By: Leandra Lederman
Usually we think of tax shelters and other tax strategies as the province of private parties. These shelters may involve accommodation parties, even foreign government infrastructure, such as transportation systems, but we tend to think of private parties as getting the tax benefits. We may not think as often about a subnational government bolstering its tax revenues at the expense of the national government, particularly via a cooperating private party’s transaction structure. But that’s what happened a few years ago in Spain.
There is a Volkswagen (VW) plant in Pamplona, a city in the autonomous community of Navarra. From 2007-2011, Navarra reportedly collected approximately 1.5 billion Euros in value-added tax (VAT) from Volkswagen for its cars manufactured at the plant there. If VW-Navarra (which is a subsidiary of SEAT) had shipped the cars directly from Navarra to Germany, presumably Navarra would have had to refund that VAT. (Cars shipped to Germany leave Spain “clean of VAT”* (translation mine)).
Instead, according to an interview with Prof. Fernando de la Hucha in this El Diario article, the basic structure was that VW-Navarra sold the cars (although without physically moving them there) to a related Barcelona company, VAESA (Volkswagen-Audi España S.A.), which is located in the Catalunya region, not Navarra. VAESA then sold them to SEAT with the very low mark-up of 5 Euros per car. SEAT, which is also in Catalunya, then sold them to VW-Germany—the transfer abroad triggering entitlement to a refund. But because the cars were sold from a city outside the Navarra region, VW’s refund claim did not go to Navarra. Instead, the Spanish national government was the one that issued the refund, which is how Navarra benefitted. (Catalunya did not issue the refund because, unlike Navarra, does not have a fiscal agreement with Spain that allows it to administer and collect taxes—only Navarra and the Basque regions do). The result was that Volkswagen was refunded the taxes it paid but Navarra profited at the expense of the Spanish government. (Spain has a credit-invoice VAT. Technically, the amount that Navarra retained was the VAT that VW-Navarra paid, which was the VAT on its sales to VAESA minus the VAT its suppliers had paid.)
Here is a simple diagram of the transaction, along with a map of Spain’s regions. (Navarra is in the north, bordering France; Catalunya—that’s the Catalan spelling—is in the northeast, also bordering France.)
The Spanish government raised the issue when it audited VAESA. It sought to collect from the Navarra government the taxes Spain had refunded. According to Prof. de la Hucha in the story linked above, VAESA had no employees, and Spain “considered VAESA a fictitious company, which only served to keep Navarra from refunding the VAT.” (Translation mine.) One source reported (as translated by Google): “To get an idea of the magnitude of the case, that amount represents 4.5% of GDP in Navarra.” The Spanish national government apparently pursued arbitration and filed an appeal from there to the Spanish Supreme Court. The parties settled when the Spanish government agreed to drop its appeal and not collect any of the 2007-2011 amounts at issue and the parties agreed to the adoption of a dispute-resolution mechanism in the Economic Agreement between Navarra and Spain.
The transaction Prof. de la Hucha described in the El Diario article linked above sounds like VAESA was acting as an accommodation party, in that VW could have sold the cars abroad from Navarra rather than from Catalunya, which would have required Navarra to refund the VAT it had collected. VW is a very important taxpayer for Navarra, and the transaction’s structure apparently benefited Navarra substantially over the 5 years in question (apparently by less than 1.5 billion Euros due to a fiscal adjustment mechanism in the law, but by at least 600 million Euros); one would therefore think it would have had capacity to make some concessions to VW. However, there is nothing in the public record that suggests that, and I have been told that VW didn’t receive anything in return—it simply did Navarra a favor.
It is interesting to consider transactions in the U.S. that are somewhat analogous. The “tax cannibalization” that David Gamage & Darien Shanske’s “Tax Cannibalization and Fiscal Federalism in the United States” discusses is similar. In the same vein, state charitable tax credits that apply to a deductible charitable contribution effectively result in a transfer of federal revenues to the states, as Phillip Bankman & Kirk Stark describe in “Too Good to Be True? How State Charitable Tax Credits Could Increase Federal Funding for California.” I have heard that in a country with equalization grants that are based on tax revenues, a sub-central government (SCG) may be able to engage in tax competition that includes giving a large tax break to a particular company to induce it to locate there, then benefit from grants from the central government, although I haven’t found a source on this. (On the other hand, the OECD explains (on p.16) that fiscal equalization can “reduce the incentives for SCGs to lower tax rates and to attract mobile tax bases since additional revenues are compensated by lower grants . . . .”)
SILO and LILO transactions involving U.S. cities transferring assets such as transit systems to private parties are done for federal tax benefits, although they’re not the same thing because the tax savings flow to the private party—the city is the accommodation party. Yankee Stadium’s use of Payments In Lieu of Taxes (PILOTs) to obtain financing that is tax-exempt under IRC section 141 arguably is analogous. The linked article explains: “This serves as a huge benefit because the bonds are exempt from city, state, and federal taxes, and have an interest rate about 25 percent below that of taxable bonds.” Thus, with bondholders getting tax-exempt payments, New York City is able to pay much less bond interest. The City also avoided charging property taxes by owning the land itself; the Yankees are a lessee. The Yankees were required to make payments (the PILOTs) to the City, in an amount calculated to pay off the bonds. The Yankees thus benefited from the lower interest rate the city could pay due to the federal tax exemption.
There are variations on these themes and other structures subnational governments can use to benefit at the expense of the federal fisc. What seems unusual about the Navarra VAT situation is that it called for a particular corporate structure by a private-party taxpayer in order to work. I would be quite interested to hear if anyone is familiar with other transactions analogous to the one in Navarra, either in the U.S. or elsewhere.
*Many of the linked sources are in Spanish. Right-clicking within the text of such a source should give an option to translate to English.