Performance Measurement and Recognition of Real Estate Assets: An International Exploration of Reporting Practices Adopted in the Real Estate Industry February 2008

Research Report for Real Property Association of Canada For the Grant Program to Fund Research Projects Exploring Issues of Importance to the Canadian Commercial Real Estate Industry 2006/2007

Performance Measurement and Recognition of Real Estate Assets: An International Exploration of Reporting Practices Adopted in the Real Estate Industry

February 2008 Steve Fortin, Associate Professor of Accounting 514-498-4021 [email protected] Desmond Tsang, Assistant Professor of Accounting 514-398-5417 [email protected] and Francois-Pierre Dionne 514-967-7228 [email protected] 1001, Sherbrooke Street West Montreal, Québec Canada H3A 1G5

Table of Contents 1

INTRODUCTION ...................................................................................................... 1 1.1 The Use of FFO or Other Non-GAAP Performance Measures .......................... 1 1.2 Reporting of Real Estate Assets.......................................................................... 2 1.3 Disclosures of Real Estate Assets ....................................................................... 2

2

DEVELOPMENT OF THE FFO CONCEPT AND THE NEW IFRS....................... 3 2.1 Non-GAAP Performance Measure ..................................................................... 3 2.2 International Financial Reporting Standards (IFRS) .......................................... 3 2.3 Related Studies – Usefulness of FFO disclosures............................................... 6

3

SAMPLE SELECTION AND DESCRIPTION OF SAMPLE FIRMS ..................... 7

4

KEY FINDINGS....................................................................................................... 10 4.1 Extensive Use of Non-GAAP Performance Measures ..................................... 10 4.1.1 Performance Measure Reporting .............................................................. 10 4.1.2 Other Non-GAAP Measures – Balance Sheet Measures.......................... 11 4.1.3 Adoption of Fair Value Reporting for Real Estate Assets........................ 15 4.2 Disclosure Related to Real Estate Assets.......................................................... 19

5

BEST REPORTING PRACTICES........................................................................... 24 5.1 Property Level Information............................................................................... 25 5.2 External Valuator Report .................................................................................. 26

6

CONCLUSIONS....................................................................................................... 26

REFERENCES FOR RELATED LITERATURE............................................................ 28 APPENDICES .................................................................................................................. 29

1

INTRODUCTION

International Financial Reporting Standards (IFRS), which will be adopted by Canadian publicly accountable entities likely by 2011, will modify the reporting of real estate firms by allowing fair value accounting for their real estate assets. Given the lack of studies on the usefulness of non-Generally Accepted Accounting Principles (GAAP) measures in an international context and the recent trend on the adoption of fair value accounting by international real estate entities (all real estate entities, whether a real estate investment trust or corporation shall be referred to as a “REIT” in this report), our study will explore the following issues: 1. Whether international REITs report Funds from Operations (FFO) and what other performance measures are reported? 2. What methodologies do international firms take in implementing the new IFRS fair value reporting? 3. What disclosures of real estate assets are made under IFRS and what is the level of detail disclosed with regards to how real estate asset values are determined? The North American real estate industry is characterized by the widespread use of a nonGAAP summary performance measure known as Funds from Operations (FFO) as an alternative to net income in measuring profitability. There have been continued debates among industry participants and standard setters on the relative usefulness of FFO compared to net income. Previous studies on U.S. REITs present mixed evidence and show that both measures FFO and Net Income are in general relevant when valuing the trusts. At present there is no conclusion on the usefulness of FFO, or any other nonGAAP performance measures, for the international REITs as well as for the Canadian REITs, a market with a capitalization of over CDN$ 23 billion. 1.1

The Use of FFO or Other Non-GAAP Performance Measures

Previous academic studies (Fields et al. (1998), Vincent (1999)) show that most U.S. REIT firms report FFO information in their annual reports. The common use of FFO in the U.S. may be due to the fact that the National Association of Real Estate Investment Trust (NAREIT) is a strong advocate of FFO and that the Security and Exchange Commission (SEC) supported the development of an “industry standard”, particularly for measures of performance. In Canada, the concept of FFO has been widely used by Canadian REITs since the introduction of the definition (see the White Paper on Funds From Operations, issued by the Real Property Association of Canada (REALpac), 2004). Rules regarding the reporting of non-GAAP financial measures by the Canadian Securities Administrators (CSA) has led REALpac (formerly known as the Canadian Institute of Public and Private Real Estate Companies or CIPPREC), to conclude that FFO may no longer be included in financial statements. Most Canadian REITs report FFO in the Management Discussion & Analysis (MD&A) report.

1

In this study, we intend to document the reporting of FFO and other non-GAAP performance measures by REITs operating in countries that have adopted IFRS. The North American real estate industry argues that net income is an insufficient measure of REIT performance as the measure includes depreciation which is based on historical cost and assumes incorrectly that real estate assets value diminishes predictably over time. Given that IFRS allows companies to recognize their real estate assets at fair value and to not recognize depreciation expense for their investment properties, it is possible that the usefulness of the net income measure has improved for firms under IFRS and that there is a lesser need for non-GAAP alternative performance measures such as FFO. In addition, by reporting investment properties at fair value, it is possible that other performance measures based on the balance sheet may increase in importance. 1.2

Reporting of Real Estate Assets

Under IFRS, a REIT has two options to report its investment properties, as defined within IAS 40 Investment Properties, on the financial statements: 1. at fair value directly on the balance sheet; 2. at cost on the balance sheet, with fair value disclosed in the notes to the financial statements. We intend to examine how companies have implemented IFRS by documenting whether firms choose to continue reporting their real estate assets using the cost method or the fair value method. Should companies decide to recognize their real estate assets at fair values, we are also interested in determining the magnitude of the impact of recognizing the associated gains and losses related to the changes in fair values on reported net income. 1.3

Disclosures of Real Estate Assets

Finally, we examine the disclosure practices of firms with regards to their real estate assets. IFRS requires firms to provide additional details on the assumptions and measurements of property, plant and equipment and investment properties when firms choose the fair value method. In our study, we intend to evaluate the quality of the companies’ disclosures, particularly for the following aspects: 1. Whether companies disclose the use of external appraisers in the determination of fair values for properties; 2. Whether companies explicitly state the assumptions or standards used in the calculation of fair values, and 3. Provide a description and examples of “practices” in the reporting of investment properties, and more particularly what seems to represent “best practices”.

2

By documenting the extent of the reporting of non-GAAP measures and the measurement of real estate assets for the international REITs, we believe our study will provide significant insights for Canadian real estate companies planning for their adoption of IFRS in Canada. 2 2.1

Development of the FFO Concept and the New IFRS Non-GAAP Performance Measure

In financial reporting, net income is recognized as the key summary measure for a company’s performance. However, the REIT industry argues that net income, computed according to Generally Accepted Accounting Principles (GAAP), does not accurately reflect the profitability of real estate assets. The industry argues that historical cost accounting implicitly assumes that the value of real estate assets diminishes predictably over time. However, since historical real estate values have fluctuated along with market conditions, REIT managers consider the use of historical depreciation inappropriate for REIT firms that invest heavily in investment properties. This perceived shortfall of GAAP becomes more severe under volatile or inflationary market environments (such that the values of real estate assets exhibit larger fluctuations leading to larger deviations from their book values). As a consequence, the concept of FFO was developed. It is generally calculated by adding back to net income the amount of depreciation and amortization related to real estate investment properties and gains and losses on the sale of real estate assets. REALpac has promoted FFO as the industry-wide standard measure of a REITs operating performance since the introduction of the REALpac Handbook in 1972. In the U.S., NAREIT clarified the concept of FFO in 1991 and the term soon became widely used in the industry. However, recent law amendments affecting financial reporting requirements in both Canada and U.S. after several high-profile legal cases have put stricter restrictions on the reporting of non-GAAP financial measures. It is questionable whether the new scrutiny on the use of non-GAAP measures has curtailed the popularity of FFO reporting. 2.2

International Financial Reporting Standards (IFRS)

The problems related to depreciating real estate assets, and more generally, of the potential lack of relevance of historical cost-based financial statements, has also been identified by the International Accounting Standard Board (IASB), although their solution is drastically different. As opposed to a restatement of a GAAP measure like the approach adopted for the computation of FFO, real estate firms are now required under IFRS to recognize the fair value of their investment properties in their financial statements (either on the Balance Sheet or in the Notes). Arguably, if the largest item omitted from FFO is accounting amortization, the fact that this expense is no longer recognized under IFRS may reduce the need for FFO. On the other hand, since fair value adjustments are still “non-cash” gains and losses, the need for a FFO-like measure may

3

remain. In our research, we explore whether FFO-like measures are reported by IFRS adopters. The convergence to IFRS in Europe first started with the European Union (EU) when the European Parliament passed a resolution in March 2002 requiring all EU publicly traded companies to prepare their financial statements under the IFRS. The standards were then approved by the Accounting Regulatory Committee (ARC). On October 13, 2003, the approved standards were published in the Official Journal of European Union and became part of European law. Around 7,000 EU publicly traded companies were required to adopt IFRS on January 1, 2005. The adoption of IFRS by the EU has prompted many other countries around the world to harmonize their national standards with IFRS. For instance, the Australian Accounting Standards Board (AASB) had been working towards converging to IFRS and since January 1, 2005, Australian companies are required to follow the Australian equivalent of IFRS. In New Zealand, the Accounting Standards Review Board (ASRB) and the Financial Reporting Standards Board (FRSB) have devised their New Zealand equivalents to IFRS and had recommended early adoption of the new standards with mandatory adoption by January 1, 2007. In Hong Kong, the Hong Kong Institute of Certified Public Accountants had fully converted the Hong Kong Financial Reporting Standards to IFRS by the end of 2004, with the new standards effective for the financial periods starting January 1, 2005. For Singapore, the Council on Corporate Disclosure and Governance (CCDG) has devised a plan of changes to the financial reporting framework since 2004 to converge its local accounting standards with IFRS. By May 2006, the Singaporean accounting standards and IFRS were almost identical despite minor differences in certain standards. As of February 2007, nearly 100 countries or jurisdictions currently require or permit the use of, or have a policy of convergence with, IFRS. (www.iasb.org). In Canada, the Accounting Standards Board (AcSB) is implementing a convergence plan with the IFRS and is anticipating total convergence by January 1, 2011. The Canadian plan of convergence is currently underway and will be an on-going process until 2011. Before 2011, AcSB will continuously phase-in with IFRS by revising the Canadian accounting rules, where possible, to make them consistent with the rules in IFRS. For example, the AcSB has recently revised the accounting rules for inventory in 2007 to eliminate the differences between Canadian GAAP and IFRS. Then in 2011, the AcSB is expecting a final conversion requiring mandatory adoption of all remaining IFRSs for all publicly accountable enterprises. For the U.S., the Financial Accounting Standards Board (FASB) and the IASB have devised a “roadmap” of convergence though there is no anticipated date yet for the total convergence of US GAAP and IFRS. From 2011 until such convergence plans are implemented in the United States, significant differences may exist in the reporting models of Canadian and U.S. REITs. Prior to the adoption of IFRS, accounting bodies of different countries generally had similar views regarding the accounting treatment of property assets but different views on investment properties. Accounting standards have traditionally required companies to carry their property, plant and equipment at cost less accumulated depreciation, a practice in existence in most if not all jurisdictions. On the other hand, accounting standard 4

setters’ views on investment properties were more varied. For example, in the United Kingdom (U.K.), the Statements of Standard Accounting Practice dictated that investment properties were to be revalued every year. The revalued amount was reported on the balance sheet, with the changes in market value of investment properties recorded as a revaluation reserve in shareholder’s equity. The New Zealand standard was based on the U.K.’s but it gave managers the option to recognize the changes in market value of investment properties as a revaluation reserve or as unrealized gains or losses on investment properties in the income statement. In Hong Kong, investment properties were carried at fair market values with changes in the fair market values of investment properties recorded in a revaluation reserve. When the revaluation reserve is calculated as an accumulated deficit, a loss must be recognized on the income statement. In Australia and Canada, there was/is no specific accounting standard to govern the reporting of investment properties, prior to the adoption of IFRS. As an effort to harmonize accounting standards across countries in the world, the IASB and its predecessor, the International Accounting Standard Committee (IASC), developed rules for the reporting of property, plant and equipment and investment properties, IAS 16 Property, Plant and Equipment and IAS 40 Investment Properties, respectively. The rules have been developed to provide guidelines on the accounting treatment of properties used in operations and properties held for investment purposes. IAS 40 and IAS 16, the current standards, were revised in 2003 and became effective as of January 1, 2005. IAS 40 requires investment properties to be recorded initially at cost but allows managers to choose between measuring investment properties at fair value or at depreciated cost subsequent to the initial recognition. If investment properties are measured at fair value, the associated gains or losses arising from changes in fair value should be included in the income statement. The standard requires the disclosure of the methods and assumptions applied in determining fair value and whether an independent appraiser was employed. If investment properties are accounted for at cost, IAS 40 nonetheless requires the assessment of the fair value and disclosure in the notes to the financial statement of the fair value and of other related information. Under the current definition of IAS 16, property, plant and equipment includes long-term assets that provide probable future economic benefits for the entity. These assets should be recorded initially at cost. Subsequent to acquisition, IAS 16 allows two methods for the measurements of property, plant and equipment: the cost model and the revaluation model. The cost model requires assets to be recorded at cost less accumulated depreciation. The revaluation model, on the other hand, requires assets to be carried at revalued amounts, namely its fair value at the date of revaluation less subsequent depreciation, provided that fair value can be measured reliably. As opposed to the standard on investment properties treating gains and losses in the Income Statement, any increase in revaluation will be credited to an equity account named “revaluation surplus” and any decrease should be applied first to any existing revaluation surplus linked to a specific asset then recognized as an expense. If the cost model is applied to properties

5

under IAS 16, it is not necessary to evaluate and disclose fair values in the notes to the financial statements. The convergence to IAS 40 and IAS 16 is obviously less dramatic for real estate firms that were previously subject to a fair value measurement standard, such as those in the U.K. and Hong Kong. These firms, having faced fair value measurement rules for years, only have to adapt to the inclusion of fair value changes in the income statement. Canadian reporters will have to implement periodic valuation systems and determine their disclosure practices. They may benefit from the practices identified in firms with many years of experience with fair value reporting. The last section of our report will document practices observed in our sample firms. 2.3

Related Studies – Usefulness of FFO disclosures

Given the continued debate of industry participants and standard setters on the role of FFO and net income, academic researchers have conducted empirical studies on the U.S. REIT market to test the usefulness of FFO versus net income in the hope of providing policy implications on the reporting and use of FFO. However, the research evidence is mixed. Fields et al. (1998) compares the predictability of FFO and net income on oneyear-ahead FFO, Cash Flows from Operations (CFO), net income and stock prices. They find that FFO dominates net income in predicting one-year-ahead FFO and CFO but not in predicting net income and stock prices. Vincent (1999) examines the relative as well as the incremental information content of FFO compared to Earnings-per-Share (EPS), CFO, and Earnings Before Interest, Tax, Depreciation and Amortization (EBITDA). However, the research results show that all four measures are associated with the trust units’ fair market values. Depending on the econometric models applied, the four measures have different levels of statistical significance on their associations with market values. Moreover, Vincent (1999) also finds that neither FFO nor EPS surprises, measured by their respective forecast errors, are related to contemporary stock prices. Vincent’s findings contrast sharply with the findings by Baik et al. (2005), who show that FFO forecast errors are positively correlated with an announcement period’s abnormal returns. Baik et al.’s study also documents a noticeable downward shift of FFO forecast errors after the implementation of Regulation G and the increased efforts of NAREIT to improve the credibility of FFO in the 2002-2003 period. Another study that evaluates accuracy of analysts’ forecasts of REIT earnings and the information content of earnings forecasts by Juergens (2000) finds the presence of significant levels of bias in the earnings forecasts and that forecast revisions lead to significant price reactions for REIT securities. In two recent studies, Tsang (2007) and Fortin and Tsang (2007) show that FFO has consistently higher accounting quality and lower forecast errors than net income for a sample of U.S. REITs.

6

3

Sample Selection and Description of Sample Firms

In selecting our sample, we begin by examining firms that are included in the FTSE EPRA/ NAREIT Global Real Estate Index. At present, there are three major indices that track real estate firms’ performances around the world. We elect to include firms that are listed in the FTSE EPRA/ NAREIT Global Real Estate Index because the constituent companies in this index are subject to stricter requirements. They must be traded on an official stock exchange, have a minimum market capitalization of $200 millions USD for Asian and North American stocks or €50 millions for European stocks, and are required to provide audited annual reports in English. As of January 2005, there are 247 real estate firms in the index representing 20 countries. We examine whether each of the 20 countries follow IFRS or have IFRS-equivalent standards and exclude countries/ jurisdictions that do not follow IFRS. As a result, our sample includes firms from 16 countries: Austria, Australia, Belgium, Denmark, Finland, France, Germany, Great Britain, Greece, Hong Kong, Italy, Netherlands, New Zealand, Singapore, Spain, and Sweden. Our research requires detailed analysis of financial statement information and handcollection of data. We chose firms with the largest market capitalization in each country for inclusion in our analysis. We also collected a supplemental sample in countries in which there is an official tax status for real estate companies comparable to the REIT status in North America. Our final sample consists of 33 firms. Table 1 lists the sample firms with their countries of operation and market capitalizations as of May 2007. We collect financial statements and other related financial information of the sample firms from the companies’ websites. Since all sample firms are public, their financial statements are audited by independent accounting firms. Previous research (e.g., Pittman and Fortin, 2004) shows that financial statement information audited by the “Big Four” accounting firms (Deloitte & Touche, Ernst & Young, KPMG, and PricewaterhouseCoopers) signifies higher auditing quality and usually implies more reliable financial information. Hence, we first examine whether our sample firms are audited by the Big Four accounting firms to assess the reliability and quality of their financial information. In some countries, firms hire a single auditor to perform their audits whereas in others, two auditors must be appointed. We find that sample firms that are required to hire a single auditor for their audit used Big Four auditors. For firms appointing more than one auditor, five sample accounting firms hire a smaller auditing firm in addition to one of the Big Four auditors and one company retained two Big Four firms.

7

Table 1: Sample Firms in Our Study Firm Country Westfield Group Land Securities British Land Rodamco Europe Gecina Hong Kong Land Liberty International PLC GPT Group Unibail Hang Lung Properties Ltd. Hammerson Klepierre Corio Sino Land Company Metrovacesa Capitaland Immobiliaria Immofinanz Wereldhave IVG Immobilien Wihlborgs Fastigheter Beni Stabili Cofinimmo Castellum Belfimmo Deutsche Wohnen Risanamento Sponda Aedes Babis Vovos International Constructions S.A. Intervest Offices Citycon Sjaelso Gruppen

Australia United Kingdom United Kingdom Netherlands France Singapore United Kingdom Australia France Hong Kong United Kingdom France Netherlands Hong Kong Spain Singapore Spain Austria Netherlands Germany Sweden Italy Belgium Sweden Belgium Germany Italy Finland Italy Greece Belgium Finland Denmark

Market Capitalization (as of May 31, 2007 in € millions) 17,128 9,817 6,698 5,697 5,113 4,916 4,781 4,715 4,361 4,341 3,611 3,388 3,054 3,042 2,897 2,801 2,126 1,792 1,726 1,678 1,566 1,409 1,188 1,096 804 675 656 598 480 418 344 332 205

8

Figure 1 shows the distribution of the Big Four auditors for our sample firms. Figure 1: Big Four Auditors Employed by the Sample Firms

30%

33% Deloitte EY KPMG PWC

17% 20%

Examining the financial statements of each of these 33 firms, we find that our sample firms concentrate on various types of property investments. Though most of our sample firms are diversified REITs that invest in various types of property investments, they tend to concentrate on one or two types of properties. From the companies’ websites and their annual reports, we are able to obtain information on their property investments. We summarize the sample firms by their major property investment types in Figure 2. Our sample firms are investing heavily in office and retail properties. It is interesting to note the low representation of residential REITs in our sample. One explanation is that REITs that concentrate on residential properties tend to require less capital investment and are more likely to be smaller or private entities. Figure 2: Major Types of Property Investments of Sample Firms

15% 30% Office

9%

Retail Residential not enough info

46%

9

4 4.1 4.1.1

Key Findings Extensive Use of Non-GAAP Performance Measures Performance Measure Reporting

We first examine whether REIT companies that are subject to IFRS report FFO and other non-GAAP performance measures. Interestingly, almost none of the firms we examine (except Deutsche Wohnen) reports FFO in their financial statements or in the notes to the financial statements. It appears that the widespread adoption of FFO in the U.S. and Canada does not translate into a world-wide recognition of the FFO concept. Unfortunately, we are not able to establish the cause of this absence of the FFO measure from annual reports. Though the companies we examine do not typically provide the industry-specific FFO, there are a significant number of companies that choose to emphasize measures such as Earnings before Interest and Tax (EBIT) or Earnings before Interest, Tax, Depreciation and Amortization (EBITDA) in their financial statements. The main difference between FFO and EBITDA is that FFO includes interest expense whereas EBITDA does not. We find that there are nine companies reporting EBITDA and another four companies reporting EBIT. Besides EBIT and EBITDA, we also find that four companies report Net Rental Income. There is one company reporting property results per share, Earning Per Share (EPS) excluding fair value gains and losses, Economic Value Added (EVA) and total passing rent respectively. Table 2 reports the definition of the key non-GAAP performance measures that are being reported.

10

Table 2: Definitions of Other Income Statement Performance Measures Panel A: GAAP-based measures Distributable Income Income available for distribution EBIT

Earnings before interest and taxes

EBITDA

Earnings before interest, taxes, depreciation and amortization Earnings per share that excludes the effect of fair value gains and losses on earnings

EPS excluding fair value gains and losses Gross operating Cash Flow (Funds From Operations)

Net profits for the years plus depreciation, amortization and write-downs plus disposals at book value

Panel B: Operational performance measures Financial Occupancy Rental income as a percentage of rental value. rate Rental income per Annualised rental income divided by lettable square metre area. Panel C: Financial Ratios Property Result per Fair value gains or losses / number of shares share outstanding

4.1.2

Surplus Ratio

Net operating income / Rental income

Total Passing Rent to Total Market Value Ratio Total Passing Rents to Total Net Book Value Ratio

Total Passing Rent / Market Value of Property Portfolio Total Passing Rent for the year / Total Net Book Value

Other Non-GAAP Measures – Balance Sheet Measures

In contrast to the Income Statement focus observed in North American REITs, financial statements reported by many firms under IFRS seem to have adopted a balance-sheet focus and seem more interested in providing financial statement users with alternative measures of the companies’ assets. We find that 24 of the 33 firms voluntarily report a non-GAAP measure known as Net Asset Value (NAV), either in their financial statements or in the notes. Some companies explicitly state that they are following the guidelines published by the European Public Real Estate Association (EPRA). By examining the financial statements of firms that provide details on the calculation of

11

NAV, we believe that other companies are also computing NAV based on the guidelines provided by EPRA. We also note that some companies provide measures supplemental or alternative to NAV. Two companies report triple net liquidation NAV. We also document reporting of measures such as new investment and disposal market value, revalued net assets, adjusted NAV, net tangible asset backing per unit, undiluted tangible asset backing per unit, net invested capital, net financial resource, and adjusted shareholders’ fund . Table 3 reports the definition of the balance sheet non-GAAP measures that are being reported. Figure 3 shows the frequency of firms reporting non-GAAP performance measures. We divide non-GAAP performance measures into two groups: 1. Measures that have a balance-sheet focus and provide supplemental information to total asset values of the companies; 2. Measures that have an income-statement focus and provide supplemental information to the reported bottom-line net income. Once again, Figure 3 reinforces the findings that companies that report under IFRS tend to have a balance-sheet focus and are more concerned in providing financial statement users with additional information on the asset values of the companies. The most popular measure of asset values is undoubtedly NAV. For firms that choose to report alternative net income measures, they tend not to report any industry-specific measure but more common measures such as EBIT and EBITDA. Recently in 2006, the European Public Real Estate Association defines Net Income (EPRA NI) in accordance with the Best Practices Policy Recommendations issued by the Association. The calculation of earnings using these guidelines excludes realized and unrealized gains or losses on investment property, intangible asset movements and their tax effects, and costs related to the entity conversion and prior year tax items. We find that among the 26 European companies, six disclosed EPRA NAV and EPRA NI using the EPRA recommendations.

12

Table 3: Definitions of Non-GAAP Performance Measures Reported With a Balance Sheet Focus Net Asset Value (NAV) Adjusted Net Asset Value Adjusted Shareholders’ Equity Dividend Yield

Equity in ratio to land portfolio European Public Real Estate Association Net Asset Value (EPRA NAV)

Leverage Property Loan to Value Ratio Net Financial Resource Net Invested Capital Revalued Net Asset (when reporting investment portfolio at cost) Triple Net Liquidation NAV

Fair value of the property portfolio less debt NAV excluding the capital gains or loss effect on the fair value of properties Shareholders' Equity excluding the capital gains or loss effect on the fair value of properties Paid interim dividend + Proposed dividend in relation to the share price at the end of the period Shareholders' equity / Fair value of investment portfolio Net assets + surplus on trading properties, excluding fair value adjustments for debt and related derivatives + deferred taxation on revaluations and capital allowances and the effect of those shares potentially issuable under employee share schemes Interest-bearing liabilities as a percentage of the carrying amount of the properties. Total carrying amount of debt / Property assets adjusted for deferred tax Total Equity attributable to equity holders Minority Interest + Net financial position (Non Current Assets + Current Assets) (Non Current Liabilities + Current Liabilities) Shareholder’s equity + unrealized capital gains Shareholders’ equity + unrealized capital gains on asset + adjustment to unrealized capital gain taxes, transfer taxes and disposal costs + market-to-market fixed-rate debt

13

Figure 3: Frequency of Firms Reporting Non-GAAP Performance Measures 40% 35% 30% 25% 20% 15% 10% 5% 0% NAV

Other

EBIT

Balance Sheet Focus

EBITDA

Other

Income Statement Focus

In Table 4, we classify the sample firms into groups and summarize the number of firms reporting other performance measures with an income statement focus and a balance sheet focus. This table shows that most companies report alternative measures having both a balance sheet and an income statement focus. However, an important number of companies only disclosed some balance sheet focused alternative measures. Table 4: Number of Firms Reporting Non-GAAP Performance Measures with a Balance Sheet and an Income Statement Focus Balance Sheet Measure Provided Not Provided Provided

16 companies

3 companies

Not Provided

12 companies

2 companies

Income Statement Measure

Overall, as previously discussed, our findings suggest that firms which report under IFRS tend to have a balance-sheet focus and are more interested in providing financial statement users supplemental information on their asset values. Interestingly, most firms reporting a balance sheet supplementary measure also report a supplementary income statement measure. The companies seem to recognize the insufficiency of net income being the only measure of their profitability. However, because EPRA does not provide FFO guidance, firms seem to rely on some non-industry measures such as EBIT or EBITDA to compare performance. We posit, although cannot be certain, that property 14

values having been recognized at fair value in a large number of jurisdictions prior to the adoption of IFRS explains this fact. Under fair value reporting, no depreciation is necessary. Thus, the need for a FFO-type performance measure is reduced. 4.1.3

Adoption of Fair Value Reporting for Real Estate Assets

One key concern with the adoption of IFRS is how companies are going to report their long-term assets subject to the requirements of IAS 40 and IAS 16. In this section, we examine the reporting practices of companies with regards to their investment properties and their property, plant and equipment under IFRS. We first examine how investment properties are reported on the financial statements. Recall that IAS 40 requires that companies determine the fair values irrespective of whether they choose to report fair values in the financial statements or in the notes to financial statements. A larger proportion of our sample companies, more than 70% in fact, report investment properties at fair values directly in their financial statements. We suspect companies elect to follow the fair value method for investment properties because they believe it directly provides information that is found useful by their investors. Figure 4 reports the frequency of companies reporting investment properties under the cost method and the fair value method. Figure 4: Reporting of Investment Properties under Cost vs. Fair Value Method 80% 70% 60% 50% 40% 30% 20% 10% 0% Cost

Revaluation

We next examine how companies report property, plant and equipment under IAS 16. We generally find that property, plant and equipment of the sample firms are reported using the cost method. Figure 5 shows the proportion of firms reporting property, plant and equipment under the cost or the revaluation method.

15

Figure 5: Reporting of Property, Plant and Equipment under Cost vs. Revaluation Method 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Cost

FV

Firms typically classify real estate assets as property, plant and equipment when they are owner-occupied for the use of the companies’ operations. Almost all firms that choose to report property, plant and equipment under the cost method recognize the associated depreciation expenses using the straight-line method. However, firms seem to have very different assumptions on the useful lives of their real estate assets but the estimated useful lives for properties do not exceed 100 years. For instance, Befimmo allocates the cost of its owner-occupied properties over 30 years. On the other hand, Sino Land Company Ltd. from Hong Kong chooses to assume a useful life of 96 years for its hotel properties and Sponda from Finland decides to allocate its company-occupied properties over a period of 100 years. While property age is a determinant of depreciation period, the fact that real estate assets have a very long economic life makes the estimation heavily subject to judgment. These findings and their large variation is interesting because Tsang (2007) shows that real estate companies depreciation choices are associated with known determinants of earnings management. Though it is beyond our scope in this project to examine the extent of earnings management exerted by real estate companies using depreciation expenses, the vast differences in the useful lives of real estate assets seem to suggest there are factors other than the economic lives of the assets that affect decisions in the depreciation policy of companies. In Table 5, we report the number of firms that choose the cost method or the revaluation method for property, plant and equipment under IAS 16, and the cost method or the fair value method for investment properties under IAS 40.

16

Table 5: Number of companies using the fair value/revaluation method or the historical cost method under IAS 16 and IAS 40 IAS 40 Fair Value

Cost

Revaluation Method

3 companies

0 company

Historical Cost

22 companies

8 companies

IAS 16

Most companies report the owner-occupied property at depreciated cost and the investment property at fair value. However, some companies using the fair value method for investment property do not include buildings in any other asset category other than investment property. As a result, owner-occupied property is reported at fair value using the revaluation method. All the other companies recognize both IAS 16 and IAS 40 properties at depreciated historical cost. An interesting issue we investigate is the impact of fair value reporting on the financial picture of the corporation. In other words, we examine if the switch from historical cost to fair value reporting has substantially affected the reported assets of companies. Though companies that choose to report investment properties under the fair value method no longer disclose the properties at historical cost, companies that choose the cost method provide the depreciated and undepreciated historical cost in addition to fair value information and enable us to examine the differences of historical cost and fair values for this sub-sample of firms. In Figure 6, we compare the depreciated historical cost and fair value of the investment properties of firms that have adopted the cost method. Since the values of real estate assets are in general rising over time, we find that fair values are higher than the historical costs for all firms. On average, fair values are 144% of depreciated historical costs and 134% of undepreciated historical cost. We suspect that firms that choose the fair value method may even have a higher ratio of fair values to depreciated historical costs relative to their peers, hence providing them with additional motivation to report fair values instead of historical costs on their financial statements.

17

Figure 6: Comparisons of Historical Costs and Fair Values for Firms using the Cost Method (in millions of €) € 14,000 € 12,000 € 10,000 € 8,000 € 6,000 € 4,000 € 2,000 €0

Historical Cost

et ro va ce sa In m ob Co lo ni al

M

Ae de s

AG

an am en to Ri s

W oh ne n

m ob ilie Im

De ut sc he

IV G

Kl

ép ie r re

n

Fair Value

Finally, we also examine how the recognition of fair value gains and losses in the income statement has affected the financial performance of the companies. Companies that choose the fair value method are required under IAS 40 to report the changes in fair values as gains or losses in the income statement. We evaluate the importance of these gains and losses to the overall performance of the companies. In Figure 7, we compare the changes in fair values for investment properties to the bottom-line net income reported by the companies. We find that, on average, the changes in fair value for investment properties represent 54.43% of net income. Moreover, 91.2% of the firms report a gain instead of a loss for the changes in fair value. The findings show that the accounting treatment of IAS 40 in general allowed companies to report higher net income in 2005 and that the improvement of net income due to the changes in fair value was also substantial. Hence, we infer companies may also have income-statement incentives to select the fair value method in accounting for investment properties. The significant positive impact to the net income of our sample companies also demonstrates the volatility such fair value changes can have on the net income GAAP measure. If the investment properties of our sample companies were to experience a period of declining values, these reductions in fair value could have an equally unfavourable impact to GAAP net income. This volatility may drive the need to develop non-GAAP measures that exclude the changes in fair value of investment properties as we have seen with the wide use of EBIT or EBITDA, or to choose non-GAAP measures not based on the income statement but rather the balance sheet, such as NAV.

18

Figure 7: Fair Value Gains and Losses as a percentage of Net Income for Firms using the Fair Value Method Net Income

46%

Fair Value Change 54%

Other

Finally, we examine whether the size of firm has an impact on the recognition of gains and losses on the fair values of assets for firms using the fair value method to account for investment properties. Table 6 divides the sample firms into three groups and reports the average fair value gains and losses as a percentage of net income for each sub-sample group. Table 6: Income Statement Impact of the Fair Value adjustment for firms of various size < €1,500 Market Capitalization (in millions) ∆FV/NI

54.96%

Between €1,500 and €4,350 67.59%

> €4,350 Average 40.74%

54.43%

There is a high variation between companies when comparing the ratio of change in fair value of investment property to Net Income and it does not seem to be related to their market capitalization. 4.2

Disclosure Related to Real Estate Assets

Next, we examine the disclosures related to real estate assets made by our sample firms. We evaluate the disclosures related to the following aspects: 1. Whether companies disclose the external appraisers that assess the fair values of their real estate assets.

19

2. Whether companies disclose the standards adopted for appraisals and valuation. 3. Whether companies disclose information on how fair values are determined. In Figure 8, we document the proportion of firms that explicitly state the appraisers who conduct the fair value appraisals. We find that most firms choose to rely on external rather than internal appraisals. Moreover, all companies, except two, explicitly state the names of their external appraisers. Figure 8: Number of Firms Disclosing the Use of External Appraisers 30 25 20 15 10 5 0 Internal Appraisal

External Appraisal (Appraiser Disclosed)

External Appraisal (Appraiser not Disclosed)

No information

In Figure 9, we document the number of appraisers the firms employ in the determination of fair values. We find that most firms employ more than one appraiser. The companies retain three appraisers on average. Figure 9: Number of Appraisers Employed 14 12 10 8 6 4 2

8+

6

5

4

3

2

1

No

In fo rm

at io n

0

We next examine whether companies disclose the appraisal and valuation standards used in the determination of fair values. We find that only 60% of companies disclose the 20

standards used in determining fair values. For companies that disclose the valuation standards that were used, we find that they most commonly follow the International Valuation Standards (e.g. IVG Immobilien) and the Appraisal and Valuation Manual by Royal Institution of Chartered Surveyors (e.g. Rodamco). Other disclosed standards include Standards of Professional Appraisal Practice followed by Immofinanz and the Real Estate Appraisal Guidelines followed by Klépierre. In Figure 10, we document the number of firms disclosing their valuation standards. Figure 10: Fair Value Standards used International Valuation Standards

RE Appraisal Guidelindes 39%

40%

Uniform Standards of Professional Appraisal Practice

3% 17%

1%

Appraisal and Valuation Manual published by the Royal Institution of Chartered Surveyors No Information

We next examine whether companies disclose the methodology applied in determining fair values to obtain some insight on how firms and appraisers determine the fair values of real estate assets. We find that 74% of the companies disclose their valuation methodologies and the three common methodologies used are i) discounted cash flow method, ii) the capitalization method, and iii) the use of market value comparables. The value that is determined with the discounted cash flow method is the present value of the expected cash flows from each property of the portfolio. The value of the property with the capitalization method is derived by dividing one single year’s income or an average income by an appropriate rate. The market value comparable method uses recently sold comparable properties that are adjusted to reflect the specifics of the subject property. Figure 11 reports the number of firms reporting valuation methodologies and the methodologies used.

21

Figure 11: Valuation Methodologies Used

26% DCF 47%

Market Value Comparables Capitalization Method No Information

9%

18%

We then examine whether larger firms have better disclosure quality with regards to their real estate assets. In Table 7, we divide the sample firms into three groups by firm size and report the percentage of firms using external appraisers, the average number of external appraisers used, the disclosure of fair value standards and the disclosure of valuation methodologies. Table 7: Number of companies disclosing details about the fair value valuation method in relation to their market capitalization Market Capitalization (in millions) Use of External Appraiser Average number of external appraisers used Disclosure of fair value standards Disclosure of valuation methodologies

100%

Between €1,500 and €4,350 100%

2

3

4

3

54.55%

45.45%

50.00%

50.00%

50.00%

72.73%

100.00%

74.24%

€4,350

Average

100%

100%

Table 7 shows that companies with a greater market capitalization tend to use more external appraisers and disclose more details about the methodologies used to determine the fair values of their property portfolio. In Figure 12, we examine the frequency with which companies using the fair value method perform the revaluation of their entire portfolio and in Figure 13, the frequency at which the companies rely on an external appraiser’s opinion for the fair value of their investment properties. 22

Figure 12 shows that most companies do not disclose the information about the frequency at which the revaluation is performed. However, for those who disclose it, the revaluation is mostly performed on an annual basis. Figure 12: Frequency of internal or external valuation of the portfolio

50.00% 45.00% 40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% Annually

Semi-annually

Quarterly

No information

Figure 13: Frequency of revaluation of the portfolio performed by external appraisers 50.00% 45.00% 40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00% every 1 to 3 years

Annually

Semi-annually

Quarterly

No information

This information is generally not disclosed but, for companies that provide it, the appraisal of the portfolio by an external expert is more generally performed once a year.

23

We then examined the level of detail of the disclosure of fair value by examining if fair value is reported in aggregate, by segment or property by property. Figure 14: Fair value disclosure

16%

globally 46%

Segment w ise Property by property

38%

Only a small amount of companies report the fair value of each property of their portfolio and most companies only disclose the total aggregate fair value of their property investments. 5

Best Reporting Practices

In our investigation, we notice that several companies go beyond the minimum reporting requirements that are imposed by IFRS. They report more or finer information than the minimum that is required by the standard. We know that some of these choices are not voluntary, but are made necessary by the regulatory rules in effect in some countries about the reporting of real estate companies. Still, as they appear in the annual reports of these entities, they may prove useful to investors that have to interpret IFRS statements. We qualify these as “best reporting practices”, mostly from our evaluation of the quality of the information from a financial statement reader’s perspective. We do not want to imply that these are mandatory disclosures under IFRS. In fact, they exceed these requirements, and while some firms disclose this information in the notes to the financial statements, for the most part, this information is presented in the annual report, outside of the financial statements proper.

24

We have identified several areas of supplementary disclosures. Our report will further extend on our observations about: 1. Property level information 2. External valuator report 5.1

Property Level Information

IFRS does not require that property information and valuation be provided on a propertyby-property basis. In fact, only aggregated information has to be provided. If a REIT’s business segments are organized geographically, they may be required to provide geographic segment operations, but there is flexibility in the definition of what constitutes a business segment. For example, a Western European firm might consider all of Western Europe to be a segment, while another may consider France, Germany, and Belgium to be three different segments. Accordingly, there is flexibility with respect to the amount of details to be provided beyond the aggregated information. For example, Westfield Group presents directly in the notes to its financial statements the detail about each property’s fair value, its carrying value in the books, its cost, the name of the most recent independent valuator retained, and the date when the latest valuation was obtained. This disclosure spans three and a half pages of the firm’s financial statements. Intervest Offices of Belgium goes further than Westfield in its presentation, providing the rental income realized on each property plus each property’s occupancy rate. Note that under regulation in effect in Belgium, the corporation must obtain an independent property valuation each quarter, and as such, it would be redundant to include the date of the latest independent valuation. IVG Immobilien of Germany provides similar information to that provided by Intervest, although it adds the EBIT realized on each of its properties. It also provides interesting operational data, such as the date each building was added to the portfolio, and the date it was last refurbished. The most striking example of detailed and enhanced disclosure we have observed relates to The GPT Group of Australia. For each property, the organization provides the date of acquisition, the physical description of the building, sales, net income, the latest valuation date, a very detailed description of the valuation method including the rate used to discount the future cash flows, and finally, the major tenants. These examples are provided in the appendices to this report. This information is helpful for financial statement users to judge both the growth potential of the corporation and also the reliability of the fair value estimates. By disclosing how recent the external estimates are, corporations allow users to judge how reliable the fair values are. More extreme even, by disclosing the property earnings or cash flows, the occupancy rate, and 25

the discount rate used in valuation, investors can easily check how conservative the valuation is compared to the current revenues of the property. A benefit from these supplementary disclosures may include a reduction in the cost of debt or equity capital of the firm. There is clearly a source of concern with supplementary disclosures since they may lead to the release of proprietary information that may expose any competitive advantages or disadvantages of the firm. For example, each tenant from a building can easily compute the average rental fee charged per square feet and assess the fairness of their rent. However, since each lease may be unique in its duration, complexity and location within the building, average values may not be applicable. Note that providing this information is beyond the requirements of the standards and that other intermediary disclosures are possible. For example, some real estate entities have aggregated their disclosure at the country level and others at the continental level. 5.2

External Valuator Report

In several corporations, we have observed a “Report by the Real Estate Expert”, also called a “Valuation Certificate” or “Report of the Valuers” in other jurisdictions. This is a report, varying in length from a thirteen line report in Singapore (Hong Kong Land) to a multiple-paged, very detailed description in Spain (Metrovacesa), that describes the valuation process and results obtained in the valuation. These types of reports would potentially benefit users of the financial statements, as they state, with more or less detail, the work performed by the valuators and their opinion. Again, this is not required by IFRS but we suspect that this was required by either the securities regulators in each of these countries or demanded by local investors. We also would like to indicate that we are not aware if the inclusion of such a report is possible given Canada’s auditing or property valuation standards. Clearly, the auditor will want to consult any valuation report in the preparation of their overall audit opinion. We do not know if financial statement readers would benefit from direct access to the valuation reports. Examples are provided in the appendices to this report. 6

CONCLUSIONS

In this report, we have surveyed the reporting practices of 33 major real estate organizations reporting in 16 different countries that have adopted IFRS. Our objective in this study is to document their reporting practices from a “performance measurement” perspective and their choices in the implementation of IFRS. We failed to identify any use of the FFO measure which is widely used in North America. In fact, only one of our sample firms reported the measure. However, we note several GAAP-based income statement or cash flow statement measures that are discussed and emphasized in corporations’ annual reports. EBIT and EBITDA are the two most heavily emphasized measures. Still, the majority of firms put emphasis on at least one supplementary

26

measure other than the typical “bottom line” net income number in reported in their annual reports. While we cannot be certain that this is the cause, we document that most sample firms were positively affected by the fair value adjustments of their real estate assets over our sample period. These real estate gains represent on average more than 50% of reported net income. This may have provided an incentive to adopt the fair value measurement basis upon adoption of IFRS. As the primary reconciling item in FFO is depreciation and since depreciation does not get recorded under IFRS, firms did not have the need to report FFO. We have little information about the frequency with which external valuations are obtained. For firms that do report, most obtain this valuation on an annual basis, with some firms obtaining it more than once a year. External valuators were retained by virtually all sample firms. However, since there is a wide variation in the frequency of these valuations, we conclude that for any given report, we probably observe property values that are the combined product of internal and external valuations. While we observe few firms reporting FFO or other adjusted performance measures, we witness much more emphasis on adjusted balance sheet measures, the most popular being Net Asset Value. While not exactly non-GAAP, this measure’s computation is similar to FFO in that it excludes some elements. A definition of the measure is provided by the industry (i.e. EPRA in Europe). It appears that the international real estate industry has embraced the fair value reporting approach. Interestingly, we observed several firms going beyond the basic requirements of IFRS in their disclosures. Some provide property-by-property information that goes beyond the key fair value requirement and include operational and financial performance and fair value methodological information. In others, an independent valuator report, somewhat similar to an auditor’s report, is attached to the financial statements. Some of these disclosure decisions may have been made voluntarily by firms but they may have also been imposed by local securities regulators or demanded by investors. While we could not test for the impact of these decisions on firms, and accordingly do not recommend their application, we believe that they are interesting practices that might be considered by the Canadian real estate industry. We have thus provided several examples that may guide preparers who may be considering extending their disclosures beyond required by IFRS.

27

REFERENCES FOR RELATED LITERATURE Baik, B., B.K. Billings and R.M. Morton, 2005, “Manipulation, Increased Transparency, and Value Relevance of Non-GAAP Disclosures for Real Estate Investment Trust (REITs),” Florida State University, Working Paper. Dechow, P.M., 1994, “Accounting Earnings and Cash Flows as Measures of Firm Performance: The Role of Accounting Accruals,” Journal of Accounting and Economics, 18, 3-42. Fields, T.D., S. Rangan, and S.R. Thiagarajan, 1998, “An Empirical Evaluation of the Usefulness of Non-GAAP Accounting Measures in the Real Estate Investment Trust Industry,” Review of Accounting Studies, 3, 103-130. Fortin, S. and D. Tsang, 2007, “GAAP vs. Non-GAAP Financial Measures: What Can We Learn from Analyst’s Accuracy in Forecasting Real Estate Investment Trust Performance?” McGill University, Working Paper. Juergens, J.L., 2000, “The Information Content of Analysts’ Forecasts of REIT Earnings,” The Pennsylvania State University, Working Paper. NAREIT, 2002, “White Paper on Funds From Operation,” U.S.A. Ohlson, J.A., 1995, “Earnings, Book Values, and Dividends in Equity Valuation,” Contemporaneous Accounting Research, 11, 661-687. Pittman, J. and S. Fortin, 2004, “Auditor Choice and the Cost of Debt Capital for Newly Public Firms,” Journal of Accounting and Economics, 37, 1, 113-136. REALpac, 2004, “White Paper on Funds From Operations,” Canada Sloan, R., 1998, “Discussion of ‘Evaluating Non-GAAP Performance Measures in the REIT Industry’,” Review of Accounting Studies, 3, 131-135. Tsang, D., 2007, “Comparing the Quality of Accruals for Alternative Summary Performance Measures in the Real Estate Investment Trust (REIT) Industry,” McGill University, Working Paper. Vincent, L., 1999, “The Information Content of Funds From Operations (FFO) for Real Estate Investment Trusts (REITs),” Journal of Accounting and Economics, 26, 69-104.

28

APPENDICES

29

FV disclosure Australia – Westfield Group Westfield Group provides, in the notes to its financial statements, detail information about each property, including the following: • • • • •

Fair Value Carrying Value Cost Name of the most recent evaluation Date of the most recent valuation.

30

31

32

33

34

FV disclosure Belgium – Intervest Offices For each of its properties, Intervest provides, in its annual report but outside of its financial statements, the following information: • • • •

Surface Area Fair and Investment Value Rental Income Occupancy Rate.

35

•

36

37

FV disclosure Germany – IVG Immobilien On a property-by-property basis, IVG Immobilien provides: • • • • • • •

Ownership and physical property data Date acquired or last refurbished Occupancy Rate Market Value Gross rental income for 2005, plus 2006 forecast Net rental income Operating profit

38

39

40

41

42

43

44

FV disclosure Australia – The GPT Group Most extensive disclosure we have observed. On a property-by-property basis, The GPT Group provides the following information: • • • • • • • • • •

Location Year of acquisition Percentage of ownership Physical data Rental Income Net Income Fair value (Internal) and External valuation Valuation method and valuation parameters used Occupancy rate Tenant Information

45

46

47

48

49

50

51

52

53

54

55

56

57

Real Estate Report Belgium – Cofinimo

58

59

60

Real Estate Report Belgium - Befimmo

61

62

Real Estate Report UK – British Land Co

63

64

65

Real Estate Report Italy – Beni Stabili

66

67

68

Real Estate Report Singapore – Hong Kong Land

69

70

Real Estate Report Spain – Metrovacesa

71

72

73

74

75

76

77

78

79