Accounting Choices: The Brazilian Case Of Real Estate Investment

ABSTRACT

The objective in this study is to show that the business model of publicly traded real estate management companies in the Brazilian market explains their accounting choices to measure the investment properties. Therefore, the five main Brazilian investment property managers were studied, based on a method that triangulated documentary research, semistructured interviews and consistency analysis. The results of the analysis of the secondary data from the documentary research of the two main Brazilian companies in the real estate exploitation sector (one of which chose the fair value and the other the cost), together with the results of the interviews with the companies’ main internal and external agents direct or indirectly involved in the measuring choices of investment properties, indicated that the companies’ distinguished business model explains their distinct accounting choices to measure the investment properties. To enhance the consistency of these findings, next, the business models of the three consecutive largest Brazilian investment property managers were analyzed (two of which chose the fair value and one the cost), confronting them with their accounting choices; consistency was observed between the business models and the accounting choices, in accordance with earlier results. Although generalization of the results is not possible due to the small number of companies studied and the specific sector, evidence indicates that an integrated set of factors, which are considered in isolation in the literature, can explain the accounting choices. Several mutually interacting factors that explain the accounting choices, to the detriment of isolated factors, can contribute to expand the knowledge on this theme.

accounting choices; investment properties; business model

RESUMO

O objetivo deste estudo é mostrar que o modelo de negócio das empresas de capital aberto do mercado brasileiro, que são gestoras de imóveis, explica suas escolhas contábeis de mensuração das propriedades para investimento. Para tal, foram estudadas as cinco principais empresas brasileiras gestoras de propriedades para investimento com base em metodologia que triangulou pesquisa documental, entrevistas semiestruturadas e análise de consistência. Os resultados da análise dos dados secundários da pesquisa documental das duas maiores empresas brasileiras do setor de exploração de imóveis (uma optante pelo valor justo e uma pelo custo), junto com os resultados das entrevistas com os principais agentes internos e externos das empresas, envolvidos direta ou indiretamente nas escolhas de mensuração de propriedades para investimento, indicaram que o modelo de negócio diferenciado das empresas explica suas distintas escolhas contábeis de mensuração das propriedades para investimento. Para dar consistência a esses achados, foram analisados posteriormente os modelos de negócios das três maiores empresas brasileiras gestoras de propriedades para investimento seguintes (duas optantes pelo valor justo e uma optante pelo custo), confrontando-os com suas escolhas contábeis. Observou-se que há consistência entre os modelos de negócios e as escolhas contábeis, corroborando os resultados anteriores. Embora a generalização dos resultados não seja possível em função da pequena quantidade de empresas estudadas e do setor específico, as evidências indicam que as escolhas contábeis podem ser explicadas por um conjunto integrado de fatores considerados isoladamente na literatura. A existência de vários fatores que interagem entre si e explicam as escolhas contábeis, em detrimento de fatores isolados, pode contribuir para a ampliação do conhecimento nessa temática.

escolhas contábeis; propriedades para investimento; modelo de negócio

1 INTRODUCTION

Questions regarding accounting choices have been the motivation for many studies since 1960. Such studies attempt to explain accounting choices in function of the opportunistic behavior of managers, results management, and information asymmetry, based on economic theories such as the Theory of the Firm and the Theory of Agency (Watts & Zimmerman, 1986; Holthausen & Leftwich, 1983; Holthausen, 1990; Fields, Lys & Vincent, 2001; Francis, 2001).

For example, Paulo (2007) argues that managers can make choices between valid accounting practices and norms, however warns that the literature shows that accounting choices can also be made from opportunistic (managers seek to maximize their own utility) and efficiency (in order to better reflect the economic characteristics of an asset or transaction, in an attempt to improve the quality of information) perspectives.

Furthermore, studies such as those by Capkun, Cazavan-Jeny, Jeanjean and Lawrence (2008), Lourenço and Curto (2010), Martínez, Martínez and Diazaraque (2011), and Osma and Pope (2011), relate explanations for accounting choices with the status quo. In accordance with these authors, those producing financial statements would tend to not alter the form of reporting, due to higher costs and a departure from a possible "comfort zone". Thus, in the case of the adoption of IFRS (International Financial Reporting Standards), for example, accounting choices would use procedures closer to existing local accounting standards.

In Brazil, studies which address this issue are still few in number. Worth highlighting are those by Costa, Silva and Laurencel (2013), Pereira (2013), and Andrade, Silva and Malaquias (2013), which studied accounting choices from the perspective of the international literature mentioned.

In this sense, this study seeks to contribute to the evolution of knowledge regarding explanations for accounting choices, by employing a methodological approach that is different from the methods applied in positivist-empirical studies, and focusing on a specific area of activity, in such a way that provides a deeper insight into the subject, in order to set the results against existing literature.

Thus, its aim is to offer explanations for accounting choices for measuring investment properties made in companies belonging to the real estate sector that operate in the Brazilian capital market. To this effect, instead of being based on a great quantity of data, it takes an approach which triangulates: (i) descriptive analysis of secondary data from five Brazilian companies from the real estate sector, obtained through documentary research; (ii) interviews with players involved in the accounting choices at two of these companies; and (iii) consistency analysis between the secondary data collected and the answers obtained in the interviews, in companies operating in the same sector.

This methodological approach allows for advancement in knowledge regarding explanations for accounting choices, despite limitations to generalization. It was found that, for companies from the real estate sector operating in the Brazilian capital market, explanations for accounting choices for measuring investment properties are related to companies' business models.

The term "business model" originated with Drucker (1954), but only gained expression and visibility during the last decade, even becoming part of business jargon (Casadesus-Masanell & Ricart, 2010). For Timmers (1998), the business model is seen as an architecture for the services, products, and information, including various benefits for all the business actors (stakeholders).

For Amit and Zott (2001, p. 511), the business model describes "the content, structure, and governance of transactions designed so as to create value through the exploitation of business opportunities." This study starts with the assumption that the business model is a set of various factors, which encompass activities and strategies adopted by companies in order to stand out, meet operational and economic needs, and create value for stakeholders. In this sense, it can be concluded that ownership structure, management model, means of funding, clientele etc. are factors that can be integrated into the business model and, consequently, have an impact on companies' accounting choices. These factors involve different alternatives for achieving competitive advantage, profitability and added value, and can thus explain managers' accounting choices.

The main contribution of this article is in adding new explanations for the implementation of accounting choices in companies, which are related to their business models. These new explanations result from research carried out using a different methodological approach from those traditionally used in positivist-empirical studies, which predominate in this area. This implies that room remains to increase the scope of the current accounting choice theory, by broadening it to a more comprehensive vision regarding the strategic positioning of companies and the role that accounting information plays in business models.

2 ACCOUNTING CHOICES FOR MEASURING INVESTMENT PROPERTIES

In accordance with Technical Pronouncement CPC 28, which follows international standard IAS 40 (International Accounting Standard 40), companies can measure their investment properties by cost or by fair value. The flexibility of the rule results in an explicit accounting choice, and companies' understanding of this choice has been the focus of studies abroad and in Brazil.

International studies regarding accounting choices for measuring investment properties involve, notably, companies listed in Europe that adopted IFRS in 2005. Such studies use a positivist-empirical approach, mainly based on statistical methods and multivariate logistical regression models. Explanations for accounting choices for measuring investment properties, using this approach, are linked to factors such as: debt, information asymmetry, accounting information relevance, results management, political cost, and change of the status quo. The empirical evidence often leads to inconclusive or even contradictory explanations, as can be seen later. And none of them relate the choice to the subject addressed in this study, which is the business model.

Christensen and Nikolaev (2013), in a study with companies from the United Kingdom and Germany, found that debt is one factor that explains the choice of fair value for measuring investment properties. According to the authors, companies prefer the adoption of fair value to maintain their debt ratios at lower levels. This result differs from those found by Quagli and Avalone (2010), which showed that debt does not appear to influence in the choice of fair value, for a sample of companies from seven European countries. In the view of these authors, the bigger a company's size (a proxy for political cost), the smaller the probability of choosing fair value, given that cost seems more efficient from a contractual perspective, since it reduces agency costs, political visibility, taxation and legal proceedings (Watts, 2003; Qiang, 2007).

For Muller, Riedl and Sellhorn (2008), companies with greater information asymmetry are more likely to use fair value for measuring investment properties, since it is investors' demands that leads them to choose fair value, and this may be preferred to historic cost because it provides users with better quality information (Barlev & Haddad, 2003; Ball, 2006; Whittington, 2008). If there is information asymmetry, managers may choose fair value with a view to informing the market of a company's "true" value (Quagli & Avallone, 2010). This comparison can be made in investment property companies, since their assets are basically composed of such properties and their fair value measurement is reported, at least in explanatory notes.

Edelstein, Fortin and Tsang (2012), as well as Devalle and Rizzato (2011), believe that the fair value of investment properties has a double effect on the results, since is allows depreciation expenses to be ignored and unrealized gains or losses to be recognized in the results for the financial period, which has a direct impact on various performance indicators. The intention to manage results would explain, therefore, the choice between fair value and cost for measuring investment properties.

Muller et al. (2008) found that European companies are more likely to choose fair value for measuring investment properties when the real estate market in which they operate has greater liquidity. Cairns, Massoudi, Taplin and Tarca (2011) showed that companies with investment properties exhibit a high adhesion to the use of fair value, which is associated to the liquidity of these properties in relation to other assets. These conclusions show that fair value information is more relevant for the market and could explain it being chosen.

On the other hand, Devalle and Rizzato (2011), Martínez et al. (2011), and Capkun et al. (2008) reported that many European companies chose historic cost for measuring investment properties in order to minimize the impact of IFRS, thus maintaining the status quo.

Collin, Tagesson, Andersson, Cato and Hansson (2009), among other conclusions, highlighted that ownership structure may explain accounting choices. For these authors, companies with fragmented property can incur greater tracking costs, since there is little direct intervention by owners and a greater use of the accounting system for accountability. Thus, companies with fragmented ownership tend to choose accounting rules which reflect more complete, more detailed and less conservative information, that is, which show better results (profits), given that their performance will be compared with other companies in the market. In the case of concentration of ownership, the owner (or few owners) does not require such detailed information and wishes to maintain control, which may lead to more conservative accounting choices (Collin et al., 2009).

International literature covers studies carried out in various countries, with different dates for adopting IFRS, which may have influenced the results. Therefore, it is essential that the relevant Brazilian literature be taken into account.

Costa et al. (2013), Pereira (2013), and Andrade et al. (2013) studied Brazilian companies with investment properties among their assets, without being concerned with the areas in which they operated nor the relevance of properties to the companies' total assets. The first two studies concluded that net revenue and company size, respectively, can explain accounting choices for measuring investment properties. The logistical regression statistical model used by Andrade et al. (2013), was no shown to be significant, in other words, it did not point to any variable that could explain such accounting choices.

In the studies mentioned, evidence that certain individual characteristics (or factors) related to companies (such as debt, level of revenue, among others) and/or related to the capital market in which they operate (information asymmetry, for example), are found that may explain accounting choices. However, explanations that linked various characteristics (or various factors) to accounting choices were not identified in the literature.

In this sense, the term business model used in this study captures various company and market characteristics, even some which are not envisaged in the literature, with the aim of relating them to choices for measuring investment properties.

3 METHODOLOGICAL PROCEDURES

The empirical stage of this study was guided by a triangulation of methodological approaches, composed of documentary analysis, interviews and consistency analysis of the results found. First, the two largest companies in the real estate sector, one choosing fair value and the other historic cost for measuring investment properties, were selected. In all, twelve Brazilian investment property companies were listed on the BM&FBOVESPA during the period in which this study was carried out (2012). The two largest companies in the sector together hold more than 23 billion reais in market value and represent 45% of the total.

With the two companies with their respective financial statements and reference forms for 2012 chosen, documentary research was then carried out, based on the published financial statements and reference forms filed with the CVM (Comissão de Valores Mobiliários), the Brazilian Securities and Exchange Commission. Following on from the documentary research, a descriptive analysis of the secondary data obtained was carried out, which allowed for an understanding of the similarities and differences in the two chosen companies' business models.

The second methodological approach consisted of interviews with players involved directly or indirectly in the entities' accounting choices, whether they were internal or external to the companies chosen. Managers, internal analysts, market analysts, an accountant and an auditor were interviewed.

The interviews were carried out in 2012 and 2013, based on a semi-structured script. At the company that chose fair value, the Investor Relations Department coordinator and the accountant were interviewed. At the company that chose historic cost, the coordinator and two analysts from the Investors Relations Department, were interviewed. It was not possible to interview the accountant. The choice of these interviewees was justified by their technical accounting abilities, their participation in company decisions and their availability to concede interviews. The auditor interviewed was, coincidently, the same one that audited both companies. The choice of market analysts was based on the classification of Thomson Reuters, which ranks those with most correct predictions for companies in the sector. Via descriptive analysis of the interviewees' answers, it was possible to understand the arguments inherent to the measurement choices for investment properties.

In order to further validate the results found in the first two approaches, a third methodological approach was adopted: consistency analysis of the results obtained. The next three companies from the same sector of activity, which made different measurement choices in relation to investment properties, were identified: two chose fair value and one chose historic cost. These three companies together have almost 15 billion reais in market value and represent 29% of the total for the sector.

As in the first two cases, documentary research was carried out for the three new companies selected. Descriptive analysis of the secondary data obtained in the documentary research revealed the characteristics of their business models, which were then set against the accounting choices regarding measuring investment properties, thus widening the results found in the interviews carried out at the first two companies.

All in all, the first two approaches allowed for the identification of the business model as an explanation for accounting choices, and the third approach allowed for consistency between business models and accounting choices to be verified, as a way of validating the results from the first two approaches.

None of the five companies studied adopted IFRS and the technical pronouncements of the CPC (Comitê de Pronunciamentos Contábeis), the Brazilian Accounting Board, before compulsory adoption in financial statements by the end of 2010. The first two companies studied were called "Fair Value Case" and "Historic Cost Case", preserving their identities, conditio sine qua non for holding the interviews. The three companies selected for the consistency analysis were called "Observed Fair Value 1 and 2" and "Observed Historic Cost".

The constructs that guided the documentary research, the interviews and the presentation and analysis of the results, are summarized in Figure 1, with the respective authors on whom they are based.

Figure 1
Study constructs

In Figure 1, "traditional variables" were considered to be those widely used in literature regarding accounting choices. Those for debt are traditionally defined as the ratios of third-party capital to equity. In this study, qualitative characteristics of debt were also addressed, such as origin (bank loans or securities issued on the market), guarantees and covenants. The net debt/EBITDA indicator was also considered, due to it being a common covenant in financial markets.

Regarding information asymmetry, the variable is traditionally the market-to-book indicator (market value per share/equity book value per share) or the difference between market values and book values of shares. In this study, the "book equity" variable was adjusted to capture the companies' equity book value, considering all investment properties of all companies measured by cost and by fair value.

The key variables normally used as measures with an impact on the result of measurements are EBITDA, earnings and earnings per share. In this study, the results (profits or losses) were readjusted by reversing the variation in the fair value of investment properties when featured in statements of results; the results were also estimated when fair value was reflected in the result (both with the respective taxation effects).

With regard to maintaining the status quo, the traditional literature captures variation in assets or even in alternations in rates of return on assets. The lower these rates, the more the status quo is maintained. In this study, maintaining of the status quo was captured by the alteration (or not) in the measurement of investment properties, from historic cost to fair value. Maintaining the status quo occurs when historic cost is maintained as the measurement for investment properties.

The variables for business models represent activities and strategies adopted by companies in order to stand out, meet operational and economic needs and create value for stakeholders. For the purposes of this study, shareholder composition was obtained via the percentage of shares under main shareholder control in relation to the percentage of shares in free float.

Company size was defined in function of non accounting variables, such as the quantity of area of investment property generating cash flow (GLA: gross lettable area), the number of ventures controlled by each company, and the fair value of these ventures.

Regarding tax, taxation uncertainty relating to the tax treatment of variations in the fair value of investment properties in companies' results was measured, in a qualitative way, without there being a specific variable.

The type of company management was also measured qualitatively: family (managed by the founder or his/her descendants) or professional (managers contracted on the market). In the same way, the businesses' clientele was identified via qualitative descriptions of target public (class A, B, C, D and E) in the management reports of each company analyzed.

Using the variables and constructs in Figure 1 as a base, there follows a presentation and analysis of the results.

4 PRESENTATION AND ANALYSIS OF RESULTS

The results are presented in three large blocks. The first involves traditional variables from the literature that can help to explain accounting choices, as shown in Figure 1 and subsequent paragraphs. In it, accounting evidence and the opinions of the interviewees are presented. In the second block, information related to the business model variables, not considered in the traditional literature, and the opinions of the interviewees, are presented. The variables and constructs can also be found in Figure 1 and subsequent paragraphs. Finally, in the third block, the consistency analysis is presented, and widens the results obtained in the initial analyses.

4.1 Traditional Explanations from the Literature

4.1.1 Debt.

Regarding debt/equity ratio, the market analysts' opinion is that this may influence the choice of criteria for measuring investment properties, if the index is analyzed "pure and simply". They argue, however, that the market is not "naive" in making this calculation, and ends up discounting fair value or even capitalizing on it, depending on its "need and convenience". For them, it would be ideal to calculate financial leverage, which corresponds to net debt divided by EBITDA. The financial leverage indicator estimates the necessary number of years for generating cash in order to pay companies' liabilities with charges that, according to the analysts, is the "most used multiple" in restrictive clauses for loans (contractual covenants). Even though the analysts are not directly involved players in investment property measurement choices, depending on how relevant the capital market is to a company, the accounting choice may also aim to reduce information asymmetries regarding the values of company assets, which directly affects analyst activities.

In the opinion of the auditor, debt can be related to accounting choice since, with fair value, companies would report "higher and prettier numbers" in their balance sheets.

The interviewees for the Fair Value Case argue that debt is not a factor that is related to the accounting choice for measuring investment properties. According to them, restrictive clauses in loan contracts are based on adjusted EBITDA, which discounts the effect of variations in fair value. They believe that, if fair value were used in debt covenants, this could even "break" the company, if the variation were negative, or covenants would "never" be breached with "constant positive variations". They argue that the way of analyzing debt is the same that the analysts recommend: net debt divided by EBITDA (financial leverage).

The interviewees for the Historic Cost Case do not believe in the assumption of the use of fair value for the "protection" of debt. According to them, if someone requires fair value in order to comply with restrictive clauses for loans, "they will be playing with fire", given that, at some moments in the economy, the variation in fair value could be negative and it will be necessary to renegotiate debts. However, according to the interviewees, the company does not deny that fair value is a way of "getting into more debt" with increases in the equity value; however, like the market analysts and Fair Value Case interviewees, the Historic Cost Case also highlights that covenants are related to multiples of net debt over EBITDA (financial leverage).

Christensen and Nikolaev (2013) concluded that debt can have an impact in companies choosing fair value, whereas Quagli and Avallone (2010), Costa et al. (2013), Pereira (2013) and Andrade et al. (2013) did not verify such an influence. These results are inconsistent with those in this study, a fact that can be explained, mainly, by the difference between the debt metrics used in the market, in companies, and in the literature, as well as the different practices and opinions of the agents involved in accounting choices. The variables from studies with a high number of observations provide lessons; however these studies do not allow the particularities of each case to be captured, as is presented in this study.

The two companies reported not having breached any contractual covenants for loans in 2012. Yet, the Historic Cost Case is more transparent and indicates the contractual clause for each one of its loans. In Table 1, the three main means of funding and contractual covenants associated with each of the companies interviewed are presented.

Table 1
Funding and contractual covenants

Company	% of Total Debt	Type	Origin	Covenants
HC	23%	BCN	Itaú	1- Net Debt/EBITDA
				2- EBITDA/Net Financial Expenses
				3- Not applying the credit entirely in the construction of the project
				4- Allocating the project in a way not envisaged in the certificate
	22%	BCN	Bradesco	1- Not transferring the current shareholder control
				2- Not meeting non-monetary creditor obligations
	13%	BCN	Banco Brasil	1- Net debt/EBITDA
				2- Not suffering legal proceedings that could place in risk fulfilling the obligations assumed in this contract
				3- Not transferring the current shareholder control
FV	26%	Perpetual Bonds	(International Subsidiary)	Not available
	13%	RERC	Bradesco	Not available
	6%	Debentures	(Brazilian Market)	1- EBITDA /Financial Expenses
				2- Net Debt/EBITDA

• HC = Historic Cost; FV = Fair Value; BCN = Bank Credit Note; RERC = Real Estate Receivable Certificate. Source: Developed by the authors based on data from the study.

Differences exist in the three main funding mechanisms for the two companies. While for the Historic Cost Case BCNs (bank credit notes), are the form of funding, for the Fair Value Case the funding options are international perpetual bonds issued in American dollars, RERCs (real estate receivable certificates), and Brazilian debentures. It is apparent that the funding strategies of the two companies are different: the Historic Cost Case relies on financial institutions, while the Fair Value Case also relies on the credit market.

What draws attention in Table 1 is one non-financial clause, which is more related to shareholder composition than to debt. This clause is present in two BCNs from the Historic Cost Case, and upholds the requirement to maintain the same controllers, showing that banks are not only placing faith in company performance, but also in the continuity of founding and controlling families at companies, when providing funding.

It is therefore established that debt is also a characteristic of companies' business models, since companies exhibit different strategies for maintaining capital structure, which do not explain on their own, however, the fair value or cost accounting choice for measuring investment properties, in the particular view of the cases studied.

4.1.2 Information asymmetry.

In this subsection the intention is to evaluate whether information asymmetry can explain the accounting choices of investment property companies. A comparison was carried out between the market value (quantity of shares multiplied by the price per share on the last day of trading of each year) and the equity value (adjusted by the fair value of investment properties contained in explanatory notes, when the cost method is chosen). The aim of this comparison is to identify market-to-book, a proxy for the quality of accounting information, which shows a possible difference in accounting information between users within and outside companies. It is expected that the original equity value from the Fair Value Case and the adjusted equity value from the Historic Cost Case are close to their market values, since their assets are basically composed of investment properties. The numbers are presented in Table 2.

Table 2
Ratio Investment Property / Total Assets and Market-to-Book

Companies	Investment Property / Total Assets (%)	Market-to-Book
	2009	2010	2011	2012	2009	2010	2011	2012
Historic Cost Case	84	89	88	90	0.8	0.7	0.7	1.0
Fair Value Case	82	92	89	90	0.9	1.2	1.1	1.4

• Source: Developed by the authors based on data from the study.

For the Historic Cost Case, adjusted equity value is greater than market value, suggesting that the market does not price assets according to the fair value information reported in explanatory notes. This may mean that the market "reads" fair value in explanatory notes with some discount, due to there being information asymmetry between managers' judgment and the market's judgment regarding the fair value of investment properties.

When questioned about this point of view, the auditor mentions that there is no "rigidity" in the rule, which may "hinder" making comparisons between companies. This occurs because users do not immediately "visualize" the differences between companies, since some choose to measure by fair value and others choose to measure by cost.

Another factor that may have an impact on this analysis is the method for obtaining fair value. In some companies, fair value is calculated internally, and in others, external specialists are contracted to make the evaluation. With regards to this, the auditor states that regardless of whether the evaluation is made by internal of external specialists, the value is "tested" not only in the historic cost companies but also in the fair value ones. The internal evaluation is related to the "ability" of the professionals to carry out the job. The auditor believes that the company's internal analysts have the "ability" to define fair value.

With respect to the differences between equity value and market value, the auditor responded that this is a "coincidence", without providing an explanation for the fact, despite believing that the fair value of investment properties "should come close" to the company's value, since 80% or 90% of the assets are composed of investment properties.

The market analysts argue that the market "likes" having their own evaluations, independent from those reported by companies. With an "independent" evaluation, it is possible for the market to use different discount rates and assumptions from those used by companies. Hence, they claim that the market "does not necessarily agree" with the fair value reported by companies, and that historic cost companies may be being "too optimistic" in their evaluations, thus explaining the differences found between market value and equity value.

The interviewees in the Fair Value Case believe that fair value is a "positive" and "transparent" way of reporting to the market. For them, the market was not "accustomed to having that" before the adoption of IFRS. They argue that it is the market that has to define the value of firms, yet they believe that the fair value of properties could be a "good indicator" for reaching the value of companies.

When questioned about the calculation of fair value by internal analysts, the initial response was identical to that of the auditor, in their believing in the "professionals' ability and competence". However, the interviewees from the Fair Value Case sometimes require "external consulting", with sampling, in order to validate the fair value calculated by internal analysts, and do not use third-party services to carry out all of the evaluations, given the delay that the process would suffer, as well as being a more "costly" job. They claim to choose fair value because it involves a form of "disclosure" that is close to European companies that are in a "consolidated market".

The interviewees from the Historic Cost Case claim not to "pass on" fair value via company results because they do not "believe it right". They report that when fair value is recognized in financial statements (balance sheets and income statements), it ends up "polluting their result". The gain from fair value "does not exist". The interviewees claim that, with cash flow statements, analysts verify a company's situation better, since the gain from fair value in income statements "is of no use at all".

When questioned about the use of internal methods to calculate fair value, the interviewees from the Historic Cost Case claimed to do the calculation "one year inside and the other outside", despite not reporting when it is done "inside" or "outside" in explanatory notes, and which independent evaluation company is contracted. The option of doing the evaluation through internal analysts is due to "economy", given that the service provided by external specialists can cost four digit values for the contracting company, and since it contains "all of the metrics" that external evaluators apply. The years that it uses an external evaluator, this is due to questions of "conservatism" and in order to have a "neutral ground" with more "governance". They even said that another point in favor of an choosing internal calculation is the six-month delay for the external specialist to report the evaluation and because the calculation, when done by external specialists, "is almost impossible" for the auditing firm to audit.

With relation to the difference between market value and equity value, they eluded the question initially, claiming not to use this metric to compare the value of the company, and that it is necessary to raise other questions, such as leverage. However, information presented by the company reports that a shopping mall was recently sold for a "value much higher than the fair value". This point shows the uncertainty that the company is passing on to the market, since fair value, by definition, is the market value of exit, regardless of the way the company is funded. However, it uses the WACC (weighted average cost of capital) as a discount rate, which depends on the capital structure of the company.

With this issue it should be noted that the Fair Value Case gives the impression of wanting to provide the market with better quality information, which is consistent with studies by Muller et al. (2008), and Quagli and Avallone (2010), while the Historic Cost Case judges that fair value information is irrelevant to the market.

In Table 3, the differences between historic cost and fair value of investment properties are shown. The values are material.

Table 3
Differences between Historic Cost and Fair Value (in R$ million)

	12-31-2009	12-31-2010
	Fair Value	Cost*	Difference	Fair Value	Cost*	Difference
Historic Cost Case	8,538,000	2,006,505	6,531,495	12,286,000	2,496,675	9,789,325
Fair Value Case	6,960,688	4,940,745	2,019,943	9,676,115	7,088,247	2,587,868
	12-31-2011	12-31-2012
	Fair Value	Cost*	Difference	Fair Value	Cost*	Difference
Historic Cost Case	13,044,000	2,987,757	10,056,243	14,938,906	4,030,575	10,908,331
Fair Value Case	12,582,924	9,218,841	3,364,083	16,100,665	10,218,547	5,882,118

• * The value of historic cost, for the Fair Value Case, was calculated from 2010 (adoption of IFRS); in subsequent periods, the appreciation in fair value and depreciation in the calculations was ignored, for not having a reliable rate and its value being negligible, since, in the sector, projects are depreciated, on average, for 50 years. Source: Developed by the authors based on data from the study.

The materiality of fair values calculated by companies was directly questioned by the market analysts. Unanimously and categorically, they said: "it is us who calculate fair value!" They claimed to ignore fair value in analyses, but did not confirm that all of the market does this. If this is not done, they agree that the discount of fair value can cause "relevant reflections" in analyses. The analysts even mentioned how they consider the publication of the fair value of companies that only report it in explanatory notes and of those that capitalize on it in balance sheets. According to them, despite it being their job to calculate fair value, the value reported serves as a "reference", is used as an "indicator" and does not diverge much from their calculations. They pay more "attention" to fair value recognized by companies in their balance sheets, that is, "they spend more time" on this analysis. When fair value is only reported in explanatory notes, they have "less obligation to understand the value", which ends up "losing force". In spite of this, they consider the number credible in both forms.

In the opinion of the external auditor, the value is more "impacting" when it is in the result and the balance sheet, since it is a "prettier" number, despite a "personal preference" for historic cost. He claims that fair value in the whole company is more useful than "baking a cake" only when evaluating investment properties.

For the interviewees from the Fair Value Case, recording fair value in the accounts is "important" for being able to "reflect" the company's worth in market value. They believe in the usefulness of fair value for analysts and readers of financial statements. One "driver" is the possibility of showing the market how much a property is worth and how much income it generates, in that it is "more transparent". Moreover, they believe in the "informative ability" of fair value.

The interviewees from the Historic Cost Case have a different opinion of fair value. They claim that it is going to "pollute" results with profits that "do not exist", since, for them, fair value reported in explanatory notes is an "important indicator" of how much their investment properties are worth, however they do not believe it to be "right to pass the value on through results".

In conclusion, it could be shown that companies that capitalize on fair value in their financial statements have less information asymmetry, and in companies that do not capitalize on it, asymmetry is greater. This implies that the fair value of investment properties of companies that capitalize on it is closer to the value that the market itself establishes for the properties. In the case of those companies that only state fair value in explanatory notes, there is a substantial difference between the stated value and that which the market evaluates, and moreover, the value stated in the notes is consistently higher than market evaluations. From this, it can be inferred that accounting choice for measuring investment properties signals more or less information asymmetry between managers and the market.

4.1.3 Impact on results.

In order to attempt to identify the effects on results using the secondary data, it was assumed from the outset that the fair value option is a way for companies to report better results.

In Table 4, the reconciliation of profits following the reversal of the variation of fair value of investment properties is shown. In this reconciliation, deferred income tax of 34% was taken into account. The result found with the reversal reveals profits hundreds of times inferior to those featured in the financial statements.

Table 4
Adjusted Profit (in R$ million)

Fair Value Case	2009	2010	2011	2012
Earnings before Interest and Taxes	1,549,349	968,351	1,445,374	3,409,403
Reversal of Fair Value	(1,244,596)	(567,925)	(776,215)	(2,518,035)
Adjusted Earnings	304,753	400,426	669,159	891,368
Financial Result	(6,584)	(94,047)	(305,693)	(433,964)
Profit before IT	298,169	306,379	363,466	457,404
Current IT	(29,262)	(42,825)	(58,934)	(100,944)
Deferred IT (without IP)	-	(4,113)	(25,406)	(92,839)
Adjusted Net Profit	268,907	259,442	279,126	263,621
Reported Net Profit	1,095,086	634,272	789,655	1,925,524
Difference	826,179	374,831	510,529	1,661,903
% Difference	307%	144%	183%	630%

• IT = Income Tax; IP = Investment Properties. Source: Developed by the authors based on data from the study.

In the interviews regarding impact on results and on manager's remuneration, the auditor and the market analysts responded that this has no relationship with fair value and that this is not taken into account in managers' remuneration.

With regards to smoothing of results, the analysts believe that this could "make sense" due to the fact that historic cost is more stable, but that it may not have been the "driving factor" for companies choosing historic cost. They said that the more volatile value may be recorded in the "accounting books history" of companies that adopt fair value, but that the market is "able" to remove this volatility, capitalizing on or discounting fair value.

The auditor claims that having an accounting record of fair value provides "punch", mainly when companies have in their portfolios projects under development, (greenfields), which are ventures with greater increase in value. In mature businesses, there is less growth and increase in value.

Regarding results management, one hypothesis raised by the analysts was that companies did not adopt fair value due to the volatility of profits, but instead due to the volatility in balance sheets. According to them, in 2009 and 2010, Brazil exhibited "prospects" of a fall in interest rates, and investment properties, being real estate with "fixed income characteristics", would have increased in value. However, companies may be "waiting" for future scenarios, in which interest rates increase again and they need to report a "loss in their property", a fact that would affect, above all, the value of their assets.

The interviewees from the Fair Value Case believe in volatility, but claim that this "does not worry them". They state that constant development of new projects "always" ends up "adding" value to the company. With stable growth, without acquisitions and development, they believe that "negative reevaluation" may occur in "some" periods, but "there is nothing to worry about", since there is no immediate effect on cash flow. Despite taking into the account the possibility of market effects on fair value variation, they believe that the Brazilian market still "lacks maturity" and requires "clearer" and more "specific" rules with regard to the issue of fair value.

The interviewees from the Historic Cost Case communicated that the possibility of volatility exists and that analysts face "difficulties" following companies, due to this volatility caused by the market. In addition, they mention that changes in the market with the "reevaluation" of investment properties "hide" investments in fixed assets (CAPEX) carried out by companies. They argue that fair value may be used for companies with negative EBITDA that report profits with "other operating revenue" resulting from fair value variations.

This evidence supports studies by Edelstein et al. (2012), as well as Devalle and Rizzato (2011), who claim that the use of fair value can have a double effect, having an impact on various performance indicators and not only on results.

The results achieved by fair value companies with "reevaluation" of investment properties can be a way achieving added value for shareholders. These are mainly international funds (see subsection 4.2.1) that invest in companies looking for a quick return, achieved with the "gains from fair value". The opposite occurs in historic cost companies, in which the investors are mainly the founders, who prefer an increase in company value and its consolidation in the market, as opposed to a return exclusively from dividends.

It can be inferred that, for companies that capitalize on fair value, managers are more worried about pre-informing the market, via results, regarding performance in managing investment properties; while those that do not capitalize on it prefer to record investment property management performance in results, only when it is realized.

4.1.4 Maintaining the status quo.

In the interviews, when the market analysts were asked whether the choice of historic cost was more comfortable for those producing financial statements, they replied that it was not. According to them, this assumption would make sense provided that the rule allowed the historic cost option, but did not oblige fair value to be reported in explanatory notes. The analysts believe that "the job is the same", whether it is recognized in balance sheets or only in explanatory notes, with the difference being that of capitalizing or not on the value. It was also asked whether the introduction of IFRS had changed the level of comfort involved in carrying out analysis work. They responded that, for carrying out the work, the adoption of IFRS did not bring "great changes", since their analyses are guided by the "business model" and, with the adoption of IFRS, there were no "great changes" in companies' business models.

The external auditor, like the analysts, believes that the segregation of investment property from fixed assets was a relevant change, since it is an income earning property, unlike fixed assets: these generate income indirectly, and the former, directly, for example through the receipt of rent.

The interviewees from the Fair Value Case highlight the difficulty of implanting IFRS and fair value in Brazil due to assumptions and judgments, and believe that the country still has to evolve a lot. They consider discussions regarding the subject as "positive", since they "seize the opportunity" to address "wider" concepts, bringing benefits not only to companies, but also to the whole market.

In the Historic Cost Case, the interviewees argue that they did not and do not have any problem with adopting the accounting rules, and that those who "request changes" are the auditors. With the computer system "working perfectly", the change can be made "smoothly", and there is no "difficulty" in following the accounting criteria. They cite examples of companies from other sectors that suffered a negative impact from fair value, confirming that historic cost maintains a "state of calm", which existed when investment property was treated as a fixed asset. According to the interviewees, investors "speak little" about the possibility of adopting fair value and, if there no "legal requirement" for the change, they will not alter the reporting method.

In this analysis, it is verified that the agents from outside the companies (market analysts and auditor) claim that maintaining the status quo does not have an impact on the choice of fair value or historic cost for measuring investment properties. However, the agents from inside the companies reveal signs of uncertainty in the use of fair value, which may lead to maintaining the status quo, as predicted by Devalle and Rizzato (2011), Martínez et al. (2011), and Capkun et al. (2008).

State of comfort is also related to shareholder profile and with each company's business model. Historic cost companies, in which Brazilian shareholders predominate, feel comfortable with maintaining it, since there is not yet a culture of evaluating non-financial assets by fair value in Brazil. Fair value companies have a strong presence among international investors who, in their countries of origin (Europe), are used to "marking for market" non-financial assets, and in the end did not experience any discomfort in measuring and reporting investment properties by fair value. This last aspect is explored later, since it is one of the differentiating characteristics in companies' business models.

4.2 Explanations Related to Business Models

In this item, other characteristics, which are different from the traditional ones already mentioned and featured in Figure 1, that relate to companies' business models, are presented. These characteristics help to understand why accounting choices are multifactoral and that these factors, in the case of Brazilian investment property companies, lie in business models. They are important factors because they complement the explanations for the accounting choices of the companies analyzed.

4.2.1 Shareholder composition.

In order to know the shareholder composition of the companies, the main shareholders and the percentage of control that they held in each company, compared with the percentage in free float, were consulted on the BM&FBOVESPA website. This information is in Table 5.

Table 5
Shareholder composition

Company	Number of Main Shareholders	Participation of Main Shareholders	Free-Float
Fair Value Case	2	13.22%	86.78%
Historic Cost Case	4	73.07%	26.33%

• Source: Developed by the authors based on data from the study.

When questioned about shareholder composition, the initial response of the auditor and of the analysts was that this business model characteristic would not have directly influenced in the choice of fair value or cost for measuring investment properties. The auditor believes that a more concentrated company may have "different policies" regarding investments in new ventures from those of less concentrated companies. Companies that are more fragmented on the market may be more "worried" about results, and, for this reason, end up buying businesses with a view to merely "earning money".

More concentrated companies are also more conservative, due to their "owners' vision" and a certain "romanticism", in the words of the auditor. Because of this, they construct the majority of their businesses instead of acquiring them already built. Also according to the auditor, investment banks are relevant investors in more "fragmented" companies, and these are "more aggressive" with regards to growth. They prefer fair value for measuring investment properties, since this measurement anticipates companies' potential for growth.

The market analysts believe that the logic for companies with a more fragmented shareholder base adopting fair value makes more sense than more concentrated companies opting to adopt historic cost. For them, companies with a more fragmented shareholder base would have incentives to adopt fair value, with a view to "showing" their fair value to the market and, for this reason, allocate more "visibility" to the fair value number. However, in less concentrated companies, they do not believe that the non adoption of fair value is due to there being other internal information that supplements it missing from financial statements, among other reasons because this value "is reported" in explanatory notes.

The interviewees in the Fair Value Case, with its fragmented shareholder base, defend the informative ability of fair value and argue that this is the way by which companies manage to "better portray" the value of their assets. They believe that historic cost does not manage to inform the market how much companies really have in their portfolios and how much these are worth.

The interviewees from the Historic Cost Case, with its shareholder concentration, affirm that minority shareholders and controllers do not have "adverse", but rather, "completely aligned" interests. However, when questioned about the use of fair value in internal company decisions, they claimed that this value passes through "much more complex" analyses than the account reported. This leads to the assumption that companies' internal controllers have more information and that fair value is not relevant for their decisions.

In accordance with Collin et al. (2009), the results of this subsection show the influence of shareholder composition over the accounting choices made by companies.

4.2.2 Company size and taxation.

In order to indentify a possible relationship between the size of a company and its political visibility and if, because of this, it does or does not adopt fair value, a comparison of the GLA (gross lettable area) and the number of businesses that each company controls is presented in Table 6.

Table 6
Comparison of Size of Companies Studied (2nd quarter/2013)

	Number of properties	GLA (m2)	Own GLA (m2)	Shopping Malls Managed
Fair Value Case	51 shopping malls	1,620,600	934,900	44
Historic Cost Case	17 shopping malls	712,000	638,959	17

• GLA - Gross Lettable Area Source: CVM Reference Form

Merely with the information from Table 6, it is not possible to accept that, the bigger a company, the less likely it is to choose fair value. The external auditor believes that this should not be taken into account, since the largest companies are visible and considered "icons" in the sector. It would not be the reevaluation of investment property that would change their visibility in the market, however, he believes that balance sheets reported using fair value feature "prettier" numbers, leading to speculation as to whether "prettier" numbers are not a form of visibility.

The market analysts also do not consider a company's size to be a "driving factor" in accounting choice, since the companies are listed on the Stock Exchange and, with this, the market eventually perceives its visibility and ends up "evaluating it every second". In spite of this, they do not deny the possibility of the existence of "vanities", for companies to appear bigger than others.

However, one point questioned by the analysts was the taxation position of investment properties. They report that, at the moment in which one of the methods for measurement was chosen, there was no "clear understanding" about what would be the tax treatment for income with fair value. The Brazilian Federal Revenue Office had not defined, at the time of making the accounting choice, what the tax treatment should be at the moment of sale if properties were "marked for market". This resulted in some companies that opted for fair value recognizing differed tax and others continuing with historic cost, since it would be "simpler", at the moment of sale, to pay tax on the difference between depreciated accounting value and sale value.

The interviewees from the Fair Value Case believe that fair value "draws attention", mainly in the beginning, and argue that attractiveness may occur as a result of the "expressive" and "more relevant" values reported in balance sheets. This fact also occurs with market analysts, who "pay more attention" to fair value when this is recognized in balance sheets. In spite of this, they believe that fair value in balance sheets is not a way of attracting more investors. The taxation question does not appear to be an important factor for companies either, since "accounting and tax balance sheets" are separated, with there being no effect of the reevaluation of investment properties in tax balance sheets.

The interviewees in the Historic Cost Case believe that fair value "does not make any difference" to the visibility of companies and are confident in "the market's ability" to evaluate explanatory notes, with no expected reaction to simply "saying" how much a property "may or is going to increase in value", Regarding the question of tax, they mentioned that, before the choice between one of the two methods was questioned, they did not obtain a "letter of comfort" from the lawyers, highlighting that that Federal Revenue Office was not going to charge tax on gains from fair value, a fact that did not make the choice of this criterion "comfortable". This situation in which there was a lack of a "clear understanding" of the position of the Federal Revenue Office, as argued by the analysts, partly resulted in Normative Ruling n. 1,397, of September 16th 2013 (which was after the interviews).

Contrary to the evidence presented here, Quagli and Avallone (2010), and Pereira (2013) concluded that company size is a characteristic that has an influence over the choice of fair value, since historic cost is linked to contractual efficiency, given that it reduces agency costs, political visibility, taxation and legal proceedings, as explained by Watts (2003) and Qiang (2007).

4.2.3 Family and professional management, business characteristics and clientele.

The Fair Value Case is fragmented throughout the stock market and the shareholders (international funds) aim for higher and quicker dividends, as a form of return on their investment in the company. Dividend is related to profit and this increased exponentially with the reevaluations of investment properties; however extraordinary dividends, paid with the "gains from fair value", started being taxed from 2014 onward.

The Historic Cost Case, with its founders in management and as the main shareholders, did not consider this immediate return from fair value interesting, and is more concerned with taxation. The relationship between these controlling shareholders and founders and their companies is more long term than the relationship of international funds with companies.

Companies with a fragmented shareholder base and belonging to the financial market have a greater need for quick results and detailed accountability to their shareholders, who are, basically, international investment funds. These funds, because they are in Europe and the United States, are used to the way of reporting by fair value, due to the fact that there is an active property market in these places. With a need for rapid development, the companies normally acquire businesses in operation and seek to serve classes C and D, which are growing exponentially in Brazil.

In the company in which the founders are the controllers (Historic Cost Case), the business model is guided by the "owner's vision". In this company, projects are constructed with the desired characteristics, in the intended location, and serve classes A and B. Due to the founders being in the business' management, fair value is much more useful in the "vision and perception" of the owners than in the reporting in financial statements.

These differences in the business models for the Fair Value Case and the Historic Cost Case are related to the companies' different organizational and management strategies to make them competitive and profitable, adding and creating value to their products, and serving the specific needs of their stakeholders.

In conclusion, the Fair Value Case has professional management (market executives), fragmented control, aggressiveness in the management of the investment properties (it seeks quick results), and a significant part of the debt is in the credit market. It is a company that is more linked to the market and uses fair value to decrease information asymmetry with respect to the market value and performance of its businesses.

The Historic Cost Case is a family company, with family management as well. Control is centralized in the family and there are few shares on the market. Debt is concentrated in bank loans that are guaranteed by the controllers (family). The management of investment properties is different, since the controllers construct the properties and select the clients. The management is more characterized by family questions than by the market, and for this reason fair value is not relevant in the financial statements.

It is possible to deduce that these management characteristics form part of the business model definitions presented by Timmers (1998) and Amit and Zott (2001), as well as them integrating into the definition of business model presented in this study, since they contribute to the establishment of strategies that are able to generate competitive advantage and, consequently, value for the companies and stakeholders.

Thus, these characteristics that are related to the companies' business models could explain the accounting choices for measuring the investment properties. As this evidence is based on secondary data analysis and interviews, another test was carried out in order to provide greater consistency to the results, as is laid out in the following subsection.

4.3 Consistency Analysis and Widening of Results

Via the analyses carried out up until here, the difference found in the two companies chosen is perceived to lie in the business model: each one has different informational characteristics and needs, which could result in different accounting choices for measuring the investment properties.

The aim of this subsection is to provide consistency to the findings that business models can explain accounting choices. This consistency analysis was carried out with the next three companies in the sector (after the first two previously analyzed), and is also based on 2012, in order to widen the results found.

In Table 7, a comparison between the business model characteristics of the previously analyzed companies (Historic Cost Case and Fair Value Case), and the business model characteristics of the three additional companies looked at are presented, for the purposes of consistency analysis.

Table 7
Business model characteristics of the companies studied

Historic Cost	Fair Value	Source
Case	Observed	Case	Observed 1	Observed 2
Concentrated shareholding	Confirms	Fragmentated shareholding	Confirms	Partial	BM&FBOVESPA website
Founder as controller	Confirms	Various investors	Confirms	Confirms	Minutes
Constructs the businesses	Confirms	Adquires the businesses	Confirms	Doesn't confirm	Corporate profile
Serves classes A & B	Confirms	Serves classes C & D	Doesn't confirm	Confirms	Standard of the businesses, leaseholders and location
Few and large businesses	Confirms	Many and small businesses	Confirms	Partial	GLA/number of businesses ratio
Low leverage	Confirms	High leverage	Confirms	Confirms	Debt/Equity ratio
Non-financial covenants	Doesn't confirm	Financial covenants	Doesn't confirm	Confirms	Explanatory notes
BCN	Confirms	RERC and securitization	Confirms	Doesn't confirm	Explanatory notes
Market to book < 1	Confirms	Market to book > 1	Confirms	Partial	Index calculation

• BCN = Bank Credit Note; RERC = Real Estate Receivable Certificate; GLA = Gross Lettable Area. Source: Developed by the authors based on data from the study.

As can be verified, almost all of the business model characteristics analyzed in this study in the companies interviewed were confirmed in the observed companies. In Observed Historic Cost, the only characteristic that was not confirmed was the existence of non-financial covenants. Despite the company having concentrated ownership, with the founding family in control and in the company's management, the funding providers did not demand the maintaining of this control in the loan clauses, as occurred in the Historic Cost Case, even though the loan mechanisms were similar (BCN).

The two companies that chose historic cost for measuring investment properties hold large businesses and aim to serve the highest classes in Brazilian society. To achieve this, they prefer to construct the businesses and maintain the standard expected by the controlling family. Moreover, they are not dependent on the financial market and hold few businesses, all with high added value.

With relation to the choice of measuring investment properties by fair value, observations were made of two companies from the sector. In Observed Fair Value 1, two characteristics were not confirmed: existing covenants, which go beyond the financial clauses with the presence of non-financial clauses, such as maintaining management; and the serving of classes C and D, since the company is much diversified and its businesses serve all social classes.

Despite the two non complete confirmations, the two fair value companies (Fair Value Case and Observed Fair Value 1) are greatly fragmented in the stock market, with shareholders strongly linked to the financial market (professional management). In this way, they are more aggressive and require a shorter period of return on their investments, preferring to buy businesses that are ready and rented, which generate immediate cash flow. In order to buy businesses that are ready, they depend on more loans and other forms of funding (RERC), compared to companies that opt for historic cost. However, there is a restriction in Observed Fair Value 1: it exploits commercial buildings and premises, and the Fair Value Case exploits shopping malls.

In order to soften the restriction of the first comparison, a second observation was carried out in a company that also exploits shopping malls. The fragmented shareholding was partially confirmed in Observed Fair Value 2 since, unlike the Fair Value Case and Observed Fair Value 1, Observed Fair Value 2 has two controllers.

In relation to the acquisition of businesses, Observed Fair Value 2 did not confirm this characteristic: the company constructs businesses. Due to the fact that it serves classes C and D, it is similar to the company interviewed and this characteristic was confirmed.

Observed Fair Value 2 exhibits a higher leverage than the others in the sector and, like the Fair Value Case, does not exhibit non-financial covenants. However, the way it captures financial resources is not via RERC, but this fact is related to another difference between the companies, which is the construction of the businesses.

The companies that buy businesses, such as the Fair Value Case and Observed Fair Value 1, require different forms of funding in order to support their investments (RERC and securitization); and the companies that construct withdraw credit in financial institutions to support the construction phase of businesses (BCN), thus exhibiting different funding structures. The market value indicator for Observed Fair Value 2 is close to 1, which indicates that it is almost at the same level as the Fair Value Case and Observed Fair Value 1.

The observations, both in the companies that opted for historic cost and in those that opted for fair value, exhibited points that are not exactly the same, but are at least similar. This shows that the business models found in the companies interviewed may make sense in the observed businesses, which extends the results of the interviews to other investment property companies.

5 CONCLUDING REMARKS

The literature presents partial explanations regarding accounting choices, almost all modeled on the Theory of the Firm and the Theory of Agency, which are based on relationships between individual (or single) characteristics (or factors) and the accounting choices made by agents. This opens up an opportunity for advancing understanding of accounting choices, since various interacting characteristics (or factors) can be incorporated to explain such choices. In the case of this study, by considering the aggregation of various characteristics (or factors) as a business model, it was possible to deduce, with some certainty (taking much into account the small sample representing basically the whole of the analyzed sector), that the business model is a predominant factor in explaining accounting choice for measuring investment properties, in publicly traded real estate management companies in Brazil.

In order to reach such a conclusion, a methodological approach that triangulated documentary research, interviews with the main agents involved in accounting choices and consistency analysis of the results found, was used. The consequences of this approach provide new explanations for accounting choices for measuring investment properties. This means that accounting choices are explained by the interaction of various factors (the business model, in this study), that include business strategy, control and management characteristics, and dependence on credit and capital markets.

Traditional explanations in the literature regarding making accounting choices (debt, information asymmetry, impact on results, and maintaining the status quo), according to studies by Capkun et al. (2008), Muller et al. (2008), Quagli and Avallone (2010), Lourenço and Curto (2010), Martínez et al. (2011), Osma and Pope (2011), Devalle and Rizzato (2011), Edelstein et al. (2012), and Christensen and Nikolaev (2013), can also be found in the five main real estate companies operating in the Brazilian capital market that are researched in this study. And the business model incorporates (totally or significantly) these individual factors that explain accounting choices as if they were a single variable, making it possible to relate business model with accounting choice.

The influence of debt over accounting choices for measuring investment properties was verified, through the existence of two new factors not previously considered in the literature on investment properties: debt calculated via net debt divided by EBITDA, and the existence of non-financial covenants.

Information asymmetry was observed in the companies analyzed, bearing in mind the divergence between the accounting values and market values for the companies, even though investment properties compose a significant part of their assets.

Regarding the impact on the results of the companies studied, the use of fair value in measuring investment properties has a double effect, and that is, the companies may choose fair value with the debt/equity ratio in mind, for example, instead of just focusing on profit or loss.

In relation to maintaining the status quo, the companies' internal agents expressed that, even with uncertainty regarding fair value, the expertise required for calculating and reporting it exists, but in the Historic Cost Case, it will be recognized in the financial statements only by legal requirement, bearing in mind that such assets were recorded by historic cost in fixed assets before IFRS.

In the case of Brazilian investment property companies, it was verified that the business model characteristics (shareholder composition, business and clientele characteristics, and management characteristics) may explain choices for measuring these assets; that is, different business models imply different accounting choices.

The company that applies fair value is controlled by international funds and uses the credit market, while the company that applies historic cost is family controlled and relies on financial institutions. Furthermore, one of the covenants of the company that applies cost is for not transferring corporate control. This situation indicates that, for the company that applies cost, a critical element in the obtainment of bank funding is the importance and influence of the controlling family. In this case, the financial statements and potential use of fair value in order to report higher results are not important factors, which justifies the choice of historic cost. The company that applies fair value seeks to follow the international standard for the sector, so as to maintain the attention of international investment funds, which may explain the choice of this method for measuring investment properties.

The consistency analysis intended to compare of the results derived from the documentary study and interviews with the two biggest real estate companies, and the results derived from the documentary research on the next three biggest companies in the same sector. The analysis revealed that the companies that made the same accounting choice for measuring investment properties feature similar business model characteristics. The final result of this study, which covered the five main Brazilian investment property companies, reveals that the business model of these companies can offer additional explanations for their accounting choices for measuring investment properties. It can thus be understood that accounting choices can be explained by a wider set of economic-financial, managerial, and strategic characteristics of companies. These explanatory factors can be added to those that are known in the literature, such as contracts, opportunistic behavior, results management, and information asymmetry.

However, it is very important to highlight that the results presented here cannot be generalized (study limitations), since they are restricted to the companies studied, which were few in number (even though the sample represents almost all of the relevant publicly traded companies that manage real estate). However, this study shows, nevertheless, that accounting choices can have deeper explanations when various characteristics (or factors) are related together. This is a clear opportunity for future studies.

In other words, this study indentified other factors (or the integration of other factors) that have still not been explored much or at all in the Theory of Accounting Choices, and that can increase the level of understanding of managers' choices. The use of different methodological approaches (like that employed here) is suggested for future studies, in order to develop the study of other accounting choices.

Referências

• Amit, R., & Zott, C. (2001). Value creation in E-business. Strategic Management Journal, 22(6-7), 493-520.
• Andrade, M. E. M. C., Silva, D. M., & Malaquias, R. F. (2013). Escolhas contábeis em propriedades para investimento. Revista Universo Contábil, 9(3), 22-37.
• Ball, R. (2006). International Financial Reporting Standards (IFRS): pros and cons for investors. Accounting and Business Research, 36(special issue), 5-29.
• Barlev, B., & Haddad, J. R. (2003). Fair value accounting and the management of the firm. Critical Perspectives on Accounting, 14(4), 383-415.
• Cairns, D., Massoudi, D., Taplin, R., & Tarca, A. (2011). IFRS fair value measurement and accounting policy choice in the United Kingdom and Australia. The British Accounting Review, 43(1), 1-21.
• Capkun, V., Cazavan-Jeny, A., Jeanjean, T., & Lawrence, W. A. (2008). Earnings Management and Value Relevance during the Mandatory Transition from local GAAPs to IFRS in Europe. Recuperado em 25 janeiro, 2014, de http://ssrn.com/abstract=1125716 ou http://dx.doi.org/10.2139/ssrn.1125716.
  » http://ssrn.com/abstract=1125716» http://dx.doi.org/10.2139/ssrn.1125716
• Casadesus-Masanell, R., & Ricart, J. E. (2010). From strategy to business models and onto tactics. Long Range Planning, 43, 195-215.
• Christensen, H. B., & Nikolaev, V. (2013). Does fair value accounting for non-financial assets pass the market test? Review of Accounting Studies, 18(3), 734-775.
• Collin, S. Y., Tagesson, T., Andersson, A., Cato, J., & Hansson, K. (2009). Explaining the choice of accounting standards in municipal corporations: Positive accounting theory and institutional theory as competitive or concurrent theories. Critical Perspectives on Accounting, 20, 141-174.
• Costa, T. A., Silva, A. H. C., & Laurencel, L. C. (2013). Escolha de práticas contábeis: um estudo sobre propriedades para investimento em empresas brasileiras não financeiras de capital aberto. Revista de Contabilidade e Organizações, 18, 25-36.
• Devalle, A., & Rizzato, F. (2011). Fair value application and disclosure of investment properties (IAS 40): an empirical analysis of European listed companies. International Journal on Business Review, 1(1), 8-15.
• Drucker, P. (1954). The Practice of Management. New York: Harper and Row Publishers.
• Edelstein, R., Fortin, S., & Tsang, D. (2012). An international exploration of financial reporting practices in the real estate industry. International Real Estate Review, 15(3), 347-372.
• Fields, T. D., Lys, T. Z., & Vincent, L. (2001). Empirical research on accounting choice. Journal of Accounting and Economics, 31(3), 255-307.
• Francis, J. (2001). Discussion of empirical research on accounting choice. Journal of Accounting and Economics, 31(3), 309-319.
• Holthausen, R. W. (1990). Accounting method choice: opportunistic behavior, efficient contracting, and information perspectives. Journal of Accounting and Economics, 12(3), 207-218.
• Holthausen, R. W., & Leftwich, R. W. (1983). The economic consequences of accounting choice: implications of costly contracting and monitoring. Journal of Accounting and Economics, 5(1), 77-117.
• Lourenço, I. C., & Curto, J. D. (2010). Determinants of the accounting choice between alternative reporting methods for interests in jointly controlled entities. European Accounting Review, 19(4), 739-773.
• Martínez, J. A., Martínez, F. G., & Diazaraque, J. M. M. (2011). Optional accounting criteria under IFRSs and corporate characteristics: evidence from Spain. Revista de Contabilidad - Spanish Accounting Review, 14(1), 59-85.
• Muller, K. A. III., Riedl, E. J., & Sellhorn, T. (2008). Consequences of voluntary and mandatory fair value accounting: evidence surrounding IFRS adoption in the EU real estate industry. [Working Paper 09-033]. Harvard Business School, Boston, MA. Recuperado em 25 janeiro, 2014, de http://www.hbs.edu/faculty/Publication%20Files/09-033_14a96a28-1bd8-4fbd-9313-b0ad89e1ea63.pdf>.
  » http://www.hbs.edu/faculty/Publication%20Files/09-033_14a96a28-1bd8-4fbd-9313-b0ad89e1ea63.pdf
• Osma, B. G., & Pope, P. F. (2011). Strategic balance sheet adjustments under first-time IFRS adoption and the consequences for earnings quality. Recuperado em 03 novembro, 2013, de http://ssrn.com/abstract=1735009.
  » http://ssrn.com/abstract=1735009
• Paulo, E. (2007). Manipulação das informações contábeis: uma análise teórica e empírica sobre os modelos operacionais de detecção de gerenciamento de resultados Tese de Doutorado, Universidade de São Paulo, São Paulo, SP, Brasil. Recuperado em 01 março, 2013, de http://www.teses.usp.br/teses/disponiveis/12/12136/tde-28012008-113439/pt-br.php.
  » http://www.teses.usp.br/teses/disponiveis/12/12136/tde-28012008-113439/pt-br.php
• Pereira, A. F. (2013). Determinantes na escolha do valor justo para propriedades para investimento no Brasil. Dissertação de Mestrado, Universidade de São Paulo, São Paulo, SP, Brasil. Recuperado em 25 janeiro 2014, de http://www.teses.usp.br/teses/disponiveis/12/12136/tde-25032013-161045/pt-br.php.
  » http://www.teses.usp.br/teses/disponiveis/12/12136/tde-25032013-161045/pt-br.php
• Qiang, X. (2007). The effects of contracting, litigation, regulation, and tax costs on conditional and unconditional conservatism: cross-sectional evidence at the firm level. The Accounting Review, 82(3), 759-796.
• Quagli, F., & Avallone, F. (2010). Fair value or cost model: drivers of choice for IAS 40 in the real estate industry. European Accounting Review, 19(3), 461-493.
• Timmers, P. (1998). Business models for electronic markets. Electronic Markets, 8(2), 3-8.
• Watts, R. L. (2003). Conservatism in accounting part I: explanations and implications. Accounting Horizons, 17(3), 207-221.
• Watts, R. L, & Zimmerman, J. L. (1986). Positive accounting theory. New Jersey: Prentice Hall.
• Whittington, G. (2008). Fair value and the IASB/FASB conceptual framework project: an alternative view. Abacus, 44(2) 139-178.

Publication Dates

• Publication in this collection
  Dec 2015

History

• Received
  12 Dec 2013
• Accepted
  22 Jan 2014