Arm's length principle misapplied | Toll manufacturing in TP

Dow Chemical Co., and Canada’s transfer pricing dispute: litigating pennies while ignoring a large toll manufacturing issue

The Tax Court of Canada, in a December 18 decision, resolved a procedural dispute between the Canadian tax authority and Dow Chemical’s Canadian affiliate involving just a few million dollars.

At issue was whether the court had jurisdiction to determine if the affiliate’s taxable base could be lowered to reflect an increase, agreed to by both the tax authority and the taxpayer, to arm’s length interest payments made by the Canadian affiliate to a Swiss affiliate. In a lengthy decision, Dow Chemical Canada ULC v. The Queen, 2020 TCC 139, the court determined it had jurisdiction.

While Dow Chemical won this round, the facts of the case reveal that the company already addressed a much larger dollar-value dispute over a different issue without even going to court.

The Canadian tax authority originally proposed to increase the Canadian affiliate’s income by more than CAD 300 million, disputing the markup applied to its toll manufacturing arrangement with the Swiss affiliate. The taxpayer resolved this issue in a competent authority negotiation between the Canadian and Swiss tax authorities. We can only speculate what the issue was and how it was resolved as the large increase in toll manufacturing income was not addressed in the court decision.

Dow Chemical Canada provided toll manufacturing services to its Swiss affiliate receiving CAD 5.93 million in profits in 2006 under their intercompany pricing policy.

The Canadian Revenue Service proposed a transfer pricing increase of CAD 307.23 million for 2006. The only mention of this wide divergence in views with respect to this toll manufacturing issue was that the taxpayer used its competent authority to resolve this issue.

A 50-fold increase in profits is a staggering level of disagreement.

Let’s imagine a situation where the toll manufacturer incurred CAD 600 million per year in labor costs for this controlled transaction. The taxpayer’s position would be consistent with a markup of only 1%. The tax authority’s position would be consistent with a 50% markup.

In my recent discussion of how the Chinese tax authority has addressed the toll manufacturing issue, I credit the approach pioneered by Dr. Ronald Simkover in his work for the Canadian Revenue Agency (“Made in China, Sold by Hong Kong: Processing Trade and Transfer Pricing”, Journal of International Taxation, October 2017).

Simkover argues that the return to total costs for a contract manufacturer that purchases components and pays for labor is not a reliable means for evaluating the appropriate return to labor costs for a toll manufacturer, even though many taxpayers pretend it is a reliable form of benchmarking.

Simkover believes that the markup over total costs (m) for a contract manufacturer should be seen as a weighted average of the return to value-added expenses and the return to pass-through costs:

m = x^.v + (1 − x)z

with x = ratio of value-added expenses relative to total costs; v = ratio of operating profits attributable to employing extensive fixed assets relative to labor costs; and z = ratio of operating profits attributable to modest working capital relative to pass-through costs.

Consider a Canadian affiliate that incurs CAD 600 million in labor costs plus CAD 5400 million in component costs, which implies x = 90%. Even if the return to total costs (m) should be a mere 1%, the premise that the return to value-added expenses (v) should be only 1% assumes that v = z.

There are actually two very aggressive assumptions inherent in this position, as we shall note.

Ascertaining what would be reasonable estimates for v and z would require understanding the composition of assets for the third-party, allegedly comparable company and providing estimates for the return to these assets.M

These markups can be seen as the product of asset intensities times the appropriate return to assets with v = R_f (fixed assets/value-added expenses); z = R_w (working capital/pass through costs); R_f = return to fixed assets; and R_w= return to working capital.

For simplicity R_f = R_w= 10%.

The 10-K filings for Dow Chemical suggest that the ratio of tangible operating assets to total costs is approximately 50%.

As such, let’s assume the Canadian contract manufacturer holds CAD 3,000 million in operating costs, including CAD 1,200 million in inventory and CAD 1,800 million in tangible fixed assets.

Even before the conversion from a contract manufacturer to a toll manufacturer, these assumptions would suggest CAD 300 million in profits or a return to total costs of 5% under arm’s length pricing. An intercompany policy that granted only a 1% markup over total costs would equate to a mere 2% return to operating assets.

The conversion to toll manufacturing reduces the cost base from CAD 6,000 million to only CAD 600 million and reduces the asset base from CAD 3,000 million to CAD 1,800 million. The taxpayer’s 1% markup over labor costs equates to a return to fixed assets of only 0.33 percent.

The position of the Canadian Revenue Agency that the markup over labor costs should be 50 percent equates to a return to fixed assets equal to 16.67%. This extreme position is consistent with a zero return to inventories.

The position of the Canadian Revenue Agency that the markup over labor costs should be 50 percent equates to a return to fixed assets equal to 16.67%. This extreme position is consistent with a zero return to inventories

A more credible position would be based on the assumption that R_f = R_w= 10%. Under these assumptions, the appropriate markup over labor costs would be 30% or profits = $180 million.

This toll manufacturing issue is a key transfer pricing issue in China as well as other jurisdictions, and the work of Ronald Simkover for the Canadian Revenue Agency pioneered the approach we have described.

The wide divergences between the taxpayer’s position and that of the tax authority often appear in other disputes.

While we do not know how this issue was actually resolved in the Dow Chemical Canada case, our discussion is an illustration of how good economics can be applied to the facts of such disputes.

By Dr. J. Harold McClure, New York City – MNE Tax