The aim of this work is to determine whether the accounting information provided by construction and real estate firms in Spain reflects the real economic activity, during upswings and downturns. A relationship is found between economic conditions and financial reporting in a continental European country such as Spain. Although there is a relationship between real housing price growth and financial accounting measures, real estate companies seem to show an artificial position, since they only present a relationship with Return on Assets (ROA) based on low quality income (accruals). Therefore, a different financial position influences a different behaviour in these two similar sectors. In fact, some signs alerting of possible risk in real estate firms are shown.

El objetivo de este trabajo es analizar si la información financiera de las empresas constructoras e inmobiliarias refleja la realidad económica, tanto en épocas de bonanza como en épocas de crisis. Los resultados muestran que existe relación entre la situación económica real y la información financiera de las empresas. Aunque hay relación entre el crecimiento real del precio de la vivienda y las variables contables, las empresas inmobiliarias parecen mostrar una situación artificial, dado que solo presentan relación con la variable rentabilidad económica, la cual está basada en resultados de baja calidad (devengos). Por consiguiente, estos 2 sectores similares que se encuentran en una posición financiera diferente se comportan también de forma diferente. Este trabajo evidencia algunos signos de alerta sobre el riesgo existente en las empresas inmobiliarias.

The research interest in the study of economic cycles has increased recently due to consecutive international economic crises and downturns ( Bertomeu and Magee, 2011 and Callao and Jarne, 2011 ). However, the revision of previous literature reveals a lack of studies focused on the relationship between macroeconomic conditions and firms’ accounting information. Only a few papers ( Brown and Ball, 1967 and Guenther & Young, 2000 ) have investigated the effects of macroeconomic features on firms accounting information. Our main purpose is to fill this gap by analysing the relationship between real economic activity and the main economic and financial measures.

The economic crisis and the influence that construction activity has on the world economy and, particularly, on the Spanish economy, make it interesting to focus this paper on both construction and real estate firms. Construction activity is responsible for an important part of the economic growth in the last decade in Spain, Europe and America. Most of the American downturns since World War II were preceded by substantial problems in housing ( Leamer, 2007 ). In Spain, according to Seopan (2008) between 1996 and 2007 the gross added value associated with the construction industry grew at an average rate of more than 5%, although 2007 was the last year of a very long phase of expansion. Even during 2007, the gross added value, directly and indirectly associated with this sector, was 18.6% of GDP (Gross Domestic Product). Furthermore, 20% of employment was attributed to this industry. Ayuso and Restoy (2007) highlight the important role played by the housing market during recent years, as it was an indicator of the general prospects of many industrialised economies. Consequently, we study construction and real estate firms in Spain during a period characterised by economic downturns and upswings. Thus, our time period allows us to cover the last year of a crisis (1996), an upswing period that includes a bubble, and the beginning of the current crisis (2007).

Since economic and financial structures are different between construction and real estate companies ( Gill & Giner, 2013 ), our paper analyses the effects of macroeconomic factors separately for each industry. Furthermore, depending on the stage of the cycle, firms may behave in a different manner. Johnson (1999) concludes that earnings response coefficients are larger in expansions (credit crunch periods) than in recessions (reliquification periods). All this can lead to different practices of earnings management, which may mislead financial reporting users.

This research is of interest due to several reasons. First, studies that consider the relationship between economic activity and companies’ financial reporting are scarce. In addition, we use several proxies for economic activity and a set of accounting variables. Second, this paper focuses on firms related to construction activity, which is one of the mainstays of the Spanish economy, as well as an indicator of the economic prospects. Furthermore, we analyse a continental European accounting country such as Spain; studies on non-Anglo-Saxons countries are less frequent in the accounting literature. Third, in studying two similar sectors devoted to construction activity, their economic conditions should be similar and, consequently, their behaviour too. However, a different economic and financial structure may result in a different behaviour. Therefore, we will be able to gain insight into this issue, by studying why they behave in a particular manner. Fourth, in order to study the strategies in terms of earning management, this research presents a broad view, since it also takes into account estimations of “real earnings management”. This methodology has not been applied to Spanish data so far. In sum, this paper provides empirical evidence to understand why the general economic development was not reflected in some financial indicators before the crisis in the property sector, and supports the importance of a contextual analysis of financial statement information. Our results are useful for several economic agents including lenders and creditors, since they allow these agents to identify the behaviour of construction and real estate firms in Spain during economic downturns and upswings. In fact, our conclusions may be interesting in order to achieve an appropriate business risk evaluation.

The remaining sections of the paper are organised as follows: Section 2 presents the background and the research questions, where the effect of real economic activity on accounting information is highlighted; Section 3 describes the methodology used in this analysis, which includes the sample characteristics, and the variables; Section 4 reports the main results and Section 5 concludes the paper.

Previous research, although scarce, indicates that there is a relationship between economic activity and firms’ accounting variables. Ball and Brown (1967) studied the association between the earnings of an individual firm, the earnings of other firms in its industry and the earnings of all firms in the economy. They find that 35–40 percent of the variation of annual earnings can be associated with the variation of all firms’ earnings, whereas a 10–15 percent can be associated with the industry average. Since then, little research has dealt with the relationship between economic conditions and firm accounting information.

Guenther and Young (2000) found that because of institutional differences, this relationship may vary across countries. They provide evidence of a strong relationship between earnings and the underlying economic reality in Anglo-Saxon countries. While in continental European countries there are incentives to reduce earnings volatility, in Anglo-Saxons countries the main aim is to satisfy shareholders’ needs in order to decrease information asymmetry in relation to managers. Johnson (1999) studied the earnings persistence through the business cycle considering 53,324 quarterly earnings announcements in the seventies for American firms. He concluded that earnings persistence is lower in recessions than in expansions (when investment opportunities are high). The importance of analysing the accruals and cash flow components of earnings has been traditionally studied by the accounting literature ( Sloan, 1996 ). Underlying this interest is that accruals are less likely to recur in the future. Along the same lines, Bernstein (1993) indicates that the accrual system involves higher degrees of subjectivity than the determination of cash flow; in other words, “ a company with a high level of net income and a low cash flow may be using income recognition or expense accruals criteria that are suspect ”. Finally, Trombetta and Imperatore (2014) considering a sample composed of 5.964 U.S. listed companies during the years 1996–2011 concluded that it is important to take into account if the firms are suffering a situation of financial distress when analysing the relationship between business cycle and accounting information characteristics; in this sense, they indicate that the dynamics of financial crises and business cycles do not fully overlap. These authors found that as the financial crisis becomes more intense, managers are more willing to engage in earnings management practices.

We study Spain, a code-law country and both listed and non-listed firms related to Spanish construction activity. These two characteristics can influence the relationship between accounting features and real economic activity. Spain is a code-law and bank-oriented country where information asymmetry problems are solved through insider communication. This fact can lead to a lower demand for disclosure, which can result in a weaker relationship between accounting measures and real economic activity. This relationship may be weaker in the case of non-listed firms. Burgstahler, Hail, and Leuz (2006) indicated that earning management is more pervasive in private firms than in public firms. Even other studies ( Arnedo et al., 2007 and Coppens and Peek, 2005 ) find that income decreasing actions are associated with private companies.