Candidly, it has taken much longer than it should for us to fully comprehend and explain the gaping distinction between financial accounting and financial reporting. Since we figured it out, though, we've looked at things differently, to the point that we now believe a company's own transactions are not valid sources of useful information.
ACCOUNTING IS 'SUPPLY-PUSH'
As we see it, financial accounting practice is deeply rooted in the need to address auditors' concerns about internal controls, with special emphasis on cash transactions. The first flowchart ("The Existing Process")[IMGCAP(1)] shows that focusing on control leads to recording transactions and maintaining account balances based on their amounts. For example, the accounts receivable balance is the sum of customers' accounts in the subsidiary ledger that also serves as the billing database. This number is reported because it's the control account balance, not because it provides useful information about the receivables' aggregate fair value or the sales revenue.
Under financial accounting, statements are mere compilations of ledger account balances, only some of which are altered by recording selected other non-transaction events. For example, simplistic systematic calculations, not observed real amounts, are used to record allocated and accrued expenses, such as depreciation, amortization and interest. Some value changes that reduce reported income are recorded, including asset impairments and contingent liabilities. Although these incomplete account balances are not useful information, they are reported because auditors' concerns about internal control and their own risk exposure have overshadowed users' needs. Preparers and auditors now push this useless information out to capital markets without fully understanding, or maybe even caring, who gets it or whether they can use it to make better decisions.
Simply put, the statements are essentially byproducts that consist of what the control system leaves in its wake, especially historical costs and book values, not what users need to know.
REPORTING MUST BE 'DEMAND-PULL'
In contrast, financial reporting must provide useful information to capital markets to support rational financial analysis and decision-making. From inception, the process should serve that objective and only that objective. The second flowchart ("The Ideal Process," page 21) describes a totally different process.
The flow goes in the opposite direction by first recognizing that its objective is distributing useful information. This goal triggers the next step, which is providing financial reports that serve the markets' needs. In turn, that leads to compiling not just financial statements but additional useful information. To produce useful statements, accounts are maintained but their contents should reflect not merely entries made to record basic transactions under a traditional internal control system but many more that capture the effects of all other relevant events, including value decreases and increases.
To provide reliable information, these entries must record verifiable empirical observations of real economic events and effects, instead of imaginary pseudo-measures for such things as depreciation and interest.
In summary, financial reporting should provide capital markets with the information they demand. This idea is consistent with the same successful commercial strategy of always serving consumers' needs.
To put it another way, financial reporting involves financial accounting, but the latter is only the means to the end, not the end in itself.
OUR VISION
We want to cast a challenging future vision for how financial reporting should be conducted. As we see it, many reforms are needed to overcome the overabundant flaws in current financial statements, including cash balances, accounts receivable, inventory, investments, income taxes, business combinations, goodwill, treasury stock, stock options, defined-benefit pensions, cash flows, and many more.
Instead of looking at those areas in detail, however, we want to explain how adopting a user-oriented objective for financial reporting will profoundly change the role played by transactions in developing the statements.
RETHINKING THE TRANSACTION
Under financial accounting, a company authorizes and monitors its own transactions with its internal control system. Achieving control requires journal entries, which, in turn, are added to or subtracted from account balances. Eventually, those balances show up on financial statements. The flawed after-the-fact rationalization is that the transaction surely provides a useful number for statement users because it has been controlled.
Once we grasped the fundamental difference between reporting and accounting, we were stunned to realize that the reporting company's own transactions are actually not very helpful. Given that financial reporting should provide useful information to help users assess future cash flows, statement numbers should communicate facts that shed light on those assessments. It then follows that the accounts should accumulate, and journal entries record, those relevant facts.
Because the statements should present information that allows users to look forward to the future, there is little to be accomplished by presenting what happened in past transactions, especially long ago in the past. Instead, meeting this goal means the statements should provide the most timely and best information by reporting fair values of assets and liabilities.
Many of you practicing accountants are probably flinching because you have always been convinced that fair values aren't reliable. Conventional education and training say so, but we now know this belief is unjustified!
Specifically, we ask which information is more reliable for predicting the future: Would it be knowing where the company used to be at some point in the past or knowing where it is now? The answer is obvious - it's now!
UNRELIABLE HYPOTHETICALS
One source of confusion is faulty definitions of fair value. Even the Financial Accounting Standards Board and the International Accounting Standards Board have gotten it wrong by defining it as the amount a company would receive if it sold an asset. That little word "if" characterizes fair value as a hypothetical number and thus renders it unverifiable. If it cannot be verified, then it cannot be reliable, and if it isn't reliable, it cannot be useful.
In contrast, we define fair values as amounts that other willing buyers and sellers of similar assets are actually giving and receiving in recent real exchanges. Because these numbers are observable facts, they are verifiable and reliable for describing an asset's ability to produce future cash flows. (Implementing this ideal will require databases of actual events; some already exist, but demand for the information will eventually lead to many more.)
INSIGHT FROM STATISTICS
We were staggered when we first figured out that a company's own transactions are sparse, even impoverished, sources of useful information for financial reporting! In fact, they are terribly deficient. Of course, we understand why others would be disinclined to agree. Please read on.
We can use statistics to explain this iconoclastic view of transactions. Specifically, a particular company's asset purchase is only one transaction out of a population of similar exchanges occurring in the same time frame. Further, the population is a distribution of various amounts because the exchanges occur at different prices.
Statistics students and auditors know they can legitimately infer anything about a population only by taking a random sample of sufficient size. They know it would be especially foolish to base any inference on a non-random sample of only one observation. For example, no auditor would confirm a client's receivables by carefully combing through the accounts and picking out one in particular.
Therefore, it is illogical to assume that a reporting company's cost for an asset always equals its fair value at the time of the transaction. Instead, the original cost is but a single data point among the population of all similar exchanges. There's no way that each of those costs could equal the expected value of the distribution!
This perspective renders invalid the conventional wisdom that a company's purchase price is a good, even the best, measure of the acquired asset's fair value at the time of the transaction.
SO WHAT?
If original cost is not a reliable measure of fair value when a transaction occurs, there is no good reason to believe this amount can ever be useful information years later or at any subsequent date, for that matter. We were flabbergasted when we realized that original transaction-based data, despite being so carefully recorded and audited, is so unreliable that it is not at all suitable for serving the objective of financial reporting!
STAYING UNDER CONTROL
Of course, original transactions absolutely must be recorded to maintain effective internal control. We're not arguing otherwise.
However, we now know that those numbers do not - indeed, cannot - provide useful financial reporting information. The key, then, is making additional subsequent journal entries to record fair values and thus allow the accounts to provide valid inputs to financial reports.
ENOUGH SAID ...
We wrote this column to explain that today's accountants and standard-setters are unfortunately hooked on maintaining sub-optimal practices that elevate internal control (and perhaps their own convenience) above the capital markets' needs for useful information.
Near as we can tell, they just haven't been challenged to see that the status quo is not worthy of their above-average abilities. Ironically, their current resistance to progress makes them keep producing information nobody wants or can use for rational decisions. Unwittingly, it also causes them to miss out on greater compensation and personal satisfaction from a job well done.
We're resolutely confident that today's mediocrity will eventually be squeezed out of the financial reporting system by powerful economic incentives to provide timely reports that tell the truth, the whole truth, and nothing but the truth. In the meantime, people should no longer believe that traditional transaction-based financial statement information meets that test.
CLOSING ON CONVERGENCE
In closing, we think FASB's smaller size, financial independence, and responsibility for only one national capital market give it a much better position than the IASB for moving beyond useless financial accounting data toward providing useful financial reporting information. The international board's complex political environment will impede the needed reforms.
Accordingly, we once more conclude that there is no benefit from having the U.S. adopt international accounting standards and endorse the IASB as the designated standard-setter. Taking that path will not lead to a bold new vision for the future.
Paul B. W. Miller is a professor at the University of Colorado at Colorado Springs and Paul R. Bahnson is a professor at Boise State University. The authors' views are not necessarily those of their institutions. Reach them at paulandpaul@qfr.biz.
