Evolution of information content from an institutional perspective: El Alcázar brewery (1928–1993)

Introduction

Historical analysis contributes to our understanding of institutional context by focusing on how changes in the organizational environment over time affect organizations and how organizations develop (Carmona et al., 1998; Gomes et al., 2011; Zald, 1990). Knowledge of the historical information produced by businesses is thus useful not only for standardization, but also for the interpretation of accounting information. This is because it points out the prism through which such information should be interpreted, as accounting can only be fully understood in its social and institutional context (Hopwood, 2000; Miller, 1994). Thus, our research is useful for users of present financial information in showing how the environment can influence organizations in general and accounting in particular (Scott, 2001). With the aim of studying the dependencies of information on its environment, the present historical study not only covers the range of company annual reports, which, according to Ditlevsen (2012), is not common, but also contrasts this external information with internal documentation in order to better appreciate the changes in the different types of information.

This article examines the information provided by a private Spanish company (El Alcázar brewery) given the institutional changes that occurred over the period 1928–1993. More specifically, we apply thematic content analysis to the evolution of external information – being the Memoriaor qualitative information contained in the annual report and which is a Management Report (MR) to shareholders – along with internal information, being the minutes of the Board of Directors (MBD) and the minutes of the Shareholder General Meeting (MSGM). We analyse these documents with the aim of determining whether their content adapts to the different institutional settings experienced by the company. All documents are compulsory, the MR through the Articles of Association (not compulsory by law until the Companies Act 1951) and the minutes under the Spanish Commercial Code 1885. However, our main reason for choosing these documents is that their non-standard content allows the company considerable discretion in what content is and is not included. This means that we can study the evolution of information over time through the changing social and institutional settings (Adams and McNicholas, 2007; Laine, 2009).

Existing studies that have linked the institutional environment and accounting have primarily focused on management accounting (Carmona and Macías, 2001; Covaleski et al., 1996). However, in the past few years, this body of research has been extended to the study of accounting change so as to include the social and institutional dimensions of organizations and their environments (Moll et al., 2006; Ribeiro and Scapens, 2006). In work that is more related to financial information, Suddaby et al. (2009) identify the possible relationship between changes in the environment and institutional change, adding that this can be tested by conducting content analysis of institutional texts, such as annual reports. In this regard, longitudinal case studies are considered the most appropriate research method for analysing changes over time (Burns, 2000; Dawson, 1994; Pettigrew, 1990). For instance, Tinker and Neimark (1987) show how changing social contexts can transform financial reporting and the ideals it reflects. Similarly, Fernández-Revuelta et al. (2002) highlight the ability of financial statements to report on the social and economic environment.

Our study includes a number of dimensions that make it especially interesting. To start with, the sample period bears witness to relevant institutional conditions, encompassing deep social, economic, cultural and political changes. In particular, political changes include the transition from monarchy to republic to dictatorship to democracy. This makes it possible to identify different contexts or stages within which to analyse whether changes in institutional pressures induced modifications in information (Meyer, 1986). In addition, we employ three long time-series of information as primary sources, namely the MR, the MBD and the MSGM. We know of no comparable methodology that analyses company minutes in this manner. Although MRs, or their equivalents containing mainly qualitative information typically found in annual reports, have been hitherto studied using content analysis (Jones and Shoemaker, 1994), a series as extensive as that found in our analysis is uncommon (Guthrie and Parker, 1989; Neimark and Tinker, 1986). Furthermore, our study expands the literature by departing from the prevailing context of accounting history in international journals and to a lesser extent from the time period most commonly analysed, namely 1850–1945 (Carmona, 2004; Carnegie and Potter, 2000), although recent reviews indicate a small reduction in this gap (Baños-Sánchez Matamoros and Gutiérrez-Hidalgo, 2010; Bisman, 2012). Our contribution is notable not merely for the context and time period covered, but also for the opportunity provided by the dramatically changing institutional conditions (Mogotocoro Guevara, 2010).

We have divided the analysis into the following sections. The second section introduces our framework with special attention to institutional change. The third section, after contextualizing the case analysed at different levels (national, regional and sectoral) defines the stages into which we divide the context. At the end of this section, the main aspects of the trajectory of the company under study are described. The fourth section discusses thematic content analysis as our chosen methodology. The fifth section conducts the content analysis in relation to the categories previously defined. In the final section, we present our most significant findings.

Framework

According to Suchman (1995), studies of legitimacy can be divided into two distinct groups: strategic (Ashforth and Gibbs, 1990; Dowling and Pfeffer, 1975) and institutional (DiMaggio and Powell, 1983; Meyer and Rowan, 1977). The first group argues that organizations react to environmental factors with the objective of legitimizing their actions and surviving (Lindblom, 1994; O’Donovan, 2002; Preston and Post, 1975). In this respect, legitimacy is a resource through which an organization seeks approval from society (Kaplan and Ruland, 1991). The second group attends to the deeper aspects of social structure and considers the processes through which structures become established as authoritative guidelines for social behaviour (Scott, 2005). In this respect, “legitimacy and institutionalization are virtually synonymous” (Suchman, 1995: 576).

Tolbert and Zucker (1983) define institutionalization as a process through which the components of the formal structure become widely accepted and are considered appropriate, whereas North (1990) identifies institutions with the rules of a society, comprising formal and informal rules. Elsewhere, Scott (2001) defines institutions as social structures that have attained a high degree of resilience. Although the institutional approach has traditionally been applied to analyse the homogeneity and isomorphism of organizations, it is also appropriate for analysing the processes of change (Lukka, 2007; Ribeiro and Scapens, 2006; Seo and Creed, 2002). ¹ Thereon, Dacin et al. (2002) argue that institutions change over time, are not uniformly taken for granted and are challenged. Thus, institutions serve both to promote change and to shape the nature of change at different levels and contexts. Greenwood et al. (2002) note that the institutional approach neither denies nor is inconsistent with change, adding that the isomorphisms involve the movement from one position to another and therefore a change. Powell (1991) states that change is neither frequent nor routine, because it is expensive and difficult, and that when it does take place it is episodic, motivated by brief periods of crisis or critical intervention, followed by longer periods of stability.

Organizational change may potentially arise not only from the internal dynamics of the organization but also from external forces (Barnett and Carroll, 1995; Dacin et al., 2002), especially the institutional environment, where socio-cultural norms and values play a crucial role (Galvin, 2002). Accordingly, Scott (1991) suggests that, in conforming to the open systems model of organizations, explanations and justifications are more likely to come from the environment than from the inside. Oliver (1992) also argues that most studies that have analysed change have almost exclusively identified exogenous factors as precipitating initial change (Boeker and Goodstein, 1991; Hinings and Greenwood, 1988).

Institutions evolve alongside the context through a series of important events. Disruptive events especially can cause sudden changes in the institutional process (Hoffman, 1999). Similarly, Oliver (1992) argues that institutional changes may be motivated by political, social or functional pressures, while Scott (2001) notes that among the external factors identified by researchers as precursors of institutional change, political changes are of great importance (Carroll et al., 1988; Galvin, 2002; Núñez-Nickel et al., 2006). Changes in power structures may lead to the introduction of new mechanisms that delegitimize the existing forms and involve new cultural-cognitive concepts that provide a basis for new policies, legal mechanisms and regulatory frameworks (Dacin et al., 2002). More specifically, Carroll et al. (1988) study the effects of different political regimes and how regime change induces organizational change. In this regard, Orrú et al. (1991) establish that the roles of the State and historical political events in each country are of great importance in explaining the development of different organizational forms.

Powell (1991) emphasizes that accounting is an activity strongly shaped by expectations and institutional pressures (Burns and Scapens, 2000; Covaleski and Dirsmith, 1988; Hopwood, 1983). Accounting and its vocabulary are identified as key elements of the isomorphism of institutions and practices around the organization, so that organizations that incorporate practices accepted by the environment increase their legitimacy and prospects for survival, to the extent that these formal structures reflect the myths of the institutional environment more than the demands of their own activities (Meyer and Rowan, 1977). Meyer (1986) examines how different environments can determine the amount of accounting information produced by a particular company, beyond the variation determined by its intrinsic technical needs. Scott (2001) then suggests that detecting the effects of institutional structures on practices requires a comparative analysis of the differences in organizational systems over long periods of time or across space.

Our longitudinal research focuses on change that occurs in accounting information in response to changes in the institutional environment. We divide the institutional environment into five distinct periods, each clearly defined by its own political, social and economic characteristics. According to Oliver (1992), longitudinal studies of institutionalized activities under changing conditions are particularly appropriate for examining the contribution of political and economic factors to the destabilization of institutional practices.

In particular, the institutional framework has been widely used in the analysis of accounting. For comparable studies, that is, longitudinal qualitative content analyses of annual reports, we identify Cámara de la Fuente (2002), Cormier et al. (2005), CasasolaBalsells (2007), Laine (2009) and Chamorro Rufián (2011). However, Cámara de la Fuente (2002) and Chamorro Rufián (2011) do not attempt to quantify the qualitative information in annual reports, instead focusing on public sector companies. Cormier et al. (2005) and Laine (2009) focus on environmental information disclosure practices. However, the first of these differs from our analysis in that it considers a much shorter time period (six years) and addresses a database comprising 55 German companies rather than a particular company. Finally, CasasolaBalsells (2007) draws on presidents’ letters for analysis, but for much of the period analysed the companies concerned were public companies.

Context

Because our objective is to study changes in the documents brought about by the effects of changes in the immediate social and institutional environment, it is necessary to divide the study period (1928–1993) into stages. To this purpose, we briefly analyse the national, regional and sectoral environment surrounding El Alcázar, a private brewing company located in Jaén (Andalucía) in the south of Spain, as shown in Figure 1. At the end of this section, we describe the main aspects of the trajectory of the company.

OPEN IN VIEWER

Figure 1.

The location of El Alcázar in Spain.

Nationally, during the twentieth century Spain experienced a range of political regimes, including monarchy, republic, dictatorship and democracy. When the company under study was founded in 1928, the dictatorship of Primo de Rivera governed Spain. In economic terms, the 1920s was a successful decade (Carreras and Tafunell, 1994). However, because of the political and economic problems of the late 1920s, Primo de Rivera resigned in 1930 and the Second Republic was inaugurated. However, the socio-economic problems and high unemployment of the last years of the dictatorship continued. This context produced General Franco’s military push in 1936, which resulted in a civil war (1936–1939) that deeply divided the Spanish people, weakened the economy and destroyed infrastructure (Comellas, 1990). The two first decades of the dictatorship were extremely difficult, characterized by the political isolation of the country and severe food shortages. The regime also controlled all matters related to the granting of quotas of raw materials, products and even pricing in many markets. Alignment with the regime and its social values was encouraged; for the workers, this was promoted with a paternalistic attitude, which only loosened with the development and liberalization of the country. Real economic expansion did not begin until 1959, when Spain joined the Organisation for Economic Cooperation and Development (OECD) and the International Monetary Fund (IMF), and when important economic and social measures of the stabilization plan were also put in practice. According to Tamames Gómez (2005), at this time Spain moved away from autarchy. The changes involved an intensification of relations with foreign countries and the abolition of intervention in the economy. Expansion continued until the energy crises of the second half of the 1970s. After Franco’s death in 1975, Spanish political life returned to normal, mainly with the passing of the liberal Constitution of 1978 which made the country a fully-fledged democracy. In economic terms, this normalization was achieved through successive adjustment plans. In 1986 Spain joined the European Economic Community (EEC), giving rise to increasing liberalization and competition, which stimulated domestic and foreign investment (BadosaPagés, 2005). From an economic point of view, this also led to extensive legislative reforms, including acts regarding VAT, audit, stock exchange and companies.

Regionally, the most important aspects of the Jaén economy were the exploitation of mineral resources, the expansion of olive tree monoculture and dramatic migration in the second half of the twentieth century (Hernández Armenteros, 1999). The vast majority of economic and social indicators have traditionally classified Jaén throughout the twentieth century as a backward region. Figure 2 plots the GDP per capita of Jaén relative to Spain and the EEC-15.

OPEN IN VIEWER

Figure 2.

Relative GDP per capita in Jaén to Spain and the EEC-15 (1955–1991).

Sectorally, in the early twentieth century, a number of major breweries began to emerge in Spain (year of establishment shown in brackets), including Mahou (1889), El Águila (1900), La Cruz del Campo (1904) and Damm (1910). However, in the first third of the century, the evolution of beer consumption in Spain was slow and consequently production growth was relatively low, as shown in Figure 3. By this time, the defining characteristics of the sector were: (a) its concentrated location in terms of manufacturing and consumption around major urban areas; (b) exotic patterns of beer consumption that did not penetrate through to the common people; and (c) the tendency for beer to be regarded as a seasonal beverage (Trujillo García, 1989).

OPEN IN VIEWER

Figure 3.

Evolution of the production of beer in Spain in the twentieth century.

Following the Spanish Civil War (1936–1939), a lack of raw materials led to stagnation in the brewing sector, a problem that did not begin to be resolved until as late as the 1950s. The 1960s witnessed spectacular growth in the industry, with production in Spain increasing fourfold, compared with an increase of only a third in the rest of Europe. This can be explained by four factors prevailing at the time: the nature of the product, the transformation of the rural geography into an urban one, the influence of prices, and the growth and distribution of income per capita (Trujillo García, 1989). This growth resulted in the first injection of foreign capital into the domestic industry and then, a few years later, over-capacity and excess supply. Consequently, according to García Ruiz and Laguna Roldán (1999), foreign investors were obliged to withdraw because they had become concerned about acquiring modern technology but had neglected market knowledge and commercial distribution. From 1975 to 1980, the industry faced another difficult time with the combination of an administered price system and the international inflation crisis (MBD, 24 March 1976). Finally, with Spain joining the EEC, a process of concentration took place in the sector, including the entry of multinationals, which led to a significant reduction in the number of breweries (Moreno, 2012; Trujillo García, 1987).

Having analysed the national, regional and sectoral environments in which El Alcázar operated, we can identify major temporal separations, resulting in five distinct stages, as shown in Table 1. Given the repeated political upheavals, including the Spanish Civil War, it appears appropriate to mark the first division in 1939 with the end of the war and the establishment of the dictatorship. For the segmentation of the rest of the period, we are supported by Fuentes Quintana (2005: 69, our translation), who states that:

OPEN IN VIEWER

Table 1.

Institutional stages used for the study.

Stages	Years	Characteristics
Political instability	1928–1939	– Difficult socio-economic situation – Increase in political divisions – Civil war: the country is divided and devastated
Dictatorship–Autarchy	1939–1959	– Country in ruins – Isolation of Spain – Severe food and materials shortages – Regime controlled all matters related to many markets – Alignment with the social values of the regime
Dictatorship–Opening	1959–1975	– Membership in the OECD and the IMF in 1959 – Stabilization plan 1959 – Expansion of the economy; increasing liberalization – Transformation of the rural geography into an urban one
Democratic transition	1975–1986	– Energy crisis in the second half of the 1970s – Programme of legislative reform (Constitution of 1978) – Plans for stabilizing the economy
Joining the EEC	1986–1993	– Entry into the EEC (1986) – Increasing liberalization and competition with foreign investment – Extensive economic reforms

Source: Authors’ compilation.

Spanish economists tried to respond positively to the changing problems of the Spanish economy by focusing on three major operations with different aims: 1) the Stabilization Plan of 1959; 2) the adjustment programmes in response to the economic crisis of the 1970s, which hit the Spanish economy particularly hard from 1975 to 1985; 3) the set of decisions that led to the integration of the Spanish economy [into the EEC].

Thus, 1959 marks another differentiating point, beginning a period of rapid development with the implementation of a stabilization plan, after which the country begins to move away from autarchy (Tamames Gómez, 2005); a period that includes membership of various international organizations. In 1975, the democratic transition process begins with important political and economic reforms. Finally, 1986 is the last turning point, caused by the impact at all levels of Spain’s entry into the EEC. Thus, the study period (1928–1993) is divided into five stages, as shown in Table 1.

The company under study, El Alcázar, was a public limited company engaged in the manufacture and sale of beer, constituted in Jaén in 1928 (Moreno, 2011). The company was important for the province of Jaén from its inception. In evidence, during the period 1925–1935, 77 companies were founded in the province; of these, El Alcázar accounted for 8.2 per cent of the total share capital raised during the period (Hernández Armenteros, 1999). In 1981, the gross value-added of the company accounted for 2.87 per cent of provincial GDP in the industrial sector. In that year, employment generated by El Alcázar, taking into account both direct and induced employment, accounted for 1.35 per cent of industrial employment in the province (El Alcázar, 1984). Employment reached its highest point in 1990, with up to 497 direct employees, making El Alcázar among the largest companies in Jaén, with respect to both the number of employees and sales volume (AEPJ, 1993; Andalucía Económica, 1991, 1992).

Following its foundation in 1928 through the acquisition of another brewery (El Lagarto, operating 1921–1928), in the first stage (1928–1939) the company settled into its environment in a context characterized by serious social problems. During the Spanish Civil War (1936–1939), the factory was placed in administration, although only during the last year of the war did production stop completely. The second stage (1939–1959) was marked by widespread shortages of raw materials caused by the autocratic regime, which created great challenges for the company’s production process; one measure taken to alleviate this situation was the installation of a maltery. In the third stage (1959–1975), and coinciding with an exponential increase in beer consumption in Spain, the factory was moved to the outskirts of the city, a new general manager with high academic qualifications ² was appointed, significant advances in accounting were carried out, and greater attention was paid to costs. In addition, two major “outsider” shareholders joined the company (Courage, Barclay and Simonds, and Corporación Industrial). In the fourth stage (1975–1986), high inflation caused by the international economic situation, together with the price system operated by the Administration, led to the first years of company losses. In other developments, some contacts were made with other breweries in anticipation of Spain’s entry into the EEC, which finally resulted in shareholders selling the company to Cruzcampo in 1985. In the fifthstage (1986–1993), Cruzcampo launched a major expansion that made El Alcázar the seventh-largest Spanish brewery in terms of production in 1990 (García Ruiz and Laguna Roldán, 1999). Finally, after Cruzcampo was acquired by Guinness in 1991, El Alcázar was legally wound up in 1993. Appendix 1 summarizes the most important events in the history of the company.

Methodology

The documentary sources that we use as the basis of this study are the MR, the MBD and the MSGM. ³ Sample pages from these documents are shown in Appendix 2. To obtain these documents, we were permitted to visit the archives of El Alcázar, SA (today Heineken España, SA ). ⁴ In addition, to complete the series of older MRs, we searched the Provincial Historical Archive of Jaén. We also checked the balance sheets and income statements from 1928 to 1993, and gained access to the rest of the archive, including such documents as deeds, empowerments, contracts, correspondence and statistics, among others.

With regard to the chosen research method, Berelson (1971) argues that content analysis can be useful for describing trends and for studying how the same issues are treated differently. Following Jones and Shoemaker (1994), most studies using thematic content analysis in accounting have focused on discriminating between firms with good and bad results (Abrahamson and Amir, 1996; Davis et al., 2012; Smith and Taffler, 2000) or on analysing the presence or absence of one or more attributes or particular topics (Kothari et al., 2009; Neimark and Tinker, 1986). The issue of corporate social responsibility has, for example, been highlighted in this way (Cormier et al., 2005; Guthrie and Parker, 1989). However, these studies have generally only codified part of the source document, in contrast to the present study, which codifies the entire document.

In terms of the methodological process, the MR, MBD and MSGM constitute the sampling units. For the recording unit (and context), we selected the paragraph. Regarding the sampling unit, existing studies in content analysis have analysed the MR or equivalent, such as the presidents’ letters, the MD&A, or more generally, the qualitative information contained in annual reports; for a review, see Jones and Shoemaker (1994). However, we have not found use of this methodology to analyse minute documents and it is also uncommon to apply it to a time-series as extensive as in our study (Guthrie and Parker, 1989; Neimark and Tinker, 1986). With regard to the recording unit, and in order to codify all of each sampling unit, we specify the paragraph (also used in CasasolaBalsells, 2007, and Karlinsky, 1981). In this respect, we note that it is more usual in content analysis to use words rather than paragraphs (Clatworthy and Jones, 2001; Henry, 2008; Tennyson et al., 1990). However, given the three long time-series of documentation to be encoded, the choice of the paragraph as recording unit is highly appropriate. A review of the literature also shows that when the recording units are words, a computerized procedure is mostly used for encoding (Davis et al., 2012; Frazier et al.1984; Kothari et al., 2009), whereas in the case of topics, paragraphs, sentences or whole documents, manual coding is generally employed (Cormier et al., 2005; Noble et al., 2002; Williams, 1999). In our case also, we opted for manual coding, which according to Wallace (1992) is more appropriate for a thematic analysis; Linderman (2001) adds that it is also preferred for complex categorization.

Regarding the construction of the category system to encode the recording units, in the first phase one of the authors conducted an initial review of the MR and progressively identified the topics that appeared, recording 71 topics in total at this stage. In the second phase, and after taking into account the theoretical literature on content analysis (Bardin, 2002; Holsti, 1969; Kassarjian, 1977; Krippendorff, 1990; Weber, 1985) and the existing empirical research (Jones and Shoemaker, 1994), we adjusted the originally identified topics, after several attempts, to a more operative and schematic system of categories, thus defining a system of seven categories, as shown in Table 2. This phase was especially influenced by studies aimed at codifying entire documents (CasasolaBalsells, 2007; Collins et al., 1993). The final category system is comprehensive and mutually exclusive. In addition, we refused to create a category such as “other” because this kind of self-protective category does not clearly define a set of phenomena, except for the exclusion of all others, and contributes little to the research findings (Krippendorff, 1990). Later, after successive pilot coding, the writing and delineation of the categories were progressively improved (a scheme of the issues encoded within each of the categories is shown in Appendix 3). Using the final system of categories, one of the authors proceeded to manually code all of the documents (i.e. MR, MBD and MSGM).

OPEN IN VIEWER

Table 2.

System of categories used for the analysis.

Collins et al., 1993	CasasolaBalsells, 2007	Our system
Others	Presentation	Formal aspects
External and market information	External events	Context
Segment information	Organization	Social aspects
	Operative	Organizational structure
Innovation and R&D	Investments	Investments
Financial position and performance	Financial references	Results
Future-orientated information	Strategy	Future-forecast

Regarding the counting unit and coding rules, we used relative frequency as the unit, taking into account the following rules: a paragraph may contain more than one category; several categories within the same paragraph have the same weight (1, ½, ⅓ or ¼); and all paragraphs in the document have the same importance, regardless of position. To check the reliability of the system (Krippendorff, 1990), we tested for stability (test-retest, reaching a coefficient of agreement of 93%) and reproducibility (test-test, using a sample from both authors of the study, ultimately reaching a coefficient of agreement of 85% and a Scott’s Pi of 81%).

Content analysis

Drawing on the results of the coding system described in the previous section, we performed both descriptive and statistical analysis, through the five periods into which we have divided the context: first stage, Political instability (1928–1939); second stage, Dictatorship–Autarchy (1939–1959); third stage: Dictatorship–Opening (1959–1975); fourth stage, Democratic transition (1975–1986); and fifth stage, Joining the EEC (1986–1993). Table 3 summarizes the paragraphs coded in the three documents

OPEN IN VIEWER

Table 3.

Summary of the number of paragraphs coded in the three documents.

	Number of paragraphs	Number of paragraphs labelled with X categories
		X = 1	X = 2	X = 3	X = 4
MR	623	526 (84%)	90 (15%)	8 (1%)	0 (0%)
MBD	2981	2433 (82%)	429 (14%)	106 (4%)	13 (0%)
MSGM	1286	1148 (89%)	109 (8%)	22 (2%)	7 (1%)

Descriptive analysis

In general for the three documents under analysis, the predominant categories are “Formal aspects” and “Results”, although in the case of the MBD the second position is occupied by “Organizational structure”. “Investments” and “Results” are the only categories that show both significant variation in tendency and the same sign (negative) across all three documents. From this, we can state that these two categories become less evident in all the documents analysed. Table 4 summarizes the relative importance and tendency of the categories.

OPEN IN VIEWER

Table 4.

Summary of the relative importance and tendency of the categories.

	Relative importance						Tendency
	MR		MBD		MSGM		MR	MBD	MSGM
Formal aspects	1st	25.2	1st	32.2	1st	47.8	↘	↗	↗
Context	4th	12.1	5th	8.0	5th	2.0	↗	→	↗
Social aspects	7th	4.2	6th	6.9	7th	0.7	↗	↘	→
Organizational structure	3rd	18.1	2nd	27.0	3rd	22.2	↗	→	→
Investments	5th	9.7	4th	8.7	4th	2.9	↘	↘	↘
Results	2nd	22.6	3rd	12.9	2nd	22.9	↘	↘	↘
Future-forecast	6th	8.1	7th	4.3	6th	1.6	↗	↗	→
		100.0		100.0		100.0

Note: Relative importance is the relative position of each of the categories in each of the three documents analysed (in bold) and the average percentage (in roman). Tendency is an indicator of the overall change for the whole period (in roman when the variation is less than 5%).

We analysed the evolution in terms of each of the documents analysed through the five stages. The results are shown graphically in Appendix 4. In the MR, in the first stage (1928–1939), coinciding with the first few years of the company, the most visible category is “Results”, which becomes less important and decreases in the second stage. During the period of autarchy (1939−1959), “Formal aspects” increases in presence, corresponding to years of scarcity of supply, whereas “Context” decreases in importance, in consonance with the dictatorial regime. During this same period, the category “Organizational aspects” mainly contains information related to provisioning because of difficulties in obtaining materials; consequently, its occurrence decreases overall. This situation is reversed in the third stage (1959–1975), when the opening of the country brings about a reduction in “Formal aspects” and an increase in “Context”. It also allows the first wave of foreign investment, finally resulting in overproduction, which in turn brings a significant increase in commercial information and, consequently, in the category “Organizational structure”. Especially noticeable in the fourth stage (1975−1986), and coinciding with the period of crisis in the sector, is the lack of “Investments”. Finally, as a result of concentration in the sector associated with Spain joining the EEC and the company’s acquisition by Cruzcampo and later integration into Guinness in 1991, in the fifth stage (1986−1993) financial information is given greater importance and thus “Organizational structure” and consequently “Investments” increase. At the same time, there is a reduction in relative terms in “Context”.

With regard to MBD, we note that “Results” is most important in the first stage (1928–1939). In the autarchy stage (1939−1959), “Organizational aspects” is at its least important. This situation is reversed in the period of opening (1959−1975) when “Organizational aspects” attains its biggest percentage, because of the increase in financial information coinciding with the development of the sector and the incorporation of non-family shareholders into the company. In turn, the participation of “Formal aspects” and “Results” falls. Similarly, in the fourth stage (1975−1986), and coinciding with a difficult period in the sector, “Investments” receives less attention in the MBD. The increase in formalism of the minutes in the fifth stage (1986−1993), as reflected in the category “Formal aspects”, is very significant, especially after the integration of the company into Guinness, as a consequence of being part of Cruzcampo. The greater need for planning in an increasingly complex environment results in an increase in “Future-forecast”. These increases are offset by reductions in the categories “Context”, “Social aspects” and “Organizational structure”.

The MSGM, as a document which reflects mainly formal aspects, is notable for the lack of such significant differences of categories over time, unlike the other two documents. Over the whole period, the categories “Formal aspects”, “Organizational structure” and “Results” represent around 90 per cent of the content of these documents, with the remaining categories accounting for the small residual. However, the first stage (1928−1939) is marked by a total absence of information about “Social aspects” and “Future-forecast” and scarcely any information about “Context” and “Investments”. In the second stage (1939−1959), more attention is paid to “Investments” and “Future-forecast”, while in the third stage (1959−1975), the presence of these categories falls in favour of “Formal aspects”. In the fourth stage (1975−1986), and coinciding at least partially with an extended period of company losses, less importance is given to “Results”, apparently justified by the negative situation in the sector and the country and leading to an increase in “Context”. In the fifth stage (1986−1993), after the acquisition by Cruzcampo, “Results” again increases and “Context”, “Social aspects”, “Investments” and “Future-forecast” are nearly entirely absent.

Statistical analysis: Hypothesis testing

To analyse the evolution of the three documents over the five stages, we conducted statistical non-parametric tests because the variables (categories) failed to adhere to the requirements of normality and homoscedasticity. First, we seek to know whether the categories vary over the five stages in the MR. According to Meyer (1986), different environments may determine the amount of accounting information that a particular company produces, beyond the variation determined by its intrinsic technical needs. Thus, we expect that the content of the MR changes through the different stages to adapt to the environment of the time. This suggests the following hypothesis:x

H1: There are significant differences in the categories over the stages in the MR.

To formally establish whether there are differences in the categories over the five stages in the MR, we employ the Kruskal–Wallis test. Using this test, we reject the null hypothesis that there is no significant difference in each category over the stages at the 95 per cent level for all but “Future-forecast”. We further analyse which particular transition (from one stage to another) exerted the greatest impact on the MR by performing the Mann–Whitney test for each of the transitions. The results suggest that the transitions between the first and second stages (1939) and between the second and third stages (1959) resulted in the greatest change in importance, in that there was a change in rank for three of the seven categories (see Appendix 5 for details).

Second, we seek to know whether the categories vary across the five stages in the MBD. According to Scott (1991), in conforming to the open systems model of organizations, the explanations and justifications for this variation are more likely to be found in the environment than inside the firm. Thus, we expect that the content of the MBD changes through the different stages to adapt to the environment of the time. This suggests the following hypothesis:

H2: There are significant differences in the categories over the stages in the MBD.

To formally establish whether there are differences in categories over the five stages in the MBD, we again employ the Kruskal–Wallis test, the results for which allow us to reject the null hypothesis that there are no significant differences in categories over the stages at the 95 per cent level for all categories except “Future-forecast”. We further analyse which particular transition had the greatest impact on the MBD by performing the Mann–Whitney test for each of the transitions. The results suggest that the transition between the fourth and fifth stages (1986) ⁵ resulted in variation in the largest number of categories (i.e. four of the seven categories; Appendix 6).

Third, and finally, we seek to know whether the categories vary across the five stages in the MSGM. As before, we expect that the content of the MSGM changes through the different stages to adapt to the environment at the time. This leads to the following hypothesis:

H3: There are significant differences in the categories over the stages in the MSGM.

To formally establish whether there are differences in the categories over the five stages in the MSGM, we again employ the Kruskal–Wallis test, the results of which allow us to reject the null hypothesis that there are no significant differences in categories over the stages at the 95 per cent level for all categories except “Social aspects” and “Organizational structure”. We further analyse whether a particular transition had the greatest effect on the MSGM by performing the Mann–Whitney test for each of the transitions. The results suggest that the transitions between the second and third stages (1959) and the fourth and fifth stages (1986) resulted in variation across the largest number of categories (i.e. two of the seven categories; Appendix 7).

Table 5 summarizes the differences in the categories in the MR, MBD and MSGM across the five stages of the study period.

OPEN IN VIEWER

Table 5.

Summary of the differences in categories in the three documents.

	Whole period			1st – 2nd (1939)			2nd – 3rd (1959)			3rd – 4th (1975)			4th – 5th (1986)
	MR	MBD	MSGM	MR	MBD	MSGM	MR	MBD	MSGM	MR	MBD	MSGM	MR	MBD	MSGM
Formal aspects	✔	✔	✔	✓			✓	✓	✓		✓			✓
Context	✔	✔	✔				✓						✓	✓	✓
Social aspects	✔	✔												✓
Organizational structure	✔	✔		✓	✓		✓	✓						✓
Investments	✔	✔	✔						✓	✓	✓		✓
Results	✔	✔	✔	✓				✓			✓	✓			✓
Future-forecast			✔			✓
Sum by document and period/transition	6	6	5	3	1	1	3	3	2	1	3	1	2	4	2
Sum by period/transition	17			5			8			5			8

Discussion and conclusion

This longitudinal study shows the relationship between the discourse of the company and the social and institutional context. Our work has examined the evolution of the MR, MBD and MSGM of a private Spanish company over the period 1928 to 1993, which was very important for the prevailing underdeveloped environment in which it was operating. We applied a thematic content analysis, with the aim of studying whether the content of the documents adapts in response to the different institutional environments in which the company operated. In particular, we analysed three long time-series of documents, of which two comprise internal information, which is uncommon in the literature. In addition, the period that our documents cover encompasses deep social, economic, cultural and political changes, which makes it possible to identify different contexts or stages within which to assess whether institutional pressures induced changes in the information (Meyer, 1986). Moreover, the study departs from the prevailing context of accounting history in international journals, and to a lesser extent from the period most commonly analysed (1850–1945) (Carmona, 2004; Carnegie and Potter, 2000).

In general terms, in the three documents under analysis, the predominant categories are “Formal aspects” and “Results”, although in the case of the MBD the second position is occupied by “Organizational structure”. However, the relative importance of the categories evolves over time in response to changing environmental pressures (Meyer, 1986; Powell, 1991). The categories of “Investments” and “Results” display the same significant variation in tendency in all three documents, with both decreasing. According to Meyer (1986), different environments may determine the amount of accounting information that a particular company produces, beyond the variation determined by its intrinsic technical needs. As discussed, the content of the MR changes across the stages identified (particularly in six categories) as an adaptation to the pressures of the environment of each time. Similarly, the content of the MBD changes across six of the categories during the stages identified to better suit the context. However, the MSGM exhibits less ability to adapt to environmental pressures, despite revealing significant change in the relative importance of five categories during the five stages. In summary, the documents that are most responsive to environmental pressures are the MR as an accounting document in essence and the MBD as an executive document in nature.

With respect to the transitions through the stages, in the case of the MR, the transitions between the first and second stages (1939) and between the second and third stages (1959) resulted in the greatest change in importance. The shift to a dictatorial regime was most evident in the MR, as an external document, as shown by an increase in the formal aspects of the document; similarly, the opening of the country was also most evident in the MR, as shown by the way it gave greater space to the context and the technical and operational aspects of the business, in line with the economic growth of the time in general and changes in the sector in particular. We relate the fact that in the MR the transition of 1959 had a greater impact than the transition of 1975, because of the profound economic changes produced in the opening, which had a greater influence than the political changes of 1975. In the case of the MBD, the transition between the fourth and fifth stages (1986) resulted in the variation of the largest number of categories. Thus, Spain’s entry into the EEC had the greatest impact on the executive document, as this resulted in concentration in the brewing industry, leading to the purchase of the company by Cruzcampo, which caused the most radical changes in the management of the company. In the case of the MSGM, the transitions between the second and third stages (1959) and between the fourth and fifth stages (1986) resulted in variation of the largest number of categories. The MSGM is most directly related to the shareholders and thus it reflects these two transitions because they are related to important financial and investment operations that the company carried out, first in order to participate in the growth that occurred at the start of the third stage and then to be acquired by Cruzcampo at the beginning of the fifth stage.

In other studies, CasasolaBalsells (2007) finds significant differences in the content of presidents’ letters of Cepsa (1930–2004) across the different stages of the environment. However, she does not find significant differences in the content of presidents’ letters of Repsol-YPF (1986–2004), although we note that the study period was later and shorter. In a different context and period, Laine (2009) analyses the environmental information for Kemira, a Finnish chemical company (1972–2005), and concludes that the company changed its discourse in response to changes in the social and institutional context. In our case, we can generally say that the three documents analysed changed significantly as the environment changed.

According to Meyer and Rowan (1977), organizations that incorporate practices and procedures institutionalized by society, and which align their practices with social values, increase their legitimacy and survival prospects. The documents analysed reflect the existence of these circumstances in El Alcázar. An example is found in the paternalistic attitude to its workers in 1941, reflecting the social values of the environment and characteristics of the prevailing political regime: “We have not neglected our workers but have kept them in their jobs with full wages, despite the financial sacrifice this represents during stoppage time; we have done so out of a genuine feeling of Christian brotherhood and considerations of humanity” (MR, 1941). ⁶ Another example of these practices is the publication of a social report (El Alcázar, 1984), which reflects the importance the company places on its operating environment in terms of employment and value-added in an attempt to legitimize the company within its environment.It is described as a “Corporate Social Report, referring to the year 1983 and prepared by the Management, which aims to provide a vision of what the company represents in its social context” (MBD, 19 September 1984). In the same sense, we note how the company remarks upon its support of the provincial agricultural policy:

The launch of the Maltery has also allowed us to collaborate crucially in the provincial agricultural policy, since … [it] has allowed us to capture an important part of our province barley production … by providing farmers with selected seeds and then purchasing their produce later. (MR, 1969)

Similarly, throughout these documents, we observe a change in the social values with respect to supporting State decisions and how the information adapts in response to this shift in institutional values (Núñez-Nickel et al., 2006). Thus, in the second stage, despite the severe shortage of raw materials caused by the autarchic regime, all documents contain multiple comments in support of State decisions, such as: “These difficulties, so natural after a war like the one Spain has just experienced, have been able to be overcome thanks to the successful efforts of the National Authorities” (MR, 1939); “on this problem, although all authorities take a great interest, there is still great insecurity in relation to continuity of factory production” (MSGM, 24 October 1940).However, it appears that, by the fourth stage, it is no longer necessary to demonstrate alignment with the official view: “as long as the prices are the result of a correct calculation of costs, and are not, as has been the case until now, arbitrarily set by the Administration in an attempt as false as it is vain to stop inflation figures officially” (MR, 1977).

Another aspect that demonstrates the company’s alignment with the values of society – and which also shows how these values change over time – is the change in the orientation of entities benefiting from sponsorship. Initially, these were very close to the regime: “the company subscribes a quantity in honour of the Armed Forces, and a subscription in honour to the Civil Guard” (MBD, 12 February 1931); “The co-operation of the Company with the parade that, in honour to our Leader, is held in Jaén, concurring with two floats decorated for that purpose” (MBD, 10 May 1961); “[we] give a donation to San Bartolomé parish church and one for the Christmas campaign organized by the Hon. Mr Civil Governor of the province” (MBD, 19 December 1961). Over time, however, they evolved: “[we] have donated two trophies for horse shows from Mengíbar and Torredonjimeno” (MBD, 7 July 1963); and “took part with Granada Savings Bank and other companies in the Foundation of the Business School of Andalucía (ESNA), which aims at creating a business training school” (MBD, 28 June 1989).

As a possible limitation of this work, we note that the content analysis process is inevitably subjective and the results are unavoidably affected by the position of the researcher (Laine, 2009) and the selection of the category system, and may even be skewed by the units of analysis chosen (Lasswell et al., 1965). However, we have attempted to minimize this subjectivity by incorporating measures to control for reliability (test-retest and test-test). Conversely, we have analysed how the different social and institutional environments can affect the information produced by the company, in terms of Adams’ (2002) general contextual factors (Burchell et al., 1985; Mezias, 1990; Oliver, 1992). However, other internal factors may be responsible for changes (Ditlevsen, 2012), such as mergers or corporate size (Adams et al., 1998), negative results (Aerts, 1994; Merkl-Davies and Brennan, 2007; Smith and Taffler, 2000) or changes in leadership (Bamber et al., 2010; Campbell, 2000). Furthermore, the statistical analysis in our study used representative values for the quantity, and not the quality, of information in each category. In this regard, factors influencing the quantity of information may differ from those that influence the quality of that information (Bravo Urquiza et al., 2010).