AgraCity’s Canada transfer pricing dispute: co-distribution and the markup for logistics

By Dr. Harold McClure, New York City

The Tax Court of Canada, on August 27, ruled in favor of the taxpayer in AgraCity Ltd. v. the Queen. This transfer pricing litigation involved affiliates owned by James and Jason Mann, headquartered in Barbados, Canada, and the US.

The key transfer pricing issues of the case concern the appropriate benchmarking for a logistics affiliate and the application of a co-distribution model pioneered by the Canadian Revenue Agency.

AgraCity and NewAgco  

The Manns established a procurement entity for farmers in the US and Canada called Farmers of North America. The specific product at issue in this litigation was ClearOut, a glyphosate-based herbicide, which is a generic version of Bayer-Monsanto’s RoundUp.

The Manns established New Agro US in 2005 as their US procurement entity. This affiliate sourced the herbicides from third-party producers and sold them to US and Canadian farmers.

AgraCity was established as a Canadian logistics affiliate for delivery to the Canadian farmers. NewAgco Barbados was established in 2006 to take over the role NewAgco US.

The overall operations were very profitable for a variety of reasons, including because the affiliates were able to buy in large volumes and because of regulatory requirements under the Pest Management Control Act which involves a detailed and complicated registration process if Canadian farmers to source herbicides from foreign suppliers. These requirements provided opportunities for the sourcing affiliate to reap high profits by navigating Canada’s restrictions to trade.   

Transfer pricing structure

The Canadian Revenue Agency did not object to the 2005 intercompany pricing between NewAgco US and AgraCity even though most of the consolidated profits resided with the US operations.

When these profits were later sourced to NewAgco Barbados, though, the Canadian Revenue Agency challenged the intercompany pricing, asserting that the Barbados affiliate had no employees and owned no assets. The Canadian Revenue Agency asserted that all profits sourced in the Barbados affiliate should be allocated to AgraCity.

The court was not convinced by this position and ultimately agreed with the taxpayer that AgraCity should be seen as a logistics affiliate.

Co-distribution

While the term “co-distribution” was not part of the court record, I discussed these considerations in an earlier paper (“Distribution Hubs, Sandwich Transactions, and the Co-Distribution Model,” Journal of International Taxation, October 2018), which provides a useful framework.

In 2005, the NewAgco US performed procurement functions, took title to the inventory, and owned any intangible assets.

The observation that the Barbados affiliate had no employees likely means that the procurement functions continued to be performed by the US affiliate, but certain assets were transferred to the Barbados affiliate.

The observation that the Barbados affiliate had no employees likely means that the procurement functions continued to be performed by the US affiliate, but certain assets were transferred to the Barbados affiliate.

The court decision specifically noted the possibility that NewAgco Barbados had secured a quasi-monopoly on supply, implying that it brought value to the transactions.

The change in the intercompany structure in 2006 affected the allocation of consolidated profits between the US affiliate and the Barbados affiliate, but it did not involve any changes to the functions and assets for AgraCity. AgraCity’s functions were limited to providing logistic services in Canada, which were compensated through a services agreement.

The court ruling narrowed the transfer pricing controversy to whether compensation for providing logistic services was adequate, noting:

The Appellant had reported approximately two million dollars of service fees (calculated at the agreed 15 and 20 cent per litre rates) earned by it under the Services Agreement in the period in question. It is that amount that the reassessments impugn as far too low for the value of the services provided and as an amount far less than arm’s length parties would have agreed to … The services provided by AgraCity pursuant to the Services Agreement were largely, though not exclusively, logistical in nature. They were also largely subcontracted by AgraCity to arm’s length third parties at arm’s length rates, eg transportation, warehouse, customs brokers, container recycling and disposal.

 Benchmarking the markup for a logistics affiliate

As a part of the court proceedings, economist for AgraCity, Brad Rolph, prepared an analysis which compared the ratio of operating profits to total costs for AgraCity to the return to total costs for third party publicly-traded logistic companies.

However, return to total costs could be misleading, and return to value-added expenses would be more convincing. I noted in a separate paper published by the Journal of International Taxation in its September 2016 issue:  

[T]he ratio of operating profits to total costs (z) overstates the ratio of operating profits to value-added costs (m) depending on the proportion of pass-through costs relative to total costs (p), while the ratio of operating profits to operating expenses (w) understates the return to value-added costs if cost of goods sold includes certain value-added expenses (s > 0): m = z/(1 – p) and m = w[(1 – s – p)/(1 – p)].

The court in AgraCity notes Rolph’s analysis:

Mr. Rolph calculated AgraCity’s return on costs on a comparable basis as approximately 30%. In his testimony, he acknowledged that it would be fair to regard it as more appropriately being about 12% because in his report he had calculated return on costs after backing out from AgraCity’s costs … Mr. Rolph’s report looked at available data on returns, margins or mark-ups on costs earned by North American corporations that are known to provide broadly similar logistics services. He looked in particular at large public corporations, many in the US, and at large Canadian freight forwarding, brokerage and agency companies. Based on these available numbers, such companies would earn in the range of 5 to 15% returns over their costs … Based on this, Mr. Rolph’s opinion was that AgraCity’s returns were at the higher end of the range of comparable cost plus returns from the available data he could obtain, and that AgraCity’s service fee under the Services Agreement was a reasonable rate of return for its primarily logistics services. This evidence is clearly not without its limitations. Mr. Rolph himself described it as “just a crude benchmarking analysis”.

Table 1 compares the financials for AgraCity to the financials for two hypothetical logistic companies.

To capture what the court record is suggesting for AgraCity, we have assumed that the intercompany payment represents 19.6 cents per liter, with 2 million liters being transported.

Pass-through costs represent 10.5 cents per liter, and value-added expenses represent 7 cents per unit. Total costs are 17.5 cents per unit, so operating profits are 2.1 cents per unit, which represents a 12 percent markup over total costs.

Since pass-through costs represent 60 percent of total costs, profits relative to value-added expenses are 30 percent.

Table 1: Financials for AgraCity and two hypothetical third-party logistics companies

	Per unit	Thousands	Third-party A	Third-party-B
Revenues	$0.196	$392	$107.5	$110.0
Pass-through costs	$0.105	$210	$75.0	$75.0
Value-added expenses	$0.070	$140	$25.0	$25.0
Operating profits	$0.021	$42	$7.5	$10.0
Return to total costs	12.0%	12.0%	7.5%	10.0%
Return to value-added expenses	30.0%	30.0%	30.0%	40.0%

Table 1 also presents two hypothetical third party logistic companies where both have pass-through costs representing 75 percent of total costs.

Third-party A receives a 7.5 percent markup over total costs, while third-party B receives a 10 percent markup over total costs.

AgraCity’s markup over total costs is higher than the markup over total costs for either third-party logistics company.

When third-party profitability is measured in terms of operating profits relative to value-added expenses, third-party A has a markup equal to 30 percent, while third-party B has a markup equal to 40 percent.

Table 2: Key financials for two logistic companies

Millions	C. H. Robinson	Knight Swift
Revenues	$15603	$5094
Pass-through costs	$13050	$1780
Value-added expenses	$1727	$2813
Operating profits	$826	$501
Return to total costs	5.59%	10.91%
Return to value-added expenses	47.83%	17.81%

Table 2 presents the recent financials for two third party logistic companies that often appear in transfer pricing analyses.

Logistics firm C.H. Robinson, from 2017 to 2019, had average annual revenues over $15.6 billion, with total costs of nearly $14.8 billion.

C.H. Robinson’s markup over total costs was only 5.59 percent. Pass-through costs represented 88.31 percent of total costs. When measured as a return to value-added expenses, its markup was 47.83 percent over this period.

Logistics companies Knight Transportation and Swift Transportation merged on September 8, 2017.

Table 2 presents the average annual financials for Knight-Swift Transportation Holdings over the 2018 and 2019 period.

Its markup over total costs was 10.91 percent, which was higher than the markup over total costs for C. H. Robinson. Its markup over value-added expenses was only 17.81 percent, as pass-through costs represented 38.75 percent of total costs.

The observed return to total costs for logistics companies depends on both the return to value-added expenses and the proportion of total costs that represent pass-through costs.

The court decision does not identify the third-party logistic companies used in Rolph’s analysis. Had his analysis examined the return to value-added expenses, it is likely that it would have supported AgraCity’s 30 percent return to value-added expenses as reasonable.

AgraCity’s transfer pricing

The key issue in the AgraCity was the functional analysis with respect to which entities bore the various responsibilities and held the key assets in a co-distribution structure that sourced herbicides from third-party suppliers on behalf of Canadian farmers.

The facts demonstrated that most of the functions and assets were attributed to either the US affiliate or the Barbados affiliate.

The Canadian affiliate’s sole responsibility was logistics services. This affiliate received compensation that afforded it with a return to value-added expenses equal to 30 percent.

While a return to total costs analysis is not necessarily convincing, an analysis based on the return to value-added expenses for North American publicly traded logistic companies would likely support this markup as being reasonable.

Editor’s note: This article was corrected on 9/4/2020 fix an error table 1.