Income Volatility and the Valuation of Investment Property

Income Volatility and the Valuation of Investment Property

Cecilia Lambert†

Christopher Lambert†

Abstract

Sustained debate on the appropriateness of adopting fair values for financial assets has ensued in the wake of the current economic crisis. Accounting regulators are addressing implementation issues in measuring financial securities when markets are illiquid. Accounting regulators are now sanctioning greater use of management assumptions as the basis for fair value, with mark-to-model valuations replacing mark-to-market valuations. However, greater the use of unobservable inputs lowers the reliability of the valuations. This study contributes to an understanding of the application of fair values to non-financial assets by examining financial statement volatility in the measurement of investment property. Moreover, prior literature suggests that weaknesses in corporate governance at the firm and national level (such as management incentives to choose less reliable fair values, implementation and enforcement of standards) do not reduce information asymmetry despite greater transparency achieved through the market valuation of investment property. This paper explores measurement error in the valuation of investment property in the context of corporate governance mechanisms adopted by real estate companies in the UAE.

†ZayedUniversity, Abu Dhabi, United Arab Emirates

1

Introduction:

Much has been written lately on the valuation of financial assets in the wake of the current economic crisis. Accounting has been accused of exacerbating the “meltdown” in the US banking system, of being pro-cyclical as financial institutions wrote-down investments, then attempted to raise capital to meet prudential requirements, sold assets at “firesale” prices which drove prices and valuations down even further (Johnson, 2008; Anon, 2008a).

An article published in the Financial Times in March 2008 commences “Accounting rulemakers have defended the use of ‘fair value’ accounting against attacks from bankers and insurers, who claim that applying it to financial instruments in the current turmoil risks undermining its financial system.” (Hughes, 2008c). Hughes (2008c) concludes that “(t)he row over fair value is pitting executives against many investors and the accounting profession. Most alternative suggestions involve heavier reliance on subjective management estimates, which investors dislike.”

However, goal of accounting regulators is not to promote financial stability. “Ultimately…responsibility for interposing a circuit-breaker between market process and banks’ capital adequacy falls on bank regulators, not accountants” (Anon, 2008a), a sentiment shared by Sir David Tweedie, Chairman of the International Accounting Standards Board (IASB), who has reportedly suggested that the nexus between accounting rules and the regulatory capital of banks be relaxed (Hughes, 2008b) as a salve to pro-cyclical tendencies.

Accounting regulators have turned, instead, to issues in the measurement of financial securities when markets are illiquid. An expert advisory panel of the IASB has recently provided new guidance to preparers on measuring and disclosing fair values in compliance with IFRS when markets are no longer active (IASB, 2008). The IASB has also proposed changes to extant regulation to effectively diminish the volatility in reported earnings while extending the disclosures for valuation techniques that “mark to model” (rely more heavily on unobservable inputs to derive fair value) rather than “mark to market” (use observable, market prices). Their solution substitutes transparency in the estimates used by management for transparency of movement in prices. For some, the modification to the relevant accounting standard to allow reclassification of some financial assets (thereby smoothing losses) was achieved at the expense of a loss of reputation for the IASB and its Chairman, accused of “caving in to political pressure” from European financial institutions (Hughes, 2008a).

In the US, the Securities and Exchange Commission (SEC) has been charged with responsibility for analyzing the impact of fair value accounting on the balance sheets of financial institutions and examining alternative methods as a response to the credit “bailout”. In the interim, in September 2008 they also issued a statement to provide interpretative guidance on the use of management assumptions in the absence of market evidence of prices, stating that “in some cases using unobservable inputs…might be more appropriate than using observable inputs”. It stopped short, however, of suspending use of the accounting method.

The IASB and FASB are currently jointly pursuing a theoretical discussion of fair value measurement which can only broaden the debate on the acceptability of fair values especially for non-financial assets. This study contributes to an understanding of the application of fair values to non-financial assets by examining financial statement volatility in the measurement of investment property.

In addition, prior literature has documented the value relevance and reliability of fair values for investment property (Owusu-Ansah and Yeoh, 2006; Dietrich, Harris and Muller, 2001; Muller and Riedl, 2002; Muller, Riedl and Sellhorn, 2008), providing evidence of the utility of fair values to investors. However, Muller, Riedl and Sellhorn (2008) show that information asymmetry is not mitigated by the mandatory provision of market values. Weaknesses in corporate governance at the firm and national level (such as management incentives to choose less reliable fair values, implementation and enforcement of standards) persist despite greater transparency achieved through adoption of IFRSs. This paper explores sources of measurement error in the valuation of investment property in the context of corporate governance mechanisms adopted by real estate companies in the UAE.

Institutional Environment:

International Financial Reporting Standards (IFRSs) governing property are contained in International Accounting Standard (IAS) 16 and IAS 40. The specific regulation governing accounting for investment property is contained in IAS 40. IAS 40 defines investment property as property held to earn rentals or for capital appreciation or both, rather than use in production and sale in the ordinary course of business. The latter type of property is referred to as owner-occupied property and is accounted for under the rules contained in IAS 16.

Following general principles for initial recognition of assets, investment property is recorded at cost. Thereafter the company may choose either to remeasure to fair value periodically (the fair value model) or maintain initial cost (the cost model) subject to depreciation and impairment write-downs. Fair value is defined as the price at which the property could be exchanged between knowledgeable, willing parties in an arm’s length transaction. Changes in fair value are recognized in income for the period.

In contrast the cost model requires investment property to be measured at depreciated cost (less accumulated impairment losses). Systematic charges are made for depreciation (as provided in IAS 16) and additional charges (or reversals) for impairment losses (gains) as appropriate. Although the cost model is the basis of valuation in the balance sheet, fair value must be disclosed as a note to the accounts.

In terms of balance sheet values, where prices are rising the fair value model will have higher asset and shareholders’ equity values relative to the cost method. Incontrast, the cost method defers gains until realized and hence the amount of any gain (loss) on disposal is likely to be larger than the fair value model (which regularly remeasures asset value).

IAS 16 prescribes the accounting treatment for property, plant and equipment used for productive or administrative purpose. Like IAS 40 it permits choice between two methods of accounting: cost or revaluation. The cost model is as previously described for IAS 40 with the asset carried at its cost less accumulated depreciation and accumulated impairment losses. The revaluation model differs from IAS 40 in one important respect: the treatment of changes in fair value.

A gain in fair value is recognized in a reserve, not directly in income. In contrast, a fall in value is recognized in income except where a balance in the reserve is sufficient to offset the loss. While the reserve forms part of stockholders’ equity the gains do not pass through the income statement (constituting a “dirty surplus” as the income statement does not articulate fully with the balance sheet). Thus a company using fair value would report similar asset and equity under either IAS 40 or IAS 16 but reported income will be higher under IAS 40.

IAS 40 provides for a two-level hierarchy of evidence of fair value, with “the best evidence…given by current prices in an active market for similar property in the same location and condition and subject to similar lease and other contracts” (IAS40.45). In the absence of such value, active markets in dissimilar properties, similar properties in less active markets and discounted cash flows may be used. To the extent that fair values are not readily observable they are approximated by using more unobservable inputs. However as unobservable inputs increase the perceived reliability of these approximations of fair value diminishes.

Fair value is a market-based measure, not entity-specific. In contrast, impairment testing requires an estimate of the individual asset’s recoverable amount, or the recoverable amount of the cash-generating unit to which the asset belongs if the former cannot be estimated, thereby introducing entity-specific estimates into asset valuation. Recoverable amount is the higher of an asset’s net selling price and value in use. Since value in use represents the stream of future cash flows expected from continuing use and disposal of the asset, management expectations and entity-specific considerations contribute to be greater variation between the values determined under either method.[1]

Financial Statement Volatility and the Valuation of Property:

As fair value encapsulates expected future cash flows, any changes in expectations will result in a change in fair value. Barth (2004) describes three sources of volatility in financial statements: estimation error volatility, inherent volatility and mixed-measurement volatility. Estimation error relates to the degree of precision in measuring the variable of interest. For investment property it may arise from, say, illiquidity in local property markets, management incentives to distort estimates and differences in monitoring (internal and external audit, regulatory enforcement) of the application of standards (Muller, Riedl and Sellhorn, 2008). Estimation error, and hence its volatility, is expected to be greater for fair values determined using estimates of fair value (unobservable inputs) than for observable market prices, given that the former are reliant on the precision of the estimates (and models) used to derive fair value.

Following this line of reasoning, the estimation error in the valuation of investment property is lower than owner-occupied property. IAS 40.07 explains that investment property generates cash flows that are largely independent of other assets of the firm while the cash flows generated by owner-occupied property are interdependent with other assets of the firm. This difference in economic attributesuggests that there is more likely to be market evidence of fair value for investment property than for owner-occupied property with owner-occupied property more likely to require unobservable inputs in estimating fair value. Moreover, the synergies between assets are not separately recognized in financial statements which, as a general rule, recognizes individual assets only (Barth, 2006)

Estimation error will also vary with market liquidity (Muller, Riedl and Sellhorn, 2008), with liquidity inversely related to estimation error. Furthermore, where a property satisfies a dual purpose the standard recommends bifurcation if feasible, otherwise the property is classified wholly as investment property if an “insignificant portion” is held for owner-occupation. Hendriks (2005) describes the valuation issues in apportioning value between land and buildings and, in the case of investment property, the attribution of differences in rental income to either component. Thus as dual-purpose property requires either apportionment of value between its uses or disregards it, estimation error is increasing in property with dual-purposes.

The reliability of fair value measurements (and the perception of reliability held by users of accounting information) in enhanced by introducing and communicating information on governance mechanisms in place. Boyles (2008) describes a range of governance arrangements that contribute to the reliability of measurements. Typical governance mechanisms for fair value measurements are the use of independent, qualified valuers together with high-quality auditors.

Inherent volatility refers to the underlying (economic) volatility of the variable. The greater the inherent volatility, that is the greater the uncertainty or timing of future cash flows the more variation in the expected realization of the asset, and hence the greater the volatility in net assets and income for firms using fair values. Property markets vary in momentum and direction. As they do so, expectations of future realizations of cash flows will also vary, and these changes will flow through the financial statements as material movements with, from time to time, reversing direction. Inherent volatility is exposed in the actual realizations of the asset. Realized gains and losses provide historical information on the variance in asset valuation.

The mixed-measurement model, in which assets are measured using a variety of measurement bases such as cost, lower of cost and current market value, fair value[2], will also produce volatility in financial statements, as will the use of fair value for assets but not liabilities as market interest rates change. Barth (2004, p. 327) explains “(a)n interest rate increase has a relatively larger negative effect on the financial statements, that is, lower net assets and lower net income, than it would if assets and liabilities were both measured either using fair values or using historical cost. Similarly, a decrease in market interest rates has a relatively larger positive effect.”

Thus it is expected that the greater the commonality in the measurement bases for assets and liabilities, the less exposed the firm to mixed-measurement volatility. Thus firms with greater use of variable-rate borrowings and adopting fair values for measuring assets will have lower mixed-measurement volatility than firms using fixed-rate debt with fair valuation of assets. Similarly firms using cost to value assets and fixed-rate debt will experience lower mixed-measurement volatility than firms using cost and variable-rate debt.

Table 1 summarizes the hypothesized relationship between mixed-measurement volatility and matching of asset-liability measurement bases.

Table 1: Relationship between the Valuation of Assets and Liabilities and Measurement Volatility

Asset/Liability / Variable-rate debt / Fixed rate debt
Fair value / Low / High
Cost / High / Low

To summarize,, high market liquidity reduces the estimation error in fair values of investment property. Estimation error is also lower for firms using independent qualified valuers for investment property and high-quality auditors. Finally, firms with single-purpose property should also experience lower estimation volatility than firms with dual-purpose property. The adoption of fair value accounting. however, produces potential volatility from mixed-measurement attributes in the financial statements. Firms with closer matching between asset and liability measurement bases minimize mixed-measurement volatility.

Application to Real Estate Companies:

To explore these propositions we examine firms with a material investment in investment property in the UAE. We choose firms in the real estate sector to ensure that investment property is a major asset of the firm and hence reporting choice has a first-order effect. The UAE provides an interesting case study for two reasons. First, the property market is experiencing intense growth providing sufficient inherent volatility for firms to experience high-powered incentives to select a specific accounting policy for investment property. Moreover, as land grants are common in the UAE, the “appreciation” of property has considerable upside potential contributing to demand for the fair value model albeit with the volatility that implies.

Sample Selection

There are three stock exchanges in the UAE—the Dubai Financial Market (DFM), the Abu Dhabi Securities Exchange (ADX) and the Dubai International Financial Market (DIFM). The first two are domestic markets regulated by the Emirates Securities and Commodities Authority (SCA)[3] while the DIFM is currently at a much earlier stage of development with its own regulatory arrangements. We restrict our sample to local companies listed on the DFM and ADX.

We confine our sample to firms expected to have a material exposure to investment property. The DFM lists ten companies as Real Estate and Construction. Of these we excluded four companies that were incorporated in Kuwait and a further company for which financial information was unavailable. This process produced a reduced sample of five companies. The Abu Dhabi Securities Exchange does not provide classification criteria to directly identify Real Estate companies; a further three companies were identified for inclusion in the study.

From these eight companies three with no existing investment property or property under construction for future use as investment property, were also excluded reducing the final sample to five companies.