Tax this. Photographer: David Paul Morris/Bloomberg
Tax this. Photographer: David Paul Morris/Bloomberg

There are lots of reasons to shop at Amazon.com Inc., though one stands out: Many customers don't have to pay state sales taxes, saving them as much as 10 percent on everything from toilet paper to televisions. New research confirms that this taxpayer subsidy is worth a lot, although the evidence also suggests that changing tax rules may not help brick-and-mortar stores compete against online retailers.

Even before the age of Amazon, some shoppers avoided sales taxes by crossing borders. It only takes about 40 minutes to drive from Boston, where sales taxes are 6.25 percent, to New Hampshire, which has no sales tax, so it isn't surprising that many consumers do so. The drive from Seattle to Oregon is a bit longer but the sales tax savings are even bigger at 9.5 percent. Scholars have found that consumers even cross international borders when the tax difference is big enough. (Here is a paper about Singaporeans shopping in Malaysia and here is one about Swedes buying alcohol in Denmark.)

Other research shows that U.S. shoppers are very sensitive to the tax impact of online shopping. In the late 1990s, Austan Goolsbee estimated that "applying existing sales taxes to Internet commerce might reduce the number of online buyers by up to 24 percent." Last year, scholars studied eBay Inc.'s online marketplace, which doesn't reveal where sellers are from until after consumers express an interest in buying a specific item. They found that "a one percentage point increase in a state's sales tax leads to an increase of just under 2 percent in online purchasing, and a 3-4 percent decrease in the volume of online purchases from home-state sellers."

Until now, however, no one has looked specifically at the benefits that accrue to Amazon. Recent changes in many state tax laws provide us with an interesting natural experiment. The new paper, from economists at Ohio State University, looks at about 245,000 households that spent at least $100 at Amazon in the first half of 2012 and tracks their daily spending through the end of 2013. Slightly more than a third of those households live in the five states that compelled Amazon to collect sales taxes during those two years (California, New Jersey, Pennsylvania, Texas and Virginia).

The average Amazon customer in affected states cut spending at the online retailer by 9.5 percent after local taxes went up, relative to spending in the other 45 states that didn't raise taxes. The effects are even bigger for big-ticket items, with purchases of $300 or more falling by a staggering 23.8 percent.

Much of that money went to Amazon competitors that already collected sales taxes. The economists found that competing retailers had sales increase of 3.4 percent, with much of those gains coming from greater online purchases rather than more buying from stores. Amazon's competitors saw sales increase by 9.2 percent when shoppers were spending $300 or more. Shoppers also shifted their spending to vendors on the Amazon Marketplace, almost none of which collect tax but all of which pay fees to Amazon. These smaller operators experienced sales increases of 15.2 percent overall and a whopping 60.5 percent for big-ticket items.

Imposing state and local sales taxes on Amazon purchases would be good for municipal budgets and help level the playing field for other online retailers, but do little for brick-and-mortar establishments. The question is whether years of tax subsidies for Amazon ultimately benefited the rest of us by nurturing a company with the resources to invest in new products, distribution channels and lower prices. Would online shopping have taken off without the tax benefit?

To contact the writer of this article: Matthew C. Klein at mklein62@bloomberg.net.

To contact the editor responsible for this article: James Greiff at jgreiff@bloomberg.net.