Since the birth of internet commerce, online transactions across state lines have not been subject to the sales taxes payable on traditional retail transactions. Following a 1992 ruling by the US Supreme Court, it is only if an online retailer has a physical presence in a state – a store, an office or a warehouse – that it is obliged to collect sales tax from customers resident in that state.
There are growing demands for this to change. With internet retail in the United States now worth well over $100 billion dollars annually and the majority of transactions occurring across state lines, forgone taxes on those sales could amount to as much as $10 billion a year. Such revenues would be particularly welcome at a time of pressure on state budgets. Taxes on internet commerce would also be popular with store-based retailers struggling to compete with their online rivals.
So what is likely to be the impact of sales taxes on online purchasing behavior? Our research seeks to estimate the potential effects by analyzing data from eBay’s online marketplace. The marketplace is large and diverse, with millions of buyers and a huge array of sellers and product categories. We take advantage of this size and diversity to observe buyers choosing across sellers located in different states, with correspondingly different tax regimes and changes in those regimes over time.
Our estimates rely on three sources of variation in sales taxes. The first is the difference, for online buyers, between in-state purchases that are taxed and out-of-state purchases that are not. But it is important to recognize that a direct comparison of intra-state and inter-state purchases may understate the effect of taxes if consumers prefer ‘home-state’ goods or sellers.
One way to address this issue is to use variation across states in tax rates, and compare customers’ relative propensity to make intra-state purchases across low tax and high tax states. This research strategy is aided by the considerable variation in tax rates: in January 2010, state sales taxes ranged from zero (in Alaska, Delaware, Montana, New Hampshire and Oregon) to 7% or more (in California, Indiana, Mississippi, New Jersey, Rhode Island and Tennessee).
The variation in tax rates becomes even greater after taking account of county and local sales taxes. A third source of variation comes from the frequent changes in state and local tax rates. We use these three sources of variation to estimate the effect of sales taxes along different decision margins.
Our first approach exploits the fact that most people shopping on eBay only observe a seller’s location – and hence the relevant sales tax – after they have clicked on a listed item. We use data from millions of these location ‘surprises’ to estimate the tax sensitivity of purchasing conditional on being interested in a given item.
This ‘micro’ approach allows us to control tightly for buyers’ preferences and the desirability of items located in different states. The results indicate that purchases by interested buyers fall by roughly 2% for every one percentage point increase in the sales tax charged by the seller.
Our second approach uses more aggregated data on sales from one location to another, which allows us to estimate how much a tax rise in one state encourages buyers to use sellers in other states. The results suggest that a one percentage point increase in a state’s sales tax leads to an increase of just under 2% in online purchasing from other states, and a 3-4% decrease in online purchasing from home-state sellers.
These finding are subject to some caveats. One is that the changes in US sales tax policy for internet commerce currently under discussion could have considerable effects on the decisions of online retailers about their pricing and the locations of their operations. These considerations are outside the scope of our analysis.
Another caveat is that the estimates come from a single online platform, which may not be fully representative. Nevertheless, we believe that eBay’s overall market share – roughly $30 billion annually – is sufficiently large that our analysis provides insights that can be extended more broadly across the online retail sector.
This article summarizes “Sales Taxes and Internet Commerce” by Liran Einav, Dan Knoepfle, Jonathan Levin and Neel Sundaresan. Liran Einav, Dan Knoepfle and TNIT member Jonathan Levin are at Stanford University. Neel Sundaresan is at eBay Research.
The preceding is re-published on TAP with permission by its author, Professor Jonathan Levin and the Toulouse Network for Information Technology (TNIT). “Sales Taxes and Internet Commerce” was originally published in TNIT’s July newsletter.