The US and Mexico’s transfer pricing approach for maquiladoras

By Dr. J. Harold McClure, New York City

The US and Mexico have agreed to renew their transfer pricing framework for US multinational enterprises that have maquiladora operations, according to a November 16 IRS announcement.

The new agreement will apply through tax year 2019, allowing a US taxpayer to avoid double taxation on the contract manufacturing and assembly functions performed by its maquiladora if the Mexican taxpayer enters into a unilateral advance pricing agreement with the Mexican tax authority according to terms negotiated in advance between the competent authorities.

The IRS notes that a new qualified maquiladora agreement with Mexico maintains the key elements of a 2016 agreement, which itself updated a 1999 agreement.

The new agreement and the ones before it are not publicly available; however, there has been some public discussion of the contents of the 1999 agreement.

Return to capital approaches for contract and toll manufacturers

Maquiladoras date back to 1965 with the introduction of the US Border Industrialization Program, which allowed for duty-free importation of intermediate components and machinery and equipment used to assemble products for exports back to the US parent corporation.

With the introduction of NAFTA, the free trade aspect of this program became less of a concern while transfer pricing issues emerged.

Anthony Barbera and John Wilkins co-authored a 1995 Coopers & Lybrand paper, which was revised in 1998 and published in the Transfer Pricing Handbook by John Wiley & Sons (“Arm’s Length Markups for Maquiladoras”).

Both versions of the paper argued that the markup over the labor costs of the maquiladora should reflect a return to capital times the fixed asset to labor cost ratio of the maquiladora. While the 1998 version claimed that this ratio was low, the 1995 paper admitted that maquiladoras employ a substantial amount of machinery and equipment relative to labor costs.

A November 14, 2001, paper by Robert Kirschenbaum and John McLees addressed the 1999 maquiladora agreement between the US and Mexico. The authors noted that maquiladoras often employ assets worth at least four times the annual operating costs recorded by the maquiladora (“New Maquiladora APA Guidelines: Are We There Yet?” BNA Transfer Pricing Report).

This asset employed to cost ratio equal to 4 includes all fixed assets and inventory, regardless of which affiliate owns each asset class relative to labor costs.

Contract versus toll manufacturing structures

Consider, for example, an electronic manufacturing services client that produces cable boxes for its US parent where: labor costs = $125 million per year; component costs = $1125 million per year; inventories = $250 million; machinery and equipment = $200 million, and plant and property = $50 million.

Table 1 presents the financials for a contract manufacturer that takes title to the components under arm’s length pricing. Intercompany (I/C) revenue is set at $1.3 billion to cover $1.25 billion in total costs and provide $50 million in profits.

These profits represent a 10 percent return to assets, which is consistent with a 4 percent markup over total costs as the asset to cost ratio is 40 percent.

Table 1 also presents two transfer pricing policies for a toll manufacturer. The toll manufacturer incurs only labor costs and holds only fixed assets.

Table 1: Contract versus toll manufacturing structures

Millions	Contract manufacturer	Toll manufacturer aggressive	Toll manufacturer arm’s length
I/C revenue	$1300	$130	$150
Component costs	$1125	$0	$0
Labor costs	$125	$125	$125
Profits	$50	$5	$25
Inventory	$250	$0	$0
Fixed assets	$250	$250	$250
Return on assets	10%	2%	10%

Some transfer pricing practitioners have suggested an aggressive approach where the markup over labor costs is only 4 percent for a toll manufacturer even though the conversion to a toll manufacturing status eliminated 90 percent of the cost base but only 50 percent of the assets.

Note this aggressive toll transfer pricing policy is consistent with a return on assets of only 2 percent.

An arm’s length transfer pricing policy would be based on a return to assets that is at least 10 percent. As such, the markup over labor costs must be at least 20 percent as the fixed asset to labor cost ratio is 200 percent.

In my discussion of the conversion of contract manufacturing structures to toll manufacturing structures, I note this example is based on the assumption that the return to inventories equals the return to fixed assets (“Made in China, Sold by Hong Kong: Processing Trade and Transfer Pricing,” Journal of International Taxation, October 2017).

If the overall cost of capital is 10 percent, but the return to inventories is less than 10 percent, then the return to fixed assets and the implied markup over labor costs would be higher than we have assumed.

Toll manufacturer with an off-balance-sheet twist

Maquiladoras often own only the portion of fixed assets that represent plant and property (P&P), with the formal ownership of the machinery and equipment (M&E) being with the US parent.

In our example, the US affiliate formally owns 90 percent of the assets that were owned by the original contract manufacturing structure and incurs 90 percent of the original total costs by incurring the component costs.

The maquiladora is effectively engaged in off-balance-sheet financing for the machinery and equipment. This off-balance sheet financing twist has created a controversy between the IRS and Mexico’s Servicio de Administración Tributaria for the past 25 years.

The maquiladora is effectively engaged in off-balance-sheet financing for the machinery and equipment. This off-balance sheet financing twist has created a controversy between the IRS and Mexico’s Servicio de Administración Tributaria for the past 25 years.

Table 2 contrasts the profits under the contract manufacturer structure, the arm’s length version of the toll manufacturer structure, and four variations of the maquiladora structure.

Under the contract manufacturer structure, the manufacturing affiliate owns all fixed assets and inventory, representing $500 million in our example. If the return to assets = 10 percent, profits = $50 million under arm’s length pricing.

Under the toll manufacturer structure, the affiliate owns all fixed assets, representing $250 million in our example. If the return to assets = 10 percent, profits = $25 million under arm’s length pricing

Some practitioners were frustrated that the appropriate markup over labor costs was 20 percent, hoping to argue for the original 4 percent markup over total costs under the original structure.

The ability to hide the machinery and equipment from the balance sheet provided an opportunity to argue for this lower markup if one is willing to ignore basic financial economics.

The 1995 Coopers & Lybrand report did just that by asserting that a maquiladora deserved none of the expected return from fixed assets formally owned by the US parent.

The third column of Table 3 presents the application of the Coopers & Lybrand (C&L) model to our example.

Table 2: Alternative profit allocations for a Mexican manufacturer affiliate

Millions	Contract manufacturer	Toll manufacturer	Maquiladora – C&L model	Maquiladora – economic model	Maquiladora – safe haven	Maquiladora – APA model
P&P	$50	$50	$50	$50	$50	$50
M&E	$200	$200	$200	$200	$200	$200
Inventory	$250	$250	$250	$250	$250	$250
Total assets	$500	$500	$500	$500	$500	$500
Return to P&P	10.0%	10.0%	10.0%	10.0%	6.9%	9.0%
Return to M&E	10.0%	10.0%	0.0%	4.0%	6.9%	1.5%
Return to inventory	10.0%	0.0%	0.0%	0.0%	6.9%	3.5%
P&P profits	$5.0	$5.0	$5.0	$5.0	$3.5	$4.50
M&E profits	$20.0	$20.0	$0.0	$8.0	$13.8	$3.00
Inventory profits	$25.0	$0.0	$0.0	$0.0	$17.3	$8.75
Total profits	$50.0	$25.0	$5.0	$13.0	$34.5	$16.25

 

The Coopers & Lybrand approach is inconsistent regarding which party bears commercial risk when one entity utilizes assets formally owned by another entity.

The role of risks is noted in the BEPS Discussion Draft on Action 9 (paragraph 63):

Risks should be analyzed with specificity, and it is not the case that risks and opportunities associated with the exploitation of an asset, for example, derive from asset ownership alone. Ownership brings specific investment risk that the value of the asset can increase or may be impaired, and there exists risk that the asset could be damaged, destroyed, or lost (and such consequences can be insured against). However, the risk associated with the commercial opportunities potentially generated through the asset is not exploited by mere ownership.

Financial economics would suggest that a maquiladora would receive a higher expected return to the assets it formally owns as it is also using an asset owned by another entity and bearing the commercial risk but not the ownership risk.

Financial economics would suggest that a maquiladora would receive a higher expected return to the assets it formally owns as it is also using an asset owned by another entity and bearing the commercial risk but not the ownership risk.

The seminal paper on the economics of leasing was written by Merton Miller and Charles Upton (“Leasing, Buying, and the Cost of Capital Services”, Journal of Finance, June 1976).

Leasing is an alternative to purchasing assets with debt financing.  In fact, financial leases are seen as equivalent to owning the asset via debt financing in the sense that the lessee incurs all operating risks whereas the lessor receives a guaranteed return over the life of the asset.

Miller and Upton note, however, that the lessor bears the risk of obsolescence when the lease term is shorter than the economic useful life of the asset.  They also provide a capital asset pricing model for the expected return to the owner of the asset:

R_j = R_f + b_l(R_m – R_f ),

where R_f = the risk-free rate, R_m = the expected return on the market portfolio of assets, and b_l = the asset beta coefficient for a pure play leasing company.

As an example, let R_f = 4 percent and R_m = 5 percent. If b_l = 0.4, then the premium for bearing obsolescence risk is 2 percent and the expected return for a leasing company would be 6 percent.

If the overall expected return for an asset is 10 percent, then the lessee deserves a 4 percent expected return on those assets for bearing commercial risk.

The economic model in Table 2 assumes that the maquiladora receives a 10 percent return on the plant and property that it owns and a 4 percent return on the machinery and equipment that it utilized via off-balance-sheet financing.

In our example, its profits under arm’s length pricing would be $13 million. These profits represent a 26 percent return to the assets the maquiladora owns but only a 2.6 percent return to all assets, including the machinery and equipment and the inventories owned by the US parent. This approach suggests a 10.4 percent markup over labor costs in our example.

The Mexican safe harbor approach

The low markups suggested by the Coopers & Lybrand approach were seen as aggressively low by the Mexico tax authority.

The Mexico tax authority proposed a variety of approaches to this issue that would result in higher markups. An approach allowed for a safe harbor if the profits of the maquiladora were equal to the greater of 6.5 percent of costs or 6.9 percent of all assets, whether formally owned by the maquiladora or the US parent.

For capital intensive operations, 6.9 percent of all assets would be the relevant safe harbor option.

While a 6.9 percent return to the fixed assets formally owned is less than most estimates of the return to capital, applying a 6.9 percent return to fixed assets financed by off-balance-sheet financing is higher than what our economic model suggests.

If this 6.9 percent were also applied to the US parent’s inventory, then the profits under this option would be considerably higher than the profits under our economic model.

In our example, the safe harbor approach results in $34.5 million in profits or a 27.6 percent markup over labor costs.

Comparison to the 1999 agreement

Kirschenbaum and McLees detailed the elements of the rather complex method for determining the income of a maquiladora under the 1999 agreement.

While they describe an involved 7-step process, the key elements are a form of return to capital approach.

The first step would be to determine an appropriate return to capital for the asset formally owned by the maquiladora. While we have been suggesting a 10 percent return, the 1999 approach suggested a 9 percent return by an examination of publicly-traded contract manufacturing firms in the electronics sector.

Steps 2 and 4 assigned a return to inventory even though it is held by the US parent as the difference between this 9 percent return and a short-term interest rate.

This difference was approximately 3.5 percent. These steps also assigned a return to machinery and equipment as the difference between this 9 percent return and the interest rate on 20-year corporate bonds with a credit rating of BBB. This difference was approximately 1.5 percent.

Step 3 considered the percentage of assets financed by debt for the third party contract manufacturing firms in the electronics sector. The other steps had minor effects and were even more obscure in their reasoning.

If we focus on steps 1, 2, and 4, the results are such that the 9 percent return on plant and property yields $4.5 million in profits, the 1.5 percent return on machinery and equipment formally owned by the US parent yields an additional $3 million in profits, and the 3.5 percent return on inventory formally owned by the US parent yields an additional $8.75 million in profits.

Overall profits implied by this approach = $16.25 million for a markup over labor costs = 13 percent.

Note that this result is higher than the result from the application of the economic model even though we have used a lower return to capital for plant and property owned by the maquiladora and a much lower expected return for the maquiladora as it utilizes machinery and equipment formally owned by another entity.

Overall profits are higher because this approach affords the maquiladora profits attributable to the inventory owned by the US parent.

It is not clear how the 2016 and 2020 agreements compare to this 1999 agreement as there are no publicly available presentations describing the newer approaches.

Concluding comments

The controversy over transfer pricing for maquiladoras has been long and, at times, unnecessarily complex.

The conversion from a contract manufacturing structure to a toll manufacturing structure is common around the world, with the enforcement of appropriate markups over labor costs being led to by tax authorities in Canada and China.

Mexican tax authorities have dealt with a more aggressive twist as maquiladoras finance their machinery and equipment with off-balance-sheet financing.

A straightforward economic model can be applied to toll manufacturing with an off-balance-sheet twist. Alas, multinationals representatives often adopt more aggressive positions based on questionable economics dressed up as transfer pricing. The Mexico tax authority struck back with its safe harbor provisions that often suggested much higher Mexican income.

The tax incentives of US parents with maquiladora operations have evolved over time.

Initially, most US parents preferred less income in Mexico even if it meant more US taxable income. Whirlpool Financial Corporation v. Comr. [154 TC. No. 9, May 5, 2020] had an interesting discussion of the tax incentives in the following years.

Mexico had lowered its effective tax rate on maquiladoras to 17 percent in the hope of encouraging multinationals to source more income in Mexico.

Whirlpool, however, relied on a structure where a Luxembourg affiliate acted formally as the contract manufacturing but assigned the actual toll manufacturing responsibilities to its maquiladora. The reasoning was that any Luxembourg sourced income would face no taxes. The IRS, however, successfully argued that this income represented subpart F income and would be taxable in the US.

In my recent MNE Tax article on the Whirlpool decision, I addressed the transfer prices paid to the maquiladora as a toll manufacturer relying on off-balance-sheet financing for its machinery and equipment.

While the Tax Court record does not provide any details to the financial information or the transfer pricing approach used, I explore the implications of both toll manufacturing and off-balance-sheet financing using standard economic models for these issues.

Similarly, we do not know the precise approaches used in the 2016 and 2020 agreements between the IRS and the Mexico tax authority. A comparison of these approaches to the results that would be derived under sound economics would be a useful contribution to hopefully providing clarity to this long-debated issue.