There is never any justification for things being complex when they could be simple." - De Bono
EXECUTIVE SUMMARY
Are we overcomplicating accounting by seeking to apply fair value principles to inherently complex transactions that may not ultimately revolve around cash?
The world of accounting has evolved significantly from its simple origins in tracking financial transactions and assessing profits or losses. Fair value accounting emerged as a response to the complexity of modern business transactions, aiming to provide a more future-oriented perspective.
We critically examines the fair value accounting approach, emphasizing its limitations, implications for assessing true profitability, and the trade-offs it makes between relevance and reliability.
In summary:
Fair value accounting, in its quest for relevance, has sacrificed reliability and comparability, particularly in transactions that lack active, liquid markets.
The complexity of fair value models and the interdependence of assumptions can lead to overly intricate financial statements, hindering their comprehensibility and usefulness.
The inability of fair value changes to reflect the true profitability of a business often compels companies to rely on non-GAAP measures to gauge their financial health.
The fair value dilemma is evident in scenarios like financial services firms using fair value accounting for loans, which makes it challenging to assess the profitability of originated and sold loans.
Complex models with numerous assumptions, often untested in real-world conditions, lead to imprecision in financial statements, raising questions about the reliability of estimates.
The undue emphasis on multiples, NAV, or analyst-quoted metrics for valuing businesses may introduce bias and undermine the reliability of financial assessments.
Concerns extend to the ability of individual shareholders and investors to comprehend the complexities of fair value accounting, raising the need for transparency and independent validation of assumptions.
This article advocates for a disciplined approach in developing, maintaining, and monitoring fair value models, especially in assessing the impact of assumptions on each other.
In the pursuit of applying fair value accounting to a wide array of assets and liabilities, we have inadvertently complicated accounting, leaving many to grapple with how to account for complex transactions that are heavily reliant on future cash estimates. It is imperative to remember that accounting should serve to 'account' for financial transactions, not artificially create their value.
Instead of perpetuating unnecessary complexity, perhaps it is time to embrace the wisdom of simplicity in accounting, as echoed by Edward De Bono: "There is never any justification for things being complex when they could be simple." Therefore, it is essential to reconsider the place of fair value accounting, reevaluating its relevance and reliability, and making an informed choice about its appropriate application in the accounting landscape.
PREFACE
Considering whether a company’s ‘fair value’ is correct is akin to Murder/Suicide… no one benefits!!!
This article is meant to stimulate thought and not somehow assume we know all of the answers. We strongly believe and state our opinions including our belief that simple is better and staying focused on the fundamentals has never resulted in failures of companies. It is the desire to complicate the simple that has resulted in the greatest company failures.
FAIR VALUE
The history of accounting was intended to be a simple way to track transactions, assess net worth and determine profits or losses. The double entry system is elegant in its construction as it allows a person to quickly determine the performance of a business, in its simplest form. The historical cost model of accounting captured this effectively, albeit past (historical) performance. Now that business transactions are more complex than those transacted in the 15th century, it seemed that fair value accounting may be the next evolution, given an appreciation for recognizing that future performance was more useful than a mere presentation of historical results.
However, are we just unnecessarily complicating the way we account for transactions due to their inherent complexity versus seeking the core of almost any transaction which revolves around cash. If it does not revolve around cash, now or in the future, then why should we spend one more minute attempting to rationalize such a complex transaction? One would argue that we have complicated accounting so severely that much of the discussions around complex transactions are focused on how to account for such transactions that the future estimated cash impact can be reasonably estimated. Herein lies the fair value dilemma, which invariably is at the heart of many complex transactions.
As in every paradigm shift one has to realize the limitations. The objective of this article is not to propose an alternative model but rather to identify the limitations of the fair value model. Since the introduction of fair value basis of accounting, the application of fair value has permeated through the accounting world to encompass many assets and liabilities. While many assets and liabilities that are actively traded (we stress the notion of ‘actively’ vs ‘inactive’ or not widely held) can have easily determinable fair values, the application to assets and liabilities, where there are many assumptions, begs the question if we have created the conditions for abuse (current and future).
While this may create job preservation for many, and quench the thirst of many who make their livelihood valuing companies, assets, or estimating whether certain inputs and assumptions are valid or reasonable, one should consider, is this what accounting was meant to do, or was it supposed to remain grounded in the fundamentals.
Advertisement
There is never any justification for things being complex when they could be simple. ~ Edward De Bono, A Maltese Physician who introduced the term lateral thinking and its application.
The International Accounting Standards Board (“IASB”) advocated the use of the fair value approach as the most relevant measurement basis, requiring a process of asset/liability recognition, initial measurement (at fair value), re-measurement (again, largely, at fair value) and de-recognition. When The Financial Accounting Standards Board (“FASB”) issued ‘SFAS 157: Fair Value Measurements’, it was to codify a basis for fair value measurement as this was previously dispersed in various accounting interpretations. IASB’s position requires that the meaning of ‘fair value’ is both clear and unambiguous, and that the fair value of an asset or liability should be determined with sufficient reliability to justify its use as the primary basis of asset/liability measurement.
However, in achieving a relevant basis of measurement we have sacrificed reliability and comparability. Fair value is relevant for instruments and assets that are actively traded in deep liquid markets. Beyond that use it just creates a level of complexity which leaves the financial statements less comprehensible and less useful.
Additionally, the aspect of accounting being able to reflect the profitability of an organization is limited with fair value changes that do not allow for the meaningful analysis of true profitability, hence the reason why many companies fail, but attempt to use non-GAAP measures such as ‘Adjusted EBITDA’ to somehow arrive at a cash measure which is void of fair value assumptions and estimates.
For example in a financial services firm that originates loans and uses fair value accounting for its loans, the time between the origination and sale of the loan may be a few months and the fair value changes do not allow the company to assess if the loan originated and sold was actually profitable? Then how does fair value accounting help one of the users of the financial statements i.e. the company itself or investors who seek to understand a company’s financials in sufficient depth to allow for informed investment decisions.
Dealing with complexity requires much attention and mental energy and is inefficient, unnecessary, and a waste of time.
FAIR VALUE - A PRIMER
All models are approximations. Essentially, all models are wrong, but some are useful. However, the approximate nature of the model must always be borne in mind… ~ George Box, Statistician
Fair value measurement is applied to assets, liabilities and/or businesses – see Graphic 1 below which shows the fair value measurement hierarchy.
Graphic 1: Fair Value Measurement Hierarchy
Fair value is easy to determine for assets that are actively traded in liquid markets. These assets typically tend to be financial instruments or commodities. Where assets or liabilities are not actively traded fair value can be based on observed market inputs or data. In simple terms this means the use of market data available in arriving at the fair value measure. This itself has some limitations, for example, if a company with one million shares outstanding which is held by unrelated parties and has four share transactions in a year where related parties buy shares, does the value of the observed transactions constitute a fair value? We think NOT!
For assets or liabilities and liabilities where there is no actively traded data or observed data, fair value is measured using models. Modeled fair values can be simple or complex, and should be based on the nature of the transactions. However sadly, many models are as old as time and those who use them, often try to force the nature of the asset or liability to fit the model.
Many assets that are fair valued using models tend to be complex. Risks associated with the business, location, asset class, and nature of operation also impact the models utilized.
WHAT MAKES THESE MODELS SO COMPLEX?
Some of the variables affecting these models include, but are not limited to, the following:
New or proven/existing assets, liabilities or businesses
While it is easier to consider existing assets, liabilities and businesses, the recent experience of the company in developing assumptions and any relative bias should be taken into consideration. Often this is overlooked or ignored (consciously or subconsciously) on the premise that somehow management’s inability to generate cash in the past has no bearing on their ability to generate cash in the future and that the future surely will be better than the past (See discussion on estimation bias in this article)
Competition
The level of exposure to competition for a particular asset will determine the degree of reliability of estimates. If an asset is within a very competitive market with low barriers to entry, the reliability of the Company’s internal estimates should be viewed with skepticism as most internal estimates for future sales or revenue are predicated on historical data. Also, in certain industries, it does not take many years for a competitor (existing or start-up) to take a sizeable share of the market, rendering certain assumptions on future value useless.
Barriers to entry
Industries that are highly capital intensive or require large sums of money to get started or require specialized skills which may be scarce, may drive higher valuations and higher risks for those trying to enter such markets or industries. However for those few that may already be established the risks may be lower, though the valuations of such businesses and their assets still depend on many facts, not all of which are quantitative. Such variables and factors include, but not limited to, the following:
Succession planning – Is the scientist aging without a clear strategy or plan for his/her replacement with similar intellect?
Are all the ‘special’ formulas and/or recipes sufficiently documented and safeguarded?
Experience of key employees and supporting staff.
What value has a third party recently quoted for the company and do you know what internal metrics (qualitative and quantitative) were used by such a third party in arriving at their valuation?
Only player in the market or are there a few other competitors, albeit fewer than other industries.
How secure is the company’s IP, is it likely that their products or technology can be easily duplicated or replicated.
You can see that there are many factors and it is difficult to apply a risk factor with great reliability to these variables; so how then does it become easier to value an asset used in such a business, or to value the business itself.
Public or private – access to information
The efficiency of capital markets is based on the available of public information. If there is limited public information or the access to this public information is controlled, the fair values will reflect this information bias.
Existence of expertise
Is there a process within the organization that allows for broad discussion of the assumptions used on the models and uses differing points of view to discuss and debate values or is the fair value determination a check list process where an analyst obtains data points, updates a spreadsheet and as long as values are within what recent market analyst reports indicate then there is no further assessment.
Duration of:
Asset life
Liability – this is normally easier than the asset life, as it would normally be contractual with terms
Business’s ability to generate cash flow – going concern
People
Experience of the leadership and those equipped to use the assets and run the business to generate cash flow:
Is a person’s or Group’s past success a reasonable indicator of future success or was the past success a matter of being in the right place at the right time or just ‘luck’.
How many successes does one need to prove themselves; could Jack Welch have being able to re-vive Apple 10 years ago, or Steve Jobs was the only one who could and did?
Social and Economic variables:
Recession
Currency
Global events
Dependency on commodities or other inputs such as ‘Oil’, Power, Natural gas, etc.
Political stability
Skills/Talent Pool – scarcity
Time Value Of Money
Discount rates to approximate the time value of money and the above risks. This area is probably the most subjective as there are many opinions and ideas but it also represents the one variable that can be most manipulated with to achieve certain results.
SIMPLE IS POWERFUL
As Occam’s Razor states “other things being equal, simpler explanations are generally better than more complex ones”. Should we not view complex models with many assumptions with some degree of concern? Especially if one cannot demonstrate the impact of one assumption on the other. E.g. Interest rate, product growth rates, economic growth rates do not operate in a vacuum. They are interdependent. Using each of these assumption selectively from varying sources for different economic scenarios will lead to error prone models; returning to our first comment, all models are wrong, and we all know it, but do we? Many who know it, hide behind the fact that they are all estimates and are prone to error or significant inaccuracies as is inherent in any future looking assessment. However, we do not highlight this fact, except in an opening or ending legalistic note which cautions investors on the risks of relying solely on the information provided in such valuations.
In some organizations there is a very strong disciplined approach towards the development, maintenance and monitoring of these models. Many organizations may not maintain such a discipline and control framework and may be more prone to modelling risks. For example the fair value of mineral rights is a complex model, it involves commodity price assumptions, currency assumptions, raw material input assumptions such as oil, quick lime, coal, electricity, natural gas, explosives, delivery/transportation, changes in local taxes and regulations, etc. Some of these can be estimated for a period of time with relative certainty. However, what happens to considerations that are 20 years (or longer) in the future?
Gold PM Fix prices averaged 1411/oz, $1,265/oz and $1159/oz in 2013, 2014, and 2015, respectively, and Oil Prices averaged $98/barrel, $88/barrel, and $51/barrel for 2013, 2014, and 2015, respectively. In 2014, the average gold price used to establish reserves was $1398/oz and $1284/oz in 2015. In the first week of 2016, oil prices hit a new low of under $40/barrel. While this won’t last, can you say exactly when it will turn around, how fast, and at what price levels? Why should we add this level of imprecision to the financial statements and complicate matters further.
Particularly when valuing businesses there is sometimes an undue emphasis on using multiples or citing analyst quoted multiples, NAV or other similar metrics. If analysts had a better understanding of markets then why are certain market events considered extraordinary. Market analysts only offer their opinion, and like everyone else in this world they are affected by their own biases.
We are sure, that many will say the users of the financial statements should understand the limitations. Well that is true for the financial analysts within the user community, however how many individual shareholders would be able to understand these complexities? Also the other element that should concern users are how do you eliminate group think from the development of these assumptions. If there are no independent sources for the development of the key assumptions, how reliable are the assumptions and the consequent models. Is there a need for an organization to demonstrate how its assumptions compare to actuals data points and adjust their models for any estimate bias? Is there a need to consider the limitations of statistical techniques?
As shown in Graphic 2 and the accompanying data set below there is a need to be cognizant of the limitation of statistical techniques.
In this article, what we have not done is attempt to arrive at a solution for all of these questions that we raise, as sometimes, the answer is to just keep things simple. Attempting to complicate things further with complex solutions to complex transactions is a futile exercise and one we do not wish to undertake.
Graphic 2: Forecast of Sales Units using Linear Regression and Estimation Bias
CAUTION TO THE WISE
“Mark to Market accounting is like crack cocaine” - Fastow
For those readers who came into the accounting profession recently, Andrew Fastow was the former CFO of Enron; perhaps one of the greatest disasters in corporate history.
In a time where there is an excessive focus on short termism and meeting quarterly targets, fair value accounting creates the opportunity to use assumptions more suitable to meet the expectations of the market. Fair value accounting has its place, it is suitable for financial assets and liabilities that are actively traded and are readily marketable.
If the most value quality of fair value accounting is relevance, how relevant is it if there is no context for the reader to understand how the valuations relates to the market value. Presumably, if assets and liabilities are measured at fair value, the value of the equity should approximate market value. Is this a test that should be applied to most public companies to determine if the asset values are appropriate?
Should assumptions used by the company in arriving at fair values be provided with contextual information? Sensitivity helps but requires context in order to be useful, for e.g. if based on metal prices, are metal prices at a historical high or has there been a selloff post the balance sheet date that would significantly impact the fair values disclosed. The aftermath of the financial crisis of 2008 exposed some failings of the fair value approach, using fair values when markets are functioning sub optimally does not provide a good indication of fair value. The question is, how does the average investor know what is sub-optimal?
If the argument for using a fair value based balance sheet is that it provides the current value of the assets and liabilities, it seems odd that in most acquisitions that we have been involved in, the starting point for the value of a business is based on discounted estimated future cash flow models (DCF or otherwise) or multiples of cash flow, revenue or EBITDA or net income. So, if the relevance of the fair value balance sheet is limited, what is benefit of using this approach? Is it just to satisfy an academic curiosity of what current value accounting would look like compared to historical cost accounting?
While fair value accounting has its limitations it is still better that historical cost accounting which uses outdated information and provides less value to the users of the financial statements. Therefore, if we all agree that this is the case, then should be done?
RECOMMENDATIONS
All assets should segregated between what can be fair valued or not
Where fair values are derived from models, then the range of the values should be presented.
Given the ranges can also be subject to biases, provide contextual information relating to assumptions, to help users of the financial statements understand the inherent bias in the data used.
For example if the asset was based on a commodity price assumption, then providing the price of that commodity over the past 12 months and a 10 year period will allow readers to understand the volatility of the commodity price and be able to determine if management’s assumption of the commodity price is overly optimistic or not.
If models utilizes several assumptions the possibility of errors multiply.
Creating a discipline around the creation and maintenance of the financial models, including regularly back testing actual results to determine if the models are still effective.
It is important for all of us to realize that fair value accounting framework as it exists today has its limitations and is not an elixir that will cure all the woes of accounting that were ascribed to the historical cost model. No business ever transacts based on fair values derived through the accounting processes; as accounting does not drive business values, instead it is only meant to ‘account’ for it, not create it. To portray it as anything else projects a falsehood and diminishes the value of the financial statements. The emphasis of fair value is on properly portraying the balance sheet values but given the issues that exist with fair value accounting framework as it exists maybe the emphasis should be on the cash flow statement and revisiting its presentation, because Cash Never Lies.