren’t arms-length deals in an open marketplace, and they raise tough questions. What price, for instance, might Ford of Germany charge a Ford division in Mexico for German-made engines that are installed in Mexican-assembled cars?

For nearly half a century textbooks in management and accounting have given a standard answer to this question based on a classic analysis by the economist Jack Hirshleifer in 1956. He showed that the best economic result occurred either at a market price for the product being shipped or,failing that, the marginal cost of the item to the division making it.

But Hirshleifer’s approach does not take account several complicating factors that are pervasive in multinational corporations: differences incorporate income taxes and limited knowledge by the firm’s central management. For example, corporate income taxes are higher in Germany than in Mexico. Ford might want, then, to put a low transfer price on its German-built engines in order to produce lower taxable profits in Germany and higher ones in Mexico. But managers in Germany might feel cheated when their bonuses suffer with the German division’s profits. They might reduce emphasis on producing these export engines.

In a prize-winning research paper, Stefan Reichelstein, the William R. Timken Professor of Accounting at Stanford GSB, with co-authors Tim Baldenius and Nahum Melumad, both of Columbia University, have developed a real-world answer to transfer pricing that balances both the economic criteria confronted by Hairsplitter the puzzle of international tax rates. “Integrating Managerial and Tax Objectives in Transfer Pricing,” published in The Accounting Review, concludes that a relatively simple weighted average is the management optimum for transfer prices. The conclusion, arrived at through series of mathematical models, promises a new management tool for thousands of firms around the world that confront these problems.

For the moment, most companies determine a single set of transfer prices,driven primarily by the goal of minimizing overall corporate taxes. The authors contend that approach has ignored other hugely important areas,from management incentives in foreign divisions to allocation of production capacities and guidance for future capital investment.

“Transfer pricing has been a somewhat neglected area,” systemizes in an interview. “People recognize that this is a big issue, but what most people think about is transfer pricing as a tax optimization issue,” he says. “Yet, transfer prices are management tools. They have an important function to facilitate decision-making, to tell certain regional or country managers what the value or price of some intermediate product is and use that information to maximize the profit of the company as a whole. That is the economic function of transfer pricing.”

The separate worlds of tax folk and management planning types “even splits the accountants,” Reichelstein notes, and creates separate industries. “The tax accountants look on pricing as entirely compliance issue,” he says. Meanwhile, management accounting consultants are preoccupied with transfer prices for both internal allocations and public reporting purposes. These are big and growing businesses in their own right: By one report (in The Economist),employment of transfer pricing experts tripled in the past few years in the top four British accounting firms.

Reichelstein favors maintaining separate prices for management purpose sand for taxation, which he says shouldn’t be done to sidestep tax laws but to aid a wide range of management tasks. It’s legal to maintain two sets of prices. But corporate lawyers and tax accountants have worried that separate management prices might be misunderstood and bring adverse results with government tax auditors and in the court of public opinion. Yet, “there are great gains through better corporate management and planning to be unlocked here by having the flexibility of different approaches,” Reichelstein says.

In their paper, which was awarded the Chazen International Research Prize by Columbia’s Graduate School of Business, Reichelstein and his co-authors demonstrate that the optimal internal transfer price for management purposes should be a weighted average of the producing division’s pretax incremental cost and the most favorable arms-length price admissible for tax reporting purposes.

For example, let’s say that a company division in Ireland, a low-tax country, has concluded that on the margin the pretax cost of producing widget is $10. At the same time, the various valuation methods allowable by the tax authorities show that $20 per widget is an admissible arm’s-length price for tax purposes. Suppose that Irish tax rate is 20percent. The optimal transfer price for management purposes is determined by putting 80 percent of the weight on the pretax cost of $10 and 20percent on the $20 tax-efficient answer. The resulting weighted average transfer price is $12 per widget.

The approach assumes that separate prices will be used for internal and tax reporting purposes. “Once the transactions have happened, the economic coordination problem no longer exists and then transfer prices are irrelevant - just an accounting construct,” says Reichelstein.”It’s just the opposite for the tax side, where we need to report value after the transaction has occurred.”

For media inquiries, visit the Newsroom.

Explore More