This speech was the keynote address at the 16th XBRL International Conference in Vancouver, B.C., on December 3, 2007.

Arthur F. Burns Fellow Peter J. Wallison

It is both a pleasure and an honor to have an opportunity today to be a keynote speaker at this conference.

One of the keynote facts of our time is constant change. Some change is brought about by the restless search of science or the pressures of competition. Not all of it is good or desirable, but change that promotes efficient use of resources and greater productivity is clearly the source of the world's prosperity, and will be in the future. XBRL is one of those changes that will have a role in this economic growth, and in this talk I want to step back a bit to provide some perspective on what we are all doing.

Despite Chris Cox's efforts, XBRL still has a long way to go in the United States. If you're like me, you are always running into otherwise informed people who offer you a blank stare--if not the look of a cornered animal--when you tell them that you're working on XBRL. One or two may have heard the term--or maybe even Chairman Cox's term, interactive data--but almost none can give it any content.

In one sense this is not surprising; XBRL is not just a product--it's an idea, and idea that requires an entirely new way of thinking about the management of information. It's a difficult concept, perhaps forever beyond the comprehension of the general media. In a recent search, I found that the term XBRL has only appeared 120 times in the world's major newspapers, and never at all in the New York Times. That's not much for thirty-seven newspapers worldwide, and a search that goes as far back as these records were kept. We probably have to get used to the fact that XBRL is not and never will be a household word, so we will always find people drifting away from us at cocktail parties after they've asked us what we do. Let's face it, like we ourselves, XBRL will always operate in the background. Even after it's in widespread use, very few people outside financial officers of corporations and analysts will even know that it's there.

But there's no need to feel sorry for ourselves. In fact, we can bask in self-satisfaction if we want to. That's because XBRL is one of those technologies that will have a profound effect on the way we all live, and that's a much better than stirring interest at a cocktail party. As our mothers used to tell us, virtue is its own reward.

Still, it helps to take account once in a while of the value we create. So perhaps you will join me in a little excursion into the future. We tend to take for granted the technological advances that have changed our lives or made us more productive, but a little thought shows that relatively small cost changes can have vast consequences. We now think nothing of gathering for a few days at a conference in Vancouver, while still keeping appointments the next day half way around the world. Because air travel has become so inexpensive--in terms of money and time--it is more efficient to meet in one place and exchange ideas than to try to do it by telephone or even e-mail. If we had to spend a week and a small fortune getting here and back, I would bet that our calculation about the usefulness of this meeting would be entirely different.

Another example of the profound effect of cost reductions is securities trading. In 1975, about 5 billion shares were traded on the New York Stock Exchange. In 2003, that number was 350 billion. The reason for an increase of seventy-fold in less than thirty years was the elimination of fixed commissions on the NYSE in 1976. The resulting competition reduced average brokerage commissions from several hundred dollars for a 1000 share trade to less than $13 at most discount brokerages. The result was a complete change in the structure of the market, with millions and millions more individuals participating the trading, and far more trading by institutions. It's also important to recall that similar, if not greater, growth occurred at the same time on the NASDAQ market.

The point of this is that when the economics of one part of a market change substantially, they produce vast changes that are seldom foreseen. In some cases, whole new industries arise. Sometimes these developments are so unexpected and startling that the line from the 1989 film Field of Dreams, "If you build it, they will come" seems more reality than dream. In the 1930s, for example, when the SEC required that all public companies report their financial results in accordance with Generally Accepted Accounting Principles, or GAAP, the resulting consistency in financial reports substantially lowered the costs of analyzing and comparing companies' financial outcomes. This new economic fact fostered the development of a whole new industry, financial analysts, who helped institutional investors make their investments and advised brokerage firm customers to decide which securities to buy or sell. The advent of a consistent system of accounting had made it possible to compare the results of companies fairly easily--and that in turn enabled financial analysts to take on this task for a cost that bore a reasonable relationship to its value to investors.

By now, you probably see where I am going with this. XBRL will drastically reduce the costs of analysis yet again. It is the next big thing, after GAAP or IFRS, that will change the fundamental economics of the business of analyzing investments. Although we are all notoriously bad at estimating the full effects of new technologies, some changes seem very likely:                                                                                                                                                                                                                                                      

• An industry of freelance analysts is likely to develop--a vast democratization of the financial advice business--offering advice to ordinary investors for a modest fee. The infrastructure for such an industry is already in place--the Internet and the various payment systems that can be used to purchase goods and services online--and all that has been missing is the ability of analysts to produce solid analysis at a reasonable price. XBRL makes this possible by reducing the search costs of analysts. For example, if you want to know the reserves of the oil companies, in today's world you have to download their financial statements, search for the footnote that contains the reserve numbers, download the information and input it into a spreadsheet. This is at least a days' work, and even then you don't know whether all the companies you are comparing define reserves in the same way. With XBRL, the search can be done in seconds, and you know that the data are truly comparable. The cost savings in resources and time are enormous.
• CFOs will be put under greater pressure to explain the underpinnings of the financial statements they release. Currently, when a company releases its financial reports, the key data are either input by hand into analysts' models or summarized and sold to analysts and others by data aggregators. In either case the data is not fully analyzed until days or longer after it is released. With XBRL, however, the data released goes instantaneously into analysts' models; they will be able to ask questions in real time, unlocking far more information about companies than is currently available to the markets.
• Small companies will find that they finally have significant analyst coverage and investor interest. One of the principal obstacles to this today is the high cost of ferreting out the information necessary to understand the financial information of companies, and to compare it to others' data. As this becomes cheaper because of XBRL, analysts and investors will move down market, so to speak, looking for bargains among the smaller companies that have not been getting much attention.
• Finally, there will likely be a flowering of retail investor activity, based on software that offers the opportunity for analyzing companies oneself, or creating a model for doing so. With useful information readily available, we can expect to see large numbers of amateur retail investors testing out their skills, and eventually offering to others through blogs and other Internet vehicles the models that have proved successful.

While all of this has the rather narrow appearance of advancing the objectives of individual or institutional investors, it has far more significance than that. The more information that is available about companies, the lower the risk premium that is built into their share prices. In other words, companies will be able to raise more capital with less dilution, and this will add to their productivity and to the availability of jobs in our and other economies.

But this is not all; better information, more easily obtained and at less cost, will also help all investors--retail and institutional--better to allocate their capital. This means that more capital and more lower priced capital will flow to the companies that can demonstrate that they are using it most effectively, again improving the material well-being of populations around the world.

So if you haven't thought about what will come from the work you have been doing on behalf of XBRL, I ask you to think of this now: XBRL is not just about better reporting of financial information; as has been true of the Internet, cell phones and a few other products of technology, XBRL will profoundly change the world and the lives of everyone in it.
But XBRL is also what I would call a flagship technology. Like asset securitization, it's a technology that has a myriad uses apart from mere financial reporting, as important as that is. When its extraordinary capabilities are better understood, it will be applied elsewhere. Reports to government are one obvious example. In the United States, all banks now report their key data to the FDIC in XBRL format, and both the banks and the FDIC are reporting extraordinary savings in time and personnel requirements. Standardization of income tax terms and income and expense categories could make the agonizing business of US income tax reporting far simpler. Standardization of the terms used by countries in economic reporting will enable investors and trade participants to understand far more fully than they do today the actual economic conditions and trends in countries, and thus to allocate capital more efficiently and with less risk.

This brings me to what we call in the United States Enhanced Business Reporting, or EBR. Today, most people who think about XBRL at all tend to think of it as useful for financial reporting. If they understand the value of XBRL, they believe that public companies should make their financial information available in XBRL format, and that this step will have the long and short-term benefits that I've described above.

But better reporting of financial data is not enough. In fact, as companies increasingly add value through the internal creation of intangible assets GAAP financial statements, and even International Financial Reporting Standards, are becoming less and less useful to investors. Let me give you an example. When companies were producing value the old-fashioned way--through plant and equipment--GAAP accounting worked well. As a financial reporting system, GAAP had two principal objectives: first, to provide some indication of a company's retained wealth, through its balance sheet, and second to provide an estimate of its operating results during a particular accounting period. It accomplished the former by recording assets such as plant and equipment at cost and depreciating or amortizing these assets over time as they were consumed in the course of production. It accomplished the latter by bringing costs--including the depreciation of productive assets-- into the same accounting period as the revenues these costs were expended to produce.

Although this effective cost-based system had survived for about five hundred years, it is gradually being rendered obsolete today by the knowledge economy that technology has produced in the United States and other developed countries. In a knowledge economy, value is not produced by plant and equipment, but by human brain power. It takes the form of patents, employee knowhow, designs, formulas, software, customer satisfaction and brands, to name a few. It even includes contractual relationships with suppliers. These assets are intangible, and they are created internally through promotional activity or the work of a company's management and employees. Under GAAP, neither employee compensation nor promotional activities may generally be capitalized, and thus these costs are written off as incurred and do not appear on a company's balance sheet. Nor does the balance sheet include employee knowhow, which is often a company's most valuable asset.

If we consider the two ways in which GAAP is supposed to do its work--estimating a company's wealth and operating results--we can see that neither provides adequate disclosure when a company's principle assets are intangible. Intangible assets frequently don't appear at all on a company's balance sheet--and so can't be amortized--and the income statement does not properly show profits because it does not capture in the same accounting period both revenues and the costs used to produce them. The employee work that produced the "killer app" may have occurred years before the software went to market and started to produce revenue.

So, in order to provide investors with the information they will need in the future about the companies in a knowledge economy, we should be thinking ahead, to the development of key performance indicators, or KPIs, that will inform investors about whether companies are adding value or competing effectively. KPIs could be part of management's discussion and analysis in a company's 10-K, or they could be disclosed separately from any formal report to the SEC, but they would be most useful if they are standardized across an industry. The important point, though, is that the same collaborative standardization process that produced XBRL will work to create KPIs, and the same process by which XBRL makes financial information readily accessible would also work for disclosing KPIs. Accordingly, XBRL is only the beginning of a much larger task--the creation of KPIs and their standardization across industries--and XBRL is integral to this task. One way to think about this is as a continuum, with XBRL as the vehicle for corporate disclosure that will begin with financial statements and footnotes but ultimately include management's discussion and analysis, and--if separate from the MD&A--KPIs.

I am hopeful that the SEC's Advisory Committee on Improvements in Financial Reporting can set a course in this direction, beginning with XBRL and proceeding to a recommendation concerning KPIs. The committee is now considering XBRL, and is wrestling with a number of issues associated with making a recommendation to the SEC--perhaps in January--that XBRL should be required for all reporting companies beginning with fiscal year 2009. Experience with Sarbanes-Oxley has justifiably made American companies suspicious--and even hostile--to mandates out of Washington. At our most recent full committee meeting, there was an overwhelming vote of support for XBRL, but on the implicit condition that the concerns of many public companies--large and small--can be satisfactorily addressed.

These concerns relate to three cost elements: the cost of reproducing the balance sheet and income statement in XBRL format, the cost of including the footnotes, and the cost of assurance. From all indications, the cost of converting ordinary financial statements into XBRL format is not great, even for small companies. There are a number of software solutions that make this relatively easy, and once the structure is in place each subsequent year is far easier than the first. I also believe that the both the footnote and assurance issue can be solved--and far from being problems are actually benefits of using XBRL. Both should result in substantial saving for companies that adopt XBRL. That's why I--a long-time opponent of government mandates and regulatory impositions--will favor recommending that XBRL be required for all public companies. Only then, I believe, will companies have the incentive to look into the savings that are available through implementation.

And that points to our most serious problem. Getting people to listen. I doubt that the SEC, by itself, would have been able to bring about the widespread adoption of XBRL. The business community in the United States, as I mentioned, is still shell-shocked by the disaster of Sarbanes-Oxley. There is also a huge embedded infrastructure for financial reporting that is not interested in change, and the practical attitude of business people that there is no reason to spend the time learning something profoundly new unless it is necessary for running the business or meeting the competition. And finally there is the traditional argument that "if it ain't broken, don't fix it"--a sensible position when we recognize that our financial disclosure system works tolerably well right now.

But XBRL is not a fix for a broken system; it is a massive improvement to an existing system, the way e-mail improves upon an ordinary letter. To understand the nature and magnitude of the savings that can come from XBRL requires that someone understand how XBRL enhances and simplifies the current reporting process, and neither most of American business nor their representatives in Congress have this understanding. This is where the SEC's advisory committee comes in. The committee is made up of people experienced in a wide-range of business and professional activity; its chair, Robert Pozen, is a respected former CEO of a mutual fund company, and frequent commentator on financial policy. If the advisory committee can be brought to understand the huge cost advantages and efficiencies that are available through XBRL, and if its report to the SEC in January successfully allays concerns among public companies about this new technology, there is a chance that a recommendation for mandatory implementation can get by business community opposition and the resulting opposition in Congress.

But that's the most favorable scenario. All of us here must work to increase understanding of XBRL--how it works, why it works, and what cost savings it can bring to business. So I am hoping to make two points today. There is a slogan used in American politics which has some applicability here: "Think globally, act locally." All of us in the XBRL community are participating in something that is much larger than a simple improvement in the process of financial disclosure. XBRL will have a profound effect on the economics of investment and thus on economic growth and development. That's thinking globally. But to reap these benefits we must first act locally, by participating in an educational process that will persuade those who must bear the initial costs of implementation that the benefits far outweigh these costs.

That's what I plan to do over the next few months, and I hope you will join me.