3.1 Corporate Law

Chile's first corporate law was approved by Congress in 1854Footnote ⁵. The law was strongly influenced by the legislation of continental Europe and was drafted in response to demands for an adequate legislation by the Chilean business community (Brahm Reference Brahm1997) and to solve the situation of share companies organised during the 1850s whose legal status was unclear under the Chilean legislationFootnote ⁶.

The corporate law of 1854 was short and straightforward. It defined a corporation as a legal person formed by a common fund divided into shares. The shareholders had limited liability. As was the case in the rest of Latin America at the time, most of the regulations present in modern corporate law were left to be decided by the charters of each firm. Nothing was said regarding the composition of the board of directors, the requirements for being a director and whether directors were paid by the firm or not. There was no mention of voting rules, whether shareholders had the right to be represented by someone else at a shareholders’ meeting, or if they had the right to call an extraordinary meeting. The only reference to governance matters was that no director could be appointed for life; rather directors should serve for specific terms that could be renewed by the shareholders.

According to the law, corporate charters must include:

• • Name, occupation and address of the founding members.

• • Location of the headquarters of the company.

• • The objective of the company.

• • Capital and number of shares, and the way and time at which the shareholders must complete their contributions.

• • Lifespan of the company.

• • Functions of the administrators and of the general assembly of shareholders.

• • The capital deficit that causes the liquidation of the company and the procedure of dissolution.

Any breach of the charters could lead to the dissolution of the corporation, and managers and shareholders responsible for a contract violation could be sued for the damage caused.

All corporations had to publish a balance sheet every 6 months. Shareholders could only have access to the books of the company at the times and in the ways prescribed by the charters of the firm.

A corporation could be created only with a special licence from the executive branch of the government; this authorisation was also necessary in the case of modification of the charter or dissolution of the company. This was a common requirement in continental Europe at the time the law was passed and was also adopted in Latin America.

The authorisation process had two parts: first, the organisers had to submit the proposal of the charter of the corporation and a list of the founding shareholders and their contributions (at least one-third of the shares must have been previously subscribed by this time). The proposal was then evaluated by the Judiciary Supreme Court, whose role was to make sure that the charter was written according to law. Once the proposal was approved the entrepreneurs were given the authorisation to organise the firm. Once this process was completed, a second authorisation allowed the company to start its operationsFootnote ⁷.

It is interesting to note that the law established that among the causes for rejection was the case where the charter did not offer shareholders «guarantees of a good administration, mechanisms to monitor the behavior of managers and rights to know the use of corporate funds». Given the fact that government officials could reject corporate charters, it raises the possibility that the authorisation process was used to filter out corporations that did not offer enough rights to their shareholders. I argue that this was not the case in Chile.

The reports of the Ministry of Finance and the Chilean Supreme Court reveal that government officials rarely raised objections regarding the governance provisions included in the charters, and the analysis of the charters shows that there were cases in different industries when a corporation was approved without including any rights for minority shareholders. Most of the objections raised by the judiciary are related to formal aspects and internal consistency of the chartersFootnote ⁸, and on the rare occasions where the Supreme Court questioned the internal governance of the firm, their recommendations were not necessarily followed by the executive officials. A telling example is the case of the «Gran Compañía Arturo Prat», a nitrate company established in 1882 with a fully paid capital of one million pesos represented by its mining properties. According to the report of the Supreme Court the valuation of these properties was far above the real market price. The report suggested denying the government authorisation arguing that it should be the government's responsibility to protect investors from potential fraudulent companies. The official answer of the Ministry of Finance was the following: «It is not the responsibility of the government to qualify the «fair price» of the mines that are exploited by corporations, but only to judge that their charters comply with the law». The charter of the company was approved without modificationsFootnote ⁹.

Moreover, it is important to notice that the law allowed a close substitute of the corporation: the limited partnership with tradable shares («Sociedad en Comandita por acciones» (LPTS)). An LPTS was formed between one or more general partners, who managed the firm and were fully liable for its obligations, and one or more limited partners with limited liability and no managing rights. The LPTS allowed entrepreneurs to raise capital from outside investors who had both the benefits of limited liability and liquidity, thereby providing a close substitute to the investment in corporate stockFootnote ¹⁰. The LPTS was used as a substitute for corporations in continental Europe before general incorporation was available. In Chile, this organisational form was included in the 1865 Commercial Code (with a statute that closely followed the French law of 1846). However, it was rarely used by Chilean entrepreneurs (Cruchaga Reference Cruchaga1929), suggesting that the obstacles to obtaining a corporate licence were not that high and that entrepreneurs organising a corporation in Chile during the 19^th century had plenty of freedom to choose the governance agreements best suited to their objectives.

The law also allowed the government to designate a special public auditor (comisario), to monitor corporate managers and their compliance with the law and charters of the company.

Did these government agents provide indirect protection to investors? The evidence provided by the «Directions for public auditors» (1887) suggests that this was not the caseFootnote ¹¹. The description of the functions of these agents seems to be only related with banking activities: «check the monthly financial statement of banks, certify their account balances, and verify that the firm complied with the procedures established in the law». For non-financial companies the role of the public auditors was restricted to tax-collection-related activities.

The law remained virtually unmodified throughout the 19^th century. More detailed regulations of corporate matters were enacted during the 19^th century in Europe and also in some Latin American countries such as Argentina, Mexico and Venezuela (Pistor et al. Reference Pistor, Keinan, Kleinheisterkamp and West2002). Although legal scholars and editorialists of the time proposed modifications in ChileFootnote ¹², there were no significant changes of the governance rules until 1918Footnote ¹³.

3.2 Corporations Chartered in Chile 1854-1902

Five hundred and seventy-three corporations were created in Chile from 1854 to 1902Footnote ¹⁴. Table 1 presents the corporations created during this period and the capital subscribed.

TABLE 1 CORPORATIONS CHARTERED 1854-1902

Source: Own elaboration, based on Boletín de las leyes y decretos del Gobierno and Braun et al. (Reference Braun, Braun, Briones, Diaz, Luders and Wagner2000).

The introduction of corporate organisational form in the Chilean business community was slow and corporations accounted for <5 per cent of the total firms created in Valparaíso between 1854 and 1870 (Cavieres Reference Cavieres1984). The creation of corporations really took off after 1870. Two factors account for this change: (1) the growth of the banking system after the banking law of 1860, which gave investors easier access to credit; and (2) the discovery of silver deposits in Caracoles, Bolivia, in 1870. Seventy-five companies were established between 1871 and 1873 to exploit these deposits and also to exploit nitrate in Tarapacá, PeruFootnote ¹⁵. During these years there was also increased activity in other sectors (including finance and manufacturing) and in the participation of investors: the average number of initial subscribers rose from 26.3 before 1871 to 34.7 during 1871-1873. Investing in stocks, until then a concern for a very small minority, became a much more common activity: teachers, public employees and artisans appear in the lists of subscribing shareholders. After 1871, stock price quotations appeared regularly in the main newspapers of Santiago and ValparaísoFootnote ¹⁶.

The decline in terms of trade led to a recession during the second half of the 1870s, which had an immediate impact on the number of incorporations. Only thirteen new corporations were chartered from 1875 to 1880 and most of the silver companies in Caracoles were liquidated. In 1879, a dispute over the control of the mining resources in the Atacama Desert led to a war between Chile and a coalition between Peru and Bolivia (War of the Pacific).

At the end of the war (1882), Chile took control of the mining-rich territories of Tarapacá and Antofagasta. The wealth generated from the nitrate mines in the newly-annexed territories spurred economic growth and increased both the interest of investors and the number of companies createdFootnote ¹⁷. As had been the case during the silver boom in Caracoles during the previous decade, discoveries of new minerals and fluctuations in international prices led to cyclical stock market booms during the 1880s.

In the last decade of the 19^th century, there was an increase in the number of banks and insurance companies. Also, the manufacturing industry that had started to develop before the war in small establishments grew in size, leading to the creation of larger companies that adopted the corporate organisational form in industries such as beer, tobacco and sugar refining.

During most of the period, stock trade was conducted by brokers in their own offices or in informal saloon meetings twice a week. The first attempts to create a formal stock market were made during the 1872-1873 boom, but none of them lasted beyond 1875. In 1892, the Santiago stock exchange was created with forty registered brokers, followed by the Valparaíso stock exchange in 1895. The majority of the corporations were chartered and based in Santiago and Valparaíso even though mining activities were located either in the northern provinces (copper, silver and nitrate) or in the south (coal). The same is true for railroads (most of them were associated with mining activities) and some of the public utility companies.

Table 2 shows the number of corporations and total capital subscribed by economic activity. I classify economic activities in six main sectors: Mining, Manufacturing, Financial Services, Maritime, Utilities, Railroads and Others (Retail, Construction Companies, Agriculture and Forestry, Racetracks and Hotels). From the table it is apparent that mining (silver, copper, gold and nitrate) and finance (banks and insurance companies) were the most important sectors both in terms of capital subscribed and in number of firms.

TABLE 2 CORPORATIONS PER ECONOMIC ACTIVITY

Source: Own elaboration, based on Boletín de las leyes y decretos del Gobierno.

Table 3 shows the market size in December 1902. Market size with respect to GDP was around 15 per cent. This figure is comparable with other Latin American countries and lower than European countries at the same time. However, in terms of the number of firms per million people, Chile compares favourably to other countries at the same time.

TABLE 3 PAID CAPITAL PER ECONOMIC ACTIVITY 1902

Source: Own elaboration, based on Boletín de las leyes y decretos del Gobierno and Anuario Wessel.

To summarise, the evidence introduced in this section shows that the lack of formal legal protection was not an obstacle to the use of the corporate form by Chilean entrepreneurs or the development of a stock market in Chile during the 19^th century, suggesting that legal protection is not a necessary condition for the development of financial markets. In the next section, I attempt to answer the following questions: Do charters provide outside investors with rights and guarantees that were absent in the law? Do the contractual provisions included in the charters reflect systematic differences between firms as predicted by the agency model?

4.1 Ownership Structure

Since the seminal study of Berle and Means (Reference Berle and Means1932), ownership concentration in corporations has been the subject of constant attention of academicians and policy makers. Available evidence shows that high levels of ownership concentration are frequent outside the United Kingdom and the United States, and Chile is no exception (see Morck et al. Reference Morck, Yeung and Wolfenzon2005).

Table 4 presents statistics regarding ownership concentration. The data indicate that, at the time of its incorporation, the average number of shareholders of a firm was forty-seven. On average, the five largest shareholders (C5) held 48.2 per cent of the stock. The ownership concentration level is lower than today's Chile, where C5 is nearly 80 per cent. These results are similar to other country level studies for the same periodFootnote ¹⁹.

TABLE 4 OWNERSHIP STRUCTURE PER ECONOMIC ACTIVITY

Source: Own elaboration, based on Boletín de las leyes y decretos del Gobierno.

The average number of shareholders needed to hold a majority was sevenFootnote ²⁰. Large shareholders were also present in companies with a significant number of investors, showing that minority investors were willing to participate in companies even in the presence of block concentration.

Both in terms of number of shareholders and the largest stake, ownership concentration was higher in the mining and manufacturing sectors. A common practice was to divide the shares between «contributive» shares (i.e. shares that were paid in cash) and «beneficial» shares (shares paid in services or kind). Two hundred and six companies (37.5 per cent)Footnote ²¹ used this mechanism, most frequently in mining (59 per cent), usually as compensation for the value of mining properties; and in manufacturing (44.6 per cent), as payment for exclusive privileges or patents. Only eleven companies (5.3 per cent) that divided their capital between beneficial and contributive shares also assigned different voting rights to each type of share. Whenever different rights were used, beneficial shares had fewer voting rights than contributive ones. Since beneficial shares were usually allocated to the organisers of the company, this shows that insiders did not allocate themselves any special voting rights (even though the law allowed them to do so), on the contrary, in some cases they were willing to restrict their voting power in order to attract outside investors.

4.2 Shareholders’ Meetings and Voting Rights

Chilean law established that «Shareholders’ assembly will meet regularly to analyze the situation of the company, modify the charter and decide in any matter of common interest to the shareholders». Most charters established two ordinary shareholders’ meetings per year on preset dates. The board of directors was required to publish both the place and time of the meeting in the press.

Regarding extraordinary meetings, the law gave the board the right to call a meeting «whenever it is in the interest of the company» but made no reference to the rights of shareholders to do so without consent of the board. Nearly all (99.3 per cent) charters included articles regarding extraordinary meetings and 96.9 per cent of them allowed shareholders to call a meeting. Each company set up a requirement based on a certain number of shareholders (221 companies, 40.3 per cent) or a certain percentage of shares (477 companies, 86.9 per cent). 45.2 per cent of the companies with a percentage of shares-based requirement allowed the shareholders to call an extraordinary meeting with 10 per cent or less of the shares.

The literature has identified voting rights as an important element in the corporate governance of a firm. Contemporary studies have shown that the use of different voting rights per share is related to minority oppression. In the original study of La Porta et al., the use of a law mandated one share-one vote (1s1v) rule is identified as one of the factors correlated with financial development.

In Chilean corporations, the use of different voting rights per share was uncommon (and, contrary to contemporary practice, tended to favour minority investors). Also relatively uncommon was the use of a pure 1s1v voting system. Only 21.1 per cent of companies had such a voting system.

While current financial literature identifies 1s1v as a pro-investor voting regime, it is not the case of Chile during the 19^th century. Most Chilean corporations had some type of restriction on the maximum amount of votes per shareholder. Three hundred and eighty-seven companies (70.5 per cent) had a ceiling on the maximum amount of votes that a shareholder could cast (including own and proxy shares). The average cap was 13.8 per cent. In 330 of these firms, the maximum vote was binding; therefore, the majority of firms had contractual provisions that were effectively curtailing the voting power of the largest shareholder.

Another voting scheme used during the 19^th century was the graduated voting scale. A voting scale is a system in which the ratio of vote to shares decreases with the number of shares, and is usually complemented with some kind of maximum vote provision (Dunlavy Reference Dunlavy2004). Fifty companies (9.1 per cent) had a graduated voting scale; this mechanism was more frequent before 1875.

Both maximum vote provisions and graduated voting scales dilute the power of the majority shareholders and consequently increase the power of minority investors. Table 5 summarises voting rules per economic activity.

TABLE 5 VOTING RULES PER ECONOMIC ACTIVITY

Source: Own elaboration, based on Boletín de las leyes y decretos del Gobierno.

To study the differences in voting rules across firms I use a linear regression model. The dependent variable is the index of voting rights (Hilt Reference Hilt2008). This index captures both the effect of maximum vote provisions and voting scales in the same measure. The index is defined as follows:

$${{V}_i}\: = \:\frac{{\rm{1}}}{N}\mathop{\sum}\limits_{n\: = \:{\rm{1}}}^N {\frac{{{{v}_i}(n)}}{n}} $$

where N is the total numbers of shares and v(n) is number of votes that a shareholder owning (or representing) n shares is entitled to. The index goes from 0 to 1, where values closer to zero mean that the voting power of large shareholders is more curtailed by the restrictions.

Table 6 presents the results of different specifications of a regression model with the index of voting rights as the dependent variable. The independent variables include capital, dummies for economic activity, use of beneficial shares, time trend and a dummy variable (location) to indicate cases when the productive activities were located far from the company investors (e.g. a water company chartered in Santiago to provide services in another town). This variable can be understood as a proxy for higher monitoring costs.

TABLE 6 REGRESSION RESULTS: DEPENDENT VARIABLE VOTING RIGHTS

*Significant at 10%, **significant at 5%, ***significant at 1%.

Source: Own elaboration, based on Boletín de las leyes y decretos del Gobierno.

The coefficient for beneficial shares is negative and significant in all cases. As beneficial shares were usually distributed among the organisers of the firm, a negative coefficient means that organisers were more willing to restrict their voting power when their contributions to the firm were made in kind. The dummy for geographical separation between investors and the operations of the firm is negative and significant, meaning that voting rules that favoured minority investors were more frequently used in cases where monitoring was more costly. The economic activity dummies are negative in most cases (the base category is «others»). Capital is only significant in the first specification, which does not include the dummies for economic activity.

The limitations to the voting rights of the majority decreased for incorporations after 1880. This result is in line with the increased use of 1s1v in Europe and the United States at the same time. However, 67 per cent of the companies chartered after 1880 had restrictions to the voting power of the majority.

4.3 Governance: Assembly, Board of Directors and Managers’ Attributions

Most firm charters identified three main groups of stakeholders with decision rights within the firm: the shareholders’ assembly, the board of directors and managers. Charters also included provisions that regulated the administration of the businessFootnote ²². The degree of detail of these provisions and of the description of the attributions of the board of directors and the shareholders’ assembly varied greatly across firms.

Board members were chosen at the shareholders’ assembly for a period between one and three years. In 526 companies (96 per cent), the charter established that in order to be a member of the board it was necessary to be a shareholder, and in 65 per cent of the companies it required a minimum amount of shares. Even though this requirement was usually low (<1 per cent of the stock on average), those who sat on the board were usually majority shareholders. Of the 330 companies for which information regarding the composition of the first board of directors is available, board members accounted for 32 per cent of the stock.

The possibility of recalling directors was included in 198 companies (38.6 per cent). The average percentage of votes required to recall one or more director was 59 per cent. This contractual clause was more frequently used in corporations chartered after 1881 (47 per cent from 1881 to 1902, 15 per cent before (t-test = 8.46) and in financial services (54 per cent).

Charters also enumerated the main functions of the board. The degree of detail of the descriptions of functions varied across companies. One important aspect of corporate governance revealed by this analysis of the functions of boards is that Chilean corporations conceived the board not solely as a supervisor of management but also as an agent with a strong role in conducting the business of the firm. In 81.1 per cent of firms, the charter defined as a function of the board either «to manage the firm» or to «conclude contracts» or to «decide in all matters related to business». When not restricted by the shareholders’ assembly, it was the board of directors, and not the manager, who decided on all purchases, sales or mortgages.

The board was in charge of appointing and removing the manager (in only one case was the manager appointed by the shareholders’ assembly). Most charters defined the manager as a paid employee in charge of the daily operations of the firm and of taking charge of the accounting booksFootnote ²³, and clearly subordinated the manager to the authority of the board. In 60.4 per cent of the charters the first obligation of the manager was to «execute all board agreements». No company gave the manager the power to sell or mortgage assets and, in the majority of cases (91.6 per cent), managers were also restricted by the board in hiring and firing employees.

Since managers had limited power regarding strategic decisions and in the majority of cases important decisions required the authorisation of the board, the following analysis focuses on the balance of power between the board of directors and the shareholders’ assembly. For the Chilean case, given that founders and majority shareholders were usually sitting on the board, the assembly-board relation is also closer to the theoretical entrepreneur-investor relation analysed in the model introduced in section 2.

Given that the degree of detail of the charters varies greatly, in order to analyse the balance of power between the shareholders’ assembly and the board of directors, I used as the variable of analysis the right to decide to sell assets or to mortgage assets above a given size or considered fundamental for the operation of the firm, which is assigned in 89 per cent of the corporate charters.

The model suggests that this right should be allocated to the shareholders’ assembly when the direct monitoring costs are high, when delay costs are low and when extraction costs are high.

In corporate governance, binary choice models are used to study the determinants of use of anti-takeover and other contractual rules (Bhagat and Jefferis Reference Bhagat and Jefferis2002). In order to make good the definition I use the following criteria: Shareholders had the right to mortgage a specific asset if the board faced restrictions on selling or mortgaging any asset, on selling or mortgaging some specific assets or restrictions on selling or mortgaging assets that accounted for more than 10 per cent of total capital. In 264 charters (54.2 per cent), this right was allocated to the board of directors; in 223 cases (45.8 per cent), the charter established that the assembly had to approve any significant transaction. The assembly was given this power most frequently in mining (63 per cent). On the contrary, in 85.1 per cent of the financial companies, the board of directors had unrestricted power to make these decisions.

Table 7 presents the results of a probit regression using the right to sell or mortgage assets as the dependent variable with a value 1 for the cases where the assembly had veto powers regarding sale or mortgage of a critical asset. The explanatory variables are: paid capital, location dummy, use of beneficial shares, dummies for time period, economic activity and place of incorporation.

TABLE 7 PROBIT REGRESSION RESULTS: DEPENDENT VARIABLE VOTING RIGHTS

*Significant at 10%, **significant at 5%, ***significant at 1%.

Source: Own elaboration, based on Boletín de las leyes y decretos del Gobierno.

For ease of exposition, I present the marginal effects of the variables on the probability of having a pro-minority voting regime instead of the direct probit coefficients. Marginal effects are evaluated at the mean of the independents variables. In the case of dichotomic variables the table reports the effect of a change from 0 to 1.

Paid capital is non-significant in the first specification, but positive (and statistically significant) in the second and third specifications when the dummies for economic activities are added. The location dummy is significant in all specifications; the value of the coefficient is between 0.2 and 0.3, meaning that companies where investors were located far from the operations of the company were between 20 per cent and 30 per cent more probable to give more attributions to the shareholders assembly. Both results are consistent with the prediction of the agency model. The dummies for Santiago and Valparaíso are not significant. The dummy for the mining sector is positive and significant when the location dummy is not included but its coefficient is close to 0 and non-significant when controlling by the location dummy. On the contrary, the dummy for the financial sector is negative in all specifications; this result is probably due to the fact that the right to veto sale or mortgage physical assets is less valuable in financial companies compared with the rest of the economic activities.

The dummy for incorporations after 1880 is positive, implying that the use of veto powers increased over time suggesting more concerns of the investors after the boom and crash of 1873. Notice that this is the exact opposite of the evolution of the pro-minority voting rules, suggesting that maximum vote restrictions and veto rights for the shareholders’ assembly were used as substitutes regarding its role as investor protection tools. This is further explored in the following section.

4.4 Factor Analysis

Factor analysis is a statistical method that makes it possible to detect patterns in a set of interval-level variables, all of which are treated as dependent. In the context of this investigation, the variables are the different contractual clauses used in corporate charters.

The methodology closely follows Danielson and Karpoff (1998). I analyse ten governance provisions and assign the value of 1 whenever this provision is defined in the corporate charter in a way that favours the minority investors and 0 if it is not. Provisions include, among others, voting rights, attributions of the assembly and board of directors, election of the board of directors, requirements to be a board member and dispute resolution mechanisms.

Table 8 presents a detailed definition of the ten provisions included in the analysis. Provisions defined in a sufficiently large number of charters were chosen, which makes it possible to compute the underlying factors (the results below are based on 367 companies for which is possible to compute all ten variables).

TABLE 8 CORPORATE PROVISIONS INCLUDED IN FACTOR ANALYSIS

For each of the contractual provisions I run auxiliary regressions using the provision as the dependent variable and dummies for economic activities and five-year period dummies as explanatory variables. I compute the factor analysis using the residuals of these regressions. This method allows me to exclude the long time span of the data and differences between economic activities as sources of variation. The three factors with eigenvalues >1 are retained. Results are presented after a varimax rotation and are summarised in Table 9.

TABLE 9 FACTOR ANALYSIS RESULTS

Source: Own elaboration.

The table shows that several provisions are observed together. The first factor is loaded with governance provisions related to ex-post control (calling extraordinary meetings, recalling directors) and the process of election of directors. The second factor is loaded with ex-ante mechanisms, such as provisions regarding the shareholders’ assembly having the explicit faculty to block some of the board's decisions (sale or mortgage of assets, charter reforms, dissolution of the company), and the third factor is related to auditing functions. In many provisions factors 1 and 2 are negatively related, suggesting that firms used ex-ante and ex-post systems of control as substitutes. This result is consistent with the agency model presented in section 2: the cost of providing ex-post or ex-ante protection differs for each firm, and therefore, each firm chooses the most cost-efficient mechanism to provide protection for their investors.

4.5 Charter Reforms

Any charter reform required the approval of the shareholders’ assembly. Four hundred and thirty-nine charters (80 per cent) included specific rules and voting requirements regarding charter reform. The average percentage of approval required for charter reform was 60.6 per cent. After a charter reform was approved by the shareholders’ assembly, it had to be submitted to the government for approval. Charter reforms were also published in the official bulletins. In 155 cases (from a total of 185), the government decree included details of the contents of the reform: the most common reform submitted for government approval was an increase of capital (45.2 per cent). Another frequent modification was on the definition of the scope of the firm's activities, for example, new policies for insurance companies, extension to mining activities by foundries or vice-versa or changes in the location of the company (22.7 per cent). Table 10 details the type of reforms and their frequency.

TABLE 10 CHARTER MODIFICATIONS

Source: Own elaboration, based on Boletín de las leyes y decretos del Gobierno.

Governance rules were also subject to modifications. There were changes in voting rights (20 per cent), attributions of the board of directors, the number of members of the board and their term of office (19.4 per cent) and rules to call an extraordinary meeting (4.5 per cent). There is no clear direction to these changes: some companies modified their charters to impose more restrictions on maximum votes while others did so to introduce a 1s1v regime, some companies increased the power of their board of directors while others took attributions from them and gave them to the shareholders’ assembly.

For the objectives of this paper, two interesting cases to consider are the reforms introduced by «Compañía de Gatico» and «Compañía Industrial». The former was a copper company established in 1900 with six subscribing shareholders, all of them belonging to the same family. The initial charter of the company had no mention of shareholder rights, no rules to designate the board and no clear distinction of the functions of the shareholders’ assembly, the board of directors and management. Two years later, the company decided to increase its capital by issuing new shares to the general public. At the same time, the governance rules were modified, increasing the number of directors from three to five to be reelected annually, giving shareholders the right to call extraordinary meetings with 10 per cent of the shares and establishing a voting ceilingFootnote ²⁴.

The case of «Compañía Industrial», a chemical company established in 1901, is an example of how governance rules were relevant from a minority investor perspective. In 1902 the majority shareholders (who controlled the majority of the board) decided to reduce the number of directors (thereby excluding representatives from minority investors), to elevate the percentage of shares required to call an extraordinary meeting to 60 per cent (at the time the directors of the company held more than 40 per cent themselves) and to establish a five-year term for board members (no other company charter approved from 1854 to 1902 had board terms longer than three years). A group of minority shareholders opposed the reform in the assembly meeting and later filed a complaint to the judiciary to block the reform. The complaint was dismissed by the courts on the grounds that the reform was approved by the number of votes required in the charterFootnote ²⁵.

Both examples show that governance rules were relevant to investors. In the first case, the desire to obtain financing from outside investors made the owning family modify the governance rules, including investor rights that were absent in the original contract, while in the case of «Compañía Industrial» the minority investors who tried to block the reform of the company charters clearly understood that under the new rules they could be easily be subjects of expropriation by the firm's insiders.