Back
to A New Way to Shop: Kentuckians Find the WebBy Peter Schirmer, Kevin O'Neil, and Michael T. Childress(*)
From Collecting Taxes in the Cyberage
p. 17-22, published 1999
The requirement to pay use tax on out-of-state Internet and catalog purchases is widely disregarded. In fact, a recent report conducted by Ernst & Young assumes that only “4 percent of taxable business-to-consumer e-commerce sales result in sales and use tax payments.”(1) Table 1 shows that an estimated $71.7 million in sales and use tax was owed on out-of-state consumer “mail order” purchases (which include Internet purchases) in 1997.(2) Yet the state collected only $712,000 in use taxes from consumers (and an unknown amount in sales tax from businesses).(3)
Table 1: Summary of Mail Order Purchases and Use Tax Collections in 1997
Internet commerce has received so much attention from the media and from policymakers that many people probably believe the Internet has surpassed more traditional forms of mail-order purchases, but that is hardly the case. According to Marketing Logistics, Inc., which conducts research for the National Mail Order Association, Internet sales accounted for less than 1 percent of total mail-order sales in 1997, which topped $318 billion.(4) Indeed, it is conventional wisdom that the bulk of lost revenue is due not to Internet sales but to mail-order sales. In late 1998, Information Week magazine ran a story entitled, “Myths and Realities,” about Internet commerce. In it, the author wrote:
Even by the most generous accounts, online retail sales remain only a tiny fraction of what’s sold in physical stores or through mail-order catalogs—even in the Web’s most popular product categories. Online book sales will account for less than 5% of all U.S. book sales this year, according to Keenan Vision. Online music sales? Less than 2%. Even online travel sales—which will reach $1.8 billion this year, leading all consumer products (except IT products) sold online—won’t even reach 1% of the $488 billion in total U.S. travel spending. Web-based advertising revenue also remains minuscule compared with broadcast and print—just 0.4% of ad agency bookings this year.(5)
But concern in Kentucky and across the country about lost sales tax revenue is not prompted by the size of the problem today, but rather by the potential size of the problem a few years from now. Estimates vary considerably (even, interestingly, when they come from the same company), but they all suggest that the value of Internet sales will be many times higher in a few short years than it is today. Forecasts from perhaps the best-known industry research firm, Forrester Research, are frequently cited, and no two are the same; but they are similar in spirit to one cited by iCompanyStore: “According to Forrester Research, online (business-to-consumer) sales in the U.S. are expected to generate over $108 billion by 2003 compared to expected revenue of $7.8 billion in 1998.”(6) Forecasts of growth rates and sales volumes from other market researchers are similarly eye-popping. Thus, there is widespread belief, if not fear, that while lost revenue may constitute a small percentage of the general fund today, it could be considerably larger in ten or even five years.
Yet any assumption that state treasuries will be barren as a result of Internet commerce is far more equivocal, for at least two reasons. First, it is unclear to what extent Internet commerce will spur economic growth that will, in turn, generate more government revenue from other sources, such as income taxes and business licenses. Second, the extent to which Internet commerce will actually replace traditional store-based commerce is unknown. Both points deserve closer examination.
Without going into too much detail, it is clear that the national and the state economies are strong. Nationally, gross domestic product growth for much of the 1990s has been at or above the long-run average; interest rates and unemployment are at 30-year lows; incomes and productivity are rising; prices are stable. In Kentucky, more people are working, and the gap between state and national per capita income is narrowing. Such economic prosperity is reflected in the relative health of the federal and state budgets. The federal budget has a surplus for the first time since 1969, and the state general fund, which is required by law to be balanced, over the past five years has had a total of $330 million more than forecasters anticipated. Table 2 shows the year-to-year underestimates of general fund revenue, dating back to FY94. Note that the consensus forecasts listed in this table are the last, best forecasts made before the respective fiscal years.
Table 2: General Fund Estimated and Actual Amounts, FY94 - FY98
Moreover, sales tax revenue, which had fallen from 38 percent of general fund revenue in 1976 to 30 percent in 1991, has rallied to 33 percent today. Part of this increase is certainly due to the rise in the sales tax rate from 5 to 6 percent in 1991, but even after 1991 sales tax revenue has grown faster than overall general fund revenue. This is somewhat counterintuitive, given that the economy is largely service-based and has become increasingly so during the 1990s. Most services are not taxed in Kentucky. One explanation for the higher-than-expected rise in sales tax revenue is the rapid increase in consumer purchases during the 1990s. As shown in Table 3, growth rates in personal income and retail sales were approximately the same between 1982 and 1992. However, between 1992 and 1996, retail sales grew much more rapidly than did personal income. If food sales (the vast majority of which are not taxed) are excluded, the growth rates in income and retail sales are even more disparate.
Table 3: Average Annual Growth Rates in Kentucky
Much of the credit for this “new economy” we are enjoying today is given to the Internet and, more generally, to information technology. The argument is that information technology makes workers and businesses more productive, and rising productivity enables businesses to hire more workers and pay higher wages without having to raise prices. Thus, although Internet commerce may cost Kentucky some sales tax revenue, it may also be responsible for raising incomes and spurring more traditional economic activity from which the state does receive revenue, as well as lowering spending on social support programs.
In today’s strong economy, people are buying more things: in stores, from catalogs, and over the Internet. Although states would surely like to be able to tax the more than $100 billion worth of consumer goods purchased by mail order, the more immediate concern for states is the extent to which mail order purchases will eat into store-based purchases in the coming years. Even though traditional retail sales are growing quickly, mail order sales are growing even faster.
Every five years, the federal government conducts an economic census that measures store-based retail sales, but presently it does not track Internet sales. Companies such as Market Statistics, which estimates retail sales for non-census years, have some fairly reliable guides. However, companies such as eMarketer have no such guides when estimating Internet sales, so estimates vary, sometimes widely. Therefore, the precise numbers in Table 4 are less important than the overall message, with which all estimates agree: Internet sales are growing much, much faster than store-based retail sales or other forms of mail-order sales.
Table 4: Average Annual Growth Rates Nationally
Although Internet sales are small today, if they continue to grow at their current rates, they will eventually make a considerable dent in store-based retail. At that point, sales tax revenue growth begins to slow and states will have to make up for lost revenue elsewhere, or else they will have to slow spending growth.
To examine the potential effect that Internet sales could have on Kentucky’s state budget, we start by looking at the estimated sales and use tax owed on catalog and Internet purchases from 1998 to 2003.(7) We juxtapose catalog sales with three growth scenarios for Internet sales, to provide a context within which to evaluate the magnitude of Internet sales.
Table 5 contains at least three key points. First, the estimated tax owed from catalog and Internet purchases amounts to millions of dollars, regardless of the growth scenario. Second, the estimated tax owed from Internet sales is considerably less than the estimated tax owed from catalog sales, particularly in the near term. Third, the growth rates for Internet sales are considerably higher than those for catalog sales. For example, in the medium growth Internet scenario, the tax owed from Internet purchases was about 1.6 percent of the tax owed from catalog sales in 1998, but it increases to 19.2 percent by 2003. This growth illustrates the extent to which Internet sales could pose a growing problem for the state budget. Depending upon how much of this money is remitted to the state coffers in the form of a sales and use tax, policymakers face a considerable loss of revenue.
Table 5: Estimating the Amount of Sales and Use Tax Owed from Mail Order Consumer Purchases in Kentucky, 1998-2003
We have estimated the amount of lost sales and use tax from consumer Internet purchases by taking into account a number of factors:
Projected total Internet sales
Kentucky’s lower retail consumer expenditures (relative to the US average)
Kentucky’s lower Internet usage rates (compared to the US average)
Substitution effect (some purchases would have taken place as catalog or telephone purchases in the absence of the Internet)
Taxability (not all items purchased over the web are taxable)
Use Tax Consumer Compliance Rate
Figure 1 shows the estimated lost sales and use tax from Internet consumer purchases from 1998 to 2003 based on three scenarios that reflect a range of estimates for the factors listed above (see Appendix B for a full explanation). Projected annual losses range from $2.2 million to $46.4 million by 2003. And these amounts range from 0.08 percent to 1.8 percent of Kentucky’s projected sales tax revenue in 2003 (see Figure 2). While the percentages seem small, the actual dollar losses are not. For example, the estimated cumulative losses from 1998 to 2003 are $6 million for the low estimate, $50 million for the middle estimate, and $115 million for the high.
Figure 1: Alternative Scenarios for Estimating Lost Sales and Use Tax Revenue from Consumer Internet Purchases in KY
Figure 2: Sales and Use Tax Losses from Consumer Internet Purchases in KY as a Percentage of Total Sales Tax Revenue
The loss of sales and use tax from consumer purchases over the Internet is a small, but growing, problem. Our estimates suggest that in 2003 the annual loss to Kentucky’s budget could range between $2.2 million and $46.4 million, which would approach 2 percent of Kentucky’s sales tax revenue. Studies by Austan Goolsbee of the University of Chicago and by the accounting firm Ernst and Young confirm our assessment. Both studies estimate that nationally states currently lose less than 1 percent of their potential sales tax revenue due to Internet commerce and that in the short term, Internet sales do not present a major threat to state budgets.(8),(9) Goolsbee estimates that even if current Internet growth rates continue, losses will still amount to less than 2 percent of expected state sales tax collections. He goes on to point out, however, that if current growth rates continue through 2007, losses could reach 10 percent of state sales tax revenues. His predictions fit with our assessment that Internet commerce will not cause serious fiscal problems before 2003, but could grow to be a problem soon thereafter.
Clearly though, significant uncertainties underlie these projections. So in the next chapter we examine these projections from another perspective. We consider the following question: how much would Kentucky consumers have to purchase over the Internet to achieve the sales and use tax losses discussed in this chapter? This analysis will allow us to ground the estimated losses presented in this chapter to actual consumer behavior and ask whether such spending behavior is plausible.
Back
to A New Way to Shop: Kentuckians Find the Web
Ahead
to Internet Commerce Estimates: A Reality Check
* Peter Schirmer was a policy analyst with the Kentucky Long-Term Policy Research Center from 1994 to 1999. While at the Center, he authored, co-authored or edited eight book-length reports on topics ranging from rural Kentucky in the global economy to the role of information technology in the delivery of governmental services. He has a B.A. in Political Science from the University of Kentucky and an M.A. in Public Policy from the University of Michigan. Currently, he is working on an M.B.A. at Georgetown University.
Kevin O'Neil got to know the world of Internet commerce while a summer intern at the Kentucky Long-Term Policy Research Center. He currently studies Economics and English Literature at Swarthmore College in Pennsylvania. Kevin’s other experiences include study abroad in Spain, summer work with the Lexington-Fayette Urban County Government, and volunteer work as a firefighter and teacher of English as a second language.
Michael Childress is the Executive Director of the Kentucky Long-Term Policy Research Center. He received a Bachelor of Arts degree from the University of Kentucky in 1984 and a Master of Arts degree from the University of California, Los Angeles, in 1986-both in political science. From 1988 to 1993, he was a social scientist at the RAND Corporation in Santa Monica, California. While at RAND, he authored numerous studies on topics ranging from demographic trends in the third world to the implications of declining budgets for the U.S. Army. In August of 1993, he became the first executive director of the Kentucky Long-Term Policy Research Center, where he oversees the Center’s numerous research activities. Return to text.
Robert J. Cline and Thomas S. Neubig, “The Sky is not Falling: Why State and Local Revenues were not Significantly Impacted by the Internet in 1998,” (Ernst & Young Economics Consulting and Quantitative Analysis, 1999): 7. The U.S. Advisory Commission on Intergovernmental Relations reports a higher rate of 16.5 percent in Taxation of Interstate Mail Order Sales: 1994 Revenue Estimates. Return to text.
According to data from Market Statistics and reported in the Statistical Abstract of the United States, Kentucky’s retail purchases per capita were 93.8 percent of the national average. Therefore, we assumed that mail order consumer goods purchases in Kentucky were also 93.8 percent of the national average, giving us the figure of $1,378 million (rounded to $1.4 billion in Table 1). After the sales tax rate was raised from 5 to 6 percent in 1991, sales tax revenue has been about 5.2 or 5.3 percent of total retail sales. The estimate of $71.7 million in potential revenue is 5.2 percent of $1,378 million. Return to text.
Some businesses collect the 6 percent sales tax from the consumer and remit it to the state. However, because of the way in which these taxes are remitted to the state, we do not know the amount related to mail order purchases. Return to text.
Only $101.7 billion were sales of consumer goods. Consumer services accounted for $67.8 billion, business-to-business sales were $85.3 billion, and sales to charities were $63.8 billion. Data available online at http://www.nmoa.com. Return to text.
Clinton Wilder, “Myths and Realities,” Information Week, December 7, 1998. Available online at http://www.informationweek.com/712/12iumyt.htm. Return to text.
iCompanyStore. Available online at http://iCompanystore.com/ecommerce.htm. Return to text.
The tax owed from catalog sales is based on a conservative estimate of 4.1 percent annual growth in per capita sales in Kentucky. The three Internet projections are based on estimates provided by DMA, Forrester Research, and International Data Corporation. Refer to Appendix B for additional details. Return to text.
Austan Goolsbee and Jonathan Zittrain, “Evaluating the Costs and Benefits of Taxing Internet Commerce,” National Tax Journal, Sept 1999: online, University of Chicago Web site, http://gsbwww.uchicago.edu/fac/austan.goolsbee/research/jzntj.pdf, accessed on 21 July 1999. Return to text.
Cline and Neubig, 9. Return to text.