FORMALIZATION OF CONTROL MECHANISMS: THE CASE OF SMALL COMPANIES OF CONTROL MECHANISMS: THE CASE OF SMALL COMPANIES.”

: Control mechanisms are the major problem for small companies such as VSEs and SMEs today. A hypothetical-deductive approach was adopted in this research to demonstrate that the more the size of companies changes, the more the degree of formalization of control systems increases. The study based on the case of 25 Malagasy companies through statistical analyses confirmed this hypothesis. VSEs still have great difficulty in implementing a control system that meets their long-term needs due to lack of resources and the use of modern management tools linked to control systems also varies according to their size. Innovation in this sense is necessary to enable small companies to better control their activities and to ensure a very good performance .


Introduction
Nowadays, in a context of free trade, several companies are being created all over the world. Private initiative is favoured in many countries where needs are growing at an exponential rate. Moreover, in the face of intense competition on the market and the constant changes in human needs, we are witnessing a real "war of entrepreneurs". Every company wants to become a market leader. Adapting appropriate marketing strategies and innovation seem to be the key factors for success in this context. However, these actions would be useless if the company is unable to control its activities, which implies the implementation of control mechanisms capable of tracking and evaluating individual actions in the company.
Control should be one of the concerns of today's entrepreneur. Indeed, for a long time, managers have focused on the four decisive tasks in business management, namely: planning, organizing, directing and controlling their affairs (Fayol, 1916). Control is at the end of this process, however, it should not be neglected as dysfunctions caused by weaknesses in the control system can influence a company's activities in both the short and long term. In order to have reliable financial data and thus not to make mistakes in decision-making, effective control mechanisms should be put in place. For this reason, the theme "Formalization of control mechanisms: the case of small [145] companies" was chosen. The resulting issue is how does size affect control mechanisms in VSEs and SMEs?
Control is defined as a "process of reducing uncertainties" (Carvalho, 2003). When we talk about this mechanism, other authors look at the fact that control is none other than the control of all one's affairs (nothing escapes the vigilance of the manager). One of the characteristics of the modern decision-maker is "the will to control the world and the will to control the world, it is to be convinced that one can (capacity), and that one must (imperative), control it" (Solé, 1996). This mastery extends over several fields: time (one seeks to anticipate, plan, anticipate and project oneself into the targeted situation), nature (microenvironment and macro-environment), space, technique and self-control (the leader should not allow himself to be influenced by his personal emotions in his decision-making; he should be objective). From this, we conclude that the greater the control, the lower the risk.
Control is often presented as a mechanism to support rational decision-making that best corresponds to the implementation of strategy and the achievement of organizational objectives (Simons, 1947). Nevertheless, following the work of the School of Human Relations, theories of motivation and studies on the behaviour of decision-makers (Simon, 1947), an approach has developed that focuses on the effects of control tools on the behaviour of decision-makers. With this perspective, it appears that the problem of control is above all that of the control of individuals (Merchant, 1998). The first decision-making facilitation mechanism (decisionfacilitating) is therefore complemented by a second mechanism whose purpose is to influence the behaviour of the organization's stakeholders (decision-influencing) in the direction expected by the organization (Demski and Feltham, 1976).
From the above, control is therefore inseparable from information. For March (1991), the main uncertainty in decision-making is ignorance of the information held by others and their likely actions. The main purpose of information is then, its role in a process of reducing this uncertainty. Information also has value because it allows us to choose, make decisions and act. Its value is thus linked to its use in the context of decision-making. Thus, for March (1991), the information gives meaning to a decision-making situation and thus modifies both the structure of options and the preferences sought. In the same vein, De Rosnay (1975) proposed defining information as the content of a message capable of triggering action. Indeed, information will only be of value, at least theoretically, if it has an influence on the decisions to be taken or the actions to be taken. As Amabile and Caron-Fasan (2002) note, attention is focused on the conditions that allow it to make sense, and to provide a useful interpretation for action.
In the organizational literature, Anthony (1988) considers control as "the way in which we ensure the implementation of strategies". Tannenbaum (1968) argues that an organization involves control and that control procedures must be put in place to ensure that "idiosyncratic behaviour" benefits the company. Organizational control is based above all on good internal control, which takes the form of a division of tasks and the establishment of written procedures. Since the introduction of the Sarbanes-Oxley Act and the Financial Security Act, internal control has taken on an increasingly important organizational role (Cappelleti, 2004). GROFFILS (2005) are among the leading authors in terms of internal control. In his opinion, two things are the reason for its existence: the management of information flows and the security of the company's assets.
On the other hand, corporate control has not only focused on the organization but also on the company's performance, and therefore on its activities. As such, management control is one of the most appropriate tools. Antony (1998) states that "management control is the process by which managers ensure that resources are obtained and used effectively (in relation to objectives) and efficiently (in relation to the means used) to achieve the organization's objectives". Robert Simons (1995) defines it as "all formal procedures, built on the basis of the information managers use to maintain or modify certain configurations of the organization's activities". Starting from a case study, Simons (1990) explains that "all organizations, large and complex, have similar management control systems (...) but there are differences in the way management control systems are used". This observation leads him to identify two categories of control systems: the first is that of systems closely monitored by managers. The second includes control systems for which supervision is delegated.
On the other hand, Simons (1995) sees "budgets" as a tool of management control. Budgetary control is also one of the instruments used to measure and monitor a company's activity. For Christian Raulet (1994) "the budget is the quantitative and financial expression of an action programme envisaged for a given period". Secondly, with regard to budgetary control, H. Court and J. Leurion (1982) define it as "a management control characterized in particular by the periodic comparison of data provided for in the budgets". It is therefore a detection system for making decisions. Inaddition, budget and budgetary control have the advantage of being able to encompass all aspects of cost control and economic profitability control at the same time (Parker, 2002).
Financial analysis also has another aspect of activity control. According to COHEN (1997), it constitutes "a set of concepts, methods and instruments that enable an assessment to be made of the company's financial situation, the risks affecting it, the levels and quality of its performance". Many other authors have written on this subject and say that the purpose of financial analysis is to ensure the company's sustainability. Thus, for HOUNKOU E. and WELE P. (2006), financial analysis (accounting approach) or financial statement analysis focuses on establishing a financial diagnosis of the company on the basis of historical data, mainly the balance sheet and the income statement, in order to assess its future performance. This financial diagnosis is a judgment made on the evolution of the annual accounts in order to ensure the value of the company and its profitability.

Following the statements of various authors mentioned above, a major research question arises:
To what extent are company control mechanisms formalized? In order to better respond to this problem, a hypothetical-deductive approach was undertaken and the following hypothesis was adopted: the formalization of control procedures depends on the size of the company. It is assumed that when the size varies, the control system adopted undergoes changes.
To this end, an empirical study was carried out on the case of 25 small enterprises, including 8 very small enterprises and 17 small and medium-sized enterprises. Statistical analyses were carried out to confirm or refute the above hypothesis. Thus, the "Materials and Methods" will be presented first, followed by the research methodology and finally the results and discussions.

Materials and Methods
In order to answer the problem raised at the very beginning of this article and to validate or refute the hypothesis that "the formalisation of control procedures depends on the size of the company", a statistical survey was carried out on 25 companies, all of which meet the small size criterion. Among them are 8 VSEs that employ between 1 and 10 employees and SMEs with less than 100 employees, the size being measured by the number of employees. The SME category has been divided into 2 homogeneous subsets: 11 companies with between 11 and 50 employees and 6 companies with between 50 and 100 employees. The maximum threshold of 100 was set because the probability that an SME in this range will meet the criteria set out in the literature review is about 80%. The survey was conducted anonymously in accordance with the wishes of certain company managers concerned. The description of these companies is illustrated in the following figure where they are referred to as company number "n". Restoration and production of "foie gras" 12 E2 Accommodation and bedroom rental 15 E3 Marketing of PPN, agri-food products, hygiene products 26 E4 Marketing of biscuits, wafers, chewing gum, 95 E5 Collection and export of medicinal plants 16 E6 Repair, maintenance and sale of computer equipment 13 E7 Marketing of solar panels and related equipment 20 E8 Multimedia services and large format printing 12 E9 Marketing of PPNs and branded alcoholic beverages 3 E10 Cyber-café and gaming rooms 4 E11 Microfinance 53 E12 Hotel business 60 E13 Import and sale of home and office furniture, sports equipment 30 E14 Sale of local products and PPNs 5 E15 Sale of PPN 6 E16 Catering and fine cuisine 10 E17 Network problem solutions 9 E18 Processing and export of essential oils 35 E19 Manufacture and marketing of Malagasy handicrafts 23 E20 Marketing of bottled mineral water 48 E21 Design and marketing of blankets and sheets 95 E22 Import and marketing of sporting goods 6 E23 Carpentry, cabinet making 10 E24 Production of bottled exotic fruit juices 70 E25 Manufacture of plastic products: pipes, containers and tubes 64 The survey questionnaire consisted of approximately 100 questions and is divided into four main sections. The first concerns general information on the company, which consists of general questions such as: the age of the company, the company name, the number of shareholders, the main activity, etc. The other three parts include questions on prior controls, ongoing controls and post-controls. By stratifying the survey population into three homogeneous groups, it was easier to track the response to the test in relation to size. Two variables were analyzed. On the one hand, the variable "T" which represents the "size of the company" whose modalities are represented by the workforce (FOLIARD, 2010): [1 ;10]; [11; 50]; [50 ;100]. On the other hand, there is the variable "F" which has the following modalities: "formalized"; "not very formalized". Basically, it was to perform a Pearson independence test on these two variables to determine if there was a correlation between them.
By choosing a dozen closed-ended questions (answer with "yes" or "no") on the formalization of control mechanisms, a first trend was established. This allowed the construction of a cross-analysis grid to verify whether there is a relationship between the formalization of control mechanisms and the size of the company. The calculation of the chi-square or the Pearson independence test rated "x²" could be performed later. This test is used to determine whether there is a relationship between two qualitative variables. It is done in stages. First of all, it was necessary to construct the actual establishment plan based on the analysis of the questionnaire (based on the first plan).
Finally, the last result to be expected will be the computerization of the control system. Computerization has a close link with the formalization of control procedures. The use of software (accounting, stock, etc.) and intranet will be particularly evaluated in this respect.

Results and Discussions
The survey questionnaire that we developed using the Sphinx software enabled us to deal with several topics, including the formalization of the control system at the level of VSEs and SMEs. The analysis of the report made it possible to establish the following table 2, which presents the respondents' responses to the 12 selected closed-ended questions asked of them. The results of this analysis make it possible to assess the general trend on which our research will focus. According to this table, it is noted that one third (33%) of the VSEs are formalized from a control point of view. In the first group of SMEs, which have between 11 and 50 employees, the situation has changed, the formalization threshold has risen to 59%. And when the threshold of 51 employees is exceeded, the degree of formalization reaches 96%. At first glance, it is therefore clear that the degree of formalization increases with the size of the company.
In order to better appreciate this relationship, Table 3 of the staffing levels observed below has been drawn up. Source : the authors It is noted that few companies are formalized in the first tranche [1; 10]. In the second[10; 50], the gap is narrowed between the most and least formalized firms and in the third tranche, control mechanisms reach a high level of formalization. Indeed, the companies in the last tranche all already have procedure manuals, concerning all management cycles, capable of tracking individual or collective actions in companies but also of empowering authors. From a control point of view, these entities use rigorous methods and all the management tools to ensure maximum security for the company: management control, budgetary control, financial analysis, formalized accounting and cost accounting. It should also be noted that they use external audit work to ensure the effectiveness of internal control, whereas this practice is almost totally absent among very small and smaller SMEs.
Before proceeding to the Chi² test itself, it was necessary to draw up Table 4 of the theoretical enrolment for the same variables. The notional headcount is obtained by multiplying the line total by the column total; all divided by the total headcount. The calculation was done using a "programmed Excel spreadsheet". The x² was then calculated. The results have been presented in Table 5 below: With α=5% and d.o.f (degree of freedom) =2, χ²calculatedis higher than the value of χ² in the Chisquare law table. There is therefore a relationship between the two variables we studied in this second section. The "nij" represent the actual numbers (Table 2) and the "eij" are the theoretical numbers ( Table 3). The hypothesis that the formalization of control mechanisms depends on the size of the company is validated.
It plays a fundamental role in all existing functions within a company and control mechanisms are closely linked to it. This is why it was decided to address the question of computerization, which is a corollary of the control system in the company, as part of this research. As such, some questions were asked to the same company managers. Table 6 below summarizes the responses provided by company size.  Positive answer Negative answer Positive answer Negative answer Positive answer

Source: the authors
The situation is different for each type of company. For very small businesses, formalization is still a real challenge. Only 50% of the companies surveyed use computer equipment. The rest of the VSEs (the other 50%) still use simple means and supports such as "paper and pencil".
When the threshold of 10 employees is crossed, computerization begins to grow. Among SMEs in the first category, the computerization rate reached 36% (compared to 18% for VSEs). Companies are already familiar with the various accounting software and especially office automation software (Excel, Access...) which allows an incredible time saving not only for management but also for control. Thanks to the computerized accounting and financial management system, the accountant can carry out all extractions independently and flexibly, enabling him to target or deepen his controls. The accountant will therefore be able to query in real time the budgetary and accounting movements of previous years.
For companies that have reached a significant size (employing more than 50 employees), the computerization rate reaches 70%. These companies have fairly formalized control mechanisms based on the previous test. In addition to the control tools used by these organizations, there are various physical security devices (100% of these companies have them). This significantly enhances the security not only of the property but also of the entity's assets. Unlike smaller SMEs, Management Control is computerized and these organizations already use developed dashboards to monitor their activities. But what is most striking about these entities is the fact that many of them have set up an intranet system to monitor individual and collective actions in real time. Despite the fact that it is a little difficult to set up, intranet saves a lot of time.
As we have seen, control procedures become much more formalized with the increase in the size of the company. It is obvious that some company managers have difficulty formalizing the way they manage their company and setting up their control systems. These difficulties may be financial in nature, but the manager of the company himself may be a barrier to the formalization process because of his reluctance towards the latter.
A good control is based above all on a good organization at all levels (purchasing, sales, storage, payroll...). The implementation of procedures is particularly important in this sense, as they indicate the processing circuit of operations by specifying: the task to be done (what), the level of responsibility (who), the different processing steps (when), the places of execution (where) and the mode of execution (how). In addition, the controller may, in the event of theft, for example, identify the perpetrator and explain how the perpetrator did it.
In VSEs, the high number of companies that do not have a manual of procedures is due to the fact that they are communicated to staff verbally. As the Manager acts intuitively, he does not attach too much importance to setting up a work organization. The staff members are, moreover, multitasking. In a VSE, when the storekeeper is sick, both the accountant and the driver could just as easily replace him. The manager should then train his staff so that each employee can meet his or her needs. In VSEs, there is actually no formal system or control mechanisms and almost all tasks are centralized around the manager. In addition, it ensures that self-checking is carried out at all levels. When procedures are given verbally, the Manager should state them as clearly as possible and ensure that everyone understands what they mean. It should also ensure that employees comply with verbal instructions given and may assist in the performance of a task by an employee (e.g., invoicing and filing invoices in chronological or alphabetical order) and correct it if necessary.
In small SMEs, we note that 4 out of 7 companies do not yet have a procedure manual. This is especially common among SMEs with fewer than 50 employees. Interpersonal management (FOLIARD, 2010) or direct supervision (STEINMETZ, 1962) still persists in these cases. The reference author mentions that interpersonal management (directly between the manager and the employees) is only applicable at a threshold of 7 employees. Beyond this threshold, a minimum of organization must be established to avoid probable losses of information. In short, these SMEs are moderately formalized and their managers implement the various control tools adapted (management control, budget, financial analysis, etc.). However, the leader remains influential despite the presence of a good organization. Its intervention will depend on the quality and effectiveness of the control mechanisms it has developed.
When the size exceeds 50 employees, the procedures are already very formal and all companies already have an organization chart. In this case, they use material resources to support the procedures (safe, surveillance cameras in warehouses, etc.). Some premises (e. g. warehouse) are only accessible by wearing a badge or holding an access code. Computerization plays an important role in this respect because it facilitates cross-checking and mutual control in order to detect anomalies, errors and fraud quickly. These SMEs use different accounting and inventory management software to automatically process information. For this category of companies, it is no longer a question of focusing on the effectiveness of procedures alone (as in previous cases) but also on efficiency (working methods generating lower costs).
In the previous cases (VSEs and SMEs with less than 50 employees), the computerization rate is quite low and "intranet" is not present in these small entities. For information purposes, intranet is an internal network within a company that facilitates the exchange of information and remote access to work applications. It has the same architecture as an Internet network, but is only private and internal to a company. It improves the exchange of information within a company. But its advantage is above all the collaborative work that makes it possible to easily make documents available to employees and to have centralized and consistent access to the company's memory, this is called knowledge capitalization. In this way, it is generally necessary to define access rights for intranet users to documents on the intranet (confidentiality, reliability, security). In SMEs with more than 50 employees, only half of the companies surveyed have an intranet network. For cost reasons, many managers of VSEs and SMEs are reluctant to set up an intranet network.
The formalization of the control system is clearly a function of the size of the companies according to the results obtained if one considers the use of certain management tools (example: budgetary control, management control, financial analysis, etc.).). Making forecasts is not customary in VSEs. According to the survey results, 87.5% of very small companies do not make forecasts (see Table 1) and among those that prepare a budget, some do not even bother to compare it with any achievements. On the other hand, SMEs seem to attach particular importance to budgeting. Indeed, respectively 63.6% (SMEs with less than 50 employees) and 83.3% (SMEs with more than 50 employees) of these companies answered "yes".
Concerning financial analysis, 87.5% of VSEs and about 36.4% of small SMEs do not carry out financial analysis, whereas for companies with more than 50 employees, this practice seems essential. According to statistics, approximately 62.5% of VSEs and 36.4% of small SMEs do not prepare financial statements (balance sheet, income statement) at the end of each financial year.
VSEs have limited resources and therefore management control does not seem to be a priority for them. However, 37.5% of respondents carry out controls by setting objectives, the achievement of which is monitored by indicators. On the side of SMEs with less than 50 employees, management control is practiced in 63.6% of cases and all SMEs with more than 50 employees do management control. At this level, it is a question of establishing a dashboard by center (purchase, production, sales...) by making the distinction between profit center and cost center. For SMEs, the costs of setting up a management control system must be optimized.
In principle, the need for management control is justified by the volume of data available internally, the complexity of the organization and the existence of a multitude of activities. In VSEs, these factors may not justify the implementation of management control. Especially since their managers do not have the appropriate knowledge of this discipline.
These results confirm the importance of size in the formalization of control mechanisms in small firms and the degree of rationality of the control system also varies in parallel with this variable.

Conclusions and Recommendations
In conclusion, the concept of the term "control system" differs from one company to another, but it depends on the size of the company. It is noted that this system is not at all formal in VSEs. Indeed, these companies still have difficulty setting up an organization that ensures a minimum level of security. On the other hand, as size increases, companies begin to be better organized despite the fact that some of them are still managed in an interpersonal way. From the threshold of 10 employees, a minimum organization is set up to help the company manager to manage his business. Beyond the threshold of 50 employees, the main concern of the manager is to keep an eye on everything that is happening in his company, but at the lowest cost. In this case, he is no longer able to take care of the elementary tasks because he is engaged in other much more important activities. He should therefore be able to find by himself the right control formula adapted to his company.
When the threshold of 10 employees is crossed, an organization should be created and it is in the entrepreneur's interest to set up what is known in everyday language as "self-checking". Despite this, many of these categories of companies are still managed in an interpersonal way. Malagasy VSEs have difficulty formalizing their control system because of the costs and the psychological implications. Nevertheless, the owners of these companies should at all costs try to follow the standard from a control point of view in order to avoid any loss of useful information and therefore likely loss of economic benefits.
The initial hypothesis was therefore validated in this study. However, the latter has had some limitations that do not distort the data. Indeed, the workforce was chosen to measure the size of a company when turnover or balance sheet total could have been an interesting option, but data in this sense were not available due to the reluctance of some respondents.
In light of what we have seen in this article, other research questions related to the case of small companies are possible. Of these, the one that focuses on the use of control tools is the most important. This issue is a corporate governance issue and another research approach is, in this case, to be anticipated.