It is a pleasure to be with you today in San Diego. As is always the case for Federal Trade Commissioners, it is important for me to note that my comments today are my own and in no way represent the views of the Commission or of any other Commissioner.
There are a number of issues before the Commission and the Country that I would like to discuss today that should be of interest to you all, especially here in California -- the Internet Economy, Online Privacy and Antitrust concerns for high tech industries.
Before doing so, I thought it might be helpful to share with you some personal insights gained from my years of experience as a federal official. First and foremost, I believe our Founding Fathers had it right: government should play only a minimal role in our lives. I believe it was Will Rogers who once observed: "All government programs have three things in common: a beginning, a middle, and no end."
The promises of electronic commerce have those in government excited: the politicians see something to tax and more money to spend, and the regulators see something with endless possibilities to regulate. In their emotion, I fear they will forget some basic rules of a society built upon private enterprise.
Given the tremendous benefits that typically flow from private markets, government intervention in these markets should be undertaken only when it is clearly necessary. We in government, responsible for regulation and for economic and tax policy, should be ever mindful of the Hippocratic Oath -- "First, do no harm."
The Internet Economy
Earlier this year, the University of Texas, backed by Cisco Systems, introduced a study of the current status of electronic commerce -- one of the very first attempts to measure the Internet economy. According to the UT/Cisco study, the Internet economy generated an estimated $301 billion in revenue in 1998 and was responsible for more than 1.2 million jobs. Last week the authors of the study projected that the Internet Economy will grow to $507 billion this year and produce more than 2.3 million jobs. That's a projected 67% and 46% increase respectively in one year my friends. (These estimates are based on worldwide sales of Internet-related products and services by U.S.- based companies.)
Let me put those figures in perspective. If the 1999 projections hold, it means the Internet Economy has become the largest industry in our country. And what is even more impressive is that the Internet is still in its infancy
The Question To Consider
As you may know, I spoke before the inaugural meeting of the Advisory Commission on Electronic Commerce this past summer. The question I posed that day is worth repeating today. "Should policymakers apply a Depression-era tax system to the economy of the 21st Century"? The answer to that question will have an enormous impact on economic growth -- the creation of wealth, jobs and prosperity -- throughout our country and the world. The economic consequences of inappropriate government actions in e-commerce will injure our country gravely, and diminish our position as the leading world economy.
Taxation is a form of regulation. The history of taxation seems to be that every time a new product, a new industry, a new form of social organization, or even a new economic concept of income or wealth has arisen, governments have moved to tax it. As President Reagan said, "The government's view of the economy could be summed up in a few short phrases: If it moves, tax it. If it keeps moving, regulate it. And if it stops moving, subsidize it."
The Internet is a competitive advantage for the United States: more than one-third of all current Internet usage is by Americans. Unwarranted taxes and regulation at a time when the technology is still rapidly evolving threaten to lock in or limit the Internet to specific technologies and modes of service that fall far short of its likely potential. Tomorrow's tax and regulatory policy will have an enormous impact in shaping the future of this burgeoning new industry of electronic commerce supported by the Internet.
The Complications of Taxing Internet Commerce
The issue of taxing the Internet is complicated by several factors:
The Internet is inherently susceptible to multiple and discriminatory taxation in a way that commerce conducted in more traditional ways is not. With approximately 30,000 taxing jurisdictions, compliance becomes a significant obstacle. If we simply required that merchants collect the relevant tax for the jurisdiction into which the product is being delivered, such legislation would produce a world that is anything but "simple."
Can you imagine the confusion that would arise in the case where a small business owner from New Hampshire (a state without sales tax) is required to collect the tax on a purchase made by a consumer living in the Atlanta area? Atlanta is a metropolitan area with numerous suburbs, several of which have different local sales tax rates, in addition to Georgia sales tax. Or even more bizarre, consider Internet sales of shoes-- a product that is tax exempted in some states but not others, depending on such factors as whether the footwear in question is tennis shoes, sneakers, or cleated athletic shoes.
How can we know how to tax it? Since Internet commerce is so new, we do not know what the basic business model will look like in a few years. There are likely many adverse unintended and unanticipated consequences lurking in the future.
How would the taxes be collected? The cost of compliance and tax collection alone for small businesses could be enough of a deterrent to keep them from participating in the marketplace, thus depriving the economy of its most vibrant stimulus.
Clearly, compelling retailers to collect tax under the current jurisdictional regime would place a significant burden on merchants; and such a burden would likely not be uniformly felt across all retailers. A recent study by the Washington State Department of Revenue indicates that small businesses would be hit hardest by compliance costs with multi-jurisdiction tax rates. A recent study by Ernst and Young has estimated compliance costs for small businesses to be close to 87 percent of the sales tax they collect, this compared to only 14 percent for large businesses to comply. (1) While these costs might be eased by employing various software packages, such software can cost well over $ 20,000. (2)
Technology is making it possible for many, many more citizens to create a business from scratch. Do we really want to place one more barrier to entry before these entrepreneurs by making them remote tax collectors for an ill-conceived tax scheme?
Who has taxing and regulatory jurisdiction? Identifying the state, country or countries that have tax jurisdiction over income generated by electronic transactions may be a problem with no solution. Electronic commerce permits a foreigner to engage in multiple business transactions with customers in the United States without ever having entered the country.
And lastly, what about personal privacy? Do we want to enact a taxation scheme that, to be effectively implemented, will likely have to systematically undermine our privacy by amassing a comprehensive database of our online purchases so that some government agency can be certain that we paid our relevant taxes? What guarantees do we, as consumers, have that such an agency, upon learning of our buying habits, will not either sell that valuable information to third parties or use it in a way that undermines our personal security?
The Lost Revenue Argument
Throughout this debate on taxing Internet use and commerce, the argument has often focused on the claim that failing to create a suitable Internet commerce tax will lead to the steady decline of state revenues, perhaps as much as $20 billion a year, (3) significantly hindering the development of state infrastructure. While such arguments conjure up a frightening vision of what could occur, are such predictions accurate? Another recent study by Ernst and Young has shown that less than $170 million of sales and use tax were not collected in 1998 on Internet sales -- only 1/10 of one percent of total state and local tax revenues. (4)
These estimates are similar to another study by scholars at the University of Chicago and Harvard who have estimated the loss to be close to 1/4 of one percent. (5) These scholars have predicted that even after five years, the average loss in sales tax revenue to states will amount to only two percent of potential tax revenues--a mere fraction of the $20 billion loss that has been predicted by the proponents of Internet taxation. Is retaining this minor loss in tax revenue worth crippling potential entrepreneurs as they strive to find a place for themselves in this dynamic new marketplace?
One is reminded of the fable about killing the goose that laid the golden egg.
Senator John McCain has introduced legislation to permanently ban Internet sales taxes by specifically outlawing any future attempts to impose a sales tax structure on Internet sales. In addition, Senator McCain's bill calls for WTO adoption of a global moratorium on Internet taxes. As Senator McCain said on the Senate floor, "This bill would make permanent the moratorium on sales and use taxes for e-commerce, and would encourage the Administration to urge our world trading partners to do the same."
I could not agree more with Senator McCain. While some Members of the Advisory Commission on Electronic Commerce seem more focused on how to tax the Internet, only Congress can authorize one state to compel sellers in another state to collect Internet taxes. It is important to move forward to ensure that the default position is not to lift the moratorium on Internet taxation. Those advocating taxing electronic commerce should be challenged to prove that doing so is a rational thing to do.
Last month I sent a letter to four groups: The Online Privacy Alliance, NetCoalition.com, the Direct Marketing Association, and the Software & Information Industry Association that said in part:
"…almost three months have passed since the FTC reported to Congress that legislation to regulate online privacy was not necessary this year. Industry is to be congratulated for the significant self-regulatory progress that has been made, but as I testified, industry should continue to lead the way if it wants to have the freedom to adopt privacy policies in response to market incentives and not government regulation. I applaud companies like IBM, Microsoft, Disney, Compaq, Novell, and Procter & Gamble for their leadership in pledging to withhold advertising from sites without privacy policies.
It was clear three months ago that passing additional laws and regulations on the Internet could produce negative results. The critical questions was, and remains, "How do we protect consumers and, at the same time, make it possible for the vast potential of the Internet to develop?" As we approach the upcoming Holiday shopping season, some estimate that consumers will spend three times as much online as they did last year. I have encouraged industry to wage an aggressive campaign to raise awareness about the need for good online privacy policies among consumers and businesses that sell consumer products on the Internet."
Responses To My Letter
Where Do We Go From Here?
Privacy has become a great political topic for discussion and cause celeb. For those who advocate regulating the Internet (and there are many,) I ask, "How would we do it?" The devil is in the details. We are coming to realize that technology and cost, not to mention the exponential growth of the online world, are serious impediments. Recent data suggest there are now almost four million commercial Web sites, and they are increasing at more than 275,000 a month. Imagine a government agency trying to regulate or control something so dynamic. This is a formula for bureaucracy-building, government intrusion, and a stifling of economic growth--in other words, many potential unintended consequences are lying in the weeds just waiting to do mischief.
Since the Web is growing so exponentially, I believe we should focus on the top 100 sites where roughly 95% of all Internet traffic goes and evaluate what visitors and consumers are finding when they get there. The matter of personal privacy is not trivial. Informed, discerning, and responsible consumers coupled with responsive, ethical business practitioners will take us a long way toward minimizing consumer concerns about online privacy…a far better solution than more government regulation. The private sector should lead the way --- market solutions should be our goal The message to you as legal counsel for your clients-- privacy is important. Believe it. CEO's need to make assuring customers privacy a corporate priority.
Common Sense Antitrust Enforcement
The high technology industry must be aware of and responsive to concerns associated with mergers and anti-competitive conduct. There is a recurring topic of discussion as to whether or not traditional antitrust concepts apply to this rapidly changing industry. Rest assured, they do.
Mergers create a wide array of economic benefits and can place the assets of the acquired firm into the hands of those who have powerful incentives to use them to their best economic advantage. Mergers can mean lower costs, better competition, economies of scale in production, research, distribution and the use of capital. Better products and services will lead to greater consumer satisfaction though better choices and lower prices stemming from a competitive marketplace. Simply stated, competition works.
The FTC is concerned with mergers that create or enhance market power or that facilitate the exercise of market power so that the merged firm or the remaining firms in a market can do just the opposite. The exercise of market power can lead to various anti-competitive effects, such as stifling innovation, higher prices, poorer choices and less customer satisfaction.
Despite uncertainties and the complexity of global businesses and economics, I believe that it is possible to discern a common sense test for evaluating a merger. Is it good for competition and for consumers? If so, then the merger is probably OK. If not, you have a business and a legal challenge ahead of you.
The Commission's action in the anti-competition arena is highly visible, well documented, and, for the most part, quite predictable. I would encourage your familiarity with our Merger Guidelines and would also call to your attention and solicit your comments on our recently published (for public comment) Antitrust Guidelines for Collaborations among Competitors. Your comments on these guidelines are very important to the FTC and the DOJ getting it right. Please let us hear from you.
Thoughts About Intel and Microsoft
The trial of the Department of Justice's ("DOJ") case against Microsoft Corporation -- the world's leading manufacturer of operating systems -- has just been completed before Judge Thomas Penfield Jackson, and it seems as if all the world awaits his findings of fact and conclusions of law.
In addition, the Federal Trade Commission (over my dissent) recently issued an order making final a consent agreement with Intel Corporation - the world's leading manufacturer of general purpose microprocessors - to resolve allegations that the company had monopolized and attempted to monopolize the market for general purpose microprocessors. (6)
These cases have brought many long-simmering disputes to a boil, and we should take this opportunity to assess where we are with regard to government antitrust enforcement in high-tech industries. You all know the facts of the cases. There are at least three issues raised by the Microsoft case which are worthy of serious discussion.
The first is: When does a firm have "monopoly power" in the market for a high-tech product? Courts have defined monopoly power to be "the power to control prices or exclude competition." (7)
Because monopoly power is difficult to measure directly, courts have used market share as indirect evidence of the existence of monopoly power. Microsoft allegedly has an 80% or more share of the market for personal computer operating systems, and courts typically have concluded that a 70% or more market share creates an inference of monopoly power absent evidence that overcomes that inference. (8)
I think that it is particularly important in antitrust cases involving high-technology to look beyond mere market share to determine whether monopoly power exists. On the one hand, in antitrust cases involving high-tech, much has been made of the theory that a firm's market power may be durable because of what are known as "network effects." Network effects essentially mean that certain products become more useful as more and more people come to use them. Simple examples are telephones and fax machines, both of which became more useful with increased use. The theory underlying network effects is that once consumers become "locked-in" to a product incorporating the current technology, they may be reluctant to switch to another product with a superior technology because of the premium associated with the widespread use of a common technology. Since network effects may make consumers less likely to switch, a firm with a high market share and a product which incorporates the current technology, is more likely to have monopoly power.
On the other hand, experience teaches that the normal operation of economic incentives in dynamic high-tech markets generally will spur advances in technology. This, in turn, causes consumers to switch to the superior new technology in spite of network effects. For example, in the early 1980s consumers were locked into the technology of vinyl records and turntables, but consumers rapidly switched technologies with the introduction of compact disks.
Perhaps the threat of similar technological switches explains why some of the titans of high-tech industry today do not appear to behave like the monopolists of old, for whom "unchallenged economic power deaden[ed] initiative . . . [and] immunity from competition [was] a narcotic." (9)
The ultimate determination of whether a firm that makes a high-tech product has monopoly power is a difficult assessment that must be undertaken in light of the particular product market at issue. This would include an analysis of how technological change may affect the ability of the firm to control price and exclude competition. For example, in the Intel case, I expressed uncertainty as to whether Intel, despite its extremely large market share in general purpose microprocessors, actually had monopoly power. To me, Intel appeared to face aggressive competition from innovative firms, especially those that were supplying general- purpose microprocessors for personal computers costing less than $1,000.
The second issue that I think the Microsoft case and other high-tech antitrust cases raise is whether the alleged unlawful conduct has caused harm to consumers, and not just to competitors. (10) Antitrust law traditionally evaluates whether the conduct at issue has caused or is likely to cause harm to consumer welfare in the form of higher prices, reduced output, and decreased innovation. Markets for many high-tech products have been and continue to be characterized by decreasing prices, increasing output, and robust innovation, often stemming from massive spending on research and development.
Such market characteristics do not confer immunity from antitrust challenge on those competing in these markets. However, they do transform the relevant questions into whether prices would have fallen even more, output would have increased even more, and innovation would have been even more vibrant in the absence of the alleged antitrust violations. While there certainly are circumstances under which such harm can be proven, we should be wary of reaching the conclusion that markets that are already creating tremendous benefits for consumers would have done even better.
For example, in the FTC's Intel case, the Commission alleged that, in the wake of intellectual property disputes between Intel and three computer manufacturers that purchased Intel microprocessors, Intel decided to curtail the supply of technical information and prototypes to those three firms. The Commission's complaint alleged that Intel's actions entrenched its monopoly position in general purpose microprocessors and diminished the incentives of firms commercially dependent on Intel to develop innovations relating to microprocessor technology. I was unwilling to support the Commission's case against Intel because "[w]hatever injury Intel might have visited on [the three computer manufacturers], I [could not] accept that it could appreciably affect -- much less stem -- the immense tide of invention and improvement that continuously drives this industry." (11)
Many high-tech markets are characterized by decreasing prices, increasing output, and robust innovation -- certainly characteristics that yield tremendous benefits to consumers. To avoid "killing the goose that lays the golden eggs" that I mentioned earlier, remedies should be very carefully calibrated to address demonstrated consumer harm and to prevent violations that are the same as or similar to the violations that caused that harm. Adherence to the Hippocratic oath -- "First, do no harm" -- seems quite applicable to the remedies that should be imposed in antitrust cases concerning high-tech.
The Microsoft case has generated a flurry of debate as to what sort of remedy the court should impose if Microsoft is found liable. Among others, the remedies that have been discussed have included breaking up the company (vertically or horizontally), mandating that the company allow access to its operating system technology, and prohibiting the company from using certain contractual provisions.
I have touched on just a small portion of issues before the Federal Trade Commission. No mention has been made of oil company mergers, deregulation of power generation utilities, consumer protection, etc. --- it is indeed a full plate. The FTC is blessed with some fine people dedicated to trying to do what is best and working long hours in the process. The issues are complex. We don't have all the answers, but we make a heck of an effort to find those pertinent to the issue at hand. And not surprisingly, we are not without those with agendas --- some with which I would disagree. As you lawyers often remind us, "Reasonable people can disagree." I am confident that we are all trying to get it right.
1. Both of these examples were taken from Cline, Robert J., and Thomas S. Neubig. "Masters of Complexity and Bearers of Great Burden: The Sales Tax System and Compliance Costs for Multistate Retailers," Ernst and Young Economics Consulting and Quantitative Analysis, September 8, 1999.
2. Harry Tennant and Associates, (1997), "Sales tax, use tax and Internet Transactions" cited in Lukas, Aaron, "Tax Bytes: A Primer on the Taxation of Electronic Commerce" (CATO Institute, typescript, 1999).
3. The estimate of $20 billion has been offered by the National Governors' Association. "Governors fear tax loss from Internet; States' surpluses treated cautiously," Boston Globe, December 31, 1998.
4. Cline, Robert J. and Thomas S. Neubig, "The Sky is Not Falling: Why State and Local Revenues Were Not Significantly Impacted by the Internet in 1998," Ernst and Young Economics Consulting and Quantitative Analysis, June 18, 1999 Cline and Neubig have offered the figure of 1/10 of one percent loss. (Goolsbee, Austin and Jonathan Zittrain, "Evaluating the Costs and Benefits of Taxing Internet Commerce"( University of Chicago Graduate School of Business, typescript 1999.) have estimated the loss to be closer to 1/4 of one percent.
5. Goolsbee and Zittrain, supra n. 4.
6. Intel Corporation, FTC Dkt. No. 9288
7. United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 391 (1956).
8. ABA Section of Antitrust Law, Antitrust Law Developments 235 (4th ed. 1997).
9. United States v. Aluminum Co. of America, 148 F.2d 416, 427 (2d Cir. 1945).
10. R. Bork, The Antitrust Paradox 89 (1978) ("the case is overwhelming for judicial adherence to the single goal of consumer welfare in the interpretation of the antitrust laws").