Welcome

INSIGHT: Pennsylvania Commonwealth Court Sustains Revenue’s Source Analysis

Sept. 17, 2020, 8:00 AM

A divided Pennsylvania Commonwealth Court en banc held that the Department of Revenue correctly sourced sales of services made in the 2011 tax year to Pennsylvania for purposes of the sales factor for corporate net income tax (CNIT), because the recipients of the services received the benefit of the services in Pennsylvania. Synthes USA HQ, Inc. v. Pennsylvania, No. 108 F.R. 2016 (Pa. Commw. Ct. Jul. 24, 2020) (5 to 2).

A taxpayer’s income from business performed across multiple states is apportioned for corporate net income taxes among those states. In 2011, business income was apportioned using a three-factor apportionment fraction, consisting of a property factor, a payroll factor, and a sales factor. To calculate the sales factor, the applicable statute required receipts from the sale of services to be sourced to Pennsylvania if:

(A) The income-producing activity is performed in this State; or

(B) The income-producing activity is performed both in and outside this State and a greater proportion of the income-producing activity is performed in this State than in any other state, based on costs of performance.

72 Pennsylvania Statutes Section 7401(3)2.(a)(17) (2011) (Section 17). The case involves two interpretations of this statute: The department treated the income-producing activity as the receipt by a customer of the benefit of a service and sourced the sales of services to where the services were delivered to the customer (the so-called “benefits-received” method), while the attorney general would treat the income-producing activity as the expense that a taxpayer incurs in producing the service and would source services to where that expense is incurred (the so-called “cost of performance” method).

The “benefits-received” method analysis is sometimes referred to as “transactional” sourcing analysis because the transaction between the customer and the taxpayer is viewed as the income-producing activity. The “cost of performance” method is sometimes referred to as “operational” analysis because it considers a taxpayer’s operations in determining its income producing activity and where the direct costs of performing that activity occurred.

Although the statute was later amended by the General Assembly to add a new Section 16.1 that expressly required service income to be sourced to Pennsylvania if the service is “delivered to a location” in Pennsylvania, the department recently interpreted the statute to source service income to the location where services are delivered both before and after the statutory amendment.

Synthes USA HQ, Inc. is a Pennsylvania-based corporation that provided services to affiliates located outside Pennsylvania. Synthes filed its 2011 CNIT return using the costs of performance method, which resulted in a significant tax liability because it incurred most of the expense in performing services to its affiliates in Pennsylvania. Synthes later learned that the department in 2011 applied the benefits-received method. It filed a request for a refund using the department’s method. The department denied the refund request due to evidentiary issues which, by the time the case reached the Commonwealth Court, were resolved.

On appeal before the Commonwealth Court, the attorney general argued that the department had misapplied the statute. The department intervened in the case to defend its interpretation of the statute. The court upheld the department’s benefits-received method. The court stated that the statute was ambiguous because it did not define the terms costs of performance or income-producing activity, and that the department’s view was entitled to deference, because it was charged with interpreting the statute, and the department stated that it had applied its interpretation consistently for a number of years.

It is not clear why the court felt that the term costs of performance is ambiguous. Costs or cost is a term that has been used extensively in both tax and accounting principles. It may be that in certain instances, there are interpretive issues. For example, for apportionment purposes, are the costs of making a telephone call the costs of just the individual transmission or a portion of the costs of operating the entire system? Courts have differed on the issue of what a cost is. In AT&T Corp. v. Dep’t. of Revenue, The Oregon Supreme Court found that sourcing cellular service revenue to the location of the customer that purchased the services to make calls on the cellular network was a cost. However, in AT&T Corp. v. Comm’r of Revenue, the Massachusetts Appellate Tax Board ruled cellular service revenue was sourced to office locations outside Massachusetts, reasoning that AT&T’s income producing activity was the operation of its global network, and the cost of performing that activity was primarily incurred at office locations outside Massachusetts where the employees operated and managed the network.

But under no circumstances should costs mean receipts. If a seller performs services and sells them to the buyer, costs of performance cannot mean the payment by the buyer to the seller because that is a receipt, not a cost. In the case of digital services, costs should not mean the expenses incurred by a seller of an actual transmission to a buyer. In the digital world, the cost of electronic transmission can be simply the push of a button. Costs in the statute is plural, implying in a particular sale costs can be in more than one state.

The department’s analysis muddles the distinction between a cost of producing a receipt and a receipt itself. Few jurisdictions have adopted the department’s view of what a receipt is. In DIRECTV, Inc. v. S.C. Dep’t of Rev., the South Carolina Supreme Court held that DIRECTV’s income producing activity was delivering a TV signal to customers in their homes.In Walter E. Heller Western, Inc. v. Arizona Dept. of Rev., the Arizona Supreme Court held that the income producing activity of interest income generated by loans made to Arizona customers was the Arizona customers’ payment of interest income.

As the Massachusetts Appellate Tax Board in AT&T Corp. suggested, this type of analysis assumes the source of income is where customers are located (i.e., where the service is delivered) instead of analyzing the cost to the taxpayer in performing its income-producing activity, as required by the statute. Services are not always performed in the same place where they are delivered, and, in fact, the failure of traditional cost of performance sourcing to capture the market where services are delivered has led many states to switch from the traditional costs of performance sourcing to market-based sourcing for sales of services. See J. Hellerstein & W. Hellerstein, State Taxation Paragraph 9.18[3][a] (WGL) (3d ed. 2001 & Supp. 2020-1). Pennsylvania had not yet made that switch in 2011.

The majority in Synthes stated that the department has consistently applied the “benefits-received” method. The history indicates otherwise. As early as 2004, the department interpreted Section 17 under the “costs of performance” methodology. The Pennsylvania Business Tax Reform Commission Report (Pa. Dept. of Rev., 11/30/2004) stated that the department’s position that “Pennsylvania assigns sales of particular services to the state in which the largest share of the costs were incurred to produce the service,” not where services are delivered, and recommended a legislative switch to market-based sourcing. Joseph Bright, one of the authors of this article, was a member of the 2004 Commission. In 2006, the department issued guidance applying this interpretation. In Pennsylvania Letter Ruling CRP-06-004 (Oct. 10, 2006), sales of intangibles (which were sourced under the same rule as services) were sourced under Section 17 to a taxpayer’s commercial domicile because that is “where a taxpayer generally manages and maintains intellectual property.” The department did not issue contrary guidance.

The court’s holding is also at odds with the legislative history of the statute. The statute was amended in 2013 to add Section 16.1 to apply market-based sourcing to sales of services. Act July 9, 2013, Public Law 270, No. 54 (Act 52). The legislative history of the amendment shows that the General Assembly intended the new statutory language to change the way the sourcing methodology for sales of services is implemented. When considering the bill that became Act 52, the Senate Appropriations Committee’s Fiscal Note described the sourcing change as one that would source receipts to where the benefit is derived by the customer, and that revenue generated from the change in the law would “primarily be from out-of-state businesses with PA-sourced service income”—i.e., those taxpayers less likely to incur the costs of performance in the state. See Senate Appropriations Committee Fiscal Note to H.B. 465 (July 3, 2013). The department, through its secretary, stated in its own testimony before the House Finance Committee that market-based sourcing will raise revenue for the Commonwealth and will benefit Pennsylvania-based service providers and increase taxes on out-of-state businesses. See Testimony of Dan Meuser, Secretary of the Pennsylvania Department of Revenue, before the House Finance Committee on April 11, 2013. The implication in the testimony is that the sourcing methodology would change with the enactment of Act 52.

The majority’s reasoning that the legislature’s amendment of the statute clarified, rather than altered, the application of the statute, is questionable. The majority relied on Gilligan v. Pa. Horse Racing Comm’n for the position that a court may give strong weight to an administrative agency’s longstanding interpretation of a statute. However, in Gilligan the statute was amended without pertinent change. The applicability of this case seems strained, since in Synthes the legislature did amend the statutory scheme to change the language applicable to sourcing services by adding a new section directly on that point. The court’s reasoning also overlooks that the department changed its interpretation of the statute, and assumes that the department’s current interpretation of the statute is correct. Further, as Judge Wojcik points out in his dissent, this interpretation is at odds with the longstanding principal that legislative amendments to statutes demonstrate legislative intent to amend their application.

The court also discussed with disapproval the attorney general’s advocacy of a position opposite to that of the department, but did not directly address the position raised by Judge Wojcik in dissent that the Commonwealth Attorneys Act confers authority on the attorney general to present a legal position contrary to a Commonwealth agency.

Exceptions were filed by the attorney general’s office to the court’s opinion. We hope that the court en banc will consider the legislative history and reconsider the department’s methodology. If Synthes stands, it would validate an incorrect interpretation of the statute, impede the attorney general, and undermine the effect of legislative changes to Pennsylvania statutes.

This column does not necessarily reflect the opinion of The Bureau of National Affairs, Inc. or its owners.

Author Information

Joe Bright is a member and Heidi Schwartz is an associate at Cozen O’Connor in Philadelphia.

To read more articles log in. To learn more about a subscription click here.