The growth of cloud computing is one of the most significant commercial developments facing modern consumption tax regimes. The growth is significant in part for the problem it presents: tax regimes designed for the consumption of goods and services transferred in a physical world struggle to adapt to virtual transactions. Often the analysis of this problem has focused on the what and the where of cloud computing. Tax authorities often have difficulty characterizing cloud computing offerings as either goods or services, and that characterization can drive tax consequences. Additionally, given the virtual nature of cloud computing, it can be difficult to figure out where the consumption takes place (or even where the offering originates from) and thus who has the right to impose tax; how many places have you accessed Spotify from?
In Cloudy with a Chance of Taxation, Bill Hader and Anna Faris invent a machine that causes cloud computing offerings to materialize and fall from the sky, solving both problems with family-friendly quirkiness and hilarity. If only. Actually, Rifat Azam and Orly Mazur’s article steps up the analysis of taxing the consumption of cloud computing. The authors expose additional challenges that cloud computing presents for modern consumption tax regimes, particularly value added tax regimes, and offer recommendations to enable those tax regimes to keep pace with cloud computing and changing consumption habits. Tax enthusiasts will find their well-written article just as captivating as meatballs falling from the sky.
In addition to the what and the where problems, Azam and Mazur detail the registration and compliance difficulties surrounding the taxation of cloud computing offerings and the fraud and avoidance opportunities presented by the uniqueness of cloud computing offerings. Because VATs are generally applied on a destination basis, imports are subject to tax but exports are not. Compliance in the case of cross-border transactions is easy enough when borders exist, but cloud computing’s virtual existence threatens to overly complicate compliance in this context. Taxpayers may lack efficient means of determining where they must register to collect and remit the VAT. Additionally, the cloud computing model offers opacity to taxpayers wishing to shirk their obligations and mobility to taxpayers wishing to route their consumption through low- or no-tax jurisdictions.
The core of Azam and Mazur’s recommendations is community. To bring VATs up to speed, nations will need to work with each other and with anyone they can reach that is connected with the cloud computing transactions. This means having nations entering into enforcement and tax-sharing “VAT Administrative Unions” with other taxing nations to address compliance and registration difficulties. It also means developing smart contracts and blockchain systems to bring taxable transactions to the attention of taxing authorities in real time. And most revolutionary, it means requiring payment intermediaries (i.e., credit card companies) to collect VAT on the transactions they process payments for.
Azam and Mazur argue that their solutions provide more benefits than they impose costs. In particular, requiring payment intermediaries to collect VATs is highly beneficial because those intermediaries have the data and connections necessary to collect and remit VATs more efficiently than other entities. The authors’ arguments for the benefits of their proposals are compelling, but it would be interesting to see the costs developed a bit further. For instance, in the case of the payment intermediaries, just how large would the economies of scale be from centralizing all of the VAT compliance to the payment intermediaries? In other words, how flat are VAT compliance costs generally? Would the proposal simply shift the same costs from businesses to the intermediaries or would there be significant net savings? In either event, who should be asked to cover the costs: the businesses, the intermediaries, or the governments? Similarly, what enticements might larger market counties need to agree to enter into VAT Administrative Unions? These questions hint at some of the political problems the proposals would likely encounter in a second (or third or fourth . . .) best world.
That said, the will to overcome any political problems may exist as governments continue to lose out on tax revenue because of cloud computing. The US states offer examples to prove this point. Over half of the states have signed onto the Streamlined Sales and Use Tax Agreement, which the Wayfair Court noted approvingly as a means of simplifying sales and use tax administration. However, the SSUTA member states represent only a third of the US population, pointing to the problem of convincing the large market states to join. Perhaps Wayfair will push those states over the edge, but the SSUTA (along with the European MOSS regime noted by the authors) does provide a potential model for the VAT Administrative Unions. In addition, many US states have begun passing laws requiring marketplace facilitators like Amazon and eBay to collect sales and use taxes on transactions by third-party vendors made on their platforms. These laws are controversial and not completely similar to the payment intermediary proposal, but they may provide a roadmap (or a cautionary tale) for how to effectively shift VAT compliance burdens to those intermediaries. As with many issues of multijurisdictional taxation, the states of the US and the nations of the world may have much to learn from each other.