With the lack of transparent pricing, it can be tricky for retail investors to know whether they are receiving the best available pricing on municipal bonds bought or sold on their behalf.
Sometime in the next 24 months, e-commerce purchases are likely to reach 10 percent of all U.S. retail sales (over $500 billion). Online sales are up from just 6 percent in 2010 and 3 percent in 2000, and forecasters expect e-commerce to represent 17 percent of retail sales by 2022.1
Given its growing market share, e-commerce is likely to impact state and local credit quality. Online transactions generate less sales tax revenue per dollar sold than do traditional sales, and sales taxes are a vital source of revenue for some state and local governments. In 2015, the sales and use tax made up 23 percent of all state and local taxes,2 and, according to Bloomberg, bonds backed by sales taxes secure over $100 billion in debt.3 Also, as they drive brick-and-mortar competitors out-of-business, e-commerce businesses are contributing to declining property values and employment in some communities, as well.
In this white paper, we explore the changing retail sales landscape in the U.S., and focus, in particular, on the rise of e-commerce. We begin with a brief update on e-commerce trends and then discuss three meaningful ways in which online sales might threaten credit quality:
- by eroding state and local sales tax bases,
- by causing retail property values to decline and retail employment to dip, and
- by shifting the location to which sales tax is owed within states.
The third threat is the most serious from a credit perspective, in our view. Finally, we note that competition from e-commerce is not the only challenge facing the sales tax: Demographic changes are likely to alter retail sales patterns, as well.
The past several months have witnessed an uptick in worry over e-commerce. Headlines like “Retail Apocalypse: 20 big retailers closing stores in 2017” (Fox Business, Sept. 19, 2017) and “Dying shopping malls are wreaking havoc on suburban America” (Business Insider, Mar 5, 2017) have dominated business reporting.
This doom-and-gloom news coverage reflects contraction in the traditional retail industry. The retail sector has lost 107,000 jobs this year, and recent estimates suggest that retailers need to downsize by 10 percent (one billion square feet) to reverse current trends in declining sales per square foot.4 Online sales are now 9.7 percent of all retail sales (Figure 1).5
No change in the current trajectory for online sales is expected. Amazon represents 40 percent of online sales. Its market position is well-entrenched, and it is forcing competitors to grow their online footprints. However, when retailers invest to expand their online presence to compete with Amazon they typically drain cash flow. For every percentage-point increase in e-commerce sales, a (non-Amazon) retailer’s margins decline by 0.5 percent.6
Impact on State and Local Governments
The changing retail sales terrain poses at least three meaningful credit risks for state and local governments. Issuers seem well-positioned to tackle the first two problems, but the third could prove more demanding for some credits.
Risk #1: Sales tax base erosion
The most immediate e-commerce challenge facing states and local governments is sales tax base erosion. Sales and use tax collected from e-commerce transactions has traditionally been lower than for brick-and-mortar sales (Figure 2). One study suggested that state and local governments lost $11.4 billion in revenue from online transactions in 2012.7
Base erosion risk is real, but it is very likely manageable. States have been successfully revamping their sales tax codes to enable more collection from remote sellers (a subset of which includes many online sellers), and previous estimates of sales tax losses may prove overstated.
Updating sales tax codes
Collection rates for e-commerce transactions have been lower than for traditional sales, in large part because state sales tax laws have prevented tax authorities from collecting from many online sellers. Under traditional rules, sellers are required to collect tax only if they have “nexus” with the states into which they sell goods. Nexus means having a “physical presence” in a state, and it is a precondition often missing for online sales, which frequently cross state borders.8 As a result, states have been forced to collect a significant portion of online sales tax not from sellers but from in-state buyers, via the “use” tax. (All states with a sales tax also impose a use tax, which requires residents to remit sales tax whenever they make an out-of-state purchase and pay a sales tax rate less than the in-state rate.9) The use tax complements the sales tax and is designed to protect in-state retailers from out-of-state competition. But it is notoriously difficult to collect. A 2012 study by the State of Minnesota revealed that use tax was paid by only 1.7 percent of taxpayers who were asked to report it on their state income tax returns.10
To remedy the collection issue, states have been actively reforming their sales tax codes. During the 2017 legislative session, lawmakers in 31 states introduced 80 bills designed to update their laws.11 Most of these took aim at remote sales. Generally, the strategies fell into at least one of three categories.
One approach is to reimagine what “nexus” means, enlarging the concept from the current “physical presence” standard to a more nebulous one known as “economic nexus.” Economic nexus bills generally establish sufficient nexus whenever a seller crosses a minimum threshold of in-state sales. They are legally questionable given Supreme Court precedent, but states are passing laws grounded in economic nexus theory, anticipating that the Court might revisit its current thinking.12
A second strategy is to expand the existing definition of “physical presence” to include concepts like “affiliate nexus,” “click-through nexus,” and “drop ship” nexus. These concepts expand sales tax collection liability to retailers that sell into a state via a third party.13 New York State’s click-through nexus law was passed in 2008 and its affiliate nexus law in 2009.14 Since then, other states have written similar laws.
A third technique is to give remote sellers a choice: voluntarily collect sales tax or provide reporting information about in-state buyers and the volume of sales made in-state. This approach was pioneered in Colorado in 2010, and it has since been upheld in the federal court system. The Multistate Tax Commission has developed model language so that other states can enact their own versions of Colorado’s law.15
Legislative action at the state level has spurred Congressional interest. The federal government can empower states to collect tax from remote sellers, and many state officials have lobbied for such federal authorization. Congress has given two bills serious thought in recent years: The Marketplace Fairness Act (which passed the U.S. Senate in 2013) and the Remote Transaction Parity Act, which was proposed in 2015.16 Both bills would enable states to collect tax from most online sellers.
New estimates of sales tax lost to online sales
Apart from legislative activity, there is growing evidence that the sales tax lost from e-commerce might not be as significant as first imagined. Notably, Amazon’s physical footprint has grown in the past few years to include brick-and-mortar stores, more warehouses, and even grocery stores (it recently purchased Whole Foods Market Inc.). In April, Amazon announced that it was collecting sales tax from online purchases in all 50 states. So, a growing share of internet sales are now subject to collection.
Looking closely again at Figure 2 above, one will notice that the collection rate for online sales began ticking up in 2013. This may relate to Amazon’s decision to expand its physical footprint in more states, in combination with states’ changing sales tax rules. Recent estimates of lost online sales tax in California, New York, and Wisconsin suggest that lost taxes may be as low as 20 to 25 percent of previous estimates.17
Risk #2: E-commerce as a threat to retail property values and employment.
Although e-commerce may end up having only a modest impact on aggregate sales tax collections, it seems likely to impact property values and employment trends and, therefore, value in some property tax-backed bonds. Online retail has left thousands of brick-and-mortar stores unoccupied and has reduced assessed values and retail employment in communities throughout the country. But general obligation bonds represent about 39 percent of the bond market, and property taxes account for almost $500 billion in local revenue, annually.18 Large jurisdictions with vibrant and diverse economies are likely to withstand the right-sizing of the retail sector, but in smaller, rural areas the changing retail landscape poses more serious credit challenges.
Local governments with large and diverse tax bases are generally well-positioned to manage through today’s retail industry stress. Vacant retail space generally makes up a small proportion of the tax base in these areas, and when a tenant departs it is likely to be replaced by a restaurant, entertainment-oriented business, or by another quality tenant.19 Employment prospects for displaced retail employees are also generally brighter in these places. Notably, transportation and warehouse employment tend to infill some of retail job-loss in these jurisdictions, as online retailers build the capacity to ship their goods to homes and businesses in the area (Figure 3).
The reverse is true for many small and rural jurisdictions. A big-box retailer can be a significant component of the local tax and employment base in a small community. For example, in tiny Marquette Township, MI, with a population of roughly 3,000, the reassessment of several shuttered big-box stores recently led to a 22 percent decline in property tax collections.20 Several states, including Michigan, Wisconsin, Indiana and Texas are now considering legislation to change how abandoned retail stores are assessed to prevent similar outcomes in their jurisdictions.21 Fortunately for investors, small localities typically issue a limited amount of debt, and, when they do, it is often via bank loans or through bond bank structures. Still, there is no question that retail store closings have negatively impacted many rural communities.
Risk #3: Shifting the location to which sales tax is owed.
Apart from its impact on property values and employment, e-commerce is likely to cause a portion of local sales tax revenue to be redirected from some governments to others. In our view, this is likely the most significant e-commerce credit risk facing local governments.
For background on today’s sales tax laws, the 45 states that impose a sales tax do so using either a “destination”-based or an “origin”-based system. Under the destination-based approach, sales tax is owed based on where the buyer takes possession of a purchased item. In an origin-based system, it is owed based on where the item is sold.22
Figure 4 illustrates that for a traditional in-store sale, the distinction between a destination-based or origin-based sale is meaningless for tax purposes. The buyer takes possession at the checkout counter, so sales tax is paid to the local government in which the store is located under both systems.
However, e-commerce changes the traditional setup because the location at which possession occurs (or in an origin-based state, where the sale is made) can differ from the location at which the traditional in-store sale happens. Towns where retailers have physical locations may not receive tax revenue if an item is purchased online.23
Redirecting sales tax revenue among local jurisdictions poses material credit risk, especially for smaller communities. Smaller issuers have less flexibility to respond to changes in sales tax collections than do their larger peers. They tend to have more concentrated tax bases, sometimes with a substantial concentration of retail centers or malls. And like other local governments, smaller issuers are also generally prohibited from raising their sales tax rate or expanding its base.24 Rate and base decisions are usually reserved for states and imposed on local governments.
It’s unclear how many local governments have been impacted by the redirection of revenue from e-commerce sales, but it’s likely substantial. The sales tax is a meaningful portion of local revenue in just over one-half of the states (Figure 5). Debt-conscious Iowa has 136 different sales tax bond issuers and Louisiana has 63.25
Smaller issuers may also be disadvantaged relative to larger peers in terms of their ability to collect tax on e-commerce sales. E-commerce sellers often complain about the complexity associated with sourcing online sales (especially under the destination-based approach). There are nearly 10,000 sales tax jurisdictions in the United States.26 To the extent that online retailers’ compliance with updated sales tax laws falls short, it is unclear whether modest-sized local governments have the resources to collect what they are owed. The latest version of the Marketplace Fairness Act requires states to collect sales taxes at the state level. This is meant to streamline sales tax payment for retailers, but it would also be helpful for local jurisdictions.
Notwithstanding concern about e-commerce’s negative impact on retail sales and tax collections, it is likely that demographic factors will play at least as large a role in driving sales tax growth (or lack thereof) in the future. Demographic factors are likely to influence sales tax patterns in at least three key ways.
First, the millennials are expected to spend a bit less than prior generations did at the same age. Millennials are delaying key life markers, like marriage and children, that are associated with higher spending. They generally carry more student debt than similarly aged persons in prior generations, and their mean earnings are less than in previous generations, at the same age.27
Second, baby boomers are also likely to pare back spending on goods and services as they age. Research suggests that retirees meaningfully reduce consumption after they retire.28 Certainly, recent data on age and consumption spending at the state level supports this conclusion. Figure 6 below shows that consumption declines as a state’s median age increases. Presumably, this dynamic is at play at the local level, as well. If so, the data suggests that rapidly aging local jurisdictions that rely on sales tax for operations or to secure bonds may be more susceptible to credit deterioration than their peers.
Third, it may be difficult to adapt sales tax codes to mitigate aging-related sales tax losses. Many of the items that retirees purchase, like prescription drugs, healthcare services, food and transportation, are deemed essential and are politically hard to tax. The chronic inability of states to transition their sales tax bases to include more service expenditures is instructive in this regard. During the 2017 legislative session, 23 states considered bills to expand their sales tax base to include more services, but none of the bills became law.29
The rise of e-commerce presents a variety of new credit challenges for state and local governments. In particular, e-commerce threatens to alter long-standing consumer behavior, and, in turn, sales tax collections, property values and employment patterns. Credit risks seem most profound for smaller, typically rural, issuers whose tax bases are more easily depleted when a large retailer closes or when sales tax is redirected to a neighboring jurisdiction. In addition to e-commerce risks, demographic change is likely to alter the sales tax experience for state and local governments. Consumption behavior among millennials and the aging baby boomers may keep sales tax growth tepid in coming years. Changes in states’ sales tax codes may temper the impact of these changes in consumer behavior.
 Forrester Data: Online Retail Forecast, 2017-2022 (U.S.), August 1, 2017.
 U.S. Census of Governments, 2015.
 Breckinridge analysis of Bloomberg data on October 11, 2017. Figure includes bonds with a “muni source” described as “sales tax.” Bonds listed under other “muni source” labels may also be secured by sales taxes.
 Bureau of Labor Statistics. Total U.S. retail jobs peaked in January 2017. For retail downsizing data, see Costar Portfolio Strategy Group’s January 2017 retail outlook.
 U.S. Census.
 See “The decline of established American retailing threatens jobs,” The Economist, May 15, 2017.
 Study conducted by Donald Bruce and William Fox, Center for Business and Economic Research, University of Tennessee, based on data from 2009. See: cber.utk.edu/ecomm/ecom0409.pdf.
 Quill v. North Dakota, 504 U.S. 298 (1992). Note that a state must establish “nexus” under two tests if it seeks to impose collection responsibilities on an out-of-state seller. The first test is a due process requirement. To meet due process concerns, “nexus” merely requires that the seller have “minimum contacts” with the state in question. Usually, an e-commerce seller will have sufficient minimum contacts to meet this threshold. The second test is a dormant interstate commerce clause requirement. To meet this test, the seller must have a “physical presence” in the state. Under current law, many online sellers lack a physical presence.
 Michael Mazerov, “States Should Adopt a Version of Colorado’s Remote Sales Tax Law,” Center on Budget and Policy Priorities, August 3, 2017.
 Nina Manzi, “Use Tax Collection on Income Tax Returns in Other States,” Research Department, Minnesota House of Representatives, April 2012.
] Liz Malm, “Sales and Use Tax Compliance Legislation was a Big State Tax Trend This Year,” Multistate Insider Blog, July 11, 2017.
 For an example an economic nexus bill, see Georgia HB 61 (2017). It is unclear if the economic nexus will pass constitutional muster. A South Dakota law is pending in the courts along with an Alabama regulation. The Alabama case involves Alabama Department of Revenue administrative rule 810 6-2-.90.03 which requires out-of-state sellers to collect and remit sales tax if they sell more than $250,000 of goods in-state. The case involves online retailer Newegg Inc. States have become more aggressive in passing economic nexus laws, in part because they want to appeal adverse rulings to the U.S. Supreme Court. There are signs that Justices Kennedy and Gorsuch might be amenable to expanding the definition of “nexus” to include in-state connections beyond mere physical presence. In a 2015 concurrence in the case Direct Mktg. Ass’n v. Brohl, Kennedy essentially asked states to provide the Supreme Court with an opportunity to revisit existing precedent. Likewise, Justice Gorsuch found Colorado’s notification law constitutionally permissible when he had occasion to consider it as a judge on the 10th Circuit Court of Appeals in February 2016.
 The precise definitions of each of these concepts are beyond the scope of this commentary. For a description of each concept, see: http://www.salestaxinstitute.com/resources/remote-seller-nexus-chart. As an example, click-through nexus can be established when an in-state company directly or indirectly refers customers to an out-of-state retailer through a website link and in return receives a commission for the referral. New York State enacted a click-through nexus law in 2008. See: http://www.salestaxsupport.com/blogs/state/new-york/new-york-tax-nexus-for-out-of-state-sellers/.
 Tom Mazurek, “New York Nexus: For out-of-state sellers, no news is good news,” SalesTaxSupport.com, September 28, 2017. Available at: http://www.salestaxsupport.com/blogs/state/new-york/new-york-tax-nexus-for-out-of-state-sellers/.
A third bill, the Online Sales Simplification Act of 2015, includes an origin-based approach, and it does not appear to have the same level of support as the other two. Under an origin-based sales tax system, the tax rate from the selling state is applied rather than the tax rate in the state into which goods are sold.
 Joseph Bishop-Henchman, “Internet Sales Tax Collections Falling Far Short of Experts’ Estimates,” Tax Foundation, March 18, 2013, and Rebecca Helmes, “Sales Tax Slice: How Much Revenue do State (Really) Lose From Remote Sales?” Bloomberg BNA, April 3, 2014.
 The Bond Buyer and U.S. Census of Governments, 2015.
 Barbara Denham, “Is the retail real estate industry doomed,” REIS, June 2016.
 Liz Farmer, “Big-Box Stores Battle Local Governments Over Property Taxes,” Governing, September 2016.
 Mary Ann Barton, “Counties fighting ‘dark store’ math by big box retailers,” National Association of Counties, May 12, 2017. Available at: http://www.naco.org/articles/counties-fighting- percentE2 percent80 percent98dark-store percentE2 percent80 percent99-math-big-box-retailers.
 Of the 45 states that impose a sales tax, 12 impose sales tax based on the origin of the seller. These states are known as “origin-based” sales tax states. They include Arizona, California (partially), Illinois, Mississippi, Missouri, New Mexico, Ohio, Pennsylvania, Tennessee, Texas, Utah and Virginia. For a plain language description, see: the Taxjar.com blog, by Mark Faggiano, “Origin-based and Destination-based Sales Tax Collection 101,” September 5, 2017.
 Note that if the seller is an out-of-state entity, destination-based rules typically apply, even in origin-based states.
 Helen Samuelson, Robert Schulz, and Lisa Schroeer, “Sea Change in the U.S. Retail Sector Could Shake Up Long-term Stability of Sales Tax Revenue Bonds,” Standard & Poor’s, August 7, 2017.
 Based on Bloomberg data from 10/23/2017.
 Joseph Bishop-Henchman, “State Sales Tax Jurisdictions Approach 10,000,” Tax Foundation, March 24, 2014.
 Per research by Goldman Sachs. Brief summary available at: http://www.goldmansachs.com/our-thinking/pages/millennials/.
 Sudipto Banerjee, “Change in Household Spending After Retirement: Results from a Longitudinal Sample,” Employee Benefit Research Institute, November 2015.
 Elain S. Povich, “Why States are Struggling to Tax Services,” Stateline, June 27, 2017.
DISCLAIMER: The opinions and views expressed are those of Breckinridge Capital Advisors, Inc. They are current as of the date(s) indicated but are subject to change without notice. Any estimates, targets, and projections are based on Breckinridge research, analysis and assumptions. No assurances can be made that any such estimate, target or projection will be accurate; actual results may differ substantially.
Nothing contained herein should be construed or relied upon as financial, legal or tax advice. All investments involve risks, including the loss of principal. An investor should consult with their financial professional before making any investment decisions.
Some information has been taken directly from unaffiliated third party sources. Breckinridge believes such information is reliable, but does not guarantee its accuracy or completeness.
Any specific securities mentioned are for illustrative and example only. They do not necessarily represent actual investments in any client portfolio.