Stephen H. Penman is the George O. May professor in the Graduate School of Business, Columbia University. Prior to this he was the L. H. Penney professor in the Walter A. Haas School of Business at the University of California at Berkeley, and has served as a visiting professor at London Business School and Stockholm School of Economics. He has a first-class honors degree in commerce from the University of Queensland, Australia, and MBA and PhD degrees from the University of Chicago. His research is on equity valuation and the role of accounting information in security analysis. He is managing editor of the Review of Accounting Studies and is on the editorial board of the Schmalenbach Business Review, and in 2002 he was awarded the Wildman Medal for his book, Financial Statement Analysis and Security Valuation.
In your recent book, Accounting for Value, you take the accounting standards-setting bodies to task for bringing speculative values into the accounts. Where do you see them going wrong?
I approach accountancy by asking what it measures that is of use to an investor. Basically, I am looking for accountancy to be a set of tools for measuring value—the real value of a business. My book was first and foremost a book on valuation, aimed at investors and those to whom they entrust their savings. This last group includes investment advisors, analysts, and portfolio managers. Beyond them, of course, I hope that it will also be useful to businesses that have an interest in ensuring that investors and their advisors are able to grasp the value of a company and its operations.
With this as my starting-point, I distinguish between accounting techniques for measuring value and models that are aimed at predicting, say, exit values on assets that the business is not actually exiting. I want to use known value as my anchor and then add to that a speculative component about future performance where value is derived from the best information available to me. In investing you are always taking a view of future performance, so a speculative dimension is inevitable. However, this should come only after you have anchored yourself on actual valuation. If your starting-point in appraising a particular company is also speculative, where are you?
With this approach it is clear that historic cost accounting is very useful to investors who like to base themselves on fundamental values, since it grounds itself in prices actually paid. My problem with the accountancy bodies begins at the point where they start to move away from historic cost accounting toward what they define as balance sheet accounting. With balance sheet accounting they are trying to put a present value on all the assets on the balance sheet, including assets such as stock, which the business has no intention of selling, but which, rather, is designated for production purposes. As such, external prices for that stock are completely irrelevant until and unless the company is in the process of winding up—at which point you would be highly unlikely to want to invest in it anyway.
There are very few companies where balance sheet accounting is actually useful for a potential investor. With a pure investment company that holds a large equity portfolio, for example, knowing the present value of that portfolio is excellent information. The company’s worth really does go up and down with the market, so you need to know where it stands. A bank’s portfolio of mortgage loans, on the other hand, is not, I would argue, something to which you need to apply a moment-by-moment “fair value.”
That mortgage portfolio is what drives revenue for the bank, and it is, by its nature, going to be held to termination. Where the bank conducts a securitization of its mortgage book and sells that part of the book, of course you have a present value for the securitized bundle, but that is a special case where the bank is literally turning part of the book into cash and there is nothing speculative about that cash value.
What you would want to know about the bank’s mortgage portfolio is not its fluctuating present value, but rather that there has been appropriate due diligence in the granting of the loans that comprise the portfolio. You want to know, in other words, that the bank’s business processes are sound and under control. If you conclude that its processes are hugely flawed, then the present value of the bank’s book is largely irrelevant since you can assume that the bank is highly likely to experience a much higher rate of default than its competitors—in other words, that the value of its book is likely to be considerably impaired over time, irrespective of what its present value might be—and you would be a fool to invest. This point, clearly, has to do with the importance of information and nothing to do with fair value, or the supposed present value of the bank’s mortgage portfolio.
You contend then that taking a “fair-value” view of the assets on the balance sheet does not generate much useful information for investors?
In my view this is a crucial point to make. Fair-value accounting, which is what the accounting standards bodies are now nailing their colors to, by definition sees value as being communicated through the balance sheet. The US Financial Accounting Standards Board (FASB) and the International Accounting Standards Board (IASB), in developing their “conceptual framework” for accounting—which is designed to harmonize the standards-setting process between the two bodies—appear to be committed to this balance sheet focus.
In their view, if you measure value in the balance sheet, then earnings will fall out as simply the change in balance sheet measurement. My contention is that, for anyone interested in a company’s value and its actual and potential earnings capability with a view to investing in that company’s shares, this approach is misguided. Accounting for value looks not to the balance sheet but to the company’s income statement for an assessment of value, and then adds to this to the information in the balance sheet.
It should be clear from what I have said that fair-value accounting is not the same thing as accounting for value except in the case of the example just given of an investment company. An accountant cannot hope to capture value by listing assets and liabilities at their fair value, defined as their present price in the market.
The role of the accountant, then, is to avoid all speculation? Is this possible?
As I say in the book, accounting defines reality by bringing specificity to what would otherwise be speculative generalities. It takes concepts in common use by economists, such as “revenue,” “cost,” “income,” and “assets,” and applies measurement to them. “Cost of production” is, fundamentally, an accounting measure, and the figure that emerges depends on how one does the accounting. “Economic profit” is just a phrase, a pretentious label, until accounting gives it content. There is much in accounting that is an art rather than a science, and that relies on professional judgment. How “profit” is determined, for example, depends very much on the accounting conventions that are used, but there is a solid, agreed body of best practice here as far as historic cost accounting is concerned and investors, by and large, are comfortable with this best practice.
Without accounting, concepts such as “profitability,” “financial position,” and “growth” are just speculative ideas in the mind of the beholder. Accounting forces concreteness—not simply concrete numbers, which are important, but also concrete thinking. It gives investors the chance to ground their analysis of an organization on something solid. In the heyday of “efficient market” theory, accounting was held not to matter. It was said that the market “sees through” the accounting to the future cash flows of the company concerned. When we analyze this idea, however, it turns out that what the market is “seeing” is some form of information that allows it to forecast future cash flows, and ultimately the information that the market is seeing flows from accounting. We can, of course, see factories, employees, the movement of goods, and the delivery of services, but accounting produces a representation of these realities that is appropriate for valuation.
Proponents of the efficient market have a tendency to dismiss accounting as unconnected to reality, an archaic system that is unrelated to cash flows. This is a gross misconception. Accounting forces management to face the numbers when reporting to shareholders, rather than simply delivering platitudes about plans and prospects. Similarly, sound government accounting forces politicians to be straightforward in reporting to taxpayers—to view borrowing as a debt, for example, rather than as a miraculous revenue stream.
For a few quantitative investment techniques that focus purely on stock volatility and the correlation between stocks to build a portfolio, accounting “facts” may not be part of the data sets that the funds look at to pick the stocks to make up their portfolios. This might seem as if accounting has been cast aside, but accounting is still there, of course, in that the vast number of investors who are not “quants” use accounting facts in their investment decisions. Their decisions, informed in part by accounting “facts,” set the prices of stocks, and the changes in those prices constitute the volatilities used by quant funds. Volatility, then, is simply a measure of investor and market uncertainty. It is a very useful measure, but it certainly does not supplant accountancy, without which uncertainty would be rampant!
How do you see the interaction between accounting for value and financial engineering, or the building of mathematical models to forecast cash flows and to determine where investment can profitably flow?
Modern finance, and modern financial engineering, begin with the no-arbitrage principle. This principle is the cornerstone of modern finance. Its insight is that prices are set in relation to each other such that there is no profit to be gained from selling at one price and buying at another. Prices cannot be arbitraged. Oil should trade in Rotterdam and New York at the same price, adjusted for transport and other transaction costs. Oil futures trade relative to spot prices such that there is no advantage to arbitraging the two. And the price of a call or put option on a stock must bear a no-arbitrage relation to the stock price.
The power and usefulness of the no-arbitrage principle is that it enables theorists, particularly mathematical modelers, to move much more cleanly towards their goals, unencumbered by a whole collection of assumptions about consumer or investor likes and dislikes. Price and risk can be put into direct relation to each other, with additional risk being rewarded by additional price.
However, although this relationship between price and risk is very useful for fundamental investors, it is important to realize that value investing picks up the no-arbitrage principle from a different standpoint to mathematical modeling. For the value investor, what the no-arbitrage principle speaks to is value, not price, and the principle is really a function or summary of the role that information plays in decision-making. This is what drives it. Prices obey the no-arbitrage rule, as far as fundamental investors are concerned, not because of their relationship to other prices, but because of their relationship to information. As I say in my book, information is the arbitraging mechanism in the market. Prices gravitate to fundamentals as information on which value is based is recognized by the markets.
One of the most important contributions of modern financial engineering is the Black–Scholes formula for pricing derivatives, which was formulated by Fischer Black, Myron S. Scholes, and Robert C. Merton. This rests on the no-arbitrage principle and has been hugely useful as it has enabled financial engineers to develop a wide range of hedging instruments which reach through to the underlying prices of whatever is in question, be it share prices, indexes, commodities, or mortgages. This has had enormous benefits since it has enabled the development of markets in those instruments, enabling enhanced risk-sharing in the economy and greatly improving our understanding of risk.
For the value investor, the positive outcomes of all this are many and various. The investor can now focus on the alpha performance they are after, while specifically identifying and hedging out the risks that they do not want to be exposed to. Of course, there is a cost to buying insurance against those other risks, but evaluating this cost against the benefits of the hedge is now part of the investor’s skill set.
Where it all gets tricky, however, is that these new financial instruments can be used as speculative tools as well as being pressed into service as hedging tools. The price at which insurance can be bought can become the prime area for speculation.
There is a more basic problem for the fundamentalist, however, based on the distinction between price and value. The Black–Scholes formula might reach through to the underlying price of the equity in an option contract, but it is silent on its value, assuming that the two are identical. From this standpoint, for fundamentalists the no-arbitrage principle is no help at all.
So the question then is how should fundamental investors regard financial engineering innovations? It is clear that they should welcome the risk-sharing opportunities that financial engineering products offer, for now (in the language of active investing) investors can be exposed to “alpha” (manager outperformance) while buying insurance against risk factors to which they do not wish to be exposed. But fundamentalists have reservations. They see arbitrage opportunities, so no-arbitrage engineering goes against the grain. If you are buying a business rather than a share, then you recognize that businesses are fundamentally about arbitrage. They trade in input markets (assets, labor) and in output markets (customers) on the basis that they sell high (to customers) and buy low (from suppliers). That is arbitrage. When you buy a share of a business, you are essentially expressing your confidence in its ability to operate the arbitrage principle! The no-arbitrage principle comes in one level above this, as it were, and for the fundamental investor what it states is that information about that business’s ability to operate the arbitrage principle is available to the market and therefore determines that business’s value to investors. It is that value which the principle is stating cannot be arbitraged. Put like this, the principle is seen as no more than a probability statement about the likelihood of investors all reaching a similar conclusion on value based on identical access to information. Differences between price and value have to do with which kinds of information a market is mainly paying attention to. If fear is driving the markets, then value information often has little traction and whole sectors are marked down. This misalignment of price and value creates opportunities for value investors. What is being arbitraged here is not price, but the misalignment of price and value.
Looking again at the information content, what is clear is that a value investor needs an accounting model that accounts for how the shareholder’s value is increased (or diminished) by the business’s ability to arbitrage inputs and outputs. This brings us to the negative use of financial engineering. It is clear that for a given accounting system, say US GAAP (Generally Agreed Accounting Principles), financial engineering will be used to structure transactions to get around the requirements of accounting. Lease transactions can be arranged to take leasing contracts off the balance sheet. Assets and liabilities can be moved into “special-purpose vehicles” (SPVs), and borrowing can be restructured to look like an option on stock. Fundamental investors look for ways to finesse these masking activities by bringing that information back into the accounting model. The energies and expertise of the accounting standards bodies would be much better employed focusing on undoing these negative impacts of financial engineering, rather than continuing with their current enthusiasm for balance sheet accounting.