Figuring out transfer pricing: How to set terms of negotiation between departments

Research offers insight into balancing competing interests inside a company.
By Andrew Kenney

Figuring out transfer pricing: How to set terms of negotiation between departments
Image by tacktack/iStock

The business world has been fixated lately on the question of how to put a price on a product. Countless pricing strategists, consultants, and even chief value officers are employed now in search of the numbers that sell products at a profit.

Clara Chen, an associate professor of accountancy at the College of Business at Illinois, University of Illinois at Urbana-Champaign, and her research team examined a unique pricing function: What happens when a company sells something to itself?

Transfer pricing comes into play when the output of one division becomes the input of another division of the same business. It's perhaps best known as a tax-minimization strategy, but transfer pricing also is a battleground for competing interests within the company. And when those interests aren't balanced properly, the company as a whole loses out.

The research group surveyed 210 divisional managers and built a predictive model to show how companies in their sample are handling this challenge. Their work focuses largely on the question of how companies distribute power, asking how leadership can minimize conflict between managers while maintaining an efficient and effective process. (The full research results are available in the article "Determinants and Consequences of Transfer Pricing Autonomy: An Empirical Investigation," by Clara Chen, Shimin Chen, Fei Pan, and Yue Wang, Journal of Management Accounting Research, Fall 2015, Vol. 27, No. 2, pp. 225—259, available at, login required to view the full article.)

In a recent interview, Chen talked about how companies can enable better decisions on pricing and how that advice applies to the business at large. The following transcript has been edited for length and clarity.

What can go wrong in transfer pricing?

Chen: This is a very interesting setting because there are conflicts of interest between the buying division and the selling division.

The division managers are often compensated based on divisional profit. The selling manager has an incentive to charge a higher price, and the buyer has an incentive to get a lower price. Sometimes, if they can't reach an agreement, that internal transfer doesn't happen, and the buying division may buy externally, which could decrease the profit of the business as a whole.

How can we tell if companies are setting these prices correctly?

Chen: To answer that question, we need to first understand what companies want to achieve with transfer-pricing policies. The first objective is to make sure that the policy is going to provide information that corporate headquarters can use to evaluate whether a divisional manager is doing a good job.

We also want the policy to encourage division managers to make decisions that are in the best interest of the firm as a whole, not just in the best interest of the division. We want to provide information to division managers. If the transfer price is distorted, it will also trickle down to the division managers' other decisions.

Finally, companies also want to minimize tax obligations.

How did you do your research?

Chen: To test our predictions, we collected data using a survey of general managers from divisions belonging to multidivisional firms in China. We asked them questions related to our variables. Our main variable of interest is how much autonomy the divisional managers have in transfer-pricing decisions. We also asked them about how fair they think the process is, how effective they think the pricing decisions are, among other questions.

Can we apply these findings outside of China?

Chen: Although our results are based on data from China, our theory should apply to other areas such as the U.S. ... However, the right level of autonomy in transfer-pricing decisions and the specific factors that go into determining the optimal level of autonomy in transfer-pricing decisions may vary across regions.

What patterns did you find in how companies distribute the power to set prices?

Chen: Agency theory, which we draw upon in our study, generally shows that there are two broad sets of factors that influence how companies decide on the level of centralization.

The first is information asymmetry: What is the difference in information between the firm and the divisional manager? How much does the divisional manager know that the firm's leadership does not know? The bigger that asymmetry, the more decision rights you should give to divisional managers because they have the information to make a good decision.

The second set of factors is goal congruence between the divisional manager and the firm. How aligned are the interests between the individual manager and the top management? When the individual managers are also compensated based on firm-level performance, or stock price, then there is a greater goal alignment between the divisional manager and the firm because the division manager also wants the firm to do well. And when we see that happen, we see a greater delegation of decision rights to the divisional manager.

How much autonomy should leadership delegate to management to set these prices?

Chen: It depends. For example, in a firm where there's greater goal congruence between the middle manager and the top manager, it's OK to grant autonomy. But in other situations, where you're not so sure about the congruence, it's probably not a good idea to give too much autonomy to the middle manager.

How much autonomy should be given will depend upon the other elements of the management control system, which includes the design of the performance measurement and control system. The whole measurement and control system should be viewed as a system with many interdependent practices.

How do these findings apply beyond transfer pricing?

Chen: We do have several predictions that can apply to other settings. When you have too much or too little autonomy for divisional managers, divisional managers are going to perceive the process as unfair. And, as a result, the decisions are going to be less effective.

Sometimes, if the firm management intervenes too much in those transfer-pricing decisions, divisional managers may feel that it's unfair; they feel that their decision rights are undermined. So, they may be upset that top management is intervening too much in their decision. The reason we have those divisions is because we wanted to allow the managers to have some autonomy to make decisions.

The divisional managers have the best knowledge and the best information about how to operate their divisions. But when they have too much autonomy, the consequence is the divisional managers may not make decisions in the best interest of the firm.

What was the final product of your research?

Chen: We have a prediction model about how much autonomy should be provided, given the firm's circumstances. The circumstances include factors such as: What is the degree of standardization of this intermediate product? What is the degree of interdependence between the divisions? How much uncertainty is the firm facing? How much weight is put on firm-level performance?

So we include all of those variables, throw those into a model, and based on that model, we can predict how much autonomy should be given at certain firms. The difference between this predicted level of autonomy and the actual level of autonomy in a firm is our empirical measure of mismatch. We show that mismatch reduces transfer-pricing effectiveness.

What practical methods can companies use to resolve price conflicts between departments?

Chen: Sometimes, there is a market price for that intermediate product, and you can observe whether the transfer price diverges from the market price. But when the market price is not available, you can still look at the product itself: What is the cost? What are the true costs of that product?

Also, if a company wants to assess whether they've been doing this right, whether they've been granting the right amount of autonomy, they could survey their divisional managers periodically to see whether they're happy about the process, whether they have sufficient autonomy or too much autonomy. This would be a great approach to get some feedback from the divisional managers.

About the author

Andrew Kenney is a JofA contributing editor based in Denver.

About the researchers

Clara Chen is an associate professor of accountancy at the College of Business at Illinois, University of Illinois at Urbana-Champaign. The rest of the team is Shimin Chen, the Zhu Xiaoming Chair in Accounting and associate dean, China Europe International Business School; Fei Pan, professor and deputy dean, School of Accountancy, Shanghai University of Finance and Economics; and Yue Wang, an associate professor at the School of Accountancy of the Shanghai University of Finance and Economics.

To comment on this article or to suggest an idea for another article, contact Jack Hagel, editorial director, at or 919-402-2111.

AICPA resources

CPE self-study

  • International Taxation (#732012, text)
  • Transfer Pricing (#165376, online access)

For more information or to make a purchase, go to or call the Institute at 888-777-7077.

Where to find June’s flipbook issue

The Journal of Accountancy is now completely digital. 





Better decision-making with data analytics

Data analytics has become a hot topic, but many organizations have not yet managed to understand its potential, let alone put it to work. This report will take a deep-dive on how to best introduce or enhance the use of data in decision-making.